Technical and Military Imperatives: A Radar History of World War II

Technical and Military
Imperatives
A Radar History of World War II
Related Titles
Cockcroft and the Atom
G Hartcup and T E Allibone (out of print)
Operation Epsilon: The Farm Hall Transcripts

Edited by Sir Charles Frank
Radar Days
E G Bowen
Echoes of War: The Story of H2 S Radar

Sir Bernard Lovell
Boffin: A Personal Story of the Early Days of

Radar and Radio Astronomy and Quantum Optics
R Hanbury Brown
Heavy Water and the Wartime Race

for Nuclear Energy
P F Dahl
Electronic Inventions and Discoveries

G W A Dummer
Technical and Military
Imperatives
A Radar History of World War II
Louis Brown
Carnegie Institution of Washington
Washington DC, USA
1999
Institute of Physics Publishing
Bristol and Philadelphia

c 1999 IOP Publishing Ltd
All rights reserved. No part of this publication may be reproduced, stored
in a retrieval system or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without the prior permission
of the publisher. Multiple copying is permitted in accordance with the terms
of licences issued by the Copyright Licensing Agency under the terms of its
agreement with the Committee of Vice-Chancellors and Principals.
The author has attempted to trace the copyright holders of all the figures
reproduced in this publication and apologizes to them if permission to publish
in this form has not been obtained.
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library.
ISBN 0 7503 0659 9
Library of Congress Cataloging-in-Publication Data are available
Production Editor: Jacky Mucklow

Production Control: Sarah Plenty and Jenny Troyano
Commissioning Editor: Jim Revill
Editorial Assistant: Victoria Le Billon
Cover Design: Kevin Lowry
Marketing Executive: Colin Fenton
Published by Institute of Physics Publishing, wholly owned by The Institute of

Physics, London
Institute of Physics Publishing, Dirac House, Temple Back, Bristol BS1 6BE, UK
US Office: Institute of Physics Publishing, Suite 1035, The Public Ledger Building,
150 South Independence Mall West, Philadelphia, PA 19106, USA
Typeset in TEX using the IOP Bookmaker Macros
Printed and bound in the UK by Cambridge University Press, Cambridge
CB2 2BS
CONTENTS
PREFACE WITH ADVICE TO THE READER ix

ACKNOWLEDGMENTS xiii
1 PRELIMINARIES 1
1.1 Radio vision for war 1
1.2 Electromagnetic waves 6
1.3 Perceptions of air power, 1919–1939 11
1.4 Navigation in 1939 19
1.5 Anti-aircraft artillery, 1914–1939 23
Photographs: Preliminaries 30
2 ORIGINS 33
2.1 Electronic component development 33
2.2 Beginnings, 1902–1934 40
2.3 Britain builds an air defense system 49
2.4 Americans and Germans build prototypes 64
2.5 Five other nations 83
Photographs: Air warning—Europe 92
3 FIRST CLASHES 97
3.1 War in Europe 97
3.2 The Battle of Britain and the Blitz 107
3.3 The Atlantic, 1941 120
3.4 Friend, foe or home? 129
3.5 The Japanese realize they are behind 135
Photographs: Air warning—Pacific 141
4 NEW IDEAS 145
4.1 Microwaves 145
4.2 The Tizard Mission 159
4.3 The Radiation Laboratory 166
4.4 The proximity fuze—the smallest radar 174
4.5 Greater and lesser microwave sets 186
Photographs: Land and naval fighter control 198
Technical and Military Imperatives
5 YEARS OF ALLIED DESPAIR AND HOPE 205

5.1 The Mediterranean, 1940–1942 205
5.2 War in the Pacific 215
5.3 The Channel, 1942 224
5.4 Carrier warfare defined 235
5.5 The South Pacific, 1942 246
5.6 The Eastern Front 262
Photographs: Naval radar 270
6 THE GREAT RADAR WAR 279
6.1 The destruction of German cities initiated 279
6.2 Countermeasures 290
6.3 An air war of attrition 300
6.4 Arbeitsgemeinschaft Rotterdam 310
6.5 The destruction of German cities completed 317
Photographs: Radar for AA artillery 326
7 ALLIED VICTORY IN SIGHT 334
7.1 The Battle of the Atlantic, 1939–1945 334
7.2 Radar in arctic waters 348
7.3 The Mediterranean, 1943–1945 353
7.4 Japanese shipping destroyed 359
7.5 The wide Pacific 367
Photographs: Airborne radar 382
8 THE END IN EUROPE 386
8.1 Invasion 386
8.2 Flying bombs 392
8.3 The battlefield transformed 398
8.4 Post mortem 403
Photographs: Tubes 409
9 THE END IN ASIA 411
9.1 The Philippines and Okinawa 411
9.2 The destruction of Japanese cities 420
10 THE MEASURE OF RADAR 427
10.1 Navigation transformed 427
10.2 Science and the electronic age 433
10.3 Secrecy and the technical imperative 444
10.4 An evaluation 456
APPENDIX A: A FEW RADAR ESSENTIALS 466
A.1 Electromagnetic waves 466
A.2 Reflected signals 468
A.3 Antennas 469
A.4 Lobes 471
A.5 Vertical lobe structure 471
vi
Contents
A.6 The Yagi antenna 473

A.7 Electron tubes 473
A.8 Transmitters 475
A.9 Receivers 475
A.10 Indicators 476
NOTES AND SOURCES 479
BIBLIOGRAPHY 524
NAME INDEX 533
SUBJECT INDEX 544
vii
PREFACE WITH ADVICE TO THE
READER
This book is about science and war. These are unquestionably the two most
dissimilar manifestations of the behaviors that distinguish man from beast.
Science—allow it here to mean all elements of our search to understand
the universe and put that understanding to use—is the most remarkable
of human occupations. It is also the most important. For us not to strive
to understand the universe would be as foolish as for castaways not to
explore the island on which they find themselves. Science, like art and
literature, draws from the mind those inner, intangible feelings which we
flatter ourselves give us our station on Earth. It discloses to us things of
unparalleled beauty and wonder—the mathematical beauty of the laws of
physics, the wonder that life is understandable. War is the negation of all
this. It is a disease, like smallpox, that must be eradicated.
Yet war is almost as unique to man as is science. Other than ourselves,
only ants organize their violence so that it can be called war. Violence is
ubiquitous to the biological world, but it is man’s doing to make it an inner-
most part of civilization. War is the great embarrassment of civilization, yet
all great civilizations have maintained themselves by their military skills
and have vanished when those skills no longer sufficed to protect them
or sustain internal order. Moreover, from the dawn of civilization science
and war have been inseparable companions, locked in a partnership that
neither desires and that neither is capable of dissolving.
No one setting out to read this book needs the recitation of a catalog
of invention and discovery that bears on military history to be convinced
of science’s significance, but if this relationship is so well known, what is
the purpose of this book? It has a void to fill, for an important element
in this story is not well told. Years ago I was struck by the absence of
a comprehensive and international history of radar, of the kind that has
appeared many times about the atomic bomb. Not that there were no
books on the subject; on the contrary, there is almost an over-abundance,
as the numerous references cited herein attest. There are among them some
extremely good accounts of various parts of the story, and it is my hope
that this study will impel many to read them for the things I have had to
ix
leave out. However, these histories are all parochial, for they invariably
found their origins in the personal involvement of the authors in radar or
on material gleaned from national or laboratory sources. The secret nature
of radar did not give the engineers, scientists and military much chance
to learn what took place on the other side of the hill, so national versions
tell one little about other countries and all too frequently leave the reader
with a distorted impression of the whole subject. This does not imply
dispute about the facts. The German and British accounts each emphasize
the development of their own equipment, of course, but describe the great
radar struggle between the Luftwaffe and RAF Bomber Command without
significant disagreement. They leave the curious impression, however, that
World War II was a private fight between Britain and Germany.
There were in America four large radar development laboratories of
comparable importance. The histories written about them are quite un-
equal and show poor understanding of the contributions of others. Three
years into the project I was still encountering major surprises that changed
my understanding in a substantive way. Most radar histories trace its ori-
gins to a few relatively well known and oft reported events. It has not
taken any great depth of scholarship to recognize that radar was just in
the air during the 1920s and 1930s. This is obvious when one learns that
eight nations had radar projects in 1939. There is evidence that leads one
to suspect that hundreds of communication engineers observed reflection
and interference phenomena causing them to think about radio location
without recording it. The strong similarities of the German and American
programs alone speak volumes for its invention being a technical impera-
tive.
‘The bomb may have ended the war but radar won it.’ This was
the comment of many radar workers in August 1945, when the news of
the atomic bomb upstaged the release of public knowledge of the MIT
Radiation Laboratory, planned as a cover story for Time. This statement
has been repeated many times since then, often with categorical statements
concerning some engagements, but has never been examined for the war
as a whole.
It is the object of this study to attempt to improve on these perceived
deficiencies: (1) to describe the wide simultaneous development, (2) to
examine the idea of it resulting from a technical situation that made it
inevitable, (3) to determine how the severe restraints imposed by secrecy
affected design and use and (4) to approach with the objectivity that should
come with the passage of time the question of its effect on the war, both in
deciding battles and altering the military leaders’ approach to war. I shall
have succeeded if my effort presents in balanced form the main elements
of the whole story.
An appendix explains some of the rudiments of radar. Many will
choose to skip it because their technical knowledge equals or, more likely,
greatly exceeds what is presented. It is possible to gain a fair understanding
x
Preface with Advice to the Reader
of radar history by judiciously skipping technical matters because most of

the story is about individuals and organizations, so readers for whom the
terms megahertz, transmitter, antenna, bandwidth and ionosphere have
only vague meanings need not set this book aside. They may prefer to skip
certain sections whose absence will not detract greatly from understanding
the general development of radar. These sections are 1.2, 2.1, 4.1, 4.5,
6.4 and 10.2. Other parts have technical matters that may perplex a lay
reader, but I suggest adopting an attitude much as one must assume for
science fiction—which cannot be understood by anyone. The lay reader
will obviously appreciate that many readers will find the technical details
meager.
In discussing these things with young people I have come to realize
that events burned into the memory of my generation are no longer current,
lying further back for them than the wars in South Africa or Cuba for us.
Comments from readers of various draft sections have disclosed that many
well read persons are not familiar with parts of the conflict. The British
and Germans often have an uncertain idea of what happened in the Pacific,
and many Americans know little about the air war against Germany other
than it involved strategic bombing. This has required sketching in outline
much of what happened from 1939 to 1945 in order to make the radar
contribution coherent.
The subtitle, A Radar History of World War II, indicates that only equip-
ment that was widely used and provided important results in the conflict
is described in any detail. These were also the sets that established the
basic principles. The mix of technical and military events is essential to
understanding, for no weapon was ever designed with such intimate col-
laboration between inventor and warrior. Engineers frequently tried out
ideas in combat, and some died there. No adequate history of radar can re-
sult from a knowledge of what transpired in the laboratories alone. Many
radar sets were designed toward the end of the war—superb examples of
the art—that had trivial effects on the outcome and so will either not be
mentioned at all or be described only because the extent of the development
project was so great or the device so unusual. None fits this description
better than Cadillac, which could rightly have a book devoted to it alone,
yet is a minor notation in the chronicle of the war even though it consumed
a large part of the effort of the MIT Radiation Laboratory.
The same unevenness applies to the military evaluations. The presen-
tation of World War II is highly skewed with campaigns treated according
to the radar involvement. Thus the Pacific War is treated in detail whereas
the titanic clash on the Eastern Front receives a single, brief section, and the
Blitzkriegs, east and west, are hardly mentioned at all. Substantial space
is given to battles in which radar was a unique and decisive weapon, es-
pecially early engagements in which commanders struggled to master this
new technique and devise tactics to go with it. Little space is given to bat-
tles in which huge amounts of radar were employed routinely but without
xi
novel or decisive effect. Because of the lessons revealed, certain actions

are treated in so much detail as to magnify them out of all proportion to
the war as a whole or even to radar itself. Two examples of this are the
radar aspects of the American carrier raids in early 1942 and the Dieppe
Raid of the same year. One provided an invaluable, last-minute school for
carrier operations that altered the way in which the US Navy conducted
the Pacific war; the other showed the ineptitude of both sides in their use
of radar during a major landing, lessons that had great impact on radar
usage two years later in Normandy.
Sources properly belong to a section of acknowledgment, and they are
listed there with many, very grateful words of thanks, but the nature of the
sources has much to tell us about radar history. I began this study because
the book I would have liked to read did not exist. As my studies progressed
the amount of published material startled me and, more amazing yet, there
seems to be no end to it. A great deal of very valuable information was
to be found in publications of extremely limited circulation, much of it
published privately. With the help of our librarian, skilled in the art of
making inquiries by computer, many such books were located. He and his
assistant found others by methods that must have been based on instinct
supplemented by an intimate knowledge of the book trade.
There was, however, much more that was not accessible by such
an approach. Early attempts to ferret these things from archives simply
showed that, although they provided much help in following specific clues,
they would have quickly become swamps from which I should have never
extracted myself. Fortunately, I found another path on hard ground. There
are radar-history enthusiasts to be found all over the world. They are gen-
erally electrical engineers or physicists and, more often than not, radio am-
ateurs; they have squirreled away documents, books, photographs, journal
articles and artifacts as well as written histories, and communicated with
one another about their common interests. This rich source was opened
through the old method of writing letters—many letters. These correspon-
dents, found in eight countries, have the judgement and understanding to
know what is important and have supplied me with a great part of the un-
published items used. Through their letters they have shown themselves
to be astute critics possessed of a deep understanding of the background of
events. A correspondence was also developed among military personnel
with radar experience, but radar is not the common thread that links them,
so this circle has not had the continually expanding radius that marks the
other.
Louis Brown
July 1999
xii
ACKNOWLEDGMENTS
From the beginning I decided against using interviews as a research tool,

partly from memories of my past faulty attempts to gain information that
way, but primarily from the certainty that such a course would skew the
source material toward those groups for whom published memoirs and
interviews were already available in significant quantity. In retrospect this
has shown itself to have been a wise decision but for a different reason: I
have come to find the interview a flawed method. Recollections of events
half a century ago, called up in conversation with an interviewer uncertain
of what needs to be asked and with neither party given sufficient time for
reflection, generally extract little that is new and much that is misleading.
There are numerous transcripts of such interviews and, though they cer-
tainly have value, they also disclose many errors, the result of the passage
of years, the secrecy of the early times and the subsequent development
of mythology. One must use these sources with caution and with em-
phasis on the subject’s personal experiences and not his impressions of
history.
My postal substitution for interviews generally began as a probe to
follow a clue, asking the recipient to clarify some bit of puzzlement. The
answers and subsequent exchanges frequently brought more than was re-
quested, such as personal memoirs, documents, articles published in ob-
scure magazines, out-of-print books and the like. The oft decried slow-
ness of this method of communication has proved to be of value, as it
assured both parties time to think over questions and answers. Informa-
tion obtained in this way, including some acquired by less formal means,
came from: H-J ALBRECHT, Niels Chr BAHNSON, Lori BERKOWITZ,
David B DOBSON, John DUGGAN, Gerhard EBELING, Anna Maria Runge
ELSTNER, George A EMERY, Pekka ESKELINEN, Helmut FAUDE, the late
John FINDLAY, Ivan A GETTING, John M GODFREY, Charles R
HABERLEIN, Frank HEWITT, Marvin HOBBS, Joseph HOPPE, Yasuho
IMAI, F Y ITO, the late R V JONES, Harumasa ITO, Robert KAUFMANN,
Gordon M KLINE, John B LUNDSTROM, Colin MACKINNON, John F
MANNING, Allan ANEEDELL, Karl NEUMANN, Mark PEATTIE, Carsten
PETERSEN, William R PIGGOTT, Charles T PREWITT, Henry PROBERT,
the late Robert RALSTON, H-J RICHTER, J R ROBINSON,
xiii
Hans SORGENFREI, Rudolf STARITZ, Jane E TUCKER, Oliver VOLLAND

and H W WICHERT.
In addition to providing information, some correspondents were
kind enough to read draft portions of the manuscript, comment on the
content and all too frequently point out errors. For such efforts I am
indebted to Ralph B BALDWIN, David K BARTON, Ray BOWERS, the
late Adam BUYNOSKI, Russell S CRENSHAW, Paul FORMAN, Richard
B FRANK, Hugh G HENRY III, Axel NIESTLÉ, Hans PLENDL, Jr, the late
Ernest C POLLARD, Peter J PRICE, Max SCHOENFELD, Merrill SKOL-
NIK, P G SMITH, Harry STREET, John STREET, Kathleen WILLIAMS, the
late Tom WINCHCOMBE and Thomas Y’BLOOD. Before any copy left
the department it was subjected to careful reading by Merri WOLF, who
removed many clumsy sentences and grammatical blunders and raised
questions about clarity.
Radar history is by no means devoid of serious scholars, identified
to me as the circle of my correspondence expanded. This resulted in ex-
changes of information as well as discussions in person, by telephone and
mail. The late John BRYANT and I spent many happy hours arguing our
varied viewpoints and exchanged large amounts of material. Michael S
DEAN served through his maintenance of the Historical Radar Archive
and in the organization of Royal Air Force radar reunions; his reading of
the manuscript identified numerous errors, and he located many valuable
photographs. F A KINGSLEY helped greatly in clarifying the contribu-
tions of the Royal Navy, some of which I had attributed to the Air Min-
istry. Russell BURNS and Sir Edward FENNESSY have not been content
to accept some of my theses, which has provided lively correspondence, a
few changed opinions and wonder that an otherwise rational man persists
in error. Colin LATHAM and Hans JUCKER have provided important
technical criticism, JUCKER adding an unexpected continental point of
view. Frederick SEITZ proved a valuable companion in our combined
efforts to trace early aspects of the German work. In addition to com-
ments on the manuscript Robert BUDERI taught me most of what I know
about postwar radar. Arthur O BAUER and Bernd RÖDE sent me their
technical studies of the Würzburg of which I made frequent use. Aus-
tralia and South Africa had their share of radar activity, and their veter-
ans went beyond reunions to ensure a historical record, Ed SIMMONDS
and Geoffrey MANGIN supervising. Andrew BUTRICA sent me his his-
tory of radar astronomy in draft, which helped remove the narrowness
of my World War II point of view, and J Michel BLOM always seemed
to find some new reference. Failing health and death prevented Fritz
TRENKLE from providing the ever-ready answers to questions that the
community had become accustomed to receiving, but his published vol-
umes continued to guide us. Charles SÜSSKIND began a comprehen-
sive study of radar history at a time when much less material was avail-
able, resulting in an unpublished manuscript that terminated at the time of
xiv
Acknowledgments
America’s entry into the war. This limited study proved to be extremely
useful.
At the beginning of the work the very important German side of
the story was unsatisfactory. There were important published books and
articles, but all suffered from the destruction of postwar records and the
isolation imposed on the authors by the severe secrecy that characterized
the Nazi state, and three years into the project had shown me no way of
correcting this. Letters to museum curators and historians proved fruitless
until Joseph HOPPE suggested that Conrad H von SENGBUSCH, an elec-
trical engineer with a small electronics museum, might know some who
had radar as a hobby, a suggestion that brought unparalleled riches as
he introduced me to Hans Ulrich WIDDEL, physicist, radio propagation
scientist and wartime radar operator. Not only did he have a wealth of
documents, books and miscellaneous papers in his own files, he was con-
nected through amateur radio to a group of like-minded individuals who
began to discuss my needs during their routine contacts. Their individual
efforts are included in the second paragraph and have completely trans-
formed the German picture. The size of my correspondence with WIDDEL
is larger than any of the others and has a corresponding breadth in tech-
nical and historical content. Equally important was the ten-year study by
WIDDEL’S lifelong friend, Harry von KROGE, who was determined to
learn details of GEMA, whose story had become distorted and much of it
lost over the years.
Knowledge of Japanese radar had come to us primarily from the
wartime and occupation technical intelligence reports and from articles
by Shigeru NAKAJIMA, leader of their magnetron development. These
sources, valuable as they are, miss many important historical details. For-
tunately, Yasuzo NAKAGAWA wrote two books in 1987 that contained
large portions about radar history. Don Cyril GORHAM translated rel-
evant sections and thereby provided a more secure basis of study. Not
surprisingly, this opened many questions, and I was greatly helped by
Naohiko KOIZUMI, a retired electron tube designer with a wide knowl-
edge of radar and a colleague of both NAKAGAWA and NAKAJIMA. He
entered enthusiastically into primary historical research, uncovered many
new items and explained the inevitable mysteries. He joined Bryant and
me in editing the translation of Nakagawa’s writings into a published book,
which contains most of what is known about Japanese radar.
The modern library system with its computer-linked catalogs has
been of such great help that one wonders how historical studies were done
before. The librarian at the Department of Terrestrial Magnetism, Shaun
HARDY, made the Inter-Library Loan a key research device, drawing on 28
libraries. Local libraries have been an important resource and their staffs
courteous and helpful: the District of Columbia Public Libraries, the Naval
Historical Center Library, the Bender Library of American University, the
libraries of the Smithsonian Institution, the library (now called a Research
xv
Information Center) of the National Institute of Standards and Technology

(known by most as the National Bureau of Standards) and the Library of
Congress.
Richard RHODES and Alfred PRICE resisted my urging to have them
write this book; this is a pity, said even now that I have gained the pleasure
of having done it, but they provided encouragement, something for which
I shall ever be grateful, and PRICE supplied documents that only his long
activity in the field of electronic warfare could have uncovered. William
ASPRAY and Frederik NEBEKER at the Center for the History of Electrical
Engineering provided early guidance and found competent critical readers
at the half-way point.
At the beginning of the project there were tentative plans for an in-
ternational museum exhibit on radar that involved me in meetings with
E J S BECKLAKE, Oskar BLUMTRITT and Bernard FINN. Although the
planned exhibit came to naught the discussions were productive. Thomas
BALLARD, H Warren COOPER and Michael CROSS guided me through
the archive and collection of the Historical Electronics Museum, a substan-
tial resource for anyone studying electronic history.
By granting me emeritus status Dr Maxine SINGER, President of the
Carnegie Institution, and Dr Sean SOLOMON, Director of the Department
of Terrestrial Magnetism, significantly eased my labors by providing access
to the important features of the modern office. Dr SINGER went beyond
this in helping provide translations from Japanese.
To all of these persons and organizations my sincerest and most heart-
felt thanks, extended also to the many persons from whom help has been
received through casual discussion and whose names have slipped from
an imperfect memory.
xvi
CHAPTER 1
PRELIMINARIES
1.1. RADIO VISION FOR WAR

In winter the Denmark Strait, the stretch of water between Greenland and
Iceland, is hardly a favorite passage for mariners. Heavy fog is common,
sustaining little or no vision during the very short days, and ice presses
down from Greenland with bergs a danger in or out of season. Storms
hardly count as a rarity. These unpleasantries seldom trouble a shipmas-
ter as little commerce need pass these waters, but they were well lodged in
the mind of Vice Admiral Günther Lütjens, who was passing the strait in
February 1941 with the two German battle cruisers Gneisenau and Scharn-
horst. His object was the destruction of British merchant shipping, but to
do this he had first to bring his ships from German ports into the Atlantic.
The trickiest parts of his task were caused by the geographical position of
Great Britain and the pugnacity of the Royal Navy that compounded his
navigational problems with cruisers, pickets set out for just this event.
Lütjens now felt confident because of his experiences of the previ-
ous few days with the electronic devices mounted on the forward and aft
fire-controls of the two vessels—radar designed for the Kriegsmarine and
bearing the name Seetakt. This was the best radar in the world for search-
ing the ocean out to the horizon, and it had already paid for itself before
they reached the Denmark Strait. They had sailed up the coast of Norway
and headed southwest to pass south of Iceland, but the radar had picked
up too many ships whose speed identified them as warships, and after
several course changes Lütjens had turned back and headed north to ren-
dezvous with a tanker to replace expended fuel. The motion of the British
ships gave no indication that they had seen him, allowing him to make the
correct assumption that they had no radar, at least none usable for surface
search.
Meeting the tanker was no mean trick. With scant opportunity for a
stellar fix in the fog, both he and his supply ship had been depending on
dead-reckoning for position, using course and speed to estimate position
and hoping ocean currents and wind had not made the inevitable errors
1
too large. Without radar, finding a tanker on the fog-covered northern

ocean might well have failed; instead it became a trivial exercise.
Now he was safely passing north of Iceland and had avoided ice
and cruisers. His position was easily determined by observing Iceland’s
mountains with the radar. The way was free into the Atlantic where he
ranged north and south, meeting supply ships and sinking convoys not
fortunate enough to have capital ships for escort.
The early mist had cleared and the sunny prospects had alerted the
Ventnor operators, in the forward-most station of England’s system of
radar early warning on the south coast of the Isle of Wight, that an air
attack was probably in the offing. At 1100 hours small wiggles in oscillo-
scope traces, insignificant to any but the young women observing them,
confirmed their expectation. Across the Channel formations of bombers
were assembling and would soon be heading toward England. The opera-
tors were puzzled that the Luftwaffe adhered to the tactic of forming over
Calais and Boulogne in full view of radar, as it gave an extra margin of
warning. Indeed, indications kept coming that they simply did not know
how the Royal Air Force was countering them. They seemed to know the
significance of the huge towers that lined Britain’s south and east coasts,
because they had opened the heavy raids against England with attacks
on several radar stations on 12 August 1940, a few weeks before. Fifteen
Ju-88s of Kampfgruppe-51 had attacked Ventnor, destroying the majority
of the buildings and creating circumstances that led to one operator earn-
ing the Military Medal—the first awarded to a woman. A delayed action
bomb exploded later at the thick concrete wall of the transmitter block and
severed the feeders to the antenna. Four days later another attack by six
Ju-88s finished off the destruction and put the station out of operation for
two months.
Ventnor’s crew switched to a Mobile Reserve Unit at nearby Bem-
bridge that replaced the more powerful main station and preserved most
functions with shorter range. They saw the day’s build-up, although later
than adjacent stations that had full power. A tension reiterated itself. Had
the Germans finally got the point? Were they finally going to take out the
stations? But the objective was neither the radar stations nor the bases
of Fighter Command, Churchill’s ‘Few’, which had been severely mauled
during the preceding weeks, many of which could have been put out of
service completely by one or two more raids. No, their destination was
London and decisive defeat, for this was 15 September 1940, remembered
as Battle of Britain Day.
The three men in the open cockpits of the biplane seemed in that
summer of 1941 to be in the wrong war. Their aircraft was slow even by
the standards of the previous war and was armed even more poorly, having
only a single Vickers gun firing forward and a single Lewis for someone
in the rear cockpit, which was half the guns a comparable machine would
have had in 1918. It had, however, design elements that compensated for
2
Preliminaries
such remarkable deficiencies. There was the 690 hp Pegasus engine that
allowed it to carry 1000 kg loads, generally either of bombs or a torpedo,
for in addition to reconnaissance the Swordfish was the Royal Navy Fleet
Air Arm’s dive and torpedo bomber. To perform such duties required an
extremely strong airframe. To attain this strength yet allow the wings to be
folded for storage on a carrier required struts and wires in an arrangement
that defied perfunctory understanding.
On this particular night one of the men in the rear cockpit was study-
ing the traces of an oscilloscope, for their plane had been outfitted with
radar, and they were hunting for convoys of Axis ships bound from Italy
to Tripoli with supplies for Rommel, who was menacing Egypt. They were
skilled at night operations and appreciated this new electronic eye. A con-
voy duly appeared on the scope, and the pilot followed the operator’s
instructions so as to drop a flare that illuminated it for the other aircraft
of the flight to launch their torpedoes. The ships not hit during the night
could look forward to a reception by submarines after daybreak, when the
Swordfish’s infirmities made a return to Malta prudent before the Luft-
waffe began to seek them out. Rommel was never to secure his supply
line.
The United States Navy was having a difficult time in the Solomon
Islands during summer and fall of 1942. They had suffered a humiliating,
and what could have been a decisive defeat during the night of 8/9 August,
had the Japanese followed their victory with the destruction of the trans-
ports and supply ships unloading marines onto Guadalcanal. As night
closed on 11 October circumstances were similar: a Japanese squadron
from the northwest was intent on destroying vessels reinforcing the island
that a screen of American cruisers was equally intent on protecting. Night
action was the specialty of the Imperial Fleet, and their mastery of it had
led to the earlier loss of three American and one Australian cruisers. Radar
was supposed to have given the Allies compensating night fighting skills,
but clumsy use of it in August had, if anything, contributed to the disaster.
There had been subsequent improvements, but command understanding
of the new technique was still marginal. The skipper of USS Boise, Captain
Edward J Moran, appreciated the power of radar and had studied it. He
also had a competent radar officer in Lieutenant Philip C Kelsey to look
after the fire-direction radar for his main battery of fifteen 6 inch guns and
a recently installed microwave surface-search radar that allowed the scene
of battle to be presented as a maplike presentation.
Naval battles are often confusing, and night actions are almost never
capable of accurate reconstruction after the event. The naval actions in the
Solomons were night-time gun battles where it was difficult to establish
who was friend and who foe, but in the fight of 11/12 October the begin-
ning at least stands out in remarkable clarity: the Boise initiated the battle
by opening fire in the darkness with radar control alone, and her first salvo
scored direct hits on an astounded enemy. Kelsey had first sighted them,
3
and Moran had maneuvered to obtain the best firing position. The loss of
six warships and the death of the commanding Admiral sent a strong mes-
sage to Tokyo about the importance of their lagging radar development.
The nights of late fall 1943 were particularly hard for the men who
fought one another in the sky over Germany. The Royal Air Force pressed
attacks on German cities to break the will of the civilian population and
injure the means of producing and transporting the goods on which a
modern war depended. Air Chief Marshal Arthur Harris, chief of Bomber
Command, had initiated a series of enormous bombing attacks on Berlin in
November with the objective of destroying the war potential of this sprawl-
ing metropolis. Key to this effort was a radar system called H2S that gave
the navigators of the Lancaster four-engine bombers something purported
to be a radar map of the ground. Its purpose was to allow the formations
to find their way and aim their bombs. Both attacker and defender were
continually faced with the consequences of not being able to see in the dark.
Both attempted to ‘see’ with radar. Neither was conspicuously successful.
After a year of war the RAF and the Luftwaffe had come to realize
that large daylight bombing attacks could not be sustained, and both had
changed to night bombing. When Hitler made war on the Soviet Union,
strategic bombing, which is the euphemism given to bombing civilian tar-
gets, became primarily a British function. German antiaircraft (AA) guns,
first with searchlights and optical sighting, then supplemented with radar
control, had forced the bombers to their maximum altitude. By the end of
1943 the bombers carried one, sometimes two kinds of radar plus devices
to warn them of the enemy’s radar and to identify themselves electroni-
cally to their countrymen and allies. The defending night fighters carried
radar sets for locating the bombers as well as homing devices that allowed
them to locate either the enemy aircraft by their radiations or their own
landing fields by radio beacons. They were aided in their search through
the darkness by an extensive ground radar control system that tried to
place them close enough to the bombers for their airborne radar to pick up
a target. Success then turned on the skill of the air-crew radar operator in
guiding the fighter to a position close enough for the pilot, depending on
whatever light might be present from the stars, Moon or burning cities to
identify it and open fire.
These technical tasks, each requiring analytical thought best attained
with a relaxed mind, had to be performed in extreme cold and noise while
breathing oxygen through masks and in the presence of electronic inter-
ference designed to thwart them. The bombers threw out bundles of alu-
minum foil cut to resonate at the frequency of the AA and older airborne
radars. Both sides transmitted electronic noise intended to overwhelm
their opponent’s radar. There were elements of humor in the attempts to
confuse the night fighters by interjecting false information and distractions
into the instructions of the ground controllers, but the extreme casualty
rates and the apparently unending number of sorties required of the par-
4
Preliminaries
ticipants made humor a rarity and dampened any thrill of combat in all
but the most devoted warrior.
When war broke out in September 1939, South Africans had not yet
been informed by Britain about radar. They quickly received a briefing
covering the technical essentials but learned that the Dominions could
not expect to receive any equipment of British manufacture for the in-
definite future. A group at the Bernard Price Institute in Johannesburg,
under the direction of Dr Basil Schonland, built a set of their own design
from available components and by December were tracking planes with
it. Events had not yet established South Africa’s part in the growing con-
flict, other than the responsibility for protecting the valuable sea routes
around southern Africa, but modern weapons would be essential. This
home-made radar, called JB for Johannesburg, allowed training, and its
designers formed a cadre for future radar needs, needs that became fixed
when Italy sent forces into recently annexed colonies to attack adjacent
British and French territory. South Africa countered these moves and JB
radars went north to help protect Nairobi, the capital of Kenya and nerve
center of the vast region. These sets, each somewhat different from its
predecessor and accompanied by one or more of its builders, moved on to
help in the defense of the Suez area after the Italians were disposed of.
In 1942 when submarine activities began to take a toll on Cape ship-
ping, the JBs returned to join other locally manufactured sets along the
southern coast. Their presence forced the U-boats out of range, which
in turn kept merchant shipping dangerously close to an unmarked shore,
and radar frequently prevented steamers from running aground. Although
they seldom saw the conning towers of U-boats on their screens, they did
see much of the debris that covered the war-racked ocean and sent aircraft
to investigate, occasionally forcing down or sinking a submarine. Sorties
that did not lead to an attack on a U-boat were compensated many times
over when the target proved to be neither a conning tower nor floating
wreckage but a lifeboat full of exhausted survivors.
On 30 July 1943 American warships began the preliminary bombard-
ment of Kiska, one of the western islands of the Aleutians that had been
occupied by Japan more than a year earlier. Retaking the other island, Attu,
had required the better part of the preceding May and had led the attackers
of Kiska to expect the worst, hence the extravagant expenditure of explo-
sives that became the standard prelude to a Pacific invasion. But when
the 40 000 American and Canadian soldiers went ashore no one was to be
found. The Navy was confident the garrison could not have been evacu-
ated, so they must still be on the island. It was obviously some grand trick,
and patrols cautiously sought them out. But no one was there. All had
been evacuated.
Japan had come to realize there was nothing to be gained by holding
this Aleutian real estate, which had become difficult to supply. Experience
on these islands had also convinced them that they were unsuitable bases
5
for bombing the homeland, one of the causes for the original occupation,
so why hold them longer? Rear Admiral Masatomi Kimura had entered
the harbor with two cruisers and six destroyers, loaded all personnel in less
than an hour and departed. What the US Navy had ruled impossible had
been made possible through use of the microwave radar that had allowed
Akiyama to find the island and its harbor in the eternal Aleutian fog. A
bit of luck in the dispositions of the blockading American fleet helped, but
being able to see in the fog and thereby move rapidly was the secret. The
Allies did not suspect that Japan had such shortwave radar. Such beliefs
naturally influenced the actions of the blockaders.
The search for variety of experience has sent more men to war than
patriotic zeal, ideological fervor, high pay or hopes of plunder. The preced-
ing incidents, to which many more could be added, have sufficient variety
in them to gratify this longing in all but the most voracious appetites, but
they have military experience that confounds anything ever encountered
before. In the endless struggles of the biological world, vision is by far the
faculty of greatest value. Certainly hearing and smell contribute advan-
tages, but their power is almost trivial in comparison with sight. Other
than a few crude uses of infrared, the First World War saw no improve-
ments in this all-important sense beyond the telescopic equipment that had
evolved since Galileo. The introduction of radar, a completely new way to
see, in the Second World War altered the basis of warfare more profoundly
than any of the inventions that had marked the industrialization of combat.
To be able to see an adversary in the dark, in fog, at distances difficult or
impossible even with the best optical equipment under the best conditions
was unprecedented. This was radar, and it is its story that I propose to tell.
1.2. ELECTROMAGNETIC WAVES

Ask an electrical engineer or a physicist to name the most momentous
occurrence of the 19th century, and the answer could well be ‘Maxwell’s
formulation of electromagnetic theory’. More likely, the answer would
be the clipped ‘Maxwell’s equations’—four equations that one can write
without crowding in the message portion of a picture post card. In them
reside all our knowledge of electricity and magnetism except that lying
in the domain of quantum theory, and even there they have furnished
reliable guidance. They are to the engineer Truth. They are to the physicist
matters of veneration that form an ever present model in the search for the
description of subatomic forces. Maxwell’s formulation was the second
great fundamental law of physics, following Newton’s laws of mechanics
by two centuries.
Newton’s laws united the mechanics observed on Earth as described
by Galileo with the motion of the planets as set forth by Kepler, the work
of just two men. By way of contrast Maxwell’s theory united the experi-
mental and theoretical work of many investigators, great and small, and
6
Preliminaries
concluded a striving to understand mysteries that had intrigued our ances-

tors back into prehistory. Amber, the fossilized resin whose Greek name,
electron, has so enriched our vocabulary, has been found in the remains of
European lake dwellings where the certain presence of fur must have led
to the discovery of its strange, attractive powers. Amber was an object of
trade in the ancient Mediterranean civilization and is well remembered in
its literature. The discovery of magnetism probably does not lie so far back,
as it must have taken place early in the iron age, very likely by a smelter
who encountered a lodestone in his ore. Legend has it that a shepherd,
Magnes, discovered the lodestone. If he made this discovery as reported
by finding the nails of his boots magnetically extracted, it reflects badly
on his cobbler. But the legend improved, and soon had nails pulled out of
ships to the consternation of the crews.
The ancients pondered these two forces and found it puzzling that
amber and magnets were so similar and yet so different. Centuries passed
with knowledge of the two forces stagnant, generally in the custody of wiz-
ards and charlatans. From the 12th century, two millennia after the earliest
records of magnetism, one finds confused reports of the north-pointing
characteristics of a magnet, and a century later Peter Peregrinus invented a
compass by which a mariner could steer, with immediate consequences for
navigators and the basis for the opening of the Age of Exploration. Queen
Elizabeth’s physician, William Gilbert, published an exhaustive treatise
on magnetism, based in good part on his own experiments, that gave the
subject a sound foundation. Henceforth magnetism could be considered a
science.
Electricity, more correctly electrostatics, advanced during those years
no further than separation of charge by friction, and knowledge of it was
left to industrious mechanics, who devised ever cleverer ways of building
up charge. The discovery of electrostatic influence and the capacitor, go-
ing by the descriptive name of the Leyden jar, gave rise to much improved
electrostatic generators that allowed some rather spectacular demonstra-
tions in courts of the 18th century. The discharge of a capacitor charged to
tens of kilovolts gained the respect that it retains to this day. This was the
state of the subject to which Benjamin Franklin, a Philadelphia publisher,
applied his remarkable skills. Out of his experiments and study came
our complete qualitative knowledge of electrostatics, but his most notable
and well remembered discovery was the identification of lightning as an
electrical discharge.
The flow of electric current was discovered from its physiological ef-
fects, proceeding from the twitching of frogs’ legs through chemistry to
the Voltaic cell, which allowed the chemically generated potentials to be
identified as electric. In 1819 Hans Christian Oersted, a professor of natu-
ral philosophy in Copenhagen, made the great discovery that the current
from a chemical cell passing through a wire deflected a compass needle.
The suspicion, so long held, although without basis, that electricity and
7
magnetism were in some way related had been proved, and the search for
an electrical effect generated through magnetism led to Michael Faraday’s
discovery of the magnetic induction of electric currents a dozen years later.
The end of the 18th century saw the initiation of quantitative mea-
surements on matters electric with the inverse square force law being the
first established. From it grew a system of units for electrostatic quantities,
followed quickly by a similar system for electrodynamic phenomena. It
was, of course, realized that quantities measured in units defined by elec-
trostatics could have their values expressed in units defined by electrody-
namics. A conversion factor, c, quantified the conversion, and although its
value was unknown it was known to have the dimensions of velocity. The
system of units now universally used by engineers, which includes the fa-
miliar ampere and volt, eliminates the need of the conversion factor and its
dimensionality by adding current to the three mechanical quantities taken
to be fundamental: length, mass and time.
In 1856 this ratio, c, was measured experimentally by W Weber and
F Kohlrausch [1]. A Leyden jar of known capacity was charged to a po-
tential determined with an electrometer, thereby establishing the charge in
electrostatic units. The jar was then discharged through a galvanometer,
a current-measuring device calibrated in magnetic units. The ratio of the
charge calculated for the Leyden jar in electric units to the charge mea-
sured by the galvanometer in magnetic units proved to be the same as the
velocity of light, which had been known accurately for only a few years.
The historical records do not tell of the psychological effect of this on the
theorists of the day, but it must have been strong. Faraday had shown a
little earlier that polarized light was affected by magnetism in some exper-
iments furnishing a hint that light and magnetism were related, but this
new result certainly gripped the mind. Its significance was hardly lost on
James Clerk Maxwell, a professor at King’s College, London, who wrote
‘we can scarcely avoid the inference that light consists in the transverse un-
dulations of the same medium which is the cause of electric and magnetic
phenomena’ [2].
Maxwell’s achievement was not only to unify the laws governing
electromagnetic fields but to change entirely the way in which physicists
look upon the universe. Maxwell wrote sets of partial differential equa-
tions that stated the three laws determined earlier by Gauss, Faraday and
Ampère plus the mathematical statement that there are no isolated mag-
netic poles analogous to electric charges. From this he obtained no mathe-
matical indication of the wave motion so necessary if light was of electro-
magnetic nature, as strongly suggested by the recent experimental value
of the unit conversion factor, c. He noted a curious lack of symmetry in
his formulation for Ampère’s law, which had no time derivative to match
the magnetically similar statement of Faraday’s law. He remedied this
by boldly postulating the existence of a quantity, named the displacement
current, that gave his equations the desired form. When applied to space
8
Preliminaries
having neither static nor moving charges his equations then reduced to the
wave equation with a velocity equal to c.
The effect of this formulation, published in 1864, was in no way com-
mensurate with its importance. Physicists found the introduction of the
displacement current artificial and unnecessary and the mathematics im-
penetrable, primarily because the equations were not given in the crisp
notation familiar to modern readers. Several eminences of physics went to
their graves without understanding. Nine years later and five years before
his untimely death, Maxwell published his treatise on electricity and mag-
netism, which attempted to clarify his ideas; it was generally respected but
generated little enthusiasm, and waves of electric and magnetic fields were
not immediately produced in the laboratory. Two champions appeared on
the scene who altered the situation dramatically.
When the treatise was published a copy came into the hands of a
telegrapher, Oliver Heaviside, with no education beyond what had been
available to a child of the London lower class. Heaviside decided that to
understand electricity he would have to read the book and set about to
teach himself calculus and differential equations, both ordinary and par-
tial. In order to comprehend the material he invented a new branch of
mathematics (vector analysis) and in so doing reduced Maxwell’s page of
equations to the four recognizable by the modern reader. In the course of
this remarkable achievement, during which he lived off his kin, he invented
another branch of mathematics (operational calculus, transformed by rig-
orists to a subject honoring Laplace), derived the telegrapher’s equation,
which finally solved the problems of transmission lines to the great relief
of the owners of the new telephone systems, and engaged in delightful and
acrimonious controversy. He lived and died in poverty.
Heinrich Hertz also found inspiration in Maxwell’s treatise and
fought his way through to an understanding using mathematical notation
not greatly improved over the original. This understanding guided him
to the production of electromagnetic waves in his Karlsruhe laboratory in
1887. He produced waves of length from a few decimeters to meters and
demonstrated reflection, refraction and polarization. He measured their
velocity and found it as predicted. The response of the scientific world was
rapid and positive with numerous confirmatory experiments undertaken.
German scientists and engineers began to use his name for the unit of fre-
quency, and it was finally taken up by the International System of Units
in 1960. The Nazi regime attempted to have it replaced because of Hertz’s
want of racial purity but surprisingly failed. He died before the Marconi
experiments showed the communication possibilities of his waves.
The scientific basis of radar was not complete by the turn of the cen-
tury, but the fundamental understanding of what have come to be called
radio waves was complete. Small libraries were to be written about their in-
teraction with antennas, wires, transmission lines, waveguides, dielectrics,
the ionosphere, the sea, clouds, meteors, nuclear explosions, aircraft, ships
9
and more, but the fundamental principles were well established and the
dramatic predictions verified. By the turn of the century wireless telegra-
phy found ever wider application for marine communication, and in 1905
a major naval battle was fought in which wireless had much to do with the
outcome. The fundamental advances leading to radar and all of modern
electronics were now to be made in other parts of science.
Hertz’s waves were not originally seen as suitable for communica-
tion, and the first experimenters concentrated on demonstrating properties
that were recognizably similar to light. The virtues of much longer wave-
lengths than Hertz had used were noted by Guglielmo Marconi, an Italian
with strong family ties to Britain through his mother. Marconi became in-
terested in these waves as a student and began increasing the dimensions
of the dipole, giving him ever greater range. Family connections and capi-
tal allowed him to form a company for providing communication between
ships, an important application that he recognized and dominated com-
mercially for the first decade of the century. The subject naturally attracted
the mentally active, and invention followed invention all through this cen-
tury bringing rapid changes in the design of transmitters and receivers.
Its story becomes the history of electronics from which we shall be able to
examine only a small part of the riches.
Hot filaments of carbon and refractory metals were found to emit
electrons, which could be made to flow from a hot filament to a positive
electrode, a vacuum diode. It was John Fleming of Marconi’s company
who applied the vacuum diode as a sensitive detector for wireless signals
and to which Lee de Forest added a grid between cathode and anode to
make the triode, an element capable of amplifying an electric signal. By
1920 huge arrays of triodes were generating easily controlled kilowatts
of radio frequency power and rapidly replacing mechanical alternators as
well as arc and spark transmitters with vacuum tube oscillators. The tri-
ode oscillator had been invented at about the same time by de Forest and
Edwin H Armstrong, who fought for a score of years for a court judge-
ment of a priority long since granted both by their colleagues. They also
fought about the regenerative receiver that came about when the inductive
coupling of the oscillator was reduced below some critical value.
During the 1920s the vacuum tube transformed ‘wireless’ into ‘ra-
dio’, and broadcasting burst on the public as a new and almost universal
form of entertainment. The new circuit elements quickly replaced the am-
ateur’s spark equipment, and the demands of commercial wireless com-
panies for low frequencies drove them to wavelengths shorter than 200 m,
with which they soon began attaining astounding ranges with tiny frac-
tions of the power required by the big stations. In November 1923 a French
amateur in Nice established a two-way connection with two amateurs in
America [3]. Within a year amateur contacts were worldwide. These spec-
tacular feats were recognized to result from multiple reflections between
the ionosphere and the earth’s surface, but success in making a connection
10
Preliminaries
depended on the frequency used and time of day with other apparently
random effects thrown in. An empirical understanding of the ionosphere
and the way it affected long-distance transmissions was not long in com-
ing, with commercial and government use not far behind. The very long
wavelengths, which had been quickly overcrowded, were left for transmis-
sions for which the location of the receiver relative to the transmitter was
not well known, generally maritime. This pioneering by amateurs forced
them to relinquish some desirable frequencies when the commercial value
of ‘short wave’ became known. It was not the last time they would be
‘rewarded’ in this manner for their contributions to radio science.
1.3. PERCEPTIONS OF AIR POWER, 1919–1939

Jan Christian Smuts rode onto the pages of history during the third or
guerilla phase of the South African War as the leader of a particularly
successful Boer commando, one which rampaged through Cape Colony,
even menacing Capetown. With defeat Smuts, along with several other
prominent Boers, made peace with Britain and worked out a satisfactory
position for the Transvaal within the Empire. By the outbreak of war in
1914 he had become a trusted imperial advisor, and it was in this capac-
ity that Prime Minister Lloyd George called on him for recommendations
about a highly political military matter. On 13 June 1917 a group of Ger-
man bombers killed 162 (including 16 children at school) and wounded
426 in London and departed without loss despite the efforts of a much
larger group of defending aircraft [1]. A large number of air squadrons
and AA guns had already been deployed for defense against the attacks
of airships, which no longer dared daylight attack and had begun to suf-
fer losses high enough even at night that their end as an effective weapon
against Britain was at hand. The bombers had struck in daylight. That the
air defense of Britain already made use of a force much larger than was
reasonable given the possible damage that the raiders could do, and that
the civilian deaths, however regrettably distributed among women and
children, were insignificant compared to the daily attrition of the Western
Front, had no calming effect on public reaction, which showed signs of
panic. The Cabinet met to consider the problem, and Smuts was asked to
study it.
A conventional answer was obvious: more defending aircraft, more
AA guns, more ground observer stations and above all unified command,
all of which were incorporated in Smuts’s July report. Smuts had a broad
education, a penetrating mind and wide experience in war and peace and
followed the first recommendation with another a month later that was
certainly not conventional. He recognized that the air attacks, though not
militarily serious at the moment, were a completely new form of warfare,
unrelated to either the Army or the Navy, and proposed that the Royal
Flying Corps and the Royal Naval Air Service be combined and reorga-
11
nized as the Royal Air Force with status equal to the other two services,
including having a Secretary of State for Air in the Cabinet [2]. This would
remove the subordination of aviation to the Army and Navy and allow it to
organize air offense as well as defense. He further recommended forming
an independent bomber command to carry attacks to Germany. It was his
opinion that direct bombing of cities would become the principal means
of waging war in the future. The plan had enough political appeal to be
enacted quickly. After all, if sturdy Englishmen had been thrown into a
panic by a few bombs, the Germans would be so terrorized as possibly to
make peace. In 1918 cities in the Rhineland and beyond were bombed and
the first four-engine bombers capable of reaching Berlin began operation,
although the Armistice postponed that city’s experiencing air attacks for
a quarter of a century. The German defense against these attacks had not
been successful and had left a residual belief that the bomber could always
get through [3]. The attacks on London were stopped by mid-1918 but at a
cost of deploying 376 airplanes, 469 AA guns, 622 searchlights, 258 height
finders, ten sound locators and a balloon apron, requiring 13 400 men and
women [4].
When British fliers set about in the post-war world to protect their in-
fant and rapidly shrinking RAF from being devoured by its parent services,
Americans of the US Army Air Service returned from France filled with en-
thusiasm for the future of military and civilian aviation. Their leader was
Brigadier General William Mitchell, who had commanded them in France
and who had ideas about air power that went far beyond reasonable ex-
trapolations of his wartime experiences. As the son of a US Senator he in-
herited natural political instincts to further his cause, which he enveloped
with evangelism. His cause was the formation of an American Air Force,
independent of Army and Navy. His extravagant claims, demands and
charges soon led him to face a court martial and to enter the hagiology of
aviators.
British and American airmen came to accept similar views about
aerial warfare. Simply stated these views reduced to the idea that strategic
bombing would decide future wars. Exactly what strategic bombing meant
was never entirely clear but what it did not mean was clear. It did not mean
attacking the enemy’s armed forces as had been done in 1914–1918, some-
thing to be resisted at all costs. It did, of course, include attacking enemy
air power as a vital part of the ‘knock out blow’ delivered at the opening of
hostilities, but strategic bombing generally meant attacks on the enemy’s
industry, transport, communication and cities. It was assumed that attacks
on cities would lead to panic and immediate demands for peace. It would
have an ugly aspect that the airmen regretted but would insure a rapid end
to the war with far fewer casualties than even a few days of trench warfare
would accrue. Sir Hugh Trenchard, Mitchell’s British counterpart, valued
the morale effects of attacks on cities as 20 times the material.
Civilians began to expand on the ideas of strategic bombing during
12
Preliminaries
the inter-war years. There was a curious alliance of thought in Britain be-
tween airmen and pacifists on the terror and death that would come from
the sky, with the airmen emphasizing the terror and the pacifists the death.
Both assumed the usage of liberal dosages of poison gas. From a widely
read book [5] of the time: ‘But first we must make up our minds on one
very important point, namely, that gas will be used. Let there be no mis-
take about that!’. A spectacular movie issued in 1936 by Alexander Korda
entitled ‘Things to Come’ foretold war beginning on Christmas 1940 with
the dreaded bombs falling on London. The screen writers had missed the
teaching of Mitchell and Trenchard about a quick decision and prophesied
instead a quarter-century of war that almost brought an end to civilization.
For whatever value it had as prophesy, the scenes of attacking waves of
bombers and of destroyed cities showed remarkable prescience. The film
did little to calm the fears of society. The resulting expansion of Fighter
Command was the result of political pressure and viewed by the RAF as a
sop for the public.
If British and American airmen thought in terms of strategic bomb-
ing, their air comrades on the continent thought in terms of army support.
This is curious because they showed great interest in the writings of Giulio
Douhet, commander of the Italian air service during World War I, who
stressed strategic bombing in a book that had been published in transla-
tion in French, German and Swedish by 1936 but not into English until 1942
[6]. Thus the writing that best expressed the dogma of the Anglo-American
fliers was absent from their bookshelves during the 1930s, although trans-
lations had been studied at the US Air Corps Tactical School. The founders
of the new, independent Luftwaffe often mentioned Douhet’s effects on
their thinking, but in 1919 the German air service began an analysis of
military aviation, concluding that their ground support had been valuable
but their strategic bombing not. By their evaluation the RAF’s strategic
bombing of Germany in 1918 had cost the British more in aircraft lost
than in damage inflicted [7], and the Luftwaffe became, in fact and spirit,
subservient to the army. Unlike the RAF, which organized according to
mission—Bomber Command, Fighter Command, Coastal Command—the
Germans organized Luftflotten, air fleets, that had a balance of all types of
machine and even included anti-aircraft and airborne infantry units. Ger-
many did not reject strategic bombing as such and made extremely good
use of British, French and Czech fears of it during the months before the
outbreak of war, but neglected it during the pre-war planning. There was
some interest in long-range bombers, and two prototypes flew in the mid-
1930s, but the costs of such aircraft let them slip out of the production plans
for the new air force [8]. When development resumed, a terrible design
resulted, the Heinkel 177, the worst production airplane of the war.
The new Luftwaffe had few aviators in its top ranks. Albert Kessel-
ring, who commanded Luftflotte II in the Battle of Britain and in the inva-
sion of the Soviet Union, was transferred to air service against his will in
13
1933 and learned to fly at the age of 48 [9]. It is hardly remarkable that he

and his comrades carried with them an army point of view. Kesselring even
commanded ground forces again as the war proceeded, holding the Allies
at bay in Italy with very little air power. The superb army support tech-
niques that characterized the Blitzkrieg came from the experiences of the
members of Legion Condor in Spain, not from higher levels of command
[10]. Germany’s professional military caste also looked on war as a clash
between military forces, not as a clash between nations. Their ideals of ser-
vice grew out of studies of Frederick the Great and the War of Liberation
against Napoleon. ‘The professional soldier was very particular about the
proper observance of the conventions—the usages and disciplines of war.
These conventions were rather like a set of trade union rules, designed to
make the profession of soldiering tolerable’ [11]. Unnecessarily involving
civilians repelled them. That their careers became tied to a government
entirely without moral scruple is one of the ironies of the Second World
War.
Thus we find a most curious situation as war crept closer. Britain,
both her military and civilian components, put every available resource
into protecting herself against strategic bombing, which Germany was not
seriously planning. That the Luftwaffe lost the forced and unforeseen
Battle of Britain came about in no small part because they fought it ill
considered, whereas RAF Fighter Command had planned it for four years.
While Britain and America were specifying long-range heavy
bombers in the mid-1930s the Luftwaffe was testing prototypes of excel-
lent army support planes. Particularly valuable in that role was the Ju-87,
the Stuka dive bomber—valuable, that is, so long as the Luftwaffe main-
tained command of the air. Dive bombing had evolved in ground support
aviation on the Western Front in 1918 where it had found limited special
application. The American Navy and Marine Corps developed it in the
1920s but the RAF and the US Army Air Corps (upgraded from Air Service
in 1925 but not to be independent of the Army until after the war) were
not interested in techniques or equipment that deflected them from their
ability to win the next war alone. In 1939 dive bombing was standard for
ground support only in the Luftwaffe [12]. This had come about in part
from the accuracies demonstrated by it in comparison with the results of
the inferior German sight for horizontal bombing [13].
Japanese attitudes toward the use of air power were better expressed
by their actions during 1938–1941 than by theory. The war with China
that began on 7 July 1937 soon bogged down along the coast and railway
lines. The army could take land but was unable to force the Chinese army
into a decisive battle. This exasperation caused them to attempt to reduce
the Chinese will to fight by bombarding cities, beginning mid-1938. In
May 1939 Chungking, the Nationalist capital, suffered 8000 killed in two
days and by 1941 experienced ‘fatigue bombing’ in which aircraft were
overhead all day, there being no air opposition left [14]. The Japanese
14
Preliminaries
army and navy maintained separate air arms that were not known for
inter-service cooperation.
The legal and moral aspects of strategic bombing were widely dis-
cussed in Britain during the inter-war years but essentially ignored in Ger-
many because of the attitudes of the professional military and later by
the suppression of critical thought by a dictatorial regime. In the United
States any strategic bombing was expected to be accomplished with the
miraculous Norden bombsight, hence precisely dropped on key targets of
the enemy economy, thereby minimizing civilian deaths—the combustible
cities of Japan perhaps excepted [15]. But strategic bombing was not the
official mission of the Army Air Corps. It had gained the responsibility of
coast defense in 1931, allowing it to order the much wanted four-engine
bombers, which American fliers were sure could sink surface ships un-
der operational conditions and be ready to fulfill the unexpressed strategic
mission if war came [16]. But British thought was thoroughly exercised
by the legal and moral problems of bombing, and many articles and books
appeared.
The legal case for strategic bombing was perhaps best made by
J M Spaight in 1930 [17], who based his arguments on the examination of
the history for the previous two centuries and on international law. Spaight
considered an air force to be similar in function to a navy, and one of the
time-honored functions of navies had been to raid enemy ports. The Royal
Navy made several raids of this sort during the War of 1812 with towns
being bombarded and destruction carried out by shore parties. Other port
cities had been bombarded during the 19th century with severe damage
to the civilian population as a result of inaccurate gun fire and the desire
to destroy civilian property of value to the war effort. These depredations
had been found to fit international law, such as it was, and to be accepted
practice. Spaight noted that during the American Civil War destruction of
cities and the civilian economy became routine and quoted General Philip
Sheridan: ‘Reduction to poverty brings prayers for peace more surely and
more quickly than does destruction of human life, as the selfishness of man
has demonstrated in more than one great conflict’ [18].
Efforts were made around 1932 at Geneva to place some kind of con-
trol on air war with proposals generally running toward either banning
the bombardment of cities or prohibiting air forces outright. None of the
proposals was adopted, and September 1939 saw no international law con-
cerning the employment of air power at all. There were Hague rules for
warfare on land and sea but nothing for the air. The use of gas was banned,
and it was not employed significantly during World War II. It is an open
question whether such a ban on air power would have been effective.
The airplane was not the only novelty to have emerged in 1918. The
tank had done more to gain victory for the Allies than the airplane. It was
not a particularly radical device. A mobile soldier protected with armor
had been the reason behind the knight of the Middle Ages, brought to an
15
end of his dominance by firearms. It did not require a lot of imagination

to replace the horse with a tractor that could carry sufficient armor to stop
bullets.
The first employment of tanks in 1916 and 1917 had not been ex-
traordinarily successful, but by 1918 their tactics had been mastered and
the effect at the opening of the Amiens offensive was dramatic—‘the black
day for the German army’ that convinced Ludendorff that the war was lost.
The Germans had recognized by then the danger of the new weapon and
had fielded anti-tank guns that quickly helped reduce the rate of advance
to one typical of the Western Front. The Air Force attempted to counter
this, as a report from the beginning of the campaign describes: ‘The tanks
suffered heavily from gun fire, and one of the lessons of the battle was
that it was imperative to allocate a special fighting squadron to the Tank
Corps in order to develop liaison and offensive plans for dealing with the
anti-tank guns in the next attack’ [19].
The lesson was carefully studied but the results were not greatly
improved. The aircraft attacking battlefield targets were unarmored and
took heavy losses from ground fire. An analysis by Wing Commander John
Slessor questioned whether air actions had had any effect on the Amiens
offensive at all other than preventing with an overwhelming air superiority
German reconnaissance observing the build-up and thereby gaining the
crucial surprise. Specifically, the Air Force was unable to cut the rail and
road supply lines, which allowed the penetration to be contained within
a few days [20]. That was the lesson, a consequence of primitive aircraft,
that was fixed in the minds of the RAF after the war. It meant they required
a mission other than army support to justify their existence. That mission
was to be strategic bombing.
Most tank use in 1918 had had them moving forward at the speed of
advancing infantry, eliminating the dreaded machine gunners and often
terrorizing the enemy, although there had been more imaginative and suc-
cessful tactical variations that led to local breakthroughs. Post-war military
doctrine saw no reason to change the direct infantry support mission, and
it was still the doctrine of the British and French commands in 1940. It was
certainly not the doctrine of younger officers, especially those in the Royal
Tank Corps. They, speaking frequently through the voice of B H Liddell
Hart, saw matters from a radically different point of view. That they were
overruled led to the great disaster for France and Britain in 1940.
Hart had been gassed on the Somme while serving as an infantry
officer. He wrote the Army’s official Infantry Training manual in 1920 and
began a long period of military scholarship after being invalided out of
the service in 1924. He was military correspondent for the Daily Telegraph
and later for the Times, functions that kept him in close association with his
old comrades. He observed and criticized military exercises and gained
acceptance into the ruling stratum of British society.
Liddell Hart recognized the dominant effect of mechanized forces
16
Preliminaries
on the outcome of the past war, which he formulated into a widely dis-
cussed book, Paris and the Future of War [21], playing on Paris as the Greek
hero, the city as a military objective and as a capital. Out of these gener-
alized concepts grew his ideas to maximize mobility and surprise with a
fast moving armored breakthrough supported by low flying ground attack
planes and accompanied by motorized infantry. These tactics were tested
by J F C Fuller with the Royal Tank Corps on the Salisbury Plain and con-
vinced the tank men of their validity. The aging, horse-loving leadership
of the Army rejected them, and the RAF wanted nothing to do with army
support. The writings of Hart and Fuller were translated into German and
found a strong disciple in Heinz Guderian, who put them into practice in
Poland and in the Ardennes breakthrough that defeated the British and
French armies with such startling suddenness. He generously gave credit
for the foundation of his victories to Hart and Fuller but noted that he had
had first-hand instruction at Cambrai [22]. The RAF had to learn—or was
it relearn—those lessons over the desert battlefields of North Africa.
The aircraft carrier emerged out of the First World War as well as
the tank, but its contribution to the struggle was close to zero. Planes had
landed on and taken off from platforms added to warships even before
the war, and during the war Britain converted a merchant ship to the first
aircraft carrier, HMS Argus, which was used to make a raid on a German
coastal air station. The part the carrier was to play in future warfare was
much less well defined than that of the tank. Indeed how air power itself
would affect sea power in the future was a wide-open question, one that
greatly agitated naval thinking during the immediate post-war years. The
mental ferment about tank warfare found its great debate in England, but
the equivalent about carrier warfare took place in America. The contrast in
the style of the discussions was characteristically different. The British dis-
putes took the form of articles in service journals and newspapers, books,
heated talk at table over brandy and cigars, and exercises carried out on
the Salisbury plain, the interpretations of which left the opposing parties
as far apart as before. The American disputes were much less tidy, with
factions contending within the Army and Navy, superimposed on fights
between the two services. It saw Admiral William Moffett, ardent builder
of naval aviation, the fierce enemy of General Mitchell, to the extent of
testifying against him at his court martial. It saw General John J Pershing,
Commander of the American Expeditionary Force, announcing that the
battleship was still the backbone of the fleet and Admiral William Sims,
Commander of the US Atlantic Fleet in the recent war, asserting that no
more battleships should be laid down. It involved Congressional com-
mittees and sensational newspaper headlines. At the middle of it stood
Mitchell, whose rhetorical craft was unfortunately limited to hyperbole
[23]. Few of the participants can be considered, given the vantage point of
time, to have demonstrated great sagacity. There was no American Liddell
Hart for naval aviation, but Hart’s message was ignored by his own people
17
whereas sound naval doctrine emerged from the American cacophony of

the 1920s.
Could aircraft sink surface warships? This was the question about
which the controversy turned. Scarcely returned from France, Mitchell
called for a chance to sink a battleship with Army bombs. Equipment for
the tests was at hand: a number of warships ready for scrapping and a few
impressively large bombing planes. All parties called for tests, but each
had its own view of how the tests were to be carried out. When all the
explanations and interpretations for the results of the tests, which began
in November 1919 and ended five years later, had been voiced, one fact
remained uncontested: a battleship of the most modern design could be
sent to the bottom by a well placed bomb carried by an airplane.
What did not and could not come from these tests was the reality of
the situation. The ships were at anchor; their watertight integrities were
not maintained; they were undefended by other planes or AA fire (which
Mitchell dismissed then and forever as hopeless [24]) and there had been no
opportunity for damage control. But he and his Army comrades concluded
that land-based heavy bombers could dispose of any surface vessel and that
the nation’s coasts could be so protected. Surface vessels were obsolete,
expensive and no longer needed for national defense. Regardless of the
protests voiced by the admirals about the meaning of the tests, they made
the Navy conscious of air defense.
While these shouting matches raged Moffett called for carriers. The
first, USS Langley, the conversion of a collier, was commissioned in 1922.
Her function with fleet maneuvers was so impressive that two uncom-
pleted battle cruisers were finished as the carriers Saratoga and Lexington,
commissioned in 1927. Tactical doctrine began to take shape, and the
three kinds of carrier plane that were to be decisive were recognized: dive
bombers, torpedo bombers and fighters. Mitchell had only scorn for dive
bombing and aircraft carriers [25]. Others saw with insight that their thin
decks made carriers very fragile, and the British soon began to build them
with armored decks.
Britain and Japan launched carriers in the 1920s and set about similar
armament programs, although without the level of contention that marked
America’s entrance. Britain commissioned HMS Eagle in 1923, Furious in
1925 and Glorious in 1930; Japan commissioned Akagi in 1927 and Kaga
in 1928. At the outbreak of war the Japanese carrier force was the best,
whether measured in number of ships and aircraft or in the quality of
equipment and training. The American force was smaller, had poorer
planes and crews that had not been pushed in their capabilities to the level
of the Japanese—but it had radar, the ability to read Japanese signals and a
comprehensive training program. Britain was not free of carrier acrimony.
The Royal Navy had lost airplanes and pilots to the Air Force in 1917
and found this increasingly exasperating. When the Air Force suffered the
inevitable cuts, the carriers lost much of their bite [26]. The Navy continued
18
Preliminaries
to press for full control over the Fleet Air Arm and succeeded in August
1937. Force of argument had eventually combined with a diminished fear
in the RAF of being eaten alive by its parent services to decide the outcome
[27]. Their carriers had obsolete aircraft but fortunately were to have only a
brief, if bruising, encounter with their Japanese counterparts. The German
Navy never gained control of an air arm.
If any of the navies gave thought about how aircraft might affect
the conduct of submarine actions, it is not apparent. In so far as Britain
and America were concerned the U-boat problem had been solved by con-
voys in 1918, and any residual difficulties would be taken care of by asdic
or sonar, the underwater sound detection methods that the two navies
trusted. All navies had to re-learn the submarine lessons of 1918. In 1938
only Dönitz understood, and he had not yet convinced his superiors.
The Second World War was to put to the test all of the theories con-
cerning air power. All had been conceived without so much as an inkling
of radar or its possibilities. In strategic bombing radar would dominate all
aspects of the conflict, and almost everything was to proceed differently
than expected, the changes ever confounding the planners as the conflict
evolved. Radar would favor the defender one month, the attacker the next.
Radar would affect naval warfare to just as high a degree. It would pro-
vide fragile carriers with electronic armor. Admiral Dönitz was to write a
completely new chapter on the tactics of submarine warfare, but aircraft
and electronics would be employed to negate it. Aviation’s use for ground
support would, under the force of circumstances, evolve from the tactics
established in 1918, and radar would not affect it until late in the war.
1.4. NAVIGATION IN 1939

Navigation is now a commodity, something that can be purchased for a re-
markably small sum. The mariner who wishes to know his place upon the
sea need only glance at the liquid crystal display of latitude and longitude
to be found on the bridge, and it is irrelevant whether he runs in fog or even
a storm. A few hundred dollars can provide him his position at any time to
an accuracy of tens of meters. Signals from orbiting satellites are received
by radio and interpreted with an attached computer, replacing at one the
skill and art that had always set navigators apart from the rest of mankind.
For years now aircraft have followed courses set by devices sensing their
accelerations and accurately calculating whence they have come and how
they must steer. The completely automatic pilot is a technical reality.
The contrast between the present state of things and that of 1939 is
greater than that between 1939 and the age before the invention of the
ship’s chronometer. Just the changes brought about during the Second
World War completely remade the world of air and sea travel, changes
resulting from radar and its companion electronic techniques. To appreci-
ate the accomplishments of radar and how they affected the nature of the
19
war requires a knowledge of the state of navigation at the opening of the

struggle.
In 1939 the word ‘navigation’ usually carried the modifier ‘celestial’,
for it was by observing the stars that sailors and more recently airmen had
found their way. Latitude was the easiest to obtain; with an ephemeris
one could have it quickly from a measurement of the altitude of the sun or
any other listed star as it crossed the observer’s meridian, the north–south
great circle in the sky that lies directly overhead. Longitude was equally
simple in principle but far more difficult in practice. One must know the
time at which a given star passes through the meridian and compare it
with the time when it had passed over the location of zero reference, taken
internationally as the Royal Observatory at Greenwich. In addition to
the ephemeris that gives the time a given star passes over Greenwich one
needs a clock that runs on Greenwich time. It was the absence of a clock that
would run accurately aboard a rolling, generally damp and uncomfortable
ship that led to countless vessels breaking up on unexpected rocks.
A voyage from England to Jamaica and return in 1762 with a
chronometer manufactured by John Harrison showed an accumulated er-
ror in longitude of less than 2 minutes, which met the standards set forth
by the Board of Longitude for the award of a prize of £20 000. The skills
of horologists soon provided seamen with excellent instruments at afford-
able prices, and the ticking of three chronometers became one of the many
sounds to be heard in the cuddies of deep-water sail. The substantial prize
for a practical means of determining longitude had not been intended for
a practical man, however; the Board had had an astronomer in mind, and
Harrison waited ten years to receive his reward [1].
The period before 1939 saw a number of practical aids for navigators
in the forms of tables easing calculation and the issuance in 1914 of the
Nautical Almanac abridged for the use of Seamen and in 1933 the Air Almanac.
Agreement to begin the day of Greenwich Mean Time at midnight rather
than noon ended a source of unnecessary confusion [2].
Celestial navigation presented the navigator with a sufficient variety
of difficulties as to transform what appears to be scientific into the realm
of cunning. Foremost was the need to see celestial objects and the horizon
at the same time. When the horizon cannot be seen, as is often the case
in an airplane, a spirit level had to be attached to the sextant. When the
sky was overcast one had to rely on the unfortunately named method of
dead reckoning, which predicts successive positions from the course and
speed of the vessel or aircraft. This involved also knowing the effect of the
motion of water and wind on the craft. For a steamer this is not so difficult,
although days without a fix could produce tension on the bridge. For an
aircraft moving at 20 or 30 times a steamer’s speed and much affected by
unknown, possibly high-speed winds the results could be very serious.
The need to make observations and calculations correctly and rapidly in a
noisy and bitterly cold airplane did not contribute to their reliability.
20
Preliminaries
Fliers navigated during the day by comparing their view of the

ground with their map while keeping an eye on the compass, and they
especially liked to follow the railroads. Unexpected clouds often meant
serious trouble. In 1939 Royal Air Force bombers were unable to find their
targets in daylight 40% of the time [3]. Night flying became sufficiently
common in the 1930s that beacon lights began to add their rotating beams
to the evening stars that rural people watched as they relaxed at the end
of day. Flying over the open ocean presented an obvious extension of
difficulty.
Radio brought some degree of help after the First World War in the
radio direction finder and the radio compass. Direction finders for low-
frequency signals proved to be rather good, and various shore stations
could determine the direction to whatever ship or aircraft made request.
The information from two or more such stations yielded an accurate fix.
Ships and planes with loop antennas could determine the direction to
known stations and establish their positions without the need of time-
consuming interrogations. This was much more difficult for aircraft be-
cause only small loop antennas could be used, and adjusting the loop to
minimum or maximum signal was difficult in the noisy conditions that
usually prevailed in the aircraft of the time. The high frequencies that
were convenient for communication had poor directional qualities, owing
to the altered polarization of the wave reflected from the ionosphere and
its irregular reflecting surface.
During the 1920s a system of radio navigation based on a 1907 pro-
posal by Otto Scheller came into wide use in equipment designed by the
German firm of Lorenz that was manufactured abroad under license and
known as the ‘four-course range’. (Here ‘range’ is aviation parlance and
does not infer range as a measured distance, as the four-course range pro-
vided only direction or course.) In one configuration widely used before
the war, four vertical antennas located at the corners of a square, about
100 m on a side, received the same audio-modulated radio frequency on the
100 m band. Alternately feeding opposite antenna pairs with the Morse-
code dit–dah or dah–dit produced two overlapping figure-eight radiation
patterns that identified one pattern with ‘A’, the other with ‘N’. Four di-
rections emanated from the station where the intensities of the of the two
patterns were equal, and the signal received by the pilot was a pure tone.
If he strayed to the left, he got an ‘A’, to the right an ‘N’, or the reverse,
depending on the course [4]. By 1939 radio ranges had led to the phrase
‘on the beam’ entering the language of the young and technically minded.
The known imperfections of aerial navigation do not seem to have
troubled those planning strategic bombing and the wildly inaccurate
bombing of Britain by the Germans in the First World War motivated no
intense navigation studies [5]. The traditional methods were taught, but
few exercises attempted to evaluate this crucial aspect of American and
British war plans. Before war broke out both thought in terms of day-
21
light bombing. If clouds obscured vision, navigation suddenly became

quite problematic and bombing impossible. Navigation over a completely
blacked-out enemy land left the attacker entirely dependent on the stars
or whatever radio signal he might devise. The Germans, who were not
planning strategic bombing, were nevertheless concerned about being able
to guide their planes over the enemy at night. The firm Lorenz AG had
marketed a beam navigation system and were given the task of construct-
ing one that would be good enough to drop bombs accurately at night or
through clouds. An excellent beam system was ready at the outbreak of
war, but its similarity to known systems was to make countermeasures
relatively easy.
Aircraft introduced a third dimension into navigation, altitude. It
was not so difficult to determine as latitude and longitude, but the accu-
racy of its value was extraordinarily critical. Uncertainties by amounts
trivial for the other two coordinates could be fatal to a flier. The method
used initially was the decrease in atmospheric pressure as one ascends,
discovered by Galileo’s secretary, Evangelista Torricelli, the first person
to create a sustained vacuum. Aneroid barometers were installed in the
earliest aeroplanes and were unrivaled during the first four decades of
flight.
Such instruments have two serious flaws. In an ideal atmosphere they
give the height of the aircraft above mean sea level, but fliers must deal with
real atmospheres in which weather dominates and not with ideal ones.
Discrepancies can run to several hundreds of meters as a result of local
barometric pressure variations. This can be corrected with an adjustment
at the time of take-off, if the altitude of the field is known, but such a
correction does not necessarily apply to a location far removed from the
origin of the flight or even at the origin hours later. More important than
these difficulties is the uncertainty about where the ground is. Height
above mean sea level is useful, but it is height above the surface that is
critical. When flying over poorly mapped terrain or when lost, height
above the ground can be a matter of life or death. A bombardier must
know his altitude above the ground accurately, if he is to hit the target. In
either case the inaccuracies inherent in the aneroid altimeter can be serious.
As a consequence of these problems inventors sought other meth-
ods, and radio formed a favorite approach. During the 1920s and 1930s
patent clerks were busy sorting out the tenuous differences in the many
suggestions for radio altimeters that were proposed. Heinrich Löwy was
apparently first, suggesting in 1923 the use of pulsed high-frequency waves
by timing the arrival of the reflections from the Earth’s surface. The elec-
tronic techniques of the time did not allow realization of his suggestion, but
that did not prevent it being followed by many others employing reflection
phenomena, either with pulsed or continuous waves [6].
The Aircraft Radio Laboratory of the US Army Air Corps and the
Naval Research Laboratory worked on the problem with a Navy design,
22
Preliminaries
which came out of their radar research. A contract was given to the Ra-
dio Corporation of America, and a satisfactory pulse altimeter prototype
was demonstrated in 1937. Its price of $20 000 caused much Air Corps un-
happiness, but by haggling and scouring ledger books for money two sets
were bought. By April 1940 a 40 kg set of good accuracy was designated
SCR-518 and put into production, ultimately reduced in weight to 12 kg
and functioning to an altitude of 12 km [7].
1.5. ANTI-AIRCRAFT ARTILLERY, 1914–1939

Anti-aircraft artillery dates from 1871 when the Prussian forces besieging
Paris countered the French balloons, used to communicate with forces out-
side the city, with a hurriedly designed Ballon-abwehr-kanone, a Krupp
36 mm breech-loading gun allowing high elevation and complete traverse
[1]. By 1914 ballistics had made great strides since that beginning but
aiming had not changed and was quite incapable of hitting the fast, ag-
ile aircraft that presented themselves as targets. The problem was similar
to that confronting naval and coast defense gunners, a target that would
change its position significantly by the time a projectile fired at it arrived.
But it was vastly more difficult, for the speeds were much greater and the
targets much smaller; furthermore, ships are confined to the surface of the
ocean, making the calculation of gun orders a two- rather than a three-
dimensional problem. The small size of the target meant that a direct hit
by an artillery projectile was unlikely to an impossible degree, so the shell
had to be made to burst with a time fuze set to explode when the gunner
calculated it should be close enough to be destructive.
If the target could be seen, either in daylight or illuminated by search-
lights, a series of three independent measurements had to be made of its
present position to predict its future position. The target had to be tracked
in the cross- hairs of telescopes to obtain its present azimuth (horizontal di-
rection) and elevation angle, relatively simple to do unless angular speeds
were very high or visibility poor. The third and key measurement was the
slant range to the target, which was harder to obtain. It required either
a stereoscopic optical range finder operated by scarce observers with ex-
ceptional vision or triangulation from a base line of at most a few meters.
Neither could determine ranges accurately enough for calculating firing
data. These three measurements had to be converted into gun orders of
elevation, azimuth and fuze time, and the calculations had to be made
simultaneously with the observations. The computer, called a director or
predictor, was a mechanical analog computer that used cams of extraor-
dinarily high precision. When all of this was working well and the target
conditions were ideal, the results were good enough to make the fliers
worry, but the conditions were seldom good, and extravagant amounts of
ammunition were consumed with little effect.
Raids at night required that the gunners find and hold the attacker in
23
the beam of a searchlight, a difficult problem in its own right. Sweeping the
sky was of little use, but if the weather was favorable, an array of listening
horns could give clues to put skilled searchlight operators onto the target.
In fog or above the clouds nothing worked, but then the same was true for
the fliers.
During a high-altitude bombing run there was a period when the
pilot of the attacking airplane had to set a level, straight course to allow
the bombardier to aim. This course was also a perfect opportunity for
AA gunners, as close to ideal as they would find, and provided a strong
inducement to the fliers for reducing this time to a minimum, with de-
creased bombing accuracy generally the result. Good fire could drive the
bombers off the selected path. One of the reasons the Luftwaffe favored
dive bombers was the efficiency demonstrated by their Flak on these kinds
of run.
An alternative to directed fire was used by the British well into the
war: barrage fire. A region of bursting shells was placed through which
the attacking aircraft had to pass if they were to reach their target. What
it lacked in accuracy it sometimes made up in morale effect on the pilots
encountering the curtain of shell bursts. Scientists in operational research
had little good to say about this technique. Professor A V Hill dismissed it
‘as based on sloppy thinking and bad arithmetic’ [2].
Anti-aircraft gun fire became ineffective at close range because of the
guns’ inability to track rapidly, and this region fell naturally to machine
guns. In the 1914–1918 conflict these were adapted from existing types
and included the 37 mm Maxim pom-pom. During the inter-war years
more powerful automatic weapons appeared. John Browning contributed
the 0.50 inch and a high velocity 37 mm. The 20 mm Oerlikon came from
Switzerland and the 40 mm Bofors from Sweden. The US Navy developed
a 1.1 inch. All contributed to the streams of tracers that filled the air during
World War II, but few were to shoot with radar direction.
1.5.1. Germany
The German Army entered the First World War with six truck-drawn and
twelve horse-drawn 75 mm Ballon-abwehr-kanonen, organized as sepa-
rate guns, not as batteries. A year after the outbreak of war the number of
AA guns had increased by a factor of ten in addition to field guns set up in
various improvised mountings. By fall 1916 air units and air defense units
were organized into a single branch and anti-aircraft received the abbre-
viation that was to become internationally recognized, Flak, standing for
Flugabwehrkanonen. By 1918 there was a total of 2900 AA guns in service.
Fifty batteries [3] had computing directors of some kind, none entirely sat-
isfactory, but the Schönjahn-Gerät was good enough to become the basis
for the director used in World War II. Evidence for greatly improved tech-
nical capability is found in the numbers of Allied planes brought down
24
Preliminaries
by AA gun fire: of the 1590 total for the war, half fell during the last ten
months [4]. A new 88 mm gun saw service before the end of the war.
German anti-aircraft left the war with a tactical experience denied
to the Allies that was to prove of value in the later conflict. When tanks
encountered AA guns in 1918 they came out the worse for it, the result of
the high velocity and ease of traverse of the AA guns. Revisions of the
design in 1936 and 1937 turned the 88 into the ‘triple threat’, useful for
anti-aircraft, field or antitank artillery. The last capability would prove
extraordinarily useful in the steppes of Russia and the deserts of North
Africa, places where long, unrestricted fields of fire made such a weapon
very effective.
The Treaty of Versailles forbade the German Republic both an air and
an anti-aircraft arm, although clandestine development of both took place:
in Russia for air, in Sweden for anti-aircraft. Whether this had a reverse
effect during the rearmament of the 1930s is a psychological question, but
for whatever reason Flak was given high priority and held in high esteem
as part of Hermann Göring’s new Luftwaffe. By the fall of 1935 Flak had or-
ganized 15 heavy and three light battalions; 12 months later these numbers
had doubled. Naturally there was a special Flak school [5].
German anti-aircraft profited by experience in the Spanish Civil War
as did other branches. General der Flieger Sperrle led an air contingent for
Legion Condor that contained Flak units. By the end it had grown to nine
batteries of 88 mm, which were credited with destroying 61 Loyalist planes.
The versatile nature of the 88 proved itself, especially as field artillery in
Franco’s artillery-poor forces. On return to Germany the Legion Condor
veterans were distributed throughout their branch [6].
Assigning Flak to the Luftwaffe brought problems with the Heer
(Army) because this arrangement suggested poor protection for front-line
troops. This led to the organization of heavy machine gun battalions in
the fall of 1938 that were armed with the 20 mm automatic gun and under
Army control.
When war broke out in the summer of 1939 Germany had far and
away the most advanced anti-aircraft force of any nation, with 107 000
men on the rolls [7].
1.5.2. United States

The US Army not only had no anti-aircraft units in 1914 but had none when
America entered the war in April 1917, although the matter had been stud-
ied as early as 1913, airships being considered the possible raiders. In 1915
the Ordnance Department began the design of a 3 inch AA gun, which
reached prototype stage by 1917 but never reached troops in France. Ex-
perience with Curtiss JN-2s in the expedition to Mexico in 1916 ‘indicated
that airplanes had a sufficient tendency to fall unaided out of the sky’ so
that an air defense arm hardly seemed to be urgent, but the nature of the
war in Europe quickly dispelled such ideas [8].
25
That anti-aircraft artillery was assigned to the Coast Artillery Corps

came about in part from the success of the Royal Navy in holding the Ger-
man fleet in its harbors, making the possibility small that the heavy guns
defending the east coast cities would be needed [9]. The Coast Artillery
also had had experience shooting at moving targets even if an order of
magnitude slower and confined to two dimensions. Thus it was a detach-
ment of Coast Artillerymen who embarked for France to study European
AA techniques, establish a training school and organize American units.
They found that French gunners had evolved a more successful technique
than the British, who mistrusted ‘technical shooting’ and preferred to shoot
by eye [10]. With French instruction and French guns the Americans de-
ployed eight batteries in combat, credited with having downed 19 enemy
aircraft. Eight machine-gun batteries were credited with having downed
41 [11].
The National Defense Act of 1920 assigned air defense to the Coast
Artillery, which organized four battalions in 1921, expanded three years
later into four regiments of two battalions each. These were followed by
the organization of AA regiments in the National Guard and the Organized
Reserves. The Coast Artillery pressed for modern equipment and secured
from the Ordnance Department in 1925 a satisfactory plan, which came to
nought because of the budget restrictions of the 1930s. The Coast Artillery
School at Fort Monroe emphasized technical quality.
In 1937 tension between the United States and Japan resulted in the
War Department giving top priority to air defense. The number of 3 inch
guns was to be increased from 135 to 472, which would equip 34 regiments.
As the characteristics of the four-engine heavy bombers became known in
the mid-1930s the inadequacy of the 3 inch gun to reach them became
apparent, and by 1938 a successful prototype 90 mm gun was ready as a
replacement for the 3 inch. As part of a general strengthening of the arm a
37 mm automatic gun, which had been under development for more than
a decade, was to replace machine guns [12].
Despite the rejection of the Air Corps thesis that land-based bombers
sufficed to dispose of ships, the Navy took anti-aircraft defense seriously
and as funds became available in Fiscal Year 1938–1939 began improving
these capabilities in existing vessels. A memorandum to the Chief of Naval
Operations summarized AA firing as generally ineffective, resulting in the
designation of Admiral Ernest King as head of the Navy Department Anti-
Aircraft Board. The dual purpose, semi-automatic 5 inch/38-caliber gun
had been introduced in 1934 on the Farragut class destroyers. It was judged
to have adequate range and power and was to be mounted wherever space
allowed on all heavier ships as well. Fire control of matching excellence
was absent. The earlier decision to rely on the combination of 1.1 inch and
0.50 inch machine guns for close defense was rejected for the 40 mm Bofors
and the 20 mm Oerlikon [13].
26
Preliminaries
1.5.3. Great Britain

The preparation of the anti-aircraft defense of Britain fared substantially
worse than its American counterpart during most of the pre-war years. At
the time of the Armistice the homeland had 286 guns and 387 searchlights
in position; by 1920 there were none. Soon the gunners were reduced to a
small Regular Army cadre and an AA school. Such troops as existed were
in the Territorial Army, which had occasional evening drills and summer
exercises using inadequate equipment. Anti-aircraft suffered, then and
later, from a social sensibility the Americans would not have understood. It
was not a branch of service to which an Englishman, defined by his club and
his regiment, wished to be assigned, whereas graduates of West Point often
selected the Coast Artillery for assignment because of the technical nature
of anti-aircraft. That the men and women who fired on the Zeppelins,
Gothas and Giants received no War Medal tells us much of the attitude
toward this arm. Among some there was even the romantic idea that by
not facing their foes the gunners were distastefully comparable to guerillas
attacking from ambush [14].
Many in Britain shared Billy Mitchell’s opinion that AA guns were
useless and would remain so, there being no immediate cure for their
deficiencies. Vocal in this was Frederick Lindemann, later Lord Cherwell,
Churchill’s friend and scientific adviser. During 1914–1918 attacks on cities
had generally been made at night, and the AA guns of both sides had not
shown themselves particularly lethal—but then neither had the raids. In
America the Coast Artillery on assuming the duty of anti-aircraft sought
the help of the Ordnance Department and the Signal Corps for methods
of locating aircraft at night, yielding active, if modest research in both
services; no such pressure was felt in Britain, where radar for AA gunners
came along as an afterthought in the radar project, as the poor quality of
its gun-laying equipment was to show.
There were committee studies, however. In April 1923, when there
was uneasiness about the size of the French Armée de l’Air, the Steel–
Bartholomew Committee reported that Britain’s air defenses were non-
existent. The effect of the report was negligible. By 1934 the absence of air
defense began to cause serious concern, this time with the worry directed
towards Germany, and Sir Robert Brooke-Popham, the Air Force officer in
charge of the Air Defence of Great Britain, began serious study of the matter.
The committee that he headed recommended a first stage of preparation be
attained by 31 March 1940 of 17 AA batteries and 42 searchlight companies,
and this part of Britain’s arms build-up slowly began [15].
In the Army exercises during the two-decade armistice General Sir
Frederick Pile was one of the few Army officers who gave evidence of
understanding the significance of the changes that technology had brought
to the profession of arms. A hereditary title interfered not at all with his
enthusiasm for things mechanical, which brought him to the Royal Tank
Corps where he was remembered by Liddell Hart: ‘One of the few bright
27
spots in the exercise was the activity and ubiquity of the reconnaissance
group of armored cars and tankettes under Tim Pile’ [16]. As mobilization
began to accelerate in 1937 there was talk of creating an armored division,
and Pile wanted very much to command it. One tends to speculate on the
course of the future had it been formed then and Pile made its commander,
but it was not to be. Instead he was asked to accept command of the 1st
AA Division, which was being formed. He had had no experience with
anti-aircraft and came to the assignment because of the unpopularity of
his tank warfare ideas with the upper levels of the Army and because of
his obvious technical ability. There is reason to believe that Pile accepted
the command with small pleasure, but he was a fine soldier and devoted
himself with all the drive and ability that he possessed to what was in the
British Army a military backwater. The job was expanded in April 1939 to
leadership of AA Command, which shared the Battle of Britain with RAF
Fighter Command; but where Fighter Command received the praise and
honor it so richly deserved, AA Command more often received the abuse
of a public that did not understand. Pile’s nature was not one to show
resentment, but he was bitter about criticism that touched the skill and
devotion of his people, whom he commanded until the end of the war.
Pile took over a pitiful group of Territorials and appeared before town
meetings trying to bring his division to authorized strength through ap-
peals for volunteers. Only after Munich, which delighted Pile for the time it
gave him, did recruits begin to join in substantial numbers [17]. The 3 inch
guns from 1918 were unable to reach the altitudes of modern bombers, and
the new 3.7 inch guns did not begin to come off production lines in quan-
tity until after the Munich crisis. The old predictors required six skilled
operators and were not satisfactory at best. On one target course flown un-
der conditions ideal for the gunners but tactically unrealistic, only two hits
could be scored out of 2935 rounds fired. Most of Pile’s comrades looked
on AA fire as a way of making the enemy fliers nervous and making noise
to reassure the civilians. Tactically it was seen to have value in breaking up
the bomber formations, thereby giving the fighters a better chance. For the
first years of the war Pile had to soldier with this kind of technical support
and for most of the war with these kinds of attitude.
The Royal Navy also lagged their Yank cousins in anti-aircraft. A
report of January 1939 described the alarming vulnerability to air attack of
carriers and cruisers and found destroyers almost defenseless. The situa-
tion was so serious that foreign guns were ordered and licenses for man-
ufacture purchased in March, the 40 mm Bofors from Sweden, the 20 mm
Oerlikon from Switzerland [18].
Of all the weapons in general use in 1939 none were improved in such
a startling way by radar as was anti-aircraft artillery. When war broke out in
1939 the effectiveness of anti-aircraft fire ranged from thousands of ‘rounds
per bird’ for daylight shooting to tens of thousands for night. Automatic
weapons firing at low-flying planes did much better. The accuracy of heavy
28
Preliminaries
guns began to improve rapidly as gun-laying radar and better directors

came into use. By 1945 an airplane caught in the range gate of an SCR-
584 radar feeding the data to an M-9 director controlling an automatic
tracking 90 mm gun using proximity fuzes simply meant that the plane
was finished. It might save itself by violent evasive maneuvers, but if the
first rounds were not placed accurately enough to produce bursts, there
was no warning.
29
PHOTOGRAPHS: PRELIMINARIES
A 1931 transmitter–receiver
antenna pair from an exper-
imental 17 cm communica-
tions link across the English
Channel. It failed because the
circuitry to allow the simul-
taneous transmission of many
telephone channels had not
been developed, leaving it infe-
rior to submarine cable. Such
wavelengths were the basis of
early American and German
radar experiments but had to
be given up because no trans-
mitters of suitable power were
at hand. Photograph courtesy
of Bernd Röde.
Prototypes of the SC and SG

radars mounted on USS Semmes
in May 1941. The dipole ar-
ray at the top of the foremast is
the SC, a 1.5 m air-warning set
based in design on the XAF or
CXAM but smaller for use on
destroyers. The dark cylinder
just below the crow’s nest con-
tains the paraboloid antenna of
the Radiation Laboratory’s 10 cm
surface-search radar. Final de-
sign replaced the full with the cut
paraboloid soon to become famil-
iar to seamen. National Archives
photograph 80-G-700311.
30
Preliminaries
The antenna of an experimental 1.5 m radar being tested aboard USS Leary by the
Naval Research Laboratory in April 1937. Mounting it on the destroyer’s deck
gun allowed pointing the beam at targets. The next stage of development was the
XAF, which used a dipole array rather than the Yagi antenna shown here. The
success of XAF in the fleet exercises in early 1939 put the Navy on a strong radar
development program. National Archives photograph 80-G-700710.
Failed intelligence. In August 1939 Colonel Wolfgang Martini, Chief of Luftwaffe

Signals, chartered the airship LZ-130 for a flight along British coastal areas in the
hope of locating possible radar defense. There are various published versions of
this flight, each with abundant error. Determining the true course of events and
the reason for the failure to observe the radar that followed the ship during the
entire flight have engaged the detective instincts of four investigators in Britain
and Germany. The airship broke through the clouds in the vicinity of Aberdeen,
causing people to call the Dyce Station of the Auxiliary Air Force from which a
plane with a photographer took off and made this photograph. Royal Air Force
Museum, Hendon, London, photograph P26. Crown Copyright.
31
Failed intelligence. A German publisher issued annually a pocketbook describing

the ships of the world’s navies. When the 1939 volume appeared it caused conster-
nation in German radar circles because of a photograph dated 1938 of the black
Torpedo School Ship G 10 displaying a prominent Seetakt antenna just forward
of the foremast. The photograph was passed for publication by naval authorities,
all kept in the dark about the new technique and, of course, unable to recognize
the apparent mattress as the mark of a secret weapon. There is good reason to
assume that the British naval attaché in Berlin purchased the book and that naval
intelligence in London studied it, but there is no record of them having grasped
the significance of the antenna either, very likely for the same reason that the pic-
ture had escaped in the first place—radar was too secret. Photograph through the
courtesy of Hans Sorgenfrei.
32
CHAPTER 2
ORIGINS
2.1. ELECTRONIC COMPONENT DEVELOPMENT

So much of modern technology owes its origin to military requirements
that it is somewhat startling to learn that the pace of radar development un-
til 1939 was set instead by civilian electronic use. Except in Great Britain
the pre-war radar engineers worked on restricted, sometimes very tight
budgets. They could not have considered such extravagances as having
special vacuum tubes or other devices designed for their needs and had to
make do with the components available from civilian electronics suppli-
ers. But electronics was a strong industry even during the depression, for
radio broadcasting had boomed in the 1920s, and the public spent lavishly
buying radios, paying to have them repaired, replacing rapidly aging sets
with new models and in the United States buying the products that broad-
casting advertised. When hard times hit, the family radio replaced other
pleasures that had to be foregone and kept farmer and rancher advised of
the latest market fluctuations. Radio was good business throughout the
interwar years, and broadcast companies dreamed of the coming market
for television, and television—like radar—required high frequencies.
Military services did improve their communication sets during the
inter-war years, but these projects, with rare exception, came from the nor-
mal evolution of communication electronics of the civilian world, adapted
by the small military service laboratories to their special needs. Devel-
opment contracts to private corporations, now the principal method of
designing new equipment and the source of many post-war electronic mar-
vels, were completely unknown.
2.1.1. Cathode-ray tubes

Ferdinand Braun, Professor at the University of Strassburg, was one of the
eminences of early wireless. He saw the need for waveform diagnostics
and applied the newly discovered cathode rays to this purpose [1]. In the
anatomy of a radar set the cathode-ray tube is the heart. Cathode rays—
streams of electrons—had been observed when high voltages were first
33
applied to electrodes in vacuum tubes with effects dependent on the degree

of vacuum (by present standards scarcely a vacuum at all), the voltage
applied and the configuration of the electrodes. Julius Plücker discovered
them in 1859 when he succeeded in reducing the pressure in his discharge
tube to new limits and demonstrated that they could be deflected with a
magnetic field. The next two generations of physicists spent many happy
hours in the laboratory experimenting with them. In 1897 J J Thomson
produced cathode rays in well defined beams and demonstrated that they
were made up of grains of electricity, electrons, by deflecting them with
electric and magnetic fields. At about the same time Braun constructed a
tube with a cathode at one end followed by a positive electrode with a small
hole in it. When high voltage was applied and the gas pressure adjusted
to the right value, electrons passed through the hole and struck the glass
at the far end of the tube, causing the glass to glow. By extending the path
from the positive electrode to the target glass, Braun was able to deflect the
beam with magnet coils applied at the outside. Waveform analysis made
use of rotating mirrors to view the tube or of an external magnet rotating
in synchronization with the waveform being studied [2]. The electronic
time base or sweep circuit lay in the future.
Use of the technique was slow in establishing itself because the oper-
ator needed to be a master experimentalist, and just operating the tube and
its peripherals was a major undertaking, certainly not the kind of thing for
which the wireless men had time. In 1922 J B Johnson and H J van der Bijl
of the Bell Telephone Laboratories [3] produced a low-voltage cathode-ray
tube that Western Electric manufactured as the WE-224. It used two sets
of internal electrostatic deflection plates and had a modern appearance.
Its success resulted from something called gas focusing, which also set its
limits.
The need for the extreme high voltages in the early tubes was pri-
marily to produce a tight beam that made a small, bright spot on the glass
target, but the high voltages made the tube less sensitive and troublesome
to operate. If one used lower, more convenient voltages the beam diverged
and produced less light from the glass, resulting in a large, dim spot. John-
son produced a bright spot by having a copious electron beam taken from
a thermal filament impinge on a phosphorescent coating. By introduc-
ing some gas into the tube he could produce a counter-effect to the space
charge of the electrons that caused the beam to diverge by forming a core of
positive ions at its center that canceled the effect of the negative electrons.
The problem was that the electron beam was no longer light and nimble
but heavy and sluggish, as it had to carry the positive ions with it to re-
main focused, making the response time of gas-focused tubes about 80 000
times slower than high-vacuum tubes, argon being a favorite gas. Such
performance was out of the question for the future television receivers, so
major efforts went into correcting this while preserving the simplicity of
the low-voltage tubes. Such tubes were also what the radar men needed,
34
Origins
because the gas-focused tubes were much too slow to display a pulse of
only a few microseconds in width, a minimum requirement. The ease of
operation of the gas-focused tubes was strong motivation for correcting
the gas problem. Focusing the beam of a high-vacuum tube required a
new branch of engineering called ‘electron optics’.
Two names are deservedly associated with these developments:
Vladimir Kosma Zworykin in the United States and Manfred von Ardenne
in Germany. By 1930 the crucial element for radar and television was at
hand and by 1935 was incorporated into handy laboratory oscillographs
with frequency responses limited only by the amplifiers driving the cath-
ode ray tubes.
Zworykin was one of the many refugees who have enriched Ameri-
can life. He fled the new Soviet state at the age of 30 and became Director of
Research at RCA in 1929, the first of a series of leading positions that he held
with the company. At the same time Zworykin demonstrated his proto-
television [4] there were almost identical activities in Germany by von Ar-
denne [5]. Indeed, Zworykin and von Ardenne duplicated one another’s
inventions with near simultaneity for more than a decade: the electrostat-
ically focused cathode-ray tube, the flying-spot television camera and the
electron microscope. But whereas Zworykin worked for the growing giant
RCA, von Ardenne had his own private laboratory at Berlin-Lichterfelde,
which he had established at the age of 21 with capital loaned to him on
the basis of his youthful successes in radio. He was 23 when he publicly
demonstrated his all-electronic television system. To exploit the market
for cathode-ray tubes and oscillographs that was about to overwhelm his
laboratory he entered into partnership to form the firm of Leybold und von
Ardenne; to exploit the greater potential market for television he entered
into an agreement with the firm of C Lorenz [6].
2.1.2. Multielectrode tubes

Electron tube designers of the 1920s soon encountered the parameter of
the new triode that checked their plans: the capacity between the control
grid and the positive electrode, called the anode or plate. If one incor-
porated this capacitance into the tuned circuit of an oscillator or tuned
amplifier, one could work frequencies of tens of MHz, but if one wished to
build a high-frequency broad-band amplifier, this inter-electrode capaci-
tance drastically limited the frequencies attainable. The envisioned televi-
sion sets required radio-frequency amplifiers with broad pass bands, and
triodes alone could not do the job. The solution was obvious: drastically
reduce the grid–anode capacitance.
In 1919 Hiroshi Ando applied for and was eventually granted a
Japanese patent for a tube with a fourth electrode, another grid, intro-
duced between the control grid and the anode specifically intended to
prevent ‘the undesirable influence of the inter-electrode static capacity be-
35
tween electrodes almost entirely or substantially’ [7]. No use was made of

this specific property, however. In 1926 Albert Hull and N H Williams of
the General Electric Company described a similar tube [8]. (Double grid
tubes had been around since 1916, the invention of Walter Schottky, but
for them the second grid was connected to the anode in order to neutralize
space charge with the object of being able to use lower supply voltages.)
In another paper Hull presented examples of circuit design with tetrodes,
as the tubes came to be called [9]. At about the same time Bernard Tellegen
of Philips Research Laboratories patented a similar design [10].
If used for large-amplitude signals, the tetrode showed an unpleasant
characteristic caused by secondary electrons released at the anode being
attracted to the new screen grid whenever the anode potential went below
the screen–grid voltage. Tellegen quickly solved this problem by inserting
yet another grid between screen grid and anode that was connected to the
cathode and that reflected the low-energy secondaries back to the anode. In
his patent he acknowledges Ando’s work. The pentode reigned supreme
as amplifier for three decades, although marketing arrangements kept it
out of the United States for three years [11]. The multi-element tubes made
possible the necessary broad-band amplifiers, and completely functioning
television broadcast systems of current-use definition were tested at the
beginning of the 1930s [12].
Electronics engineers found a multitude of uses for pentodes in al-
most every kind of electronic circuit. Radar would have been impossible
without the pentode.
2.1.3. Electron velocity effects

The speed of an electron accelerated by a potential difference of a few
volts is pretty fast. For voltages typical of electron tube usage it can be
calculated as a fraction of the speed of light to be 0.001 98 times the square
root of the potential difference in volts. If the signals being generated or
amplified have wavelengths below one meter, these speeds, slow relative to
the speed at which electric potentials are propagated, cause complications
and limits in electron tubes that rely on control grids. Sluggish electrons
are not the only problem. Reactance of, radiation from and resonances of
the conductors making up the circuit cause problems of their own, and
designers of the 1930s found themselves blocked at almost every turn in
their quest for the highest frequencies.
The first tube capable of working centimetric wavelengths was the
Barkhausen–Kurz tube [13]: a triode with a grid placed at a positive and an
anode at a slightly negative potential relative to the cathode. The electrons
are attracted toward the grid but generally miss it, are reflected by the
anode and head back toward the grid, resulting in oscillations about it.
Changes in direction by the electrons require radiation of energy, which will
occur primarily at wavelengths favorable to a resonant structure formed
36
Origins
by conductors connected to the grid. The radiating electrons spiral down

onto the positive grid. Such tubes generate microwaves but at very low
power because the electrons spend their excess energy in heating the grid,
which can radiate only small amounts of heat, owing to its tiny surface
area. The Barkhausen–Kurz tube was at once both a step forward and an
end.
Another approach to the velocity effect began in 1916 at the General
Electric Company as a way of circumventing de Forest’s patent on the
triode. This invention by Albert W Hull, called a ‘magnetron’ at General
Electric, controlled the electrons leaving the filamentary cathode with an
axial magnetic field. Its function was to replace de Forest’s grid and had
nothing to do with high frequencies; it was capable of high power, and
the complicated trajectories followed by the electrons on their way to the
anode attracted theorists. Alterations in the anode, specifically dividing it
into more than one piece and using a static magnetic field, soon brought the
production of very-high- frequency oscillations in experiments in Europe
and Japan, the latter the result of work by Hidetsugu Yagi, a student of
Heinrich Barkhausen, and his own student, Kinjiro Okabe [14], a discovery
shared with August Zacek of Prague [15]. Magnetrons were capable of
generating very short waves at power levels typical of small transmitter
tubes and much greater than the Barkhausen tube because the anode had
sufficient area to radiate the heat produced. Large tubes required large
magnets, putting a practical end to growth. A formidable literature grew
concerning magnetrons.
A direct approach to the problem of electron velocity effects was the
reduction in dimensions between electrodes, which was undertaken in
1934 by RCA in three miniature tubes called ‘Acorns’ intended for experi-
mental use at extremely high frequencies. A triode, type 955, was followed
by pentodes, types 954 and 956. To reduce inductance and capacitance
the electrode connections were placed at the top, bottom and sides of the
tiny glass envelopes. These tubes were quickly prized by radar men and
copied by German and British manufacturers [16]. They were also difficult
to manufacture, requiring highly skilled personnel [17].
2.1.4. Transmitter tubes

Initially radar engineers worked with commercially available transmitter
tubes. American designers, both Army and Navy, saw great merit in Eimac
tubes, a product intended exclusively for radio amateurs, who wanted
tubes that could take punishment [18]. It was often said they wanted to
be able to read from the glowing anode. The reputation of Eimac tubes
spread during the war, remembered even in an Australian poem [19]; they
were produced for decades after the war. Early British radar work relied
on silica tubes manufactured by His Majesty’s Signal School. Fused silica
has remarkable physical properties, specifically its extremely low thermal
37
High-frequency operation of transmitter triodes. Inter-electrode reactance and

electron velocity limit the maximum frequency of conventional vacuum tubes. This
figure plots the maximum continuous output power of selected transmitter triodes
against the frequency in MHz with wavelength shown on the lower scale. The
WE316A was typical of a number of triodes used for decimeter-wave equipment,
such as the British NT99 or the German LS180. When operated in a pulsed mode,
the peak power was several kilowatts. From Karl Spangenburg, Vacuum Tubes.
expansivity, high melting point and strength that allowed much smaller
high-power transmitter tubes to be made. The Navy incorporated them
into their equipment but, there being no demand for them outside the
Navy, had to make them at the Signal School [20]. By 1938 special tubes
for radar were being made in the United States, Britain and Germany.
The very-high-voltage operation in radar transmitters caused seri-
ous positive ion bombardment of the cathode, which rapidly destroyed
oxide cathodes. Tungsten cathodes were sufficiently robust for this service
but required heating to 2500 K, consuming a lot more power and adding
38
Origins
significantly to the heat load. Thoriated tungsten, the discovery of Irving

Langmuir in 1922, came to the rescue. It was equally robust and functioned
at 2000 K [21]. Driven to the limits of what one could do with triodes these
tubes had very short lives, usually less than 100 hours.
2.1.5. Polyethylene
A troublesome problem for electronic engineers throughout the 1930s was
the lack of an insulating material for high-frequency high voltages that was
flexible and could be shaped. Ceramics insulated well enough but were
certainly not flexible, and machining ceramic was sufficiently difficult to
remove it from consideration for production. There were, of course, many
insulators that were flexible or machinable or both—rubber, resin, amber,
paraffin, wax but especially the then new plastics.
The industrial manufacture of plastics can be said to have begun in
1869 with the invention by the American John Hyatt of Celluloid, made
from nitrocellulose but unfortunately not losing its inflammatory relation-
ship to gun cotton in the processing. By the turn of the century advances
began to come ever more rapidly with polyvinyl chloride in 1912, Bakelite
in 1918, acrylates in 1927 and nylon in 1928 [22]. But all had relatively
high dielectric loss, i.e. at high frequencies they absorbed energy, which
attenuated any signal passing them and, given enough power, seriously
overheated.
Expedients had to make use of glass and ceramics. Transmission lines
for early high-power sets had balanced stiff copper conductors, which al-
ways radiated a bit, were a hazard and were clumsy. Coaxial cables, now
ubiquitous in the electronics laboratory, had to have the central conductor
held in place by beads and were generally unsatisfactory both mechani-
cally and electrically. A German co-ax used cup-shaped ceramics linked
together in a daisy chain that was remarkably flexible and did not allow
the center conductor to become misaligned, something that causes unfa-
vorable transmission characteristics [23].
Important help came by accident, at least from radar’s point of view.
In March 1933 E W Fawcett and R O Gibson at Imperial Chemical Industries
in Northwich, Cheshire found a thin layer of white waxy solid in a vessel in
which ethylene had been subjected to 1400 atmospheres at 170 ◦ C, one of a
series of high-pressure experiments being carried out [24]. They were not
particularly pleased, as they had hoped to produce a lubricant [25]. The
repeated experiment took a violent end, and nothing more was done until
May 1935 when better high-pressure apparatus yielded 8 g of white solid.
An improved compressor, not constructed for the project but available to
it, allowed larger amounts of the substance to be produced by December
1935, but the question of its use had not been considered. They named it
polythene, called polyethylene in America.
Someone noticed its similarity to gutta percha, long used for insulat-
39
ing submarine telegraph cables but which had troublesome losses even for
the audio frequencies of telephony, so polyethylene’s electrical properties
were soon determined and found very satisfactory. It proved to have excel-
lent high-voltage and high-frequency properties. Coaxial cables holding
in excess of 50 000 volts became available to the radar men. On the day
Germany invaded Poland a plant began producing hundreds of tons per
year, and use of it in field installations is reported within months [26]. In
1941 DuPont and Union Carbide equipped themselves to produce huge
quantities of the wonder insulator [27], and pre-polyethylene electronics
went the way of spark wireless.
The German electronic use of polyethylene came about through radar
equipment taken from a British bomber downed near Rotterdam in Febru-
ary 1943, an incident much more famous for the discovery of a cavity
magnetron, the microwave generator that so greatly altered radar. It was
quickly ascertained that the material used in the British high-frequency ca-
bles was a plastic manufactured since 1938 by I G Farben under the name
Lupolen H [28]; they had missed its remarkable electric properties.
After the war the United States sent investigators to evaluate the Ger-
man plastics industry and found a plant in construction hidden away at
Gendorf, Austria to escape the air attacks that made production in Lud-
wigshafen impossible [29]. The Germans were manufacturing the sub-
stance by a more advanced process than the ICI–DuPont method. Further
investigation showed that this was the same process developed by the
Liquid Nitrogen Division of what is today Union Carbide and transmitted
to their affiliate, I G Farben [30]. Given the secretiveness of the chemical
industry, one must assume Union Carbide accepted the task of making
polyethylene from ICI without comment.
2.2. BEGINNINGS, 1902–1934

For convenience the mind likes to associate inventions with inventors,
even though most readers realize that things are often too complicated
for such a simple correspondence. Most new devices have long pedigrees
with persons other than the titular inventor furnishing important elements.
Radar may provide the extreme example of multiple origins. Certainly
one cannot attribute radar to any single person. Indeed one cannot even
honestly say who first thought of building such a device, for it was an idea
that was incipient in Hertz’s experiments, which demonstrated reflections
of short waves quite prominently, as well as the disturbing effects of nearby
objects. A description of relevant technical work in the years preceding
the first prototype radar sets should remove the idea of a single inventor
without much further evidence.
Most remarkable of the early suggestions was in fact an experimental
demonstration. Christian Hülsmeyer patented and built a device that he
demonstrated to numerous witnesses. It used all of the elements of Hertz’s
40
Origins
experiments: a spark-dipole transmitter with a cylindrical parabolic re-

flecting antenna mounted next to a similarly constituted receiver. The
assembly was to be mounted aboard ships and give indication of objects
with which the vessel might be in danger of colliding. Hülsmeyer tried to
convince ship owners to buy it, but the primitive state of the technology of
the day, especially the receivers of 1902–1904 and above all the absence of
the range data, which required accurate timing, easily explains why such
a thing would not have appealed to mariners [1].
The Hertz parabolic reflector used by Hülsmeyer was not a bad radar
transmitter, although a modern engineer would find design elements to
criticize. Each spark created a short wave train for which the peak power
was very high, and the time between sparks was of the same order as
the time between typical radar pulses. The receiver, of course, was woe-
fully inadequate, having orders of magnitude less sensitivity than modern
equipment and no time resolution.
Those familiar with early wireless history will find it no surprise that
Nikola Tesla made an early suggestion for using reflected radio waves
to detect distant objects, but given the complete lack of guiding detail
it is hardly surprising that the idea did not inspire the engineers of the
time. That Tesla was already noticeably on the path from brilliance to self-
delusion did little to advance interest, and the idea was lost among the
myriad proposals coming from his fertile imagination.
The World War of 1914–1918 brought wireless techniques to many
projects, but the closest things to radar were attempts to locate aircraft
from the radiation generated by the ignition systems of their engines.
Edwin H Armstrong, serving as a major in the American Signal Corps,
worked unsuccessfully on the project. The interest generated in him for
sensitive receivers led him to the fundamentally important superhetero-
dyne principle [2].
The World War brought various experiments for locating aircraft by
observing the infrared radiation produced by the motors, and this ap-
proach was continued during the postwar years but with insufficient im-
provement for use in aircraft detection. In a world in which the doors of
public buildings are opened, television sets are switched, the deaf are al-
lowed to hear the actors at the theater, intruders are detected and missiles
are guided by infrared, one may well question the lack of application in
the 1939–45 conflict. Infrared detection depended then on its heating of a
substance, thereby changing its electrical properties. Postwar semiconduc-
tors extended the sensitivities of these effects and added the photoelectric
effect, which provided startling improvements in sensitivity. Infrared is
unsuitable in any case for many of radar’s functions because of its inabil-
ity to penetrate fog and rain.
Probably the most often cited suggestion of radar is from a review
paper on radio telegraphy delivered by Guglielmo Marconi at the meeting
of the Institute of Radio Engineers in New York in 1922 [3]. The idea pro-
41
posed was Hülsmeyer’s without spark and with vacuum tube. The earlier
work was not mentioned, most probably being unknown to Marconi. For
completeness one must note that King George V made a similar suggestion
to the Admiralty Director of Scientific Research in 1925, probably without
knowledge of any previous proposals. He asked whether a radio method
for locating aircraft analogous to the acoustical methods used against sub-
marines could not be devised and was told ‘no’ [4]. The Royal suggestion
evidently caused afterthoughts because the Navy’s Signal School entered
a comprehensive patent a few years later for radio location [5]. While
these experiments of the mind were going on, electrical engineers were
beginning to bump into radar as they mastered the techniques of tens of
megahertz. All through the 1920s and 1930s many papers were written and
patents issued describing methods of determining distance by radio [6].
Engineers examining the propagation of high-frequency waves dur-
ing the late 1920s, especially those planning television, found much to tickle
their imaginations. Their experiences were much the same as those of the
reader who has noted the effect of human movement in a room in which
a television or fm radio is receiving a weak station. Strong reflection and
interference are easily observed, and with care one can note the passage of
an airplane. Such disturbances are reduced in modern equipment by auto-
matic gain control and were far more prominent with the early detectors.
In 1931 Marconi experimented with a 50 cm transmission link between the
Vatican City and Castel Gandolfo and noted a disturbance in the signal
that he traced to the motion of a steam roller, initiating an interest that led
to Italian prewar development [7]. It was the kind of knowledge that must
have spread informally and widely among high-frequency experimenters.
2.2.1. The United States

In September 1922 Hoyt Taylor and Leo Young, engineers from what soon
became the US Naval Research Laboratory (NRL), were studying various
characteristics of 5 m equipment with which they hoped to design sets that
might be less easily overheard than the long-wave sets that dominated
maritime communication at the time [8]. They noticed interference and
standing waves resulting from reflections from buildings, followed by an
observation guaranteed to arouse the interest of naval men: a steamer
passing on the river produced strong variation in the signal recorded by
the receiver from the transmitter on the opposite shore. Memoranda were
written, ideas for harbor defense discussed.
Three years later NRL was to furnish a transmitter for an experiment
with pulsed radiation. Fifteen kilometers north by west of NRL is the
Department of Terrestrial Magnetism, Carnegie Institution. At that time
one of its staff, Gregory Breit, and a Johns Hopkins student and future staff
member, Merle Tuve, devised a successful demonstration of the existence
of the ionosphere by measuring its height. They had an NRL transmitter
42
Origins
modulated with a sine function and measured the phase lag between the
modulation of the wave received directly at their laboratory and of the
wave reflected off the ionosphere. The transmitter was soon changed to
pulsed modulation, which greatly improved accuracy. The aircraft traffic
of a nearby field often disturbed their measurements [9].
In June 1930 Young and Lawrence Hyland carried out experiments
at NRL observing reflections from airplanes with 9.1 m equipment, and
succeeded in attaining Navy authorization for two low-priority radio de-
tection programs. One was to develop superfrequencies, as most thought
that very short wavelengths were highly desirable, possibly necessary, be-
cause the radio searchlight that was envisioned meant the antenna would
have to be large compared with the wavelength. If anything, all the work
around 10 m emphasized how difficult direction determination was going
to be with long wavelengths. The other program was to concentrate on
aircraft detection with designs that relied on circuits known to work. There
were more reflections, big ones from the airship Akron.
Young and Hyland devised with Taylor a system of widely spaced
transmitters and receivers based on these observations that would signal
the passage of aircraft into the area. It was at best a qualitative effect
with little or no localization of the plane possible, something of a ‘radio
screen’. It was also obviously of little value at sea, so Taylor tried to interest
the Army Signal Corps in the idea as a means of protecting cities. The
Director of the Signal Corps Laboratories at Fort Monmouth was Major
William Blair, who was quite interested in a radio or any other technique
for the detection of aircraft and was invited to a demonstration of the
method at NRL. Blair’s comment that there was really nothing new being
demonstrated—which tells us something about the general perception of
the principles involved at the time—followed by his inability to see any
use for it in air defense so long as position was not determined led to
Taylor breaking off the meeting abruptly. Blair was never invited to NRL
again and remained bitter over the incident [10]. In late 1935 the Bureau
of Standards suggested the same idea to the Signal Corps [11].
Blair’s and Taylor’s discordant points of view about there not having
been anything new in the NRL work are not revealed in the source of this
dispute, but can be reconstructed with some assurance. Blair worked at
Fort Monmouth in metropolitan New York where Bell Telephone and RCA
had their laboratories. In the 1920s engineers from Bell had observed strong
reflection and interference from New York’s tall buildings and noted that
the patterns altered greatly with slight changes of frequency [12]. In July
1931 engineers measuring signal strength in the second floor of the RCA
Building produced by all-electronic television transmissions from the top
of the Empire State Building found they could monitor the motion of the
elevator in their building and the automobile traffic in the street below,
stop and go conditions being clearly discernable [13].
No engineer working with meter-wave amplitude-modulated televi-
43
sion could have possibly failed to observe the effects of airplanes, and Bell
had seen them too. All of this was obviously the subject of informal discus-
sion among communications engineers in the New York area, discussions
in which the NRL men seldom, if ever participated. That communication
existed between the Signal Corps Laboratory and the television engineers
is evidenced by a Fort Monmouth engineer having published a paper on
television matters in 1933 [14]. One is inclined to suspect that the RCA
people did not add aircraft to their reports of elevators and automobiles
out of deference to their colleagues in the Signal Corps and the obvious
defense aspects of radiolocation.
In March 1933 men at Bell Labs published a paper describing how
their high-frequency work had been affected by airplanes [15]. In July
1934 the Chief of the US Army Signal Corps recommended a radio echo
technique for the Army, and a modest program got under way at Fort
Monmouth. Blair thought microwaves, wavelengths significantly shorter
than the dimensions of antenna or target, were the reasonable approach
to the problem. At RCA Irving Wolff had built 10 cm equipment with
paraboloid antennas that he demonstrated to a meeting of the Institute of
Radio Engineers in 1934, and Blair invited him to demonstrate it to the
Signal Corps on Sandy Hook, where numerous ships entered New York
harbor regularly. He demonstrated the ability to determine the angular
location of objects from which reflections were received, but the reflected
energy was much too weak for any practical purpose. Their continuous-
wave reflections showed the Doppler effect produced by moving motor
vehicles. They were ready for traffic police duties if not for war1 .
Wolff continued his work with microwaves and by 1937 had con-
structed a remarkable radar set, which he called ‘radio vision’. It used
10 cm waves generated with a split-anode magnetron2 and modulated to
1 µs pulse width. A transmitter and receiver were set up on the roof of
the RCA Laboratory in Camden, New Jersey that received reflected signals
from ships in the Delaware River and from the Philadelphia skyline, but
the short range still left it inherently uninteresting to military designers
[16]. RCA made no further significant contributions to microwave radar
during the war.
2.2.2. France
Pierre David, an engineer at the Laboratoire National de Radioelectricite,
had pressed before 1930 for a program of radio detection of aircraft, but
nothing was undertaken until impelled by the published reports from Bell
Labs of the reflection of 4 m waves from aircraft. In 1934 he observed
airplanes at Le Bourget Field when they passed between widely spaced re-
ceiver and transmitter. This system eventually became the ‘barrages’ that
1 See ‘Reflected signals’ in the Appendix (pp 468–9).
2 See ‘Electron velocity effects’ in Chapter 2.1 (pp 36–7).
44
Origins
were soon installed to protect the ports of Cherbourg, Brest, Toulon and
Bizerte from ships as well as aircraft. The equipment was simple, inex-
pensive and easily mastered with a little instruction, but the observations
were difficult to interpret [17].
Camille Gutton had experimented with 16 cm waves in 1927 using
Barkhausen tubes that had the resonant dipole as a part of the internal
electrode structure. The tubes, whether used as oscillator or as regen-
erative detector, were placed at the focus of parabolic reflectors. These
experiments encouraged Gutton to try the system on aircraft, done in the
summer of 1934 by Société Française Radio-Électrique (SFR), a company
run by his son Henri Gutton. In addition to the 16 cm set, which had less
than 1 W power, they tried 80 cm equipment that used a split-anode mag-
netron that gave a few watts. The results were failure to detect aircraft but
they caused the son to think of detecting obstacles to ocean navigation.
The owners of the new trans-Atlantic liner Normandie, probably think-
ing about the Titanic, requested that the ship be equipped with the system
for her maiden voyage. As a preliminary Gutton mounted one on the cargo
steamer l’Oregon and in July 1935 on the liner, although too late for the ini-
tial crossing. The set, which had twin-paraboloid transmitter and receiver
antennas, made only the one crossing; it did not like the sea and generated
no enthusiasm among the deck officers [18]. It did generate news reports
that attracted the attention of engineers working on secret projects.
2.2.3. Great Britain

During World War I a young physicist, Robert Watson Watt invented meth-
ods for the location of thunderstorms by triangulating the directions of the
lightning discharges detected at two or more stations. By 1917 he could
locate storms as distant as 2500 km. He continued this work, which had
obvious peacetime application, and in 1925 improved it with a technique
that would have two critical applications during the next war. The original
direction finding technique followed that which had been used success-
fully to keep track of the German fleet. Maritime wireless of the time only
used long wavelengths, which allow accurate directions to be determined
by means of manually adjusted loop antennas, and lightning generates
large amounts of long-wavelength noise, one of the disadvantages of such
wavelengths for communication. In locating thunderstorms the operator
had to contend with the various lightning strikes of a storm coming from
an ill defined area, so obvious inaccuracies resulted, with the adjustments
leading only to an average direction. Watt’s improvement eliminated the
need for the operator adjustment. At a given station he arranged two an-
tenna coils at right angles. The relative strength of the signals from the two
coils depended on the direction of the incoming signal. The trick was to
measure their ratio for an individual strike.
The element that allowed this was the new cathode-ray tube, WE-
45
224, recently invented at the Bell Telephone Labs by J B Johnson [19]. This
rather modern looking tube could be wired into a laboratory setup without
the horrible complications attendant to other cathode-ray tubes. Signals
from the receivers connected to the two antenna coils were applied to the
horizontal and vertical deflection plates of the oscilloscope. Each lightning
strike produced a streak across the screen that gave the direction to the
strike—with a 180◦ ambiguity, of course.
The WE-224 was not suited to radar because of its slow response, but
a later design was to become the heart of the radar system Watt would
later design that would change the course of history—Chain Home. It was
also to become the heart of the radio direction finders that functioned well
on an extremely short transmission. During World War II Adm Dönitz’s
signals specialists, who were evidently not well read in the literature of
lightning observation, assured him that direction finding was too slow for
locating transmitting submarines, which had reduced the length of their
messages to extreme limits.
Watt was a Scot, a tribal attribute that gave him satisfaction almost as
great as his conviction that he had invented radar, two pieces of informa-
tion that seldom escaped the attention of others. He was descended from
the inventor of the external-condenser steam engine yet wanted that illus-
trious name improved by adopting a double-barreled version on receiving
honors.
2.2.4. Germany
Research in radio engineering attracted German engineers and scientists
from the beginning. Competition between Germany, Great Britain and
the United States was keen and generally even, as the nearly equal devel-
opment of television during the early 1930s demonstrates. Hülsmeyer is
usually given credit, if late in recognition, for having first built anything
approximating a radar set. He was followed by Hans Dominik, who built
a similar set during 1915–1916 that apparently did not work as well as
Hülsmeyer’s and excited no interest among military authorities. Heinrich
Löwy, an Austrian, had a much more important effect on the thinking of the
men who made German radar. He was well known among physicists and
devoted much effort in using electromagnetic waves to probe the earth’s
interior. While studying their transmission in rock (attaining distances of
a few kilometers) he proposed in 1912 to time radio wave reflections from
conducting strata of the earth using an antenna with a motor-driven mer-
cury switch that alternately connected it to transmitter or receiver. The
switch’s function was that of a Fizeau toothed wheel in velocity of light
measurements [20]. He continued these experiments over the years but
with little success. In 1923 he made the first of the many proposals for a
radio altimeter, and the citations of his patent lead one to believe that it
was frequently discussed. Löwy seems to have been the first to try timing
46
Origins
the propagation of pulsed high-frequency waves [21].

The German Navy established the Nachrichtenmittel-versuchs-
anstalt (NVA, Communications Research Laboratory) at Kiel in 1923 to
investigate what one now calls the electronic methods of warfare. In 1928
Dr Rudolph Kühnhold, having recently received his degree in physics from
Göttigen, became the group leader for acoustics that had the goal of deter-
mining all possible uses for underwater sound: communication, listening
for vessels on and below the surface and determining direction and range
to surface and underwater targets. Experiments soon showed that under-
water sound had little chance of locating surface ships accurately, except
at very close range, and certainly did not form the basis for a method of di-
recting gunfire when optical means failed. This failure impelled Kühnhold
beyond the bounds of his assignment to consider doing the job with radio
waves above the water. A change of research direction proved no admin-
istrative problem, because he had been named scientific leader of NVA in
1933.
He envisioned doing this by forming a tight beam of a few centimeters
wavelength with a parabolic mirror and bought an oscillator and receiver
for 13.5 cm from the Berlin firm of Julius Pintsch. Attempts in fall 1933
failed to observe reflections from a large building less than 2 km distant, not
surprising when one considers he was using continuous waves for which
the reflected signals were not Doppler shifted and his transmitter power
was only 0.1 W [22]. He then began to cast about among electrical engineers
for a means of realization, which fatefully led him to two dynamic young
engineers.
Paul-Günther Erbslöh and Hans-Karl Freiherr von Willisen were two
childhood friends, at the time less than 30 years of age, who had followed
their hobby of amateur radio into the ownership of their own company
manufacturing phonograph records and recording equipment. They were
the best in the new techniques of transforming electrical signals into me-
chanical vibrations and found interest in any new electronic application.
Kühnhold had already used their little company, Tonographie, for his naval
underwater sound work. They were immediately intrigued by his ideas
and had already began to experiment with high-frequency radio for tele-
phonic communication [23].
2.2.5. The Soviet Union

The Red Army of the between-war decades cannot be faulted for ignoring
anti-aircraft defense and its attendant problem of air warning. The Main
Artillery Administration (GAU) devoted much effort to trying to eliminate
the deficiencies inherent in devices based on sound, infrared and search-
lights, coming to the same conclusions of their brothers-in-arms in other
lands. The failure of these methods left them with some sort of radiolo-
cation method as the only hope. Working carefully within the prescribed
47
administrative framework they saw their proposals rejected as unrealistic

by two organizations: the Scientific Research Institute of Communication
Engineers of the Red Army (NIIIS-KA) and the All-union Electrical Insti-
tute (V3I), which was the greatest repository of radio science in the country.
Demonstrating foresight and tenacity, the GAU refused to accept
these rejections and went to the Central Radio Laboratory (QRL), a re-
search institute for the electronics industry where they found a positive
response in late 1933. Yu K Korovin had been successful during the pre-
ceding months in designing decimeter-wave communication links and was
eager to try the ideas for radio location that came to all engineers working
this wavelength range. On 3 January 1934 using 50 cm waves of 0.2 W
(presumably from a Barkhausen tube) and a 2 m paraboloid, he succeeded
in receiving Doppler signals from an aircraft at a range of 700 m [24].
At the same time interest in radiolocation grew independently of the
GAU in the mind of a remarkable member of the Air Defense Forces (PVO),
the organization of the Red Army that employed air-defense weapons.
Pavel K Oshchepkov was a waif of revolution and civil war who received
his first schooling in 1920 and graduated from Moscow University 11 years
later as an electrical engineer. Soon after graduation he joined a research
section of the PVO where he worked on sights for anti-aircraft guns. In
August 1933 he was sent to confirm reports from the Soviet Academy of Sci-
ences about some kind of interest in a ‘radio-technical’ approach to aircraft
detection. His discussions with academician Abram F Joffe so impressed
this eminence that Oshchepkov found himself in charge of a special project.
Joffe rejected methods using very short waves, probably because of a view
held by many theorists world-wide that such waves would be reflected
but generally not back toward the interrogating device. It was a view that
was to last in some quarters beyond the 1930s and has enough truth in it
to form the basis of modern radar-evading aircraft design.
The outcome was a warning system of the kind Taylor had proposed
and the French would construct. Atransmitter of 200 W on 4.7 m was placed
11 km distant from the receiver. Tests performed in August 1934 detected
the presence of an aircraft as far away as 75 km but without revealing where.
The system, called Rapid, and referred to as an ‘electromagnetic curtain’,
produced elation among its creators and derision among others. Useful
or not, it was a start on radar research [25], and Oshchepkov was already
working on the next design step. He saw the importance of pulsed signals
both for determining range and for greatly enhancing output power, tech-
niques he saw leading to panoramic scanning and maplike display [26].
Thus by the beginning of 1934 the Soviet Union was well along the
path to producing first-rate radar. One competent group was moving in
directions that should have led to a good searchlight and gun-laying set;
another group should have quickly produced an air-warning set. Without
exaggeration one can say they were unsurpassed at the time. But it was
not to be.
48
Origins
Radar was in the air; indeed it had been almost since the Hertz ex-
periments. There is even an excellent description of radar in a 1911 piece
of science fiction [27], and numerous patents were issued, although none
reached prototype stage. Contributing to the atmosphere were reports of
‘mystery rays’ and ‘death rays’ that appeared in the popular press with
predictable regularity. A report in the New York Times of some infrared
experiments caused the Signal Corps minor difficulties in 1935 [28], and a
‘death ray’ question was to have an important effect on British radar.
The parallel approaches taken in America, France and the Soviet
Union emphasize how ideas spring up together when conditions are ripe.
The multi-station ‘radio screens’ of Taylor, the ‘barrages’ of David and the
Soviet ‘Rapid’ were clearly independent, all having been secret with no
hint of espionage. Japan put a similar system into operation, as we shall
see, at the start of the war. All were the same as Taylor’s suggestion, which
had led to animosity between him and Blair because of the latter having
rejected the idea out of hand. There is, however, nothing in the French,
Soviet or Japanese experience that contradicts Blair’s snap judgement.
The low-power microwave equipment examined during the early
1930s in the United States, France and Germany was not secret, the
techniques being available in the open engineering literature. A coop-
erative venture between International Telephone and Telegraph and Le
Matériel Téléphonique introduced a microwave communication link be-
tween Dover and Calais on 31 March 1931 using 18 cm radiation from
Barkhausen–Kurz oscillators of 0.5 W with parabolic reflectors. The mod-
ulation circuitry to allow the simultaneous transmission of many telephone
channels had not been developed, leaving the link inferior to submarine
cable, but the extraordinary ability of a directed beam to function at low
power with very low noise or interference left a strong impression on com-
munications engineers [29]. The microwave attempts at radar were all
failures, forcing everyone to longer wavelengths, and microwave research
was dropped by those actively pursuing radar in order to put scarce re-
sources behind something having a good prognosis. Microwave radar was
to come later when an adequate generator was invented. Similarly, most
of these groups had tried infrared as a means of detecting aircraft or ships
and continued research throughout the 1930s. All failed. Detection was to
be with radar, and for the immediate future it was to be half-meter waves
and longer.
2.3. BRITAIN BUILDS AN AIR DEFENSE SYSTEM

With Hitler’s coming to power the doubts that had filled the minds of
thinking Englishmen about the peace of 1919 turned into serious concerns.
The treaty had not long been signed before the tensions that it provided
became all too obvious. Any desires by statesmen to emulate the end of the
49
Napoleonic Wars had been quickly swept aside by the demand for revenge
on the defeated by the populations that had suffered much and that had
been fed steady diets of propaganda asserting that the enemy bore com-
plete responsibility for all wrongs. The model was to be Brest-Litovsk, 1918,
not Paris, 1815. All this had engendered a sympathetic counter-reaction for
Germany’s immense postwar problems among many in Britain, creating
difficulties with France and generating the most pervasive pacifist move-
ment that Europe had ever seen. The general reaction of the people to the
Nazis was initially one of distaste and indifference rather than fear; the de-
pression had given them enough problems close to home to worry about,
but there were a few who perceived the danger, regardless of the chain
of events that had brought the dictator to power, and realized the pitiful
state of Britain’s arms [1]. The difficult economic situation prevented the
Government from initiating a rearmament program had they wished to do
so—and they did not.
Britain’s vulnerability to air attack was present in the minds of many,
and out of these concerns in the Air Ministry came the Committee for
the Scientific Survey of Air Defence, almost immediately called the Tizard
Committee after its chairman, Henry Tizard. Seldom has a committee of
any kind done so well.
The first order of business was to clear out files of plans and ideas
accumulated over the years in order to be able to answer without wasting
time the questions that would be posed when the public became concerned
[2]. One of these was the ‘death ray’. The turn of the century marked
the beginning of a series of discoveries that bore on the public’s mind in
a hopelessly confused jumble: radio waves, x-rays, radioactivity, and in
1932 Hitler had had to share the headlines with the news that scientists
had split the atom. What all this meant to laymen varied, but to some it
meant that scientists could manufacture a ray, of what kind they knew not,
that would kill the King’s enemies, dared they invade the sanctity of his
realm. A casual reading of the newspapers told Harry Wimperis, Director
of Scientific Research at the Air Ministry, that he must be able to deal with
death rays, so he requested Robert Watson Watt, the superintendent of the
Radio Research Station at Slough to examine the matter of a radio death ray.
Watson Watt and Arnold F Wilkins calculated how much they could
raise the temperature of an attacking pilot’s body, if they concentrated the
most powerful wireless signals available onto him. The answer was, as
they and Wimperis knew in advance, trifling. Watt did not leave the mat-
ter there, however; he changed the question. If one were to irradiate an
airplane with a high-frequency wave, currents would be induced into its
metal structure and these oscillating currents would then re-radiate, and
the question was how much. Was it enough to offer a means of locating
an airplane? In sum, substitute radio detection for radio destruction. To
make the problem simpler an ideal approximation was assumed, which is
the physicist’s first step in approaching any problem. Watt and Wilkins
50
Origins
assumed the irradiation had a wavelength twice the wing span of the tar-
get aircraft, which was replaced in the calculation by a wire and which
was found to re-radiate a most satisfactory amount. On 28 January 1935
Watson Watt sent to Wimperis his ‘anti-death-ray’ memorandum in which
he outlined a radio location scheme capable of giving Britain warning of
air attack.
Why Wimperis approached Watt rather than Edward V Appleton,
Wheatstone Professor at University of London, chairman of the Radio Re-
search Board’s Propagation of Waves Committee and world authority on
radio matters, is not known. It may have turned on the matter of secrecy,
as Watt was a government employee, Appleton not. Whatever the reason,
it generated a fierce animosity between the two. Matters were certainly not
helped by Appleton not being told of the radar work until summer 1936.
Watt’s theoretical approach to reflections from aircraft, a phenomenon long
observed by Appleton and his students and certainly by Watt, who had
worked extensively with Appleton on the ionosphere, adds an additional
puzzling element [3].
The next step led to action by Air Vice-Marshal Hugh Dowding, Wim-
peris’s chief. Since 1930 Dowding had sat on the Air Council as Air Member
for Supply and Research, a position that had recently been split leaving him
responsible only for Research and Development. Dowding had seen the
importance of aircraft in a 1912 Staff College exercise and had learned to
fly the following year at his own expense. By 1918 he had had just about
every kind of flying experience that it was possible to survive and had risen
from subaltern to brigadier general on the basis of technical competence
and ability. As a training officer he had resisted to the bounds of direct dis-
obedience sending inadequately trained pilots to the front, which made his
relationship with Marshal of the RAF Hugh Trenchard, who ruled the Air
Force during and for the decade after the war, disputatious but eventually
respectful [4]. Trenchard was an all-out believer in the bomber dominating
aerial warfare, and Dowding believed it could be countered with properly
handled fighters. His first important work in Supply and Research was
to put all his influence behind procuring the engines and airframes that
became the ‘Hurricane’ and the ‘Spitfire’. His second important work was
radar [5].
Trenchard and most Air Force officers were convinced that there was
no effective defense against bombers. This had been amply demonstrated
in exercises. It required a minimum of 20 minutes after an alarm for a
fighter squadron to reach the expected location of the attackers, provided
one knew where to send them. So much time was not available, if the warn-
ing came from visual ground observers, and for it to come from standing
off-shore patrols would require by any calculation an enormous number of
planes, so when Dowding learned of the new idea he listened with more
than casual interest, which was fortunate because it was from Dowding
that the first funds had to be obtained.
51
Dowding was interested but wanted something more than ‘specu-

lative arithmetic’. Watt agreed to give a demonstration, on the condition
that a negative result would not be held against him, because the appara-
tus would be hurriedly assembled. (Because of Watt’s prior knowledge of
aircraft reflections, Appleton considered this ‘pure theater’ [6].) At Daven-
try, about 20 km southeast of Coventry, was the large Empire transmitting
station. Watt and Wilkins set up on 26 February 1935 a receiver a few miles
from the Daventry station with a double antenna so balanced that the di-
rect signal from the 50 m transmitter was greatly reduced, the object being
to make it sensitive to the same wavelength coming from a different di-
rection. The approach of a bomber produced an unequivocal signal, with
A P Rowe, secretary of the Tizard Committee, serving as impartial witness.
What Watt and Wilkins had produced was identical to the earlier
experiments in America, France and Russia, but what came next was com-
pletely different. Watt quickly saw how to make the system give the posi-
tion of the target aircraft, something that had escaped the others. His solu-
tion was to incorporate two electronic techniques long used at his station:
the pulsing of radio waves for determining the height of the ionosphere
with the echoes and the instantaneous determination of the direction of
lightning strikes using crossed antennas. Earlier investigators had only
thought in terms of continuous wave operation for their ‘radio screens’.
Another very important difference distinguished the British approach
to radar from that of other countries. Where foreign radar men had had to
contend with moderate interest and weak support, British radar men re-
ceived strong support from the highest level. Dowding immediately sanc-
tioned £12 300. The possible location of the enemy 75 to 100 km off the coast
would give just the amount of extra time he needed to scramble fighters,
and his fervor for radar remained steadfast. Fighter Command, a military
organization destined to join such immortals as the ‘Old Contemptibles’
and the ‘Army of Northern Virginia’, emerged out of a reorganization of
the Air Force the next year and Dowding became its first commander.
He and Tizard realized that early warning was of no value unless it
yielded a suitable response. What was needed was a system for analyzing
the information gained by the radar and ground observer stations and for
its communication to the fighters—a process that must be accomplished
in a very few minutes. While Watt assembled his design crew Dowding
began to build his control system and train his pilots to use control from
the ground. The latter task was not easy, for these young men had seen
‘Hell’s Angels’ and ‘The Dawn Patrol’ several times and had been reared
on the romances of Scott and Tennyson. Like their German cousins they
thought in terms of knights of the sky, not policemen responding to a call,
but Dowding succeeded, and the Germans, who were to enter the war with
technically better equipment, had no organization to make use of radar for
air defense and little comprehension of its tactical significance.
Although the basic design was obvious almost immediately after the
52
Origins
Daventry experiment, the creation of equipment required much detailed

work. To do this in extreme secrecy demanded a secret place, and one
presented itself. Orfordness [7], a long spit of land 15 km up the coast
from Felixstowe, had been used for secret aviation research during the
war and had been reactivated in 1929. It had the rudiments of shelter and
electric power, and Watt soon began his experiments. By the end of July
1935 they had tracked aircraft to 60 km. In September the Air Defence
Research Committee had approved the construction of Chain Home, and
in December (the year is still 1935!) the Treasury allocated £60 000 for five
stations for the Thames Estuary [8]. Rowe gave the device the code name
RDF, intended to imply ‘radio direction finding’, which was a good name
to hide behind because of Britain’s success during the First World War with
radio direction finding and Watt’s previous work.
Equipment terminology can be confusing as new designs and mod-
ifications of old ones are encountered. British radar was initiated by the
Air Ministry, which found it necessary around 1940 to designate equip-
ment as AMES (Air Ministry Experimental Station) Types. The original
design was AMES type 1, generally referred to as CH. Other AMES sets
received similar initial names, which will be used here so long as clarity is
preserved. The Army and Navy soon began to develop radar and adopted
designations of their own. As sets other than AMES type 1 began to take
their place along Britain’s coasts the term ‘Home Chain’ came to mean that
entire radar network.
The first months at Orfordness were spent in experiments with
lashed-up equipment to verify the basic ideas of the design of Watt and
Wilkins and modify it before production orders were placed. The de-
sign that eventually evolved used two arrays of horizontal dipoles slung
between two 100 m steel towers, constituting a minimum transmitter sta-
tion3 . These radiated over a wide angular front more like a floodlight than
like the searchlight that other radar designers strove for. Reflectors were
placed behind the arrays in order to prevent wasting power and generat-
ing confusing signals from the rear. The pair of arrays allowed the vertical
lobes of the radiation pattern to be altered. Initial plans had foreseen jam-
ming, hence stations had multiple towers that allowed dipoles of different
lengths to be slung, as a dipole antenna allows only slight variation of the
frequency determined by its size.
They immediately discovered what all other radar designers learned
when working with pulse transmission: that the maximum ratings for
vacuum tubes applied to the average power, determined by the time it took
to heat the anode to some critical temperature. In pulsed operation anode
voltages could be sent way beyond what was normal, from kilovolts to
tens of kilovolts. The Royal Navy manufactured a silica triode, type NT46,
that could be driven very hard, which was used in early stations, but its
3 See ‘Antennas’ and ‘Vertical lobes’ in the Appendix (pp 469–72).
53
specialized production could not meet all the demands for it, so other
output tubes had to be found. Special tetrodes, Metropolitan–Vickers type
43, were ordered that had water cooling and that were vacuum pumped
rather than sealed so that they could have hard-driven filaments replaced
or otherwise repaired [9].
Receivers used pairs of crossed dipole antennas mounted on 75 m
wooden towers with one for the E–W signal component, the other for the
N–S signal. The operator could adjust the relative amplitudes of the two
signals to determine the direction. It was a copy of the equipment used for
the determination of the directions to lightning strikes with the difference
that dipoles had been substituted for the loop antennas because of higher
frequencies. Antennas were located at three levels on the receiving towers.
Estimates of the height of the target resulted from the comparison of signal
strengths from dipoles located at different levels.
Originally wavelengths of 50 m had been planned to allow resonance
with the wing span of a typical bomber, but airplanes have complicated
structures and this special wavelength was not found to be important. In-
terferences with communication signals soon forced them to shorter wave-
lengths with a band from 7.5 to 15 m becoming typical. These wavelengths
produced quite a lot of unwanted reflections, sometimes from the iono-
sphere, sometimes from distant continents, ships and cities, the result of
ionospheric bounces, because CH was from the beginning an ‘over the
horizon’ radar, although incapable of making sense of the jumble of those
distant returns. For that one had to wait for more advanced signal pro-
cessing. (Long-wave sets having focused beams did observe and identify
targets at extreme range during the war, but these incidents came from
exceptional atmospheric circumstances.) All this meant that a much lower
pulse repetition rate had to be used in order to eliminate these extraneous
signals instead of the hundreds of Hz that were reasonable for ranges of
100 km or so. The value selected was 25 Hz, half the British power fre-
quency. This also allowed adjacent stations to be synchronized with each
other using the electric power grid [10].
For a station to be able to determine direction reliably and height at
all, aircraft were required to fly preset paths while calibration teams labo-
riously took huge amounts of data, a mind-numbing and costly procedure
that had to be repeated at regular intervals for each station. The results
of an observation were readings of dials and goniometers, which had to
be transformed into three-dimensional coordinates for the information to
be of any use to fighter directors. Not only did this require trigonometry
but had to incorporate the calibration corrections for the station: simple
but it had to be done quickly. G A Roberts of the Bawdsey staff devised an
electric analog computer for this task, soon called the Fruit Machine, that
allowed the operators to enter the data plus their estimates of the number
by pressing keys at their desks [11].
The year 1938 marked the transformation of CH from an interesting
54
Origins
radio experiment into a functioning air defense system. The previous May
the first CH station had been turned over to Air Force personnel [12] and
other stations followed as they went on the air. To Squadron Leader Ray-
mond Hart fell the responsibility for organizing the training school, and
he quickly became the leading figure in applying the new techniques to
war by methods soon to receive the name of ‘operational research’ [13].
For practice, Fighter Command began on 1 January 1938 to intercept ‘dis-
cretely’ but in the beginning not very successfully KLM and Lufthansa
airliners arriving from the continent [14]. By July five stations were ready
for the August air defense exercises.
The equipment that made up the system that began to grow in 1937
was, with three exceptions [15], unlike any other radar systems. The char-
acteristics of CH were so different from those of the German design that
it baffled completely the first attempt at electronic espionage sent against
it. It was in 1940 an obsolescent system, but it was crucial to winning
the Battle of Britain and earned the undying affection of its operators and
mechanics, who often referred to it in later years as ‘steam wireless’ [16].
While work began at Orfordness the Tizard Committee proceeded to
examine all aspects of scientific air defense and came up against a vexing
problem, one which would remain one way or another until the end of the
war. Churchill’s party was out of power and had not a great deal more
interest for rearmament in 1935 than Labour. In calling alarm Churchill was
very much alone but also very vocal and spoke for most of those engaged
in radar work. Owing to his insistence that something be done about air
defense he was added to the Imperial Defence Committee, which certainly
did not upset the growing radar community, but Churchill insisted that his
personal friend and scientific advisor, Professor Frederick Lindemann be
added to the Tizard Committee. On the surface this might have seemed
useful because he and Tizard had once been fast friends, but they had
become estranged for reasons that remain obscure and the estrangement
was complete. Lindemann, the later Lord Cherwell, had by most measures
an abrasive personality and quickly transformed a highly productive group
into one that wasted its time fighting the newcomer.
His scientific ability had initially been sufficiently great for him to
have been included among the 24 who made up the Solvay Conference of
1911, joining such lights as Planck, Sommerfeld, Jeans, Rutherford, Ein-
stein, Lorentz, Curie and Poincaré, but he lacked the soul of an engineer
and proposed various ideas for air defense, all of which he wanted seri-
ously investigated and all of which the committee rejected. His favorite,
aerial mines, was regarded as nutty by almost every technically competent
person who evaluated it. He was not enthusiastic, many say hostile to-
wards radar. Matters reached such an impasse that he had to be removed
from the committee by an administrative trick with Appleton replacing
him. In 1940 Lindemann became the Prime Minister’s advisor, and for five
years the British technical community had its hands full. The controversy
55
Transmitter towers from Britain’s CH. This radar, remembered for its service in
the Battle of Britain, retained the same basic design but had numerous antenna
configurations, some of which are illustrated in this composite of two towers. In
some cases dipoles for the 10 to 15 m radiation were mounted at the sides of the
steel structures, as shown on the left and right, but the radiation patterns were
often unsatisfactory, leading to them being suspended as a curtain array between
two. In the expectation that there would be jamming, stations usually had more
than two towers so that arrays with different wavelengths would be immediately
available to the operators. The array of eight dipoles formed a relatively narrow
radiation pattern in vertical extent but very wide in its horizontal. Reflection off
the ground and sea produced a vertical lobe structure with blind regions. These
were filled with the lower array, called the gap filler. Adapted from the Radar
Supervisor’s Handbook and taken from Sean Swords’ Technical History of the
Beginnings of Radar. IEE Copyright.
has endured. The voice of the prosecution comes from C P Snow [17], that
of the defense from the Earl of Birkenhead [18].
However all this may be, Cherwell deserves the credit for instilling
in Churchill a high opinion of the value of science in warfare, one of many
qualities that set the Prime Minister apart from his German counterpart.
As Churchill’s advisor he was able to have his aerial mines tried in 1941,
confirming experimentally the judgement of his critics.
Orfordness suited research needs in only the most rudimentary man-
ner and soon began to suffer growing pains, so the first of four changes of
station took place in March 1936, this first time to the Bawdsey Research
56
Origins
Receiver towers from Britain’s CH. Antennas for transmitter and receiver were
separated by a few hundred meters. The receiver antennas were pairs of crossed
dipoles tuned to the frequency of the transmitter with pairs mounted at two differ-
ent heights. The comparison of signal amplitudes from the two dipoles of the pair
allowed the direction of the target to be ascertained. Comparison of the amplitudes
of the signals from pairs at different height together with a third non-directional
antenna allowed estimates of the target height to be made. Wood construction
was required in order to eliminate the interferences caused by a metal structure.
Receiver towers went through minor variations as did the transmitters. Adapted
from the Radar Supervisor’s Handbook and taken from Sean Swords’ Technical
History of the Beginnings of Radar. IEE Copyright.
Station. This was a manor, suitably isolated from the overly curious, with
sufficient space for laboratories and quarters and with extensive grounds
for experimental antennas. ‘It was a grand place for the work that had
to be done and the only feature inappropriate to the new conceptions of
defence which were born there was the motto above the door of the Manor,
"Plutôt mourir que changer"’ [19]. With its timbered hall and roaring fire
it resembled an Oxford or Cambridge college rather than an engineer-
ing laboratory. The university atmosphere increased after a visit by Lord
Rutherford, whose natural and decisive support soon had students and
57
faculty from nuclear physics laboratories joining. The Bawdsey days were
long remembered for the relaxed mode and the long hours. Many an elec-
tronics problem was thrashed out in the manor house, at swimming, on
walks through the grounds or at some other recreation.
Watt proved an exemplary leader of this collection of young physi-
cists. It was probably the peak of his life, and Hanbury Brown remembered
his first year at Bawdsey as ‘one of the happiest I ever spent’ [20]. Watt
did not initially want communication men, as he expected more ingenious
ideas from scientists, but he paid a price in some early failures caused by
the lack of rudimentary radio knowledge, especially of what test equip-
ment was available and needed. In designing airborne radar Brown later
worked with EMI television engineers and was shocked to learn how much
more advanced their circuitry and equipment were [21]. One is led to spec-
ulate on whether the design of CH would have survived the criticism of
such professionals had they been present from the beginning.
It is difficult to escape wondering what the course of British radar
might have been had EMI been approached in 1935 rather than the Ra-
dio Research Station, because in 1934 they had assembled for television
‘32 graduates (nine with PhDs), 32 laboratory assistants (of first year BSc
standard), 33 instrument and toolmakers, glass blowers and mechanics,
eight draughtsmen designers, and nine female assistants’ [22]. This ex-
traordinary source of electronic talent was unaware as late as 1939 of the
radar project. EMI had declined contract work for military communica-
tions equipment, giving their deep commitment to television as the reason
[23]. Whether they would have responded differently to a radio-location
project, magnificent perhaps but ill defined and with uncertain long-term
government support, will never be known. One suspects the start would
have been time- consuming negotiations.
Watson Watt was promoted to Director of Communications Develop-
ment at the Air Ministry in May 1938 and turned Bawdsey over to AP Rowe,
an excellent administrator with intimate knowledge of the Air Ministry but
essentially no technical knowledge and little patience for the antics of his
charges. Bawdsey’s extraordinary environment was soon swept away by
the inevitable growth of ‘the official way’. Impressed by the manorial sur-
roundings, he even entertained the idea of requiring the staff to dress for
dinner [24]. Rowe had one supreme virtue that more than compensated
for his stiffness: he strongly believed in the free exchange of ideas. This
eventually led to his ‘Sunday Soviets’, meetings of scientists, engineers,
technicians, serving officers and administrators where one could say, in-
deed was expected to say anything that came to mind. They proved most
rewarding and did much to forge the strong bond between laboratory men
and fighting men. If Rowe was known as a stickler for the rules, he was
also known for his consideration for the staff.
The possibilities of radar having functions beyond those of CH had
presented themselves from the beginning, which led to the establishment
58
Origins
at Bawdsey of an ‘Army cell’ on 16 October 1936 under the leadership of

Dr E T Paris with half a dozen design engineers, including W A S Butement
and P E Pollard, who had written a memorandum on the construction of a
radar set five years earlier for the Inventions Book maintained by the Royal
Engineers [25]. It was a good design that envisioned pulsed 50 cm waves
and parabolic antennas and lacked only the elusive microwave transmitter
and administrative support.
Their first task at Bawdsey was to build a smaller, portable version of
CH, the Mobile Radio Unit (MRU) that used smaller antennas suspended
from 20 m field masts with consequent shorter range and poorer height-
determining capability. It received belatedly the designation AMES type 9
and came in various arrangements indicated by at least three sets of initials.
Further reference will be to MRU without semantic subtlety. Responsibility
for this equipment was transferred to the Air Force in 1938, its function
being early warning, and the Army cell then occupied itself with shorter
wavelengths. They set about constructing 1.5 m equipment, of a kind
very much like what was being made in America and Germany, with coast
defense fire control the object. They had been stimulated by the success
of E G Bowen, who had been using this wavelength with airborne radar
in mind [26]. They gave up the use previously made of the Yagi antenna
because of the poorly understood criticality of dimension and went instead
to a dipole array, which allowed lobe switching. By May 1939 they could
determine the range of a 2000 ton vessel at 15 km from a set placed 18 m
above sea level and determine direction to an accuracy of 15 minutes of
arc at 10 km with lobe switching4 . Practice shooting disclosed echoes from
the explosion of shells, allowing adjustment of fire [27].
Out of Butement and Pollard’s Coast Defence set grew two projects of
importance for the air war that was coming: Chain Home Low (CHL) and
Ground Controlled Interception (GCI). A characteristic of CH was the large
vertical lobe patterns that resulted from surface reflections and that were
used for height determination. Unfortunately, these patterns were such
that very-low-flying aircraft could escape detection until alarmingly close.
The angle at which maximum detection occurs can be approximated by
dividing the wavelength by four times the height of the antenna above the
Earth’s surface. A small angle, the result of short wavelength and necessary
to detect ships, gave the Coast Defence set a significant advantage over CH
in detecting low-flying aircraft, for which it became CHL. Placing it on a
high tower allowed it to perform both functions as CD/CHL. By fall of
1939 there was great pressure to deploy it, and the first station was on the
air by 1 December 1939 [28].
In parallel with this Pollard designed a rudimentary radar to allow
anti-aircraft batteries to determine range accurately and to help put their
searchlights on attackers. It had a small antenna and a long wavelength,
4 See ‘Lobes’ in the appendix (p 471).
59
from 3.5 to 5.5 m, and was ironically named Gun Laying I (GL mark I); Gun
Assisting would have been better terminology. It gave accurate range but
poor horizontal direction (azimuth), no elevation angle at all, and multi-
ple targets could completely fog even experienced operators [29]. It was
quickly designed, production was cheap and more than 400 were made; of
the slightly improved GL mark II, which provided elevation data at large
angles, over 1600 were made [30]. Elevation angle is difficult to determine
for meter-wave radar because the vertical lobe pattern is unpredictable un-
less the ground is a perfectly flat conducting surface, approximated by wire
netting. Locations by the sea, common in the Pacific but where GL mark II
was not employed, allow good use of meter-wave equipment. The meager
effort that went into the design of GL probably had its origin in the ex-
tremely unsatisfactory state of the directors available to use radar data [31],
making the expenditure of time for developing a set capable of blind fire
not worthwhile. Using it for blind fire was a waste of ammunition: if set up
under ideal conditions, it returned one hit for 4000 rounds expended [32].
Butement took on another electronics problem dealing with antiair-
craft (AA) artillery, the proximity fuze, but it failed because of low priority.
Butement did design the elegant circuit for it that would later be used in
America to make the device a success [33]. The primitive character of GL
marks I and II, and the virtual abandonment of proximity fuze work con-
trasted markedly with attitudes in the US and Germany at that time and
demonstrates the low esteem that AA artillery had in Britain. It was, how-
ever, perfectly consistent with the wooden-headed attitudes of the Army
in technical matters during the first years of the war. The War Office use
of scientists was the perfect inversion of that of the Air Ministry. They did
not wish scientists to initiate projects but to do what they were told; they
did not wish scientists to work closely with serving officers to gain under-
standing of the weaknesses of their inventions; they did not wish to hear
the opinions of scientists on how best to employ the new weapons. In the
case of GL, excellent engineers were told to provide a better range finder.
The AA predictor was incapable of using the continuous stream of data
provided by a radar set and remained so until the arrival of the American
M-9 director in 1944 [34].
The 1.5 m equipment on which the Army cell built their Coast De-
fence set came from the first attempts to construct airborne radar. Looking
forward to the success of early warning radar Tizard had predicted that the
enemy would necessarily turn to night bombing for which there was no
effective defense. If searchlights could be brought to bear on an attacker,
then fighters could intercept it or AA guns fire on it, but it was no small
trick to get a light beam on a plane and hold it. Large listening horns gave
some indication of the direction of the plane but were strongly affected by
weather and confused by large flights. The regions serviced by searchlights
would always be small, greatly reducing the amount of time for fighters or
guns to engage. What was clearly needed was airborne radar. Given the
60
Origins
gargantuan size of Britain’s initial radar equipment this seemed a tall order.
Bowen had taken to the idea of this project and had pressed hard for
it, convincing Watt to begin it after the move was made to Bawdsey. The
problem was poorly defined, and the constraints imposed by installation
in aircraft only vaguely understood, so Bowen discussed the matter with
two communications engineers at Martlesham Heath, the nearby Arma-
ment and Experimental Establishment airfield from which Bawdsey drew
aircraft cooperation. These discussions led to self-imposed limits of 100 kg
weight, 500 W power and dipole antennas not more than a meter in length.
At just the right moment appeared the EMI Company’s 6.7 m television
straight-vision receiver [35], a small but very sensitive unit that amplified
the radio-frequency signal without heterodyne action and that had been
well engineered, so the first experiments were carried out on this wave-
length.
After satisfactory ground tests a receiver and indicator were trans-
ferred to a Heyford bomber, which was large enough for the 6.7 m antenna
and had an electrically shielded ignition system. Aircraft echoes produced
by the transmitter, which was much too heavy to be carried aloft, were
satisfactorily received along with the transmitter pulses, of course, which
furnished a not particularly helpful time reference. Bowen was a strong
advocate of further developing this ground–air system, called RDF 1.5.
(RDF 1 refered at the time to CH and RDF 2 was to be the 1.5 m airborne
set.) Watson Watt could not be convinced, and the approach was dropped,
a decision Bowen always regretted.
With RDF 1.5 eliminated Bowen pushed to shorter wavelengths, set-
tling on 1.25 m. The EMI receivers became the intermediate-frequency
amplifiers that followed frequency conversion with an RCA Acorn tube.
The Handley Page Heyford bomber was replaced with two Avro Ansons,
excellent machines for flying laboratories, and with them came the person-
nel of D Flight, 220 Squadron. In August 1937 the team had clean echoes
from ships and the following month created somewhat of a sensation in
fleet exercises by locating the carrier Courageous through thick clouds.
At the end of 1937 the airborne project had two goals, Air to Surface
Vessel (ASV) and Air Interception (AI), and an important new member,
Robert Hanbury Brown. The details had to be worked out for sideways
and forward lobe patterns in the ASV set and for acceptable maximum
and minimum ranges for AI. The two ASV modes were for maximum
search and head-on attack. The need for a maximum range in AI was obvi-
ous enough, but the minimum range was even more important and much
harder to bring off because it meant very short pulse duration. Radar had
to bring the pilot close enough for him to see his target, and a 1 µs pulse
produced a wave train 300 m long, which was greater than the minimum
distance for visual contact, and sub-microsecond pulses were not so easy
to produce. Both ASV and AI were well advanced when war came but far
from ready [36].
61
The Tizard Committee had discussed air defense problems with

Charles Wright, the Admiralty’s Director of Scientific Research and
G Shearing, Chief Scientist of His Majesty’s Signal School’s Experimental
Department before disclosing to them the approach taken by Watt. Shear-
ing volunteered that radio should be carefully considered, pointing out that
the Post Office had had reception interferences caused by aircraft when us-
ing 5 m waves. Wright was soon invited to Orfordness and was able to
bring Shearing with him to a second demonstration in July 1935. In as
much as the Signal School was the only source of the silica transmitter
tubes that Watt wanted, discussions tended to be harmonious.
The Navy did not want their radar research conducted at Bawdsey,
and the Admiralty initiated work in its own existing laboratory at HM Sig-
nal School in Portsmouth, to the irritation of Watson Watt, who wanted all
radar research concentrated in one place. The Navy’s reasons were that
naval use required that designers at every level know about conditions at
sea, the effect of gunfire, the interference from masts and other structures,
the mutual interaction of various shipboard electronics and the problems
of maintaining equipment at sea. HM Signal School was no newcomer
to high-frequency work, having led the world in equipping vessels with
high-frequency direction finders. In contrast to Bawdsey, they conducted
research in microwaves [37]. The Navy won the dispute and formal Trea-
sury approval at the end of 1935 [38].
The results of the next 18 months were not impressive. R A Yeo con-
ducted experiments on 4.0, 1.5 and 0.6 m and in July 1937 produced some
not particularly impressive range-only experiments with a set designated
type 79X installed on an old minesweeper. Secrecy had much to do with
the poor progress, as Yeo could speak about the work with only two out-
side of his immediate group. He was frequently unable to get time on
the minesweeper, which was used for school functions, because he could
not tell its keeper why he needed her. The contrast with Bawdsey was
extreme. A new commander of the Signal School, Captain A J L Murray,
was appointed on 25 August 1937; he took personal interest in the radar
project, including attending their staff meetings. The change was abrupt
and very helpful.
Alfred W Ross took over the design of type 79 and immediately
changed the wavelength from 4 to 7.5 m. This had three advantages: (1)
the Navy’s silica transmitter tubes had too much inductance in the leads
for satisfactory operation at 4 m, (2) 7.5 m was in the band of British Broad-
casting Company’s television, allowing commercially developed receivers
to be used, a trick probably learned from Bowen, and (3) this wavelength
fitted well with techniques used in the Navy’s communications transmit-
ters. The result was that Ross tested a successful prototype 79 in early 1938
with copies installed on the battleship Rodney and the cruiser Sheffield near
the end of the year.
The type 79, soon improved to type 279, used synchronously rotating
62
Origins
transmitting and receiving arrays on separate masts. The antennas were

small, hence the beam broad; this was not considered serious in an air
warning set but was a problem for observing other ships. This led quickly
to the design of type 286 for 1.5 m for the same reason Butement had
selected this frequency for his Coast Defence set. The 286 had non-rotatable
antennas with the result that one had to swing ship to gain directional
information [39]. The equipment did not compare favorably with the sets
with which the American and German navies were being equipped, but
they provided essential information and had been designed by engineers
who understood what was required to make electronics work at sea.
J F Coales had proposed in February 1937 developing 50 cm equip-
ment for gunnery. After various considerations he decided to use a spe-
cially designed triode made by H C Calpine. By the end of 1938 transmit-
ter, receiver and antenna (a Yagi backed by a cylindrical paraboloid) were
ready for test. Sea trials aboard HMS Sardonyx in June 1939 observed ships
at 8 km and low-flying aircraft at 4.5 km. The 50 cm equipment was to
evolve into a series of gunnery sets [40], all characterized by pairs of Yagi
antennas, which came to be called fishbones, that used lobe switching. By
1940 designs were complete for the type 284 for main armament and type
285 for anti-aircraft; both went into production the following year. The
Navy’s enlightened attitude concerning gun-laying radar contrasted with
the Army’s near indifference.
By 1939 Dowding’s air defense system was reaching a remarkable
stage. He began a series of summer exercises immediately after taking
over Fighter Command and quickly recognized that the first problem was
confusion. Air Controllers were overwhelmed with too much informa-
tion, some of it contradictory or just wrong. The first step was to route
the various radar and ground observer sightings to a filter center at Bent-
ley Priory, Fighter Command Headquarters, to evaluate them for veracity.
Filtering these observations, which required instinctive as well as scien-
tific skill, produced the information sent on to the three Fighter Command
Groups. There the air situation was also presented by markers designat-
ing enemy and friendly units, moved on great map-tables by women and
men members of the Air Force who were connected to the filter center by
telephone. Above, watching the movement of the pieces on the table, sat
the controllers, themselves fliers, who sent orders to the pilots by radio-
telephone. Command functioned with Dowding making decisions day
by day, Group Leaders hour by hour, and squadron controllers minute
by minute. Communication between all radar stations, observers, filter
centers and squadron headquarters was through an independent wire net-
work. Communication with the pilots was through the new very-high-
frequency (VHF) radio-telephone link, the only one in the world. It gave
low-noise transmissions, provided plenty of channels and had range lim-
ited by the horizon, making eavesdropping difficult.
Women did not serve just as croupiers at the great tables but in many
63
functions throughout the Air Force. In 1937 it had been realized that Britain
would face a severe manpower shortage, and the successful employment
of women in air defense in the previous war pointed to an obvious solution,
the Women’s Auxiliary Air Force (WAAF) resulting. Three women from the
Bawdsey clerical staff were given training in operating the equipment of a
CH station, and they quickly demonstrated a keen ability, especially for the
wily nature of the receiver oscilloscope displays [41]. The value of women
as radar operators was soon widely appreciated. A succinct explanation
came from Australia: ‘Women did make the best radar operators, because
they watched the screen’ [42]. For those with insight into character it came
as no surprise that they remained calmly at their posts when stations were
attacked.
The months between the Munich conference and Chamberlain’s Dec-
laration of War on 3 September 1939 saw a remarkable alteration of atti-
tudes in Britain. After the euphoria of ‘peace in our time’ came an un-
derstanding of the price that had been paid for it. The people and their
government decided that the next aggression meant war, and Britain gave
a guarantee to Poland over an issue that even had elements of justification
for Germany. The people accepted war with a determination unmatched
in France or Germany. France would soon collapse and Germany would
not confront the grim reality until 1942.
Britain was the least prepared for war of the three powers, except
in radar. In the summer months of 1939 the physics departments and
technical schools began emptying to fill the ever-growing need for radar
men and women. The number of people working in radar one way or
the other was large by then, probably greater than the combined totals of
all other nations, and the number was growing rapidly. The engineering
quality of British sets was poorer than those of Germany or the United
States, reflecting Watson Watt’s dictum, ‘Give them the third best to go on
with; the second best comes too late and the best never comes’ [43], but in
the quality of the radar operators and mechanics it was another matter, one
that more than offset any deficiencies of equipment. They were expected
to cope with all manner of problems, and their training approached that
of electrical engineers. This core of technicians was the narrow margin of
defense.
2.4. AMERICANS AND GERMANS BUILD PROTOTYPES

2.4.1. US Naval Research Laboratory
Albert Hoyt Taylor was born nearly a decade before the demonstration
of electromagnetic waves that were to shape his life. He was born in an
age that viewed the atomic theory of matter as just an interesting hypoth-
esis. He died in an age that feared the power latent in the atom might
destroy us all. During his school years there arose in him an interest in
all things electrical that led straight to a PhD in applied electricity from
64
Origins
Göttingen. The First World War deflected him from an academic career
into the Navy where he remained after the war as a civilian employee
and directed research in radio at the newly created Naval Research Lab-
oratory. The contrast of Taylor’s German degree and Leo Young’s high
school diploma disappeared in their common love for radio; both were
active amateurs—Taylor, with call letters 9YN, and Young, W3WV. From
the time they observed the effects of the passage of a river steamer between
their 5 m transmitter and receiver they never had radio location far from
their minds.
These thoughts came to nothing during the 1920s because the nec-
essary high-frequency electronic components simply did not exist. The
combination of the Young–Hyland observations in 1930 and the availabil-
ity soon after of high-vacuum low-voltage cathode-ray oscilloscopes and
multi-element electron tubes changed things, but their preliminary results
did not provide them with a high priority by the Bureau of Engineering,
in spite of the strong support for radio location by the Laboratory Director,
Captain Edgar Oberlin. The publication in 1933 by three investigators at
Bell Telephone Laboratory of picking up high-frequency reflections from
airplanes [1] did not make the low priority sit any better with Taylor.
At the time their primary concern was an attempt to remove the Lab-
oratory’s research function and make it in Oberlin’s words ‘a glorified test
shop’ for the Bureau of Engineering, which thought that NRL could not or
anyway should not compete with commercial laboratories, a change that
would have put an end to radar work. Oberlin was not lacking in the skills
needed to protect his command and succeeded in retaining NRL’s research
freedom. He was soon backed by the appointment in May 1935 of Rear
Admiral Harold G Bowen, a strong believer in NRL research, as head of
the Bureau of Engineering [2]. That the Laboratory’s allotment for 1934
decreased by a third from the preceding year did not make matters any
easier for the birth of Navy radar.
Despite all, the radar program received a better priority in early 1934.
The first important effect of this was the assignment of Robert M Page full
time to the project, which had been the part-time job for Young and two
others. Hyland had left naval service, somewhat acrimoniously [3]. Like
Taylor and Young, Page was a mid-westerner from the kind of family that of
necessity prepared children to make their own way in the world. He came
to NRL with a bachelor’s degree in physics from a small church college in
St Paul. Navy radar was to grow around Page.
His first task was to try pulsed waves. Young had convinced Taylor
that this had to be done, for continuous waves had produced nothing but
the unpleasantness with Major Blair of the Signal Corps. The idea had
two obvious origins in NRL experience, the sounding of the ionosphere
with radio and of the ocean depths with sound, but both were in fact quite
different. Both used pulse durations measured in fractions of milliseconds
and had signals reflected from huge surfaces. The size of an airplane or ship
65
could hardly allow the use of a train of waves 300 km long that a millisecond
pulse would generate. These plans differed from Watson Watt’s Chain
Home by envisioning from the first a searchlight-like beam with target
direction ascertained by rotating the antenna for maximum signal.
The first attempt with radio waves pulsed for 10 µs took place in
December 1934 and had little to recommend it other than a clear indication
of where the problem lay—the receiver.
Page had successfully used a multi-vibrator, an electronic switch ca-
pable of turning the transmitter on or off very quickly, to modulate the
transmitter, but the receiver was a slightly modified communications set.
It had high gain and the sharp tuning for which such circuits are noted.
The transmitter and receiver had separate antennas arranged to have a
minimum transmitter signal picked up by the receiver, but enough was
picked up to cover the screen because of the ringing caused by the narrow
pass band5 . Nevertheless, there was a definite effect from signals reflected
from the test aircraft [4].
Page thus faced the same problem that faced all other radar engineers
and set about to solve it his own way from basic principles using a French
paper as his guide [5]. This problem was then being faced not only by
radar but by television engineers, who had to contend with changes in the
strength of their video signal that were just as rapid as the radar pulses.
It is curious that no one at NRL seems to have noted the significance of
the descriptions of RCA’s prototype television system that appeared with
descriptions of the receivers [6], more evidence for NRL’s isolation from
the New York electronic community of RCA, Bell Labs and Signal Corps.
Page solved the design and equally formidable construction prob-
lems by November 1935 using funds illegally diverted from another
project. Taylor realized that radio location had to have more help—indeed
Page had had to leave the work for a while during the year to help out on
another job—and in early 1935 visited Mr James Scrugham, the most influ-
ential member of the House Naval Appropriations Committee, to explain
his concerns. The result was a telephone call next day telling Taylor that
the Committee had agreed to an additional $100 000. Among other things
this allowed Robert Guthrie to work with Page, who put him to work on
the transmitter. They formed a congenial working pair as they brought the
device on line.
The next test in April 1936 achieved what was desired, with planes
observed at ranges of 8 km, quickly extended to 27 km, and Adm. Bowen
gave the project the highest priority. Taylor immediately made two de-
mands of Page: that a common antenna be used for both transmission and
reception and that a shipboard test be made at the earliest possible date.
For meter waves the antenna had to be an array of dipoles, and the
larger the array the narrower the beam and the greater the intensity of the
5 See ‘Receivers’ in the appendix (pp 475–6).
66
Origins
projected power. Using a single antenna that had the area of two separate
ones yielded improvements in angle and intensity of a factor of four. The
difficulty lay in preventing the transmitter with its enormous power from
burning out the receiver. Page arranged fractional-wavelength transmis-
sion lines in such a manner that the receiver input tube grid current, an
effective short circuit, reflected a high impedance to the transmitter. It was
a neat idea that formed the basis of several variations. Page called the
device a ‘duplexer’.
In April 1937 a 1.5 m set was installed on the destroyer Leary with
a Yagi antenna mounted to the barrel of a deck gun to allow pointing.
The results were successful in so far as shipboard service went, but the
transmitter power had to be increased. The solution to that problem was
a ring of Eimac 100TH transmitter tubes that allowed power to be built
up in phase. The Signal Corps adopted this tube and transmitter design.
A target date of 1 September 1938 was set for a prototype set, XAF, to be
mounted on a battleship and tested in fleet exercises.
In mid 1937 two electronics corporations were brought into the plan-
ning: Radio Corporation of America and American Telephone and Tele-
graph. RCA had had an interest in radar through Wolff’s 10 cm work, but
for the men from Bell Telephone Laboratory, representing AT&T, it was all
new and quite astounding [7]. RCA was interested in producing the XAF
but wanted to build a design of their own for the fleet exercises, designated
CXZ. The Bell Labs people returned home to think about these new ideas.
By January 1939 XAF was mounted aboard USS New York and CXZ
aboard USS Texas, and exercises in the Caribbean began. The results of XAF
were spectacular and made instant converts to the new weapon of all line
officers select enough to have encountered it. Ships were observed at 16 km
and aircraft at 77 km. Shells could be followed in flight and their splashes
noted. Night destroyer torpedo attacks were thwarted. An unexpected but
very important use of the new equipment soon became apparent. While
operating in the vicinity of the Virgin Islands at night the ship’s position
became poorly known, and Page located it by ranging mountain tops on
distant islands [8].
XAF liked the sea and did not mind gunfire at all, something that
could not be said for CXZ. Taylor found this useful when insisting that
RCA build exactly to NRL specifications, especially keeping Eimac tubes
in the production model. The Chief of Naval Operations immediately
asked for 20 sets in their current form. This was to become CXAM, which
would later evolve into the SK, the Navy early warning set for the entire
war, referred to by all as the ‘flying bedspring’.
2.4.2. Bell Telephone Laboratories

On returning from their radar demonstration on 13 July 1937 the Bell Labs
engineers initiated serious discussion about the part AT&T was to have in
67
this new technique. There was another visit in November after which the
directors decided to convert the field station at Whippany, New Jersey and
place it under the control of W C Tinus. It had been established in 1926 for
trans-Atlantic telephony and broadcast development, and its isolation, 50
km west of Bell’s New York City labs, made it ideal for secret work. The
company decided to enter the field at its own expense rather than solicit
a government contract because the rules of the time made it difficult to
undertake something of a highly speculative nature, and the Bell engineers
wanted to try shorter wavelengths [9].
One reason for selecting short wavelengths was the recent invention
at Bell Labs by A L Samuel of triodes for very high frequencies [10], which
placed them in a marginally better position to do something with very
short waves than their numerous predecessors. Their designs paralleled
the RCA Acorns6 but also included a transmitter tube, 316A, whose shape
caused it to be called a ‘doorknob’. Work began on 40 and 60 cm equipment
intended for gun laying or fire control as the Navy preferred to call it. A
single horizontal cylindrical paraboloid antenna, the kind used by Hertz
and Hülsmeyer, and Page’s duplexer formed the basis of the set.
Bell engineers demonstrated their first prototype, the CXAS, more
correctly a pre-prototype, in July 1939 to representatives of both Navy
and Army. They selected a position on a bluff 25 m above the sea at At-
lantic Highlands, New Jersey, which presented a rich field of targets in
the ships moving into and out of New York [11]. The 40 cm set could
pick up some ships 15 km distant. The accuracy in range was satisfactory
enough—and very important for naval fire control—but angular accuracy
was inadequate for radar-directed fire. Work was already proceeding on
eliminating this fault using the lobe-switching technique that they had
learned from the Signal Corps7 . In the meantime the Navy began to plan
radar’s integration into current fire control systems and ordered ten to
be delivered designated CXAS, then FA [12]. The first set was installed
aboard USS Wichita, but not until June 1941. The Army was interested in
the device for coast defense, but delayed ordering what they designated
as SCR-296 until late 1940 [13]. For the Army it proved to be a dead-end
design, for the Navy the beginning of all fire control, both surface and
air. For Bell Labs and AT&T it was the beginning of a gigantic undertak-
ing.
Bell Labs introduced horizontal lobe switching to alter the design for
main-battery fire control in the FC and horizontal and vertical switching
for anti-aircraft in the FD. Production began in 1941 [14]. Thus the US
and Royal Navies independently introduced similar equipment at about
the same time: FC corresponding to British type 284, FD to type 285. In
production Bell greatly improved output power over the original design
6 See ‘Multi-element tubes’ in Chapter 2.1 (pp 35–6).

7 See ‘US Army Signal Corps’ in the next section and ‘Lobes’ in the appendix (pp 71, 471).
68
Origins
that used triodes, which the British continued to use, but that story must
be delayed until Chapter 4.1.
2.4.3. US Army Signal Corps

William R Blair was about the same age as Hoyt Taylor. He too earned a
PhD degree but in physics and from the new University of Chicago rather
than from an old and famous German school. His interest in electromag-
netic waves originated in his thesis work for Professor A A Michelson dur-
ing which he generated microwaves in the range from 10 to 40 cm with
much the same techniques as Hertz. He noted, possibly for the first time,
that these waves were reflected by materials other than conductors. His
work with Michelson also included measuring the speed of light. He thus
filled his mental storehouse in graduate school with two key elements of
radar.
Radar first occurred to him in the spring of 1926 when attending the
Command and General Staff College where he heard lectures about locat-
ing aircraft by sound detectors. There were so many problems apparent
to Blair in such a method that he immediately thought of the problem
in terms of microwaves and the velocity of light. After graduating from
this preparatory school for generals he was assigned as Chief of Research
and Engineering directly under the Chief Signal Officer and soon became
Director of the Signal Corps Laboratories [15].
The Signal Corps had become responsible for Army radio research
in February 1919 and located the work at Camp Vail, soon renamed Fort
Monmouth, in central New Jersey. Their mission was differently defined
than the Naval Research Laboratory. Whereas the Navy laboratory had a
wide range of freedom in seeking out subjects for study, the Signal Corps
Laboratories were restricted to the expressed needs of the various arms. In
so far as radar was concerned this did not present a problem because the
Coast Artillery Corps, to which anti-aircraft defense had been assigned,
was very much interested in any way of locating aircraft, although the Air
Corps was not.
A great deal of effort was devoted to detection with infrared, work
not completely given up until 1939. Blair never believed infrared would
be the answer, primarily because of its inability to penetrate fog and rain,
but his approach was positive. When he learned that the country’s leading
expert for infrared, Dr S H Anderson, had lost his job as a result of depres-
sion cut-backs, he hired him to push the technique. To be the best in the
technique also fit his wish to build a first-rate laboratory. They learned as
did investigators world wide that it had no value for aircraft detection [16].
Blair was set to avoid a trap with infrared but was ill prepared to give
up microwaves, despite the poor results that Wolff’s equipment and the
Signal Corps work carried out by William D Hershberger had shown. At
the moment when important decisions needed to be made, a new player
entered the game, Lt Col Roger B Colton, as Chief of Research and Engi-
69
neering. So secret was radar that he was told he could not see what NRL
was up to, and it even took a while for him to learn that the Corps was
already on the project. His first actual encounter with radar-like apparatus
was in the summer of 1935 when he was invited to observe a demonstra-
tion by General Electric of a method for the radio detection of airplanes by
Dr C W Rice, who was experimenting with microwaves using split-anode
magnetrons. It was a failure but impressed Colton as having failed be-
cause the antenna was too small; his interest increased on visiting Rice’s
laboratory where he was shown the detection of moving vehicles at half
a mile [17], essentially the same experiment that Wolff had conducted at
Sandy Hook the year before.
Colton was quite a different personality from Blair. Blair had in a
few years made the Signal Corps Laboratories first rate, but Colton was a
radar zealot who wanted to build equipment. He was also an inspiring
leader and soon gained the respect of everyone at the Corps Lab, military
or civilian [18]. He was also known for preferring to settle an argument by
drinking his opponent under the table [19].
An unexpected incident in December 1935 marked the turning point
in the Corps’ path to radar. A proposal for the radio detection of aircraft
was sent to the Signal Corps by the Bureau of Standards and came naturally
enough into Colton’s hands. Colton requested of Blair that he send Her-
shberger to evaluate it and visit NRL, where Hershberger had worked on
sound a few years before. (The secrecy curtains that separated the two ser-
vices were not particularly hard to draw when there was a need.) It was a
propitious decision, not that there was anything in the Bureau’s proposal—
it was the Taylor–Young–Hyland ‘radio screen’ again—but it required a
report by Hershberger, who was heartily fed up with microwaves by then
and had suggested pulsed meter waves two years earlier. Hershberger’s
visit with Page allowed him to see the results of the first pulsed-wave
experiment and the preparations being made for improving the receiver.
His report gave the reasons for the Army’s lack of progress as preoccupa-
tion with microwaves and lack of support for radio detection. Blair had
lost hope for microwaves by then too and forwarded the report with a
strong recommendation to Maj Gen James B Allison, Chief Signal Officer.
Allison asked the War Department for $40 000 for radio location and was
refused [20].
Allison gave the project the highest priority, asked for the funds again
and was again refused. He then agreed that Blair should use $75 741 ap-
propriated for a top priority communications project for fiscal year 1937. It
was illegal, but Allison said he would stand behind Blair. Blair promised to
do his best to have something to show by 1 June 1937. Days followed with
many unpaid Sundays and overtime hours [21]. In August 1936 Colton
was assigned as Blair’s Executive Officer. In October Hershberger left to
take a doctorate in physics at the University of Chicago, and Paul Watson
was named chief engineer [22].
70
Origins
Events then moved swiftly. In December 1936 some lashed-up equip-

ment was transported to Princeton Junction to track the steady flow of air
traffic in and out of New York. The transmitter was placed a mile from the
receiver to keep it from being overwhelmed, and the reflections were quite
pleasing. They then built large antennas consisting of arrays of dipoles to
allow direction to be determined, and five months later they were ready to
demonstrate a 3 m set with direction and ranging capability to the Secretary
of War and the Chief of Staff. To the astonishment of the visitors the strange
arrangement of wires on wooden framework succeeded in detecting the
arrival of an unlighted night-flying bomber and putting a searchlight on
the attacker in plenty of time for AA guns to open fire. It was almost too
successful because the Chief of Staff wanted the present design put into
immediate production, something the Signal Corps was loath to do.
Allison was now offered $50 000 by the War Department but de-
manded $250 000, which required a Congressional appropriation that was
thought impossible. Colton insisted. Congress was asked and gave. The
money famine was over [23].
The next step was to prepare for a demonstration of the prototype of
what was to be SCR-268 (Signal Corps Radio-268). It was in fact only an
electrical prototype by then using 1.5 m; the mechanical design went more
slowly and could not be completed until the electrical specifications were
fixed. It was operated by a detachment from the 62nd Coast Artillery Regi-
ment commanded by Lt A F Cassevant, to learn how soldiers would adapt
to such a device; they also furnished the security for the equipment, then
on isolated Sandy Hook. With time Cassevant became an accomplished
radar engineer and transferred to the Signal Corps. Colton insisted that
SCR-268 should have lobe switching for both the horizontal and vertical an-
tennas and put James Moore on the design. Engineers from Westinghouse
and Western Electric took up residence in order to facilitate production
later [24]. Blair became hospitalized in the summer of 1938, Colton was
named Acting Director, and when Blair had to retire because of his health,
Colton became Director. The tribulations in getting SCR-268 to the tests for
the Coast Artillery Board at Fort Monroe included severe damage of the
equipment by hurricane, the near capsizing of a ferry over the Delaware
and a complete disruption of highway traffic to a degree unknown to the
innocent motorists of that day.
The tests were made at the end of November 1938 with the usual
notables plus Brigadier General Henry Arnold, Assistant Chief of Staff of
the Air Corps. The feats of the previous test were to be repeated but at
greater ranges and with greater accuracy. Almost as if following stage
direction it ended with an incident of high dramatic value. A B-10 bomber
was to make night approaches to Fort Monroe at 6000 m altitude from the
west. The 268 was to pick it up and point the searchlights, as they had done
at lower altitudes before. The Air Corps liaison announced the plane was
overhead. No radar contact. After the third failure the radar men began to
71
look at places where the plane was not supposed to be and found it—well
out over the Atlantic, for a very strong west wind had completely fooled
the experienced night navigators. The return flight took an hour, and when
the plane came within searchlight range a beam of light ended on clouds.
But the stage directions seemed well worked out, because when the errant
aircraft passed out of the clouds into clear sky, it was illuminated by the
searchlight. Arnold wanted an early warning radar [25].
The troublesome details of the mechanical design that made the 268
portable enough to accompany, to be sure rather clumsily, troops in the
field were overcome by midsummer 1940. The tests of 1937 and 1938 had
used three separate antennas: a transmitter, an azimuth receiver and an
elevation receiver. These had to be incorporated into a single unit suit-
able for field service. A contract was let to Western Electric from which
deliveries began in February 1941 [26].
Following the May 1937 tests the Air Corps had expressed interest in
an early warning set, and work started in parallel with that of the 268 on
what was to become SCR-270, a mobile set, and SCR-271, for fixed instal-
lations. Accuracy of direction was sacrificed for range so the wavelength
was increased to 3 m and lobe switching dispensed with but duplexing
incorporated using a design by Dr Harold Zahl. Whereas the 268 used
Page’s ring oscillator, initially of eight, later 16 Eimac 100THs, Westing-
house agreed to make a special water-cooled output tube, WL-530, that
could deliver the desired 100 kW power with a simple push–pull circuit,
by then a standard electronic design. The difference in wavelength, 1.5 to
3.0 m, made the difference between a 16-tube ring oscillator and a two-tube
push–pull. In June 1939 an engineering model of the 270 tracked an air-
craft consistently to 125 km and occasionally as far as 240 km and a flight
of bombers consistently to that range. A contract was let to Westinghouse
in August 1940, who delivered 112 sets before the US entered the war. The
fabrication of a 271 was completed at Fort Sherman, Panama in June 1940
and a number of 270s were shipped to Hawaii in the latter half of 1941. Six
were spotted around the perimeter of Oahu on the morning of 7 December
1941 [27].
The 268 remained the only radar for Army AA artillery until early
1944, when it began to be replaced by the incomparable SCR-584. The
Signal Corps was aware of the Bell Labs equipment and its great potential
for searchlights or gun laying but rejected it in favor of the 268 because of
the latter’s much greater range. This gave an AA battery a single piece of
equipment useful both for acquiring a target and directing guns or search-
lights onto it. In the Pacific it would serve in a function unsuspected in
1939—ground-controlled interception, about which more later. The 270
remained the standard early warning radar until the end of the war and
even saw service in the Korean War [28].
Thus by the time war had broken out in Europe the American Army
and Navy both had excellent prototype radars that would be in production
72
Origins
well before the time of the attack on Pearl Harbor. As is so well known, an
SCR-270 secured for the sleeping garrison nearly an hour’s warning, but
the organization necessary to make use of this intelligence was only just
being formed.
2.4.4. GEMA
For those persons who delight in irony the development of microwave
radar has an example. The British gave such short wavelengths the least
thought during the 1930s, yet invented the revolutionary cavity magnetron
that delivered prodigious power at 10 cm. The Germans pushed research
in the region shorter than 50 cm wavelength all through the decade and
stopped only when a wartime shortage of technical manpower and the
absence of a microwave oscillator of sufficient power forced termination,
yet they missed the cavity magnetron and ended far behind the Anglo-
Americans in this critical technique.
Serious radar work in Germany came from the zeal of Dr Rudolf
Kühnhold, the Technical Head of NVA. His work in underwater sound led
him in 1933 to the idea of using radio waves for locating and ranging tar-
gets, and his failure to observe reflections with 13.5 cm waves impelled him
to consult with a leading electronics company, Telefunken, where toward
the end of the year he discussed the problem with Dr Wilhelm Runge, Tele-
funken’s acting laboratory director. Runge was not interested because he
saw in it the need for vacuum tubes that he thought lay years in the future.
To others Runge voiced the opinion that the whole project was ‘utopian’ or
worse [29]. Kühnhold found Runge rude and did not forget. This rebuff
helped divide German radar development and production between three
companies: Telefunken, Lorenz and one not initially in existence.
By this time Kühnhold’s dealings with Tonographie concerning un-
derwater sound had become so agreeable that he dropped attempting to
gain the interest of the communication industry. He was particularly
pleased with the resolute spirit of Tonographie’s owners, Paul-Günther
Erbslöh and Hans-Karl Freiherr von Willisen, and they were equally
pleased with Kühnhold’s openness. Kühnhold loathed bureaucracy and
relished the absence of it in company operations, an appealing contrast
with the procedures for doing anything at his own NVA. Their business
in phonograph recording was prospering, which provided capital for tak-
ing risks, and, in seeking new fields for their talents, the two entrepreneurs
were experimenting with 95 cm directed-beam communication, which they
thought might interest the Kriegsmarine for secure communication be-
tween ships. It was an application for which the low transmitter powers of
the very short wavelengths were no problem and was also compatible with
Kühnhold’s interest in radio location. To aid them they sought two high-
frequency experts as consultants: Dr Hans E Hollmann and Dr Theodor
Schultes, both of the Heinrich-Hertz Institute [30].
73
In November 1933 von Willisen noticed an advertisement by Philips

for a magnetron that delivered 70 W at 50 cm and ordered one immediately,
news that raised Kühnhold’s spirits measurably. In February 1934 while
waiting for the magnetron equipment to be finished von Willisen produced
a failure of his own at 75 cm. Using the techniques being tried for directed
beam communication, he connected his transmitter to a Yagi and attempted
to observe the reflections with a dipole whose receiver was overwhelmed
with the direct wave signal, but the completion of the magnetron equip-
ment swept aside any regrets about that experiment. He designed a 48 cm
transmitter and connected it to a Yagi; Hollmann used a Barkhausen tube
as a regenerative receiver connected to a dipole antenna. All attempts to
keep the direct signal out of the transmitter failed, even substitution of a
paraboloid for the Yagi.
This failure was tempered by a successful use of the magnetron for
communication. The magnetron had proved to be extremely unstable in
frequency, especially in the push–pull form first used, so von Willisen re-
placed it with a single-output-tube design and made a careful study of the
settings for voltages and magnetic field to ensure best performance.
A successful experiment by Kühnhold in June with the Pintsch equip-
ment that had now attained 0.3 W caused von Willisen to try his improved
circuitry with a paraboloid for the transmitter and a dipole array designed
by Schultes for the receiver. This proved to be a twin success; a large
steamer that passed by was tracked by the 13.5 cm equipment to 4000 m
and by the 48 cm equipment to 2000 m [31]. An important difference was,
of course, using a moving target that generated a Doppler interference
signal.
Before these critical experiments were carried out Erbslöh and von
Willisen incorporated a new company in order to separate these military
applications from Tonographie. They called it Gesellschaft für elektro-
akustische und mechanische Apparate, a name intended to deflect the cu-
rious by its reference to previous activities of the owners and, for those who
might have known, those of Kühnhold. The acronym GEMA came into im-
mediate use, and the seven-word name seems never to have crossed human
lips again. They hired Schultes to run the high-frequency laboratory and
another Heinrich-Hertz man, Dr Walter Brandt, to run the low-frequency
lab, thereby completing the top design levels of the company for several
years [32]. Hollmann withdrew from the company when, owing to his
association with them, he was forced to delete portions of a book he had
just completed, which soured him on military work. Despite the omitted
parts, it became the ultimate authority for microwaves until the end of the
decade [33].
The GEMApeople carried out continuous-wave experiments through
October 1934 and saw that nothing useful could be done with them, al-
though a strong reflection from an airplane that happened to pass through
their beam opened their thoughts to targets other than ships and helped
74
Origins
open the pockets of the Kriegsmarine for a grant of Rm70 000 ($16 500). The
problem with continuous waves was the near impossibility of keeping the
transmitter signal out of the receiver, and on the advice of Hollmann and
Schultes they undertook to try pulse techniques. They had by then set up
their ‘GEMA tower’ on the NVA experimental grounds at Pelzerhaken near
Neustadt on the Lübecker Bucht and from it had a notable success. The
transmitting antenna had ten dipoles backed with a reflecting mesh, the
receiving antenna three dipole pairs. Transmissions were on 52 cm with
2 µs pulses at 2000 per sec. The receiver was a broad-band heterodyne that
used the new RCA Acorns. Success of a form came when the set first oper-
ated in May 1935; they were surprised by seeing the wood on the opposite
shore showing up at 15 km, but ranges on the 400 ton research boat Welle
were not satisfactory. The receiver was redesigned to use two intermediate
frequencies, and by 26 September they could follow the little boat to 8 km.
It was time to show the results to Admiral Erich Raeder, commander of the
Kriegsmarine, and lesser notables [34].
This demonstration secured adequate funding for the project but dis-
closed the first of many difficulties that were to accrue from working for
the Kriegsmarine. The demonstration had incorporated lobe switching
that had given a directional accuracy of 0.1◦ , and to their astonishment the
engineers were told this was too complicated and to leave it off present de-
signs [35]! Next they were told that a cathode-ray tube was too delicate for
use aboard ship and that a substitute must be found. The second objection
was soon removed and the decision reinforced later in a rather macabre
affair: the little test boat Welle was lost with all hands, but the cathode-ray
tube that was in a prototype set on board was recovered and still func-
tioned, which convinced the naval authorities that it was an acceptable
component [36]. The first objection was not overcome until it was too late
to provide blind-fire capabilities in any important naval action. Radar was
to provide range and search capability only.
These results led to disagreements about what to do next. Kühnhold
wanted the 50 cm work to continue, but Erbslöh and von Willisen wanted
to increase the wavelength to gain radiated power and thereby range.
Kühnhold planned to continue his 13.5 cm work and strongly favored
pushing toward shorter wavelengths, marshaling all of the valid argu-
ments for microwaves. (He and Blair would have understood one another.)
After discussions colored with asperity they decided to do all these things.
GEMA’s 50 cm work became a surface-search radar, DeTe-I, the first of the
Seetakt series, and the long-wave work became an air-warning radar, DeTe-
II, which soon settled on 2.4 m. The shorter wavelength was needed for
surface targets because the vertical lobe structure severely cut the range
to surface targets as wavelength increased; the longer wavelength was
needed for the extended range to pick up distant air targets. Von Willisen
replaced the unstable magnetron in the 50 cm set with a high-frequency tri-
ode, TS1, of GEMA’s manufacture, and Kühnhold replaced the hopelessly
75
weak Barkhausen tube in the 13.5 cm equipment with the magnetron. He

continued microwave work at NVA until a British microwave radar came
into German hands in February 1943.
The TS1 triode had to be driven very hard to meet power require-
ments, had a very short useful life and was replaced by an improved model,
TS6, which did somewhat better. It was the same story in Britain and Amer-
ica. The short lifetime of triodes in a British 50 cm set would alter the course
of battle with the Bismarck in May 19418 .
Raeder’s eagerness for the new device was tempered by a greater
eagerness for squeezing as much as possible of the generous flow of money
coming from the new regime into preparing more tangible armaments for
a surface war with Britain, to take place according to Hitler’s assurances
in 1944 at the earliest. Furthermore, he emphasized to Kühnhold that
NVA’s highest priority must be under-water sound. Nevertheless, on 18
April 1936 the Navy decided to equip all cruisers and battleships with the
Seetakt surface-search radar. The Navy specified that the sets were to be
mounted on the rotatable housing of the optical fire control with its primary
function to be the accurate determination of range, always the weak aspect
of optical systems, with detection of ships and obstacles in darkness or fog
a secondary purpose [37]. Blind-fire capabilities were certainly important
but were to be postponed until made simpler for the operators.
During the remainder of 1936 and 1937 there was a great deal of
experimentation on the naval sets, including sea trials. The inadequate
range of the 50 cm equipment forced the wavelength to 60 cm, used in
the set installed aboard the pocket battleship Admiral Graf Spee in January
1938. The next sets took the final Seetakt wavelength of 80 cm.
In adopting the first operational naval radar the German Navy ap-
pears to have been highly progressive, and surface commanders were to
make excellent use of the new weapon, but this was more the result of the
individual initiative that came to all commanders who encountered the
powerful new eye, rather than from high-level interest or understanding.
Radar had come from below, not been ordered from above. The sets were
accepted and installed but GEMA had to learn by experience about the
harsh environment of a warship with many early failures resulting. Worse
yet, the Navy’s operating and maintenance personnel were poorly trained
and instruction manuals, even circuit diagrams, were, initially at least, not
found aboard ship—the result of the obsession with secrecy and the at-
titude that equipment installed aboard ship had to be foolproof, would
not fail other than through enemy action and would not require extensive
technical knowledge by the operators [38]. No officers were assigned to
supervise the operation and use of radar on German ships until near the
end of the war [39].
Kühnhold continued to work closely with GEMA, contributing fre-
8 See ‘Bismarck and Prinz Eugen, 18–27 May 1941’ in Chapter 3.3 (p 124).
76
Origins
quently to design but never as a member of the company. He remained at

the naval laboratory, which expanded with many new functions, moved to
Kiel-Dietrichsdorf and changed its name to Nachrichtenmittel-Versuchs-
kommando (NVK) at the beginning of 1939 [40].
The feeder arrangement for a German Freya. This shows the antenna connections
of three dipole arrays such as were typically found with modification on dozens of
designs by all nations. The dipoles are spaced horizontally one-half wavelength
apart and placed one-quarter wavelength in front of a conducting screen. German
usage differed from other nations in using full-wave dipoles, which allowed them
to be mounted without an insulator to the grounded frame at the voltage null found
one quarter wavelength from the end. The receiver array is divided into left and
right arrays to allow the increased accuracy of lobe switching. The upper-most
array is for IFF; in this case the interrogation was by means of the radar signal
with the reply on a shorter wavelength to which the IFF antenna responded. Air
Ministry Air Scientific Intelligence Report No 34, April 1945. Crown Copyright.
The Kriegsmarine planned to use DeTe-I for coast-defense batteries

as well as for ships, and this proved to be where the greater number of sets
went. DeTe-II was the air-warning radar for their land installations and was
not installed in warships. This was a curious attitude when compared with
that of the US Navy, and may have had its roots in German naval thought
not having been through the bomber-ship controversy of the 1920s (which
at times gave the impression that the next war would be fought between
77
the US Army Air Corps and the US Navy). Both models for land use
were transportable. The existence of an excellent air-warning radar was,
of course, secret, even from the other branches of the Wehrmacht.
The first knowledge of it obtained by the Luftwaffe came from a visit
by Hitler, Göring and Raeder to the Torpedoversuchsanstalt (a general
naval weapons laboratory) in July 1938 where DeTe-II was demonstrated.
Göring was furious that his chief of signals, Col Wolfgang Martini, had
neither been informed of the existence of this device nor invited to the
demonstration. The response was that this was a Navy weapon; invent
your own. The Kriegsmarine did all it could to keep the Luftwaffe from
buying equipment from GEMA, a circumstance that no doubt contributed
to the widespread belief that the firm was a front for the Navy. Until
the end of the war they used all possible ways to undermine GEMA’s
growing and very pleasant relationship with the Luftwaffe [41]. Martini
was delighted by radar and immediately ordered some DeTe-IIs, which
acquired the name Freya [42].
In September 1938 Martini deployed two Freyas with GEMA per-
sonnel for possible use during the invasion of Czechoslovakia: one on
Geisinger Berg in the Erzgebirge and another on Grosser Schneeberg in Sile-
sia. There the restrictions of siting were first encountered. The mountain-
top locations chosen resulted in a clutter of reflections from nearby peaks,
the lesson being that the best location for such long-wave equipment was
at the bottom of a shallow bowl-shaped depression [43].
Martini had organized the Luftnachrichtentruppe (Air Signal Corps)
for the new Luftwaffe and would remain a competent and enthusiastic
user of radar throughout the war. After graduating from school in 1910
he became a cadet in Telegraph Battalion 1 and acquired experience in
military communications during the 1914–1918 conflict. His grasp of the
use of radar was instinctive, but he repeatedly failed to make those higher
up comprehend the full significance of such technical matters. Watson
Watt possibly explained the matter:
I have a very dear postwar friend in General Wolfgang Martini,

a shy, modest, charming, very perfect gentleman. His many
claims on my affectionate respect include his failure to endear
himself to Göring from whom the qualities I have just tried to
summarize may have concealed General Martini’s very high
technical competence, wisdom, and resource [44].
2.4.5. Telefunken
When Runge dismissed Kühnhold’s suggestions that Telefunken enter the
radar business it was not because he was not interested in very short
waves. On the contrary they had become his main interest. Shortly before
Kühnhold’s visit the management of Telefunken had been reorganized;
among other changes the Jewish director, Emil Mayer, had had to step
78
Origins
down. Jewish engineers, having marketable skills and international con-

nections, began leaving the company during the next couple of years de-
spite the protests of their less perceptive colleagues. One of their best, Otto
Böhm was designing antennas for Marconi within a year and later for the
British Admiralty [45]. The new leaders favored adding some research to
Telefunken’s consumer-goods orientation, so Runge was ordered to begin
work on decimeter waves, incorporating a group currently active in the
field. Lacking other instructions they set about trying to develop navi-
gational aids as well as communication radio. During these experiments
they noticed the disturbing effects of ships and aircraft passing into their
beams but took no great notice of the significance.
They settled instead on what was to become for the German forces in
the war a valuable communications technique, decimeter-wave relays be-
tween line-of-sight stations (Richtfunkverbindung). Such links were much
easier to set up in the field and used far less material than wire lines and
were almost as secure. These transmissions used frequency modulation
and proved especially useful in the steppes of Russia and deserts of North
Africa [46].
Wilhelm Tolmé Runge, who made Telefunken a great radar company,
was descended from a Bremen commercial family that provided him with
English uncles. He was the son of the eminent Göttigen mathematician
Carl Runge and the brother-in-law of the equally eminent Richard Courant
of New York University. The natural expectation that Wilhelm followed
the scholarly path such a family had established is mistaken. After his
father had placed him in three schools in the hope of finding one in which
he could succeed, he graduated only as the reward granted the excited
young men who volunteered in 1914 and received a war diploma. He es-
caped the fate of so many in the raw student battalions who demonstrated
the futility of rushing entrenched infantry with fixed bayonets, but his
younger brother, Bernhard, did not, and fell in October 1914 [47]. Wilhelm
transferred to signals and ended the war as an instructor at the Nachricht-
enschule (Communication School) at Spandau [48]. His wartime signals
experience steadied his interests, and his postwar studies at the Darmstadt
Technische Hochschule resulted in a dissertation that attracted attention
in the radio industry [49].
By the summer of 1935 Runge began to wonder what might be done
with reflected signals and set one of his 50 cm transmitter antennas directed
straight up with a receiver nearby. He then had one of Telefunken’s Ju-52s
fly directly overhead while he watched the signal from the receiver. The
result was a large Doppler-effect meter deflection. His immediate superior
showed no great interest, which acted as a great stimulant on Runge. The
company did make a press release, however, that was picked up in America
by Electronics as ‘Microwaves to Detect Aircraft’ and later in Popular Science
[50]. Pulsing and duplexing (common antenna) with a paraboloid reflector
were quickly introduced. Tests conducted on the Baltic coast in the summer
79
of 1936 with the help of Wilhelm Stepp, who soon assumed the primary
design responsibility, picked up an airplane but at only 5 km. The set was
named Darmstadt for Stepp’s home town and was the first in a series of
geographical code names used by Telefunken.
This work was being carried out with the rather naive idea, possibly
rooted in Runge’s droll sense of humor, that it was navigational, and he
found himself in hot water for giving an open lecture about the pulse tech-
nique at the Lilienthal-Gesellschaft in early 1939 that also caused Kühnhold
to complain to Telefunken directors about this ‘unauthorized work’, which
was futile anyway because of only 15 W of transmitter power. The result
of this scolding on Telefunken’s administration was to have their tube de-
signers provide Runge the output triode that he needed, the LS180. It was
the design that was popping up everywhere that positioned leads to min-
imize reactance at the price of inconvenience when installing the tube. It
was 140 mm long, 50 mm in diameter and gave 8 kW, extending the range
to 40 km for aircraft. Someone stuck a pin in the map, selecting in this
way the name Würzburg. It was demonstrated to Wehrmacht authorities
in July 1939. They soon used a rotating dipole feed, which Runge and
Stepp called the Quirl, at the focus of the 3 m diameter paraboloid. This
gave them a conical form of lobe switching that produced very accurate
directional pointing and that made the various Würzburg models the best
AA gun-laying radars until displaced by 10 cm equipment three or four
years later [51].
The Würzburg had all of the key elements of World War II radar: a
well defined beam with high directional accuracy, excellent range accuracy
and a common antenna.
2.4.6. Lorenz
Telefunken and GEMA were not the only German companies to enter the
radar business on their own. The Lorenz Company were not strangers
to high-frequency work and had developed in 1935 a series of very high-
frequency triodes. In 1938 Dr Hermann Berger left the laboratory Holl-
mann had founded on leaving GEMAand became head of Lorenz’s tube de-
velopment laboratory, leading to their RD12Tf becoming a better decimeter
transmitter tube than either GEMA’s TS6 or Telefunken’s LS180, the result
of having mastered the use of oxide cathodes in this demanding service
[52]. They had marketed a beam navigation method of overlapping lobes
designed by their engineer, Otto Scheller, whose remarkable 1907 patent
encompassed every element of this widely used system [53]. It had been
licensed widely abroad, even used by the Royal Air Force. These extensive
international connections caused a certain amount of hesitation to entrust
them with secret work. They had started collaboration with GEMA very
early, but the Kriegsmarine had required GEMA in April 1936 to terminate
it because of security concerns [54]; the Luftwaffe eventually overcame
80
Origins
these concerns, given that the secrets were Lorenz’s to start with.
The interest once aroused was not to be stilled, and Gottfried Müller
insisted they press on. By the beginning of 1936 they had constructed a
70 cm, 400 W, pulsed set with separate, rotatable ‘mattresses’ for trans-
mitter and receiver. From the top of their laboratory at Tempelhof they
obtained reflections from the Berlin Cathedral, 7.4 km distant. During the
course of the year the transmitter tube, DS320 (soon replaced by RD12Tf),
that generated 1 kW allowed them with 1 µs pulses to extend the range
to 14 km. In observing a windmill they noted the ‘propeller modulation’
on the returned signal, something to be of use in the not too distant fu-
ture. They replaced the mattresses with paraboloids of 2.4 m diameter,
changed the wavelength to 62.4 cm, increased the range for aircraft to 30
km, demonstrated it to Col Martini and gave it the name Kurfürst. Martini
had it tested at the Anti-aircraft Training School and was sufficiently im-
pressed to order a few sets of more advanced design for field evaluation:
20 each of Kurpfalz and Kurmark [55].
As work progressed, the design by Runge and Stepp began to show
clear superiority over the Lorenz device for gun laying as well as the des-
ignated function in air warning, although the RD12Tf gave Kurfürst a
greater range. Even without the Quirl the Würzburg A, demonstrated to
military authorities in July 1939, proved sufficiently good to be used for fir-
ing on targets obscured by clouds or fog. Telefunken’s design with rotating
dipole, common antenna and simpler circuitry [56] was clearly superior,
and Lorenz ceased production after a total of 40 sets [57]. Telefunken en-
gineers then began working on equipment intended from the start for gun
laying, which eventually became the Mannheim. In the meantime the basic
Würzburg was provided with a Quirl to become Würzburg C.
German radar was concentrated for the moment in two companies,
GEMA and Telefunken, each with a basic design. GEMA had two bands,
80 cm for Seetakt and 2.4 m for Freya. Telefunken used 50 cm for the
Würzburg and the decimeter communication relays. Both companies
brought out modifications and new models but remained true to the orig-
inal, open-end designs for which wide variety became possible, from gar-
gantuan early-warning to diminutive airborne sets. By this means they
made economical use of their limited engineering personnel. Lorenz was
not finished with radar, however, and would return with an excellent air-
borne sea-search set.
When the Luftwaffe became an independent arm of the Wehrma-
cht it obtained from the Heer (Army) the AA artillery. Initially, it did not
have an organization to evaluate new weapons other than aircraft, so it
relied on an army agency, the Heereswaffenamt (Ordnance Department)
for judgement about its AA guns and, when the matter arose, for radar.
This office arbitrarily classified the early sets into three types. For vari-
ous reasons they referred to GEMA’s Freya as A-1, to Lorenz’s Kurfürst
as A-2 and Telefunken’s Würzburg as A-3. Initially this coincided with
81
A-1 for early warning, A-2 for searchlight direction and gun laying and
A-3 for local observation and tracking respectively. Names were retained
through numerous design changes and adopted for subsequent new mod-
els. A uniform type designation was introduced indicated by FuMG (for
Funkmessgerät) followed by model number; Kurfürst became FuMG 38L
(L denoting Lorenz), for example. As the importance of radar became
more obvious, Göring wanted it added to his bureaucratic empire and had
it moved from the Heereswaffenamt to the Reichsluftfahrtsministerium
(National Air Ministry), at which time new designations were applied.
The Reichsluftfahrtsministerium was not responsible for equipment of the
wildly jealous Kriegsmarine, but the Marine nevertheless adopted the uni-
form nomenclature for radar.
2.4.7. Comments
It is scarcely necessary to point out to the reader the parallels in Ameri-
can and German radar work. The earliest work started in service radio
laboratories with heavy emphasis on microwaves. Both dropped these
wavelengths in their prototypes for want of transmitter power, although
retaining some research. This resulted in excellent meter-wave equipment:
XAF/CXAM for the US Navy, SCR-270 for the US Army, Freya for the Luft-
waffe and Seetakt for the Kriegsmarine. The approach to decimeter waves
by Bell Telephone Labs is remarkably similar to the path followed by Tele-
funken and probably came about because both had tube laboratories. The
Bell FD/mark 4 was the equal of the Würzburg, indeed in design its cousin;
it was with modification the US Navy’s AA gun-laying radar throughout
the war. One might speculate that this similarity among these industrial
engineers resulted from close association of the circuit-design engineers
with tube designers, thereby learning much earlier of the latest in tubes,
whereas the engineers at the American service laboratories worked with
generally available components. The Würzburg was a better gun-laying
set than the SCR-268, and the American equivalent, FD/mark 4, was used
only by the Navy. On the other hand the SCR-268 functioned also for dis-
tant target acquisition, which the Würzburg did not. The Germans were
generally about a year ahead of the Americans.
In 1939 the German and American prototypes were superior to the
British except for CD/CHL, which was a typical dipole array on 1.5 m. The
heart of the radar defense of Britain was Chain Home, a design not admired
abroad, but it worked, and here lies the principal difference between the
German–American and British approach. Britain knew danger in 1935 and
had the wisdom to realize that intelligence gained by radar was worthless
unless promptly interpreted and acted upon. Speed was vital—as history
has so amply shown—so a design was selected for which construction
could begin almost immediately and with it the integration with Fighter
Command. British production began in 1936, and the Royal Air Force
82
Origins
took over an operational station in May 1937. Neither Germany nor the
United States had a significant number of operational sets in September
1939 and neither had any organization to use them for air defense in any
way comparable to Great Britain. Britain had by the start of the war a
functional air defense system for the homeland.
Another distinction between the German–American approach and
the British was the relationship of the radar men to their governments.
Radar grew out of the experiments and plans of electrical engineers in both
America and Germany and came to the attention of high officials through
demonstrations. In Britain the pattern was inverted. The Tizard Commit-
tee actively sought new scientific weapons and, on finding radar and seeing
the first examples of its power, procured for them a blank check. Possibly
even more important was the interest at cabinet level from the beginning,
which had important direct and indirect effects on both the technical devel-
opment and the application. The German engineers had to contend with
Hitler and Göring at the top, both anti-intellectual and scientifically illiter-
ate. Churchill took an active, if not always helpful interest in the scientific
conduct of the war. The American Secretaries of War, Henry L Stimson,
and of the Navy, James Knox, were alert to the importance of radar and
gave it high priority.
As the new technique moved toward deployment it carried various
designations. A P Rowe named it RDF, thought to throw off the overly
curious as ‘radio direction finding’. The Signal Corps referred to it as ‘RPF’
for ‘radio position finding’, and the Air Corps, overcome with the need for
secrecy, called it ‘derax’. The name ‘radar’ was invented by S M Tucker, an
American naval officer, and adopted officially by that service in November
1940 [58], and it seems that the world was just waiting for it. It was a
composite with ra for radio, d for detection or direction finding, r for range,
and the second a for and. The British ordered its use after 1 July 1943 [59].
The Australians have their own way with the language and called the new
device a ‘doover’.
2.5. FIVE OTHER NATIONS

2.5.1. Japan
Japan left the Washington Naval Conference of 1921 with limitations that
required them to build fleets substantially smaller than those of Great
Britain and the United States and was determined to offset this by improv-
ing the quality of the Imperial Navy. One response was the creation of the
Naval Technical Research Department (NTRD, Kaigun Gijutsu Kenkyu-
sho), construction of which was begun in 1922 and completed in Septem-
ber 1930, part of the delay resulting from the great earthquake of 1923. Its
first director was Yuzuru Hiraga, a senior naval officer known as ‘the god
who creates warships’ [1].
Yoji Ito was a career naval officer who was recommended for ad-
83
vanced study in electrical engineering in Germany in 1926 where he came

under the tutelage of Heinrich Barkhausen at the Sächsische-technische
Hochschule at Dresden, an appointment that also introduced him to Hidet-
sugu Yagi, Barkhausen’s first Japanese student and by that time an interna-
tionally recognized authority on high-frequency radio. Barkhausen was
loved for his genial nature, his admirable pedagogical ways and his ex-
ceptional skills in radio matters, things that attracted excellent students.
By a curious turn of fate Ito and his brother Shigeru Nakajima visited him
in June 1937 when the son of A Hoyt Taylor, Ito’s American counterpart,
was one of the Professor’s students [2]. From the time of Yagi’s student
years Barkhausen maintained very friendly relations with his Japanese col-
leagues, culminating in he and his wife arriving in September 1938 for a
long-remembered visit. The affection of the Japanese for him is found in
Yagi proposing the ‘bark’ as the unit of amiability [3].
On his return trip to Japan in 1929 Ito chanced to meet Commander
Ryunosuke Kusaka, who was returning by sea from a flight from Japan
to the United States aboard the Graf Zeppelin (LZ-127). As a result of this
meeting Ito was assigned to the newly opened NTRD where he took up
work in radio-wave propagation [4] and soon became the Navy’s leading
electronics expert, a capability he successfully carried over into the postwar
world. In conducting ionosphere soundings he noted the effects of the
passage of aircraft, which started him thinking about radio location [5].
During the early 1930s various reports of electronic weapons came to
the laboratory. In 1935 the proposal by an American to sell his invention
for the radio detection of aircraft was discussed at length but rejected.
When Ito visited him in 1937 Barkhausen told him that the German Navy
had apparently succeeded in developing a device for measuring range at
night. At about the same time he learned that Marconi was working on
similar equipment in Italy. Ito passed this information immediately to the
military attaché in Berlin for transmittal back to Japan, where it raised
no interest that could be noticed. But some imaginations were primed; the
commander of a Japanese cruiser sent to participate in the naval ceremonies
for the crowning of King George VI observed some British exercises in
which searchlights were brought onto aircraft uncannily fast, attributed by
him to some radio location method that, of course, did not exist [6]. None
of these alarms resulted in the beginning of a radio-location program at
the NTRD.
Japan was to be the host of the Olympic Games in 1940 and wanted to
make a better international impression than had Germany in 1936, which
meant that even better television coverage would be required. In order to
bring this about Masatsugu Kobayashi, an engineer heading the vacuum
tube section of Nippon Electric (NEC, Nippon Denki), left in May 1938 on
a tour of American and European television laboratories. He judged the
British work to be the best but was not allowed to visit their laboratories.
While in London he observed the strong effects on television reception
84
Origins
caused by the passage of an airplane nearby. On returning to Japan he

learned that the Olympics were off as a result of the war with China.
Kobayashi had hardly returned home when he repeated the Lon-
don experiment unintentionally when examining signal strengths of a
very high-frequency transmitter with a portable receiver; he noted a beat-
frequency when airplanes passed between him and the transmitter. He
quickly perceived what was happening and constructed the last of the
‘radio screens’ [7] that had already come forward in the US, France and
Russia, but it was the only one that proved to be useful. It was developed
as the Army’s Type A Bi-static Doppler Interference Detector and was able
to detect the passage of an aircraft at ranges exceeding 500 km [8]. Such
equipment, which delivered neither range nor direction data, was useless
in Europe and America where the density of flights would have saturated
it but was of use in China, where few Chinese aircraft were encountered.
The Japan Radio Company (Nihon Musen) began research in mi-
crowaves in 1932 using the Barkhausen tube. The work was criticized
for want of applications as being a waste of the company’s resources and
would probably have ceased had not Ito, who sought outside help when-
ever he could, offered a Navy Lab collaboration. This led to abandonment
of the Barkhausen tube for the split-anode magnetron that Kinjiro Okabe
had invented in 1927 [9]. This work culminated in a unique magnetron
design capable of 500 W at 10 cm in April 1939. We shall return to this
astounding discovery in Chapter 4.1 dealing with microwaves. For the
moment suffice it to say that Ito finally got a radar program started, al-
though one without high priority.
2.5.2. The Soviet Union

Whoever wishes to learn how governments fail in the duties of protecting
their peoples from disaster should study the history of the Soviet Union;
whoever wishes to learn how competent engineers can best be thwarted in
their efforts to provide weapons vital for defense should study the history
of Soviet radar. It is always the case, that in large projects those close to the
details are vexed by the confusion and mismanagement they perceive in
their leaders and, above all, in the administrative machinery that attempts
to carry out their leaders’ instructions. Such was the case in all radar devel-
opment in the Second World War, but those who toiled in the laboratories
of Britain, America, Germany and Japan and who suffered in this way little
knew that their work places were ruled by reason and benevolence when
compared with their counterparts in Muscovy.
There was every reason to believe that the Soviets might have sur-
passed the west in this new craft. They began first with high-level support,
had the influential interest of academician A F Joffe, had a brilliant young
electrical engineer and veritable model of the new Soviet man, Pavel Os-
hchepkov, and a radio engineer who had proved himself with decimeter
85
waves, Yu K Korovin, as enthusiastic leaders, and had obtained financial

support in 1934 of 300 000 roubles, which dwarfed that provided by any
other power [10]. The initial work pointed toward the development of a
Freya and a Würzburg, but by 1940 the resulting radar designs were poor,
inferior to the Japanese, and left Russia dependent on Britain and America
for much of her needs during the war.
To understand the history of this place and period one must learn
the identities of the contending bureaucratic agencies, and in order to keep
this cast of characters straight it is best to know them by their identifying
abbreviations.
GAU The Main Artillery Administration, an engineering service of the

Red Army concerned with the design of weapons.
PVO The Air Defense Forces, the service to which Oshchepkov was as-
signed and that had the responsibility for the employment of AA
troops; they had interests in weapons design.
SKB (also KB-UPVO). A special construction office within the PVO to pro-
duce radar, opened in 1933 with Oshchepkov in charge.
VNOS The Aerial Observation, Warning and Communication units of the
PVO, which were to be the immediate users of radar.
VTU The Military Technical Administration, a part of Red Army head-
quarters.
LFTI (also LIPT). The Leningrad Physical–Technical Institute, Joffe’s or-
ganization, which included D A Rozhanski until his death in 1936.
LEFI Leningrad Electro-physical Institute, a GAU laboratory, led by

A A Chernishev.
QRL (also TsRL) The Central Radio Laboratory, another GAU laboratory,
led by D N Rumyanysev.
NII-9 Scientific Research Institute 9, another GAU laboratory that ab-
sorbed LEFI in fall 1935. The renowned radio engineer Profes-
sor M A Bonch-Bruevich became its director after the purges and
attracted good men. Unfortunately, he died in March 1940.
UFTI The Ukrainian Physical–Technical Institute, a laboratory organized
by Rozhanski and where research in magnetrons was conducted.
Later directed by A A Slutskin.
NKTP The Research Sector of the Commissariat for Heavy Industry, the
supervisory organization for both LIPT and UFTI.
NIIIS-KA Scientific Research Institute of Communication Engineers of the
Red Army, a group with its own program for the development of
signals equipment.
VEI All-Union Electrical Institute, a competent research organization with
a laboratory for ultra-short waves led by Professor B A Vvedenskiy.
SRI (also NII-RP). Scientific Research Institute of the Radio Industry
headed by A B Stepushkin.
86
Origins
The Academy of Sciences (mercifully seldom referred to by initials). An

organization consulted at the highest levels that concerned itself with
all manner of scientific and engineering problems relevant to the So-
viet state.
NKVD The People’s Commissariat of Internal Affairs, the secret police,
the name of whose chief, Lavrenty P Beria, carried terror to millions.
At this point one might well let the reader form in his mind the kind
of radar that was to come from the machinations of these agencies, all of
which participated, and his construction would probably be close to the
mark. Yet the poor Soviet product resulted as much from the purges that
Stalin initiated in 1937 as from clumsy, bickering agencies—which knew
how to use the NKVD for their bureaucratic ends and to take care of a
few personal matters along the way [11]. Fear concentrates thought—but
on survival, not on the subtle intricacies of electronic circuits. This no
doubt lies behind the marked deterioration in design encountered in the
second stage of Soviet radar development. Important parts of this story
will remain unknown to us.
The PVO had responsibility for early warning and had sponsored the
early work on the radio screen Rapid that was done at LEFI until that group
was absorbed into the Television Institute, the combination becoming NII-
9. Despite the objections from many that Rapid gave precious little data of
value about intruding aircraft, PVO had LEFI build under the supervision
of B K Shembel a model suitable for army deployment, which had its first
tests in July 1934 [12].
The GUA also wanted radio AA gun-laying equipment and had been
sufficiently impressed with the experiments QRL had conducted in Jan-
uary 1934 with a 50 cm set that they wished the idea exploited, and NII-
9 undertook the task of providing a suitable prototype. The work was
started under Shembel. By early fall 1936 NII-9 had produced an exper-
imental continuous-wave twin-dish set, Storm, which operated on 18 cm
using early magnetrons from UFTI that gave about 6 W of continuous wave
power. The detection range was only 10 km, and the directional accuracy
only 4◦ , neither adequate. The range problem was a compound of mag-
netrons with too little power and frequency stability and a noisy receiver
that also picked up too much of the primary transmitter signal. Shembel
devised a solution for the direction problem analogous to lobe switching
and presaging mono-pulse radar. He used four dishes, one a transmitter
and the other three paired off in horizontal and vertical coordinates [13].
The first trials failed, and he was unable to bring the concept to fruition
before being separated from NII-9 in 1937 [14].
The skepticism that had met Oshchepkov’s Rapid soon hardened
into hostility to radio location generally in the form of a report in 1935
by the Red Army Chief of Signals, which asserted on the basis of studies
by his own NIIIS-KA that radio location was unrealistic and a waste of
87
time. M N Tukhachevskii, Chief of Ordnance, had been impressed with

the possibilities of the new technique, even if it was not satisfactory at the
moment, and decided in favor of retaining the infant radar program after
a rousing fight. It was, as those familiar with the ways of bureaucracies
will recognize, not the end but only the beginning. Life was becoming
complicated—and dangerous [15].
In 1937 Army Commander A I Sedyakin conducted a large air defense
exercise using conventional acoustical and optical methods that had a most
unsuccessful outcome. He became acquainted with the new radio-location
methods during discussions with General M M Lobanov of the GAU, who
convinced him of the need to pursue this kind of work. This happy state
of affairs came to a quick end in June 1937 when the purge swept military
and technical ranks. Both Tukhachevskii and Sedyakin were quickly elim-
inated, and the NIIIS-KA instigated investigations of NII-9 and SKB with
the resulting arrest of NII-9’s chief and the dismissal of Shembel. Bonch-
Bruevich, who had attracted Lenin’s favor for his early radio work and who
stood highest in electronic prestige, appealed directly to Central Commit-
tee Member Andrei A Zhdanov, who used Party influence to preserve the
activities of NII-9. SKB was cleaned out and Oshchepkov, along with other
radio engineers, went to the Gulag for ten years; he survived, thanks in no
small part to academician Joffe, who sent him food packages and letters
[16].
NIIIS-KA stepped into this to absorb NII-9 and SKB, and once the
hated radar project was theirs their attitude changed; they completed the
transformation of Rapid into an army prototype called RUS-1 (Rhubarb).
It had a truck-mounted transmitter and two truck-mounted receivers that
were normally placed about 40 km from the transmitter. These sets, of
which only 44 were manufactured in 1940 and 1941, were technically not
significantly advanced over the way Oshchepkov had left them and proved
to be of little value [17].
Following completion of the design of RUS-1, LFTI set about building
a pulsed air-warning radar, RUS-2 (Redoubt). It was bi- static equipment of
50 kW working on the 4 m band; transmitter and receiver were mounted on
separate trucks having Yagi antennas that tracked one another in direction,
although they had to be located about 1000 m apart, the obvious result of
not having solved the common-antenna problem. The experimental set
started in 1936 was not completed until late 1939, just in time for it to be
tried in the Russo-Finnish War where it was successful enough for ten sets
to be ordered on a crash basis [18]. RUS-1 failed the same test completely
[19]. RUS-2 provides an exception to the general rule for design of meter-
wave air-warning sets of the time in using Yagis rather than dipole arrays
backed by conducting screens, the directional antenna immediately and
instinctively adopted by others.
Victorious NIIIS-KA also continued the GUA gun-laying project, re-
placing NII-9 with UFTI. The principal deficiency of Storm had been the
88
Origins
use of continuous rather than pulsed waves. Bonch-Bruevich had held the
project to continuous waves, despite having used pulsed waves in early
ionosphere studies; he even terminated pulse work at NII-9 when he be-
came director in 1935 [20]. UFTI turned their efforts to a new, pulsed wave
64 cm design called Zenith. It combined every bad feature one could rea-
sonably imagine in one set. It reported the coordinates of range, azimuth
and elevation only every 17 s, making it useless for directing an AA gun,
and had a dead zone extending out to 6 km, the result of the receiver being
unable to recover from the transmitter pulse, although it could observe
aircraft to 25 km. A pulse length of 10 to 20 µs gave correspondingly
bad range accuracy [21]. Work continued and by the middle of 1940 the
range had been extended to 30 km, but the equipment had such a cata-
logue of ills that it was given up [22]. The technical reasons for failure
are not apparent. It would appear that the designers were unable to mas-
ter the techniques of microwave electronics and thereby profit from the
magnetron that N F Alekseev and D D Malairov had invented [23]9 .
The purges had at least made one agency responsible for radar, NIIIS-
KA, but in the process had removed good engineers from the laboratories
and the most supportive top military commanders. Soviet radar entered
World War II a low priority project with equipment inferior to all the major
powers. Yet it need not have been so. The early start with high-level
support, capable engineers and the cavity magnetron could easily have
made the Soviet Union the leader in radar.
The reader must consider these simplified attempts at recounting
relevant events in Stalin’s state with suspicion. The material available is
limited and was written before the collapse of communism opened secret
files—and by men not indifferent to what history would record.
2.5.3. France
Pierre David’s experiments with the ‘radio screens’ or ‘barrages électro-
magnétiques’ left him with ideas about how to make this kind of system
work for air warning, and he devised a method of using multiple sta-
tions to determine direction and speed. For a pair of stations the observed
Doppler shift, which can be measured accurately, depends on the aircraft
course, speed, altitude and the angles from the airplane to transmitter and
receiver—assuming a single target and straight, level flight. With multiple
pairs of stations David proposed to use these data to determine direction
and speed [24]. The system necessarily covered a large area, indeed this
was considered its important virtue, but a large area meant it could be
seriously confused if more than one plane or formation was present. It is
difficult to see what was expected of a device that gave such rudimentary
data acquired with such restrictive assumptions.
9 See Chapter 4.1 (p 157).
89
Despite these restrictions Société Française Radio-Électrique (SFR)

agreed that David’s meter-wave ‘barrage’ had more possibilities for air de-
fense than their own 16 cm equipment and submitted a proposal in March
1935 for using the longer wavelengths. It received a Navy appropriation
of F70 000 in April. David tested this Doppler system, which was easy to
manufacture, in the summer of 1936 on the Loire to the south of Paris. It
detected aircraft and estimated their speeds and indicated the directions of
some, results good enough that the Army decided in September to adopt
the scheme. The result brought F60 000 from the Army for twelve sta-
tions placed along two lines around Rheims and in the Argonne [25]. The
Navy constructed stations at Cape Martin and Cape Camarat and planned
stations at Cherbourg, Brest, Toulon and Bizerta. In a manner unique to
radio location projects world-wide the equipments were made by the same
firms for both Navy and Army: transmitters by Etablissements Kraemer,
receivers by Societe Anonyme des Francaise Radiotelegraphiques (SADIR).
Twenty mobile sets were used in the 1938 maneuvers.
David was well aware of the shortcomings of the Doppler method
and proposed building a pulse echo system in October 1938. There was
not much enthusiasm for this project until the British disclosed their radar
work to their prospective French allies through a visit by Watt and Wilkins
in April 1939. Orders were placed for pulse radars with a 6 m, 12 kW
transmitter produced by SADIR in October, which detected aircraft at 60
km [26].
Henri Gutton continued the work with 16 cm radiation. From his
experience at sea on the Normandie he concluded that ranging information
was essential and set about constructing pulsed equipment, which he did
at a land station. He replaced the Barkhausen tube with a split-anode mag-
netron and by early 1938 had constructed a station on the coast using the
paraboloids from the ship with 16 cm wavelength and 6 µs pulse duration.
By March 1939 he could range a 3000 ton vessel at 6 km to an accuracy of
about 250 m. The outbreak of war canceled plans to mount the improved
equipment aboard the Normandie. It served as a component of the harbor
defense of Brest but was too weak for detecting aircraft at any reasonable
range [27].
2.5.4. The Netherlands

Dutch radar can be said to have originated from ‘death ray’ concerns much
as had the British. Such concerns had raised questions in Parliament in
1926 and had helped secure the establishment of the Physics Laboratory of
the Netherlands Armed Forces at The Hague. Such fantasies were quickly
shunted aside to be replaced by research in communication and in various
forms of aircraft detection. A radio for artillery spotting that worked on the
1 m band gave rise to what must now be considered a traditional accidental
observation: a transmitter and receiver separated by a sand dune were
90
Origins
unable to communicate until an airplane passed overhead. These 1937

observations led almost immediately to the construction of a 70 cm pulsed
set that used Telefunken type RS297 output tubes driving a 16-dipole array
with common transmission and reception. Its power of 1 kW sufficed for
the ranges needed to aid searchlights.
It was a neat little set. The antenna moved in azimuth and elevation
and was mounted on a small, easily transportable mount that allowed
elevating it with a handle and turning it with a bicycle drive, so arranged
that the operator could see the indicator tube. The first prototypes were
ready for inspection when war broke over The Netherlands.
The Philips Physics Laboratory at Eindhoven began working in 1933
on applications for the split-anode magnetron that they had invented and
that had got Kühnhold at the German Navy’s NVA thinking seriously
about radar. The result was experiments with paraboloid reflectors that be-
gan to provide reflection phenomena of diverse sorts to stir the engineers’
thoughts. Trials of pulsed radar went off and on during subsequent years
and were taken up seriously again in 1939, too late to provide anything for
use in defending the lowlands.
As soon as the invasion was seen to be fatal the two Dutch radar men,
C H J A Staal from Philips and J L W C von Weiler from the Armed Forces
Lab, were sent with critical components to England where von Weiler
joined the staff of HM Signal School [28].
2.5.5. Italy
Italy had established in 1916 the Regio Instituto Elettrotecnico e della Com-
municazioni della Marina (Royal Institute for Electro-technics and Com-
munication, RIEC). Inspired by Marconi’s work with very short waves
Professor U Tiberio submitted a report in 1935 proposing that RIEC insti-
tute a program for radio location. An experimental 1.5 m continuous-wave
set, EC1, was tried in 1936 with improvements the following year. These
tests showed that one must use pulsed waves for anything practical, which
led to such a set, EC2, that proved to be unsatisfactory. Work was discon-
tinued until the shock of conducting war without radar drove the belated
lesson home [29].
91
PHOTOGRAPHS: AIR WARNING—EUROPE
Looking up the axis of a CH trans-

mitter tower. The platforms and
gantries are evident. The ladder,
beginning at the lower left, re-
quired a bit of mental conditioning
for some of the men and women who
worked these stations but was an
exhilaration for others. Historical
Radar Archives, RAF photograph.
Crown Copyright.
Interior of a CH receiver-operations room circa 1943. The receiver is located at the

left rear with one operator, those at the right evaluate the data before transmission to
the Filter Room. Historical Radar Archives, RAF photograph. Crown Copyright.
92
Origins
Fighter control with radar: Britain. It was realized early that radar sightings
of enemy aircraft were of no value, if the information was not evaluated within
minutes and transformed into orders for fighter squadrons. It was in this that
Britain was far ahead of any other country in September 1939, when a complete
functioning air-defense system was in operation. Shown here is the Operations
Room of Fighter Command. It also illustrates how Britain used women in air
defense. Historical Radar Archives, RAF photograph. Crown Copyright.
German Fahrstuhl Freya

(FuMG 43). This version of
the Freya was devised in the
field by Lieutenant Hermann
Diehl in 1940 to study exper-
imentally by lifting the dipole
array and thereby altering
the vertical lobe structure.
‘Fahrstuhl’ can be translated
as ‘elevator’ or ‘lift’. When
used near the ocean it pro-
vided height data. National
Archives photograph 111-SC
269066.
93
German Freya LZ (FuMG 401). This 2.4 m set, known originally as DeTe-II,
appeared in numerous modifications that varied as application required, this one
was referred to by British intelligence as ‘Pole Freya’. Its electronics formed the
basis for nearly all of Germany’s air-warning radar. The basic Freya shown here
had 12 vertical dipoles on the lowest (transmitter) antenna with an equal number
on the middle (receiver antenna). At the top is an IFF antenna. Freyas quickly in-
corporated lobe switching for directional accuracy. This model is transportable by
air, LZ standing for Lufttransport-zerlegbar. Reports are that the operator’s cabin
was designed for maximum discomfort. National Archives photograph 111-SC
269043.
94
Origins
The German meter-wave phased-array radar Wassermann M II (FuMG 402), one

of various models. The series was referred to by British intelligence as ‘Chimney’.
This set used the same 2.4 m electronics as the Freya but gained range and height-
finding capability through the use of a very large dipole array. This immense array
formed a thin, horizontal fan-shaped beam that could be moved up and down by
altering the phase shifts in the transmission lines feeding the dipoles. This allowed
a direct and rapid way of determining elevation to an accuracy of 0.75◦ for aircraft
lying between 3◦ and 8◦ ; lobe switching gave an azimuthal accuracy of 0.25◦ . At
100 km this localized the target in a box-shaped region 1200 m high, 300 m long
and 435 m wide for planes above 5200 m. Aircraft 8000 m high could be detected
at 210 km, and very high fliers at 300 km. It was capable of panoramic operation.
National Archives photograph 111-SC 269031.
95
The German meter-wave phased-array radar Mammut (FuMG 52), one of various
models. The series was referred to by British intelligence as ‘Hoarding’. Like the
Wassermann, this set used the same 2.4 m electronics as the Freya but gained
range and directional accuracy through the use of a very large dipole array. This
immense array formed a thin, vertical fan-shaped beam that could be moved from
side to side by altering the phase shifts in the transmission lines feeding the dipoles.
This allowed a direct and rapid way of determining direction to an accuracy of
0.5◦ over a ±50◦ sector. There was a dipole array on both sides that allowed front
and rear coverage, which gives the antenna a complicated appearance. National
Archives photograph 111-SC 269022.
96
CHAPTER 3
FIRST CLASHES
3.1. WAR IN EUROPE

By August 1939 Chain Home was providing continuous operational cov-
erage for portions of the coast that offered probability of air attack from
Germany. At the beginning of that month, so heavy with foreboding, what
must have seemed a ghost of the previous war approached Britain. At an
extreme range a very large blip appeared on the CH screens. It was an
airship, but no air raid alarms were sounded nor fighter squadrons scram-
bled. For whatever fears it might have raised a quarter century earlier, it
was no longer a bombing threat, but its mission was military and the radar
operators knew what it was.
It was LZ-130, named Graf Zeppelin, the sister ship of the Hindenburg
(LZ-129), which had ended commercial flying for the International Zep-
pelin Transport Company in a catastrophic fire at Lakehurst, New Jersey
in May 1937. As a consequence the older Graf Zeppelin (LZ-127) had been
removed from the incredible Frankfurt–Rio de Janeiro service that had fas-
cinated the world, and passenger service that had started in 1910 came to
an end. LZ-130 was completed while arguments raged about what to do
with it and the older Graf. In any case, test flights were made, but the one
approaching England that August was on charter for Colonel Wolfgang
Martini for electronic espionage. There had been other such trips, east and
west, and now it was Britain’s turn. Evidence of radar would definitely
interest the colonel.
Martini selected this mode of transport because of its range and
weight lifting capacity, for he loaded it with radio equipment and techni-
cians under the supervision of Dr Ernst Breuning. He also liked its leisurely
pace, which allowed time to examine any intriguing signals. A gondola
with antennas could be hung far below the main airship body in order to
reduce the effects of the huge aluminum frame on the radiation patterns
and had been used on three earlier flights but does not seem to have been
used on this one [1]. After following a course well clear of the Lowlands
it turned north at a point about 45 km east of Lowestoft, a consequence of
97
the weather in the Channel, and proceeded up the coast using clouds and
overcast to stay out of visual sight [2].
Flight Lieutenant Walter Pretty was on duty at Fighter Command
Operations Room and recognized the data being reported to him as coming
from an airship. The notion that the flight was espionage occurred quickly,
but the Air Force personnel already knew a fundamental law of electronic
warfare: when you see a suspected electronic snoop on your radar, he has
already seen you, so if you turn off your set you have given him proof of
your function.
All stations within range followed the ship, which was above cloud
cover, as it proceeded up the coast, and Pretty noted it straying over land
at Hull. This was certainly unintentional, but given the difficulties of aerial
navigation at the time and the extreme uncertainties placed upon the navi-
gator by the effect of unknown winds, it was not an excessive error. When
the ship made a transmission in the clear giving the position that CH had
shown seriously at fault, Pretty amused himself with the impossible idea
of sending a helpful message to LZ-130 informing them of their true posi-
tion. Course was then corrected, presumably through a solar fix, and the
trip continued off shore [3].
A newly established station near Dundee, equipped temporarily with
a Mobile Radio Unit, picked up the large-amplitude signal in its turn at a
range greatly in excess of 150 km and well to the south on 3 August. They
first thought it to result from a large group of planes but ruled that out
because such a target would generate a highly variable return and this one
was quite steady. Their report of the strange phenomenon yielded noth-
ing more revealing than the command to keep plotting. The airship then
broke through the overcast near Aberdeen and became visible to civilians.
Squadron Leader Findlay Crerar of Dyce Station of the Auxiliary Air Force
took off with an observer and returned with an excellent photograph of
the beautiful craft (the tail insignia excepted) [4].
The trip continued far enough north for examining any signals that
might have come from the Scapa Flow naval base but none of the Thames
estuary. The original route would have covered these stations and the
entire southern coast, but the weather—England’s old ally—intervened,
causing the airship to pause near Bawdsey as it changed course. Rowe
accepted this as proof that Bawdsey was marked for destruction when
war began [5].
There have been a dozen versions of the zeppelin espionage flight, fre-
quently conflicting and that have been eliminated by subsequent research,
but all agree that Martini found no sign of British radar. The suspected
reason is that no one in Germany conceived of radar on such long wave-
lengths, of such a low pulse repetition rate or of radiating over such a wide
front as Chain Home operated rather than forming a ‘searchlight’ beam.
There is reason to believe that Breuning’s receivers were in fact overloaded
with what he thought was the German ionosphere soundings that worked
98
First Clashes
on 50 Hz, easily confused by non-linear receiver response with the 25 Hz

repetition rate of CH [6]. Indeed, the original proposal for CH intended
that its radiations be perceived as ionospheric soundings [7]. Had CH
ceased operation on sighting the dirigible, the puzzling nature of the ra-
diation received might have been revealed. When questioned about the
incident after the war Martini was startled to learn that LZ-130 had been
tracked, as Bruening had reported no radar [8]. He did observe some test
transmissions of the new VHF (greater than 20 MHz) radio equipment
being installed in British fighters [9].
When Britain declared war on 3 September 1939 CH was providing
radar coverage from Aberdeen to Southampton, but the expected air attack
did not come, although a confusion in the filter center, a consequence of
excessive back-radiation from a CH station, led to a fight between Hurri-
canes and Spitfires three days later in what became known as the Battle of
Barking Creek [10].
The greatest radar activity, however, was a precipitous evacuation
of the Bawdsey Research Station. The move had been foreseen by Rowe
and Watt as war approached because of their certainty that the activities
had been sniffed out, were considered vital and would be blown to bits.
A committee had been formed to look for a new location and suggestions
discussed, although no site inspections undertaken, but their ideas became
irrelevant because earlier Watson Watt had in a casual fashion extracted an
agreement from the Vice Chancellor of the University of Dundee (Watt’s
alma mater) to give the laboratory emergency accommodations, a place
that had never even been considered by the committee. Rowe’s arrival at
Dundee met an incredulous Vice Chancellor who had forgotten or misin-
terpreted Watt’s request and had only two rooms for hundreds of people
and many truck loads of equipment. There had been no advance party, no
proper planning, just headlong flight. The effect of this move was to set
back a number of projects seriously. Unable to work, many sat around com-
pletely dispirited. Space was eventually found in the Teachers’ Training
College, which had no alternating current power [11].
As it proved, the fear was unjustified. Bawdsey had an operational
CH station and was not the recipient of an attack until mid-1940. An admin-
istratively correct name replaced Bawdsey Research Station: the Air Min-
istry Research Establishment (AMRE), making another break with a heroic
past. Located with them was the former Bawdsey ‘Army cell’, now the
Air Defence Research and Development Establishment (ADRDE), which
avoided the Dundee problems by moving to Christchurch.
There was one important exception to the Dundee move. Activities
at Bawdsey were split between research and responsibility for the engi-
neering and construction of CH. Edward Fennessy, representing the chief
of this latter group, H Dewhurst, protested with vigor that placing the
base of their activities at Dundee would make continued construction as
well as vital maintenance and support nearly impossible, owing to the
99
constricted transport and communications that connected the north of the

island with the south. Given two days to find an alternative, Fennessy se-
lected Leighton Buzzard, a market town 80 km north of London and near
the Air Force communications center having land lines to CH [12].
The airborne radar work being carried out at the Air Force base at
Martlesham Heath had increased by summer 1939 to include ten exper-
imental aircraft with more being outfitted and a substantial complement
of civilian and Air Force personnel. To the enormous anger of E G Bowen
they too had to decamp for a civilian airfield at Perth based on arrange-
ments also made by Watt. The station manager was already hard pressed
for space, for he was giving basic flight training to Air Force men and had
no room to spare. He and Bowen soon worked something out that allowed
both to get on with their duties until a promised new station could be found
for airborne radar. The new station was the Air Force field at St Athan on
the south coast of Wales about 20 km west of Cardiff, which was occupied
at the end of October. It had space but no heat, being hangars, often open
to the elements. Fabric partitions established working areas, and overcoats
and gloves provided what warmth was to be had [13]. They escaped the
effects of a 500 kg bomb through a defective detonator [14]. So much for
the wisdom of evacuation.
It is doubtful whether Rowe’s bureaucratic personality ever clicked
with Bowen’s, but after the move from Bawdsey it reached a low from
which it continued to sink with Bowen transferring his radar work to
America. After nearly 50 years Bowen’s anger spilled over the pages of
his memoirs. ‘. . . we were the principal Air Defence Establishment in the
country. We were supposed to know when the bombers were coming and
where they were heading and we had already devised a credible system
of sending fighters against them. If we were the first to cut and run, what
about the fate of the rest of the population [15]?’ Bowen’s remarks stand
up well given the course of the war but would never have been written had
German intelligence known Bawdsey’s function and had Göring had the
understanding to act upon it. The tarrying of LZ-130 should have given
anyone pause.
The work conditions at St Athan were appalling but were marked by
one singular piece of good fortune. The first polyethylene co-axial cables
became available during those dark months, simplifying the construction
of airborne radar tremendously. A second morale boost came from a visit
by King George VI and Queen Elizabeth, made all the better by a royal
understanding and appreciation of radar.
The AI and ASV engineers also learned about that time of the first
‘micropup’ design, a high-power, high-frequency triode formed by joining
copper electrodes with metal–glass junctions (Housekeeper seals) that al-
lowed external air cooling. The first, VT90, was tried in an experimental
airborne set that generated 5 kW peak power at 1.5 m and engineering
enthusiasm. By early 1941 the most widely used model, NT99, began pro-
100
First Clashes
duction initially at a rate of a few per week. These tubes used oxide-coated
cathodes, which allowed much higher emission compared with tungsten or
thoriated tungsten. Experimenters learned a very happy and unexpected
characteristic: oxide-coated cathodes gave substantially more current for
microsecond operation. This resulted in a push–pull circuit that generated
150 kW and found wide application both in the air and on the ground [16].
During that time Bowen and his staff also trained squadrons in the
use of AI (airborne interception), the equipment to play such an important
part in the Blitz—although not effectively until May of 1941—but these first
squadrons were not allowed to hone their skills with the difficult equip-
ment or devise tactics for a completely new weapon. One particularly
adept squadron was sent to help the Finns hold off the Russians and was
wiped out. The other was thrown into the RAF sink that Belgium and
France became in 1940 [17]. The introduction of AI marks I and II also suf-
fered from having been put into production before they were adequate for
the job. Training and maintenance were even worse with no technical liter-
ature or spare parts, a condition Bowen blamed on the inadequate support
given the airborne work by Rowe. The situation caused Hanbury Brown to
compose an addendum to Watson Watt’s rule of ‘give them the third best’
that ran ‘but don’t give them the fourth best because it encourages them to
throw the whole thing out’ [18]. The design quickly proceeded to AI mark
III, which eliminated some of the problems but still had an unacceptable
minimum range of 330 m.
By 1939 the engineers at Electrical and Musical Industries, Ltd, a name
long forgotten and replaced by the initials EMI, realized that something was
going on in the matter of radio location, possibly through Bowen’s use of
their tuned radio- frequency, amplitude-modulated television receivers,
and this brought Alan Dower Blumlein into radar design. Blumlein was
Britain’s foremost and certainly most versatile electronics engineer, the per-
son most responsible for London having the first high-definition broadcast
television in 1936. On its own initiative the company quickly offered the
government a 5 m gun-laying set with characteristics similar to GL mark
I. This disposed of any constraints on EMI people designing radar, and in
December 1939 the company was asked to help with AI [19].
Blumlein saw the difficulty in attaining an acceptable minimum
range, which was the problem of producing sufficiently narrow pulses.
The method initiated at Bawdsey was to form the pulse of radio frequency
signal and then amplify it, and it was this amplification that did not al-
low sufficiently rapid rise and fall of the pulse. Blumlein’s method was to
switch the high direct voltage that supplied the oscillator, and rapid switch-
ing of high voltage was a part of the television engineer’s stock of circuits
[20]. His transmitter–modulator reduced the minimum range to 130 m and
was incorporated into AI mark IV with first installation in August 1940.
Air-to-surface-vessel radar (ASV) progressed more satisfactorily. By
January 1940 this set, using the 1.5 m wavelength band as did AI, was
101
sufficiently well advanced for three squadrons of Coastal Command to be

outfitted with equipment manufactured by E K Cole Ltd (transmitters) and
Pye Radio Ltd (receivers). It could spot a surfaced submarine at a tactically
reasonable range, although not the simplest piece of equipment for new
operators to use. Brown arrived at Leuchars, a station south of Dundee
whose planes patrolled the North Sea, thinking he would be ‘welcomed
with open arms’, but such was not the case. Military people in times of
mobilization are sufficiently harassed that anyone who attempts to add
to their responsibilities is seldom welcomed with open arms. Brown suc-
ceeded in equipping the squadron’s Hudsons and instructing the crews
in their operation, but it was the squadron commander who caused the
equipment to be valued and used.
Aerial navigation in the habitually bad weather of the North Sea made
even returning to the base problematic, and planes were frequently lost.
Squadron Leader Sydney Lugg set up at the base a radar beacon, the mod-
ification of an IFF mark II, a device designed to return a strong signal to an
interrogating radar, originally intended to identify an aircraft as friendly1 .
The result was immediate, and Coastal Command made wide use of it
with crews giving studious attention to mastering their new equipment
[21]. The beacon was immediately named ‘mother’—home to mother! By
the end of the war radar beacons were important elements of aerial nav-
igation. ASV mark I also proved very helpful in locating convoys that
were to be guarded and in finding coastlines, but its effectiveness against
submarines was poor [22].
For whatever acceptance ASV mark I had acquired it was a poorly
engineered set and St Athan was a terrible place for engineering design,
so the work was transferred to the Royal Aircraft Establishment at Farn-
borough in February 1940, the beginning of the disintegration of Bowen’s
group [23], which finally ended when Bowen went to Swanage in June to
work on microwave AI2 .
Gerald Touch at Farnsworth, helped initially by Brown, completed
the design of ASV mark II, one of the outstanding radars of the war. The
antennas were so arranged that two lobes interrogated for searching, one
to the left, the other to the right. This allowed a swath 40 km wide to be
observed. When a target was located the pilot turned 90◦ and was guided
to it by forward-looking antennas operated with lobe switching to give
accurate direction. Pye and E K Cole received orders for 4000 ASV mark
IIs in the spring of 1940, but German bombs and the pressing need for AI
sets delayed production [24].
The doctrine of the Royal Air Force required that a war be initiated by
a ‘knock-out blow’, but all that was achieved was a gesture, one certainly
not noticed by the Poles, who were fighting a desperate, radarless war.
1 For details of IFF, see Chapter 3.4 (pp 129–35.

2 For details of the move to Swanage see Chapter 3.1 (pp 99–100.
102
First Clashes
Minutes after the declaration of war a reconnaissance flight disclosed war-

ships near Wilhelmshaven, which were then attacked by three squadrons,
only two of which found the target. Bombs struck but did not explode,
and half of the attacking planes failed to return [25].
Neither side wanted to initiate bombing civilian targets, so Bomber
Command restricted itself to night flights over Germany dropping propa-
ganda leaflets and continued daylight attacks on warships in the Helgoland
Bight. This was where radar drew first blood. Leutnant Hermann Diehl of
the Luftnachrichten-Versuchs-Regiment Köthen (Experimental Air Signal
Regiment Köthen) had supervised the erection in October of a station on
the island of Wangerooge and begun experimental work, specifically exam-
ining the vertical lobe structure of his radiation patterns, which he decided
was the cause of the peculiar disappearances of a target as it moved toward
or away from the station [26]. Everyone was learning about vertical lobe
structure, but only Watt and Wilkins had put it into an original design.
On 18 December he observed bombers at 114 km, but because of ear-
lier false alarms this vital information was not taken seriously until the
22 Wellingtons were also reported by a Navy Freya on Helgoland. The
scrambled fighters then opened communication with Diehl, who success-
fully guided their pursuit [27]. Of the attacking aircraft only ten returned,
of which three made forced landings [28]. This incident made a distinct
impression on certain elements of the Luftwaffe and initiated the develop-
ment of a system for vectoring fighters into attack positions.
Diehl became very active in this work. The 25-year-old lieutenant
had an abundant supply of confidence in himself and the fighter control
method he had discovered and found Martini at his side when obstruction-
ists blocked the way [29]. He invented, after devoting a lot of thought to
the consequences of reflections off the ocean surface, the Freya-Fahrstuhl
(Freya Lift) in which the antenna could be raised 20 m on a vertical track so
as to alter the lobe pattern and allow better estimates of the target height
[30]. He also worked with GEMA to install lobe switching for accurate
horizontal direction [31]. At upper levels of the German command radar
was given little thought—it was a defensive weapon and Germany was
on the offensive—but Diehl, Martini and their associates at GEMA were
doing a lot of thinking about early warning radar. Freya was to evolve into
Wassermann and Mammut, unquestionably the best early warning radars
of either side until mid-1944 and probably for the entire war. The approach
to estimating height did not, however, make use of vertical lobe structure.
Wassermann was given an enormous vertical array of dipoles that formed
a beam so thin that it could fix the altitude of the target more accurately
and with much less confusion than by noting the change in echo ampli-
tude as the target entered and left vertical lobes. In August 1944 Diehl was
an Oberleutnant and a very successful fighter control officer in the west,
a curious advance for one so technically gifted [32]. The American Army
never put into practice any of these improvements on SCR-270 and 271,
103
which were very similar to the first model Freya.

By the end of 1939 GEMA had produced eight Freya and four Seetakt
sets [33]. By March 1940 the Luftwaffe had established a chain of 11 Freya
stations for air warning along Germany’s western frontier, starting at Stylt
in the north and ending at Freiburg in the south, and had integrated them
into the visual and acoustical observer system. These Freya stations prac-
ticed their skills during the Phoney War by vectoring fighters onto enemy
aircraft over France, little realizing how valuable this knowledge would
be during the coming years. The Freyas were entrusted to the ground
observer units in the Luftnachrichtentrupp (Air Signal Troop), which had
gained non-radar experience in the Spanish Civil War [34].
If the repulse of the December RAF attack opened a few Luftwaffe
eyes to radar, it did nothing to alert Britain that her enemy possessed
such equipment. Indeed, it was not until 24 February 1941, when aerial
photographs of a Freya had been added to electronic evidence, that Lord
Cherwell, the Prime Minister’s scientific advisor, was convinced that the
Germans had radar [35]. R V Jones, a young man assigned to scientific
espionage, had become convinced much earlier, in no small part as the
result of a mysterious document mailed after dramatic preliminaries to
the British Embassy in Oslo. This Oslo Report from a ‘German friend’ de-
scribed a number of German technical secrets including radar, although
it incorrectly ascribed RAF losses in a North Sea attack in September to
radar, a time when none was installed. Other British intelligence experts
thought it a plant to deceive, but Jones continued to use it successfully as
a guide [36]. Jones, naturally curious as to the identity of the ‘German
friend’, finally met him years later through a chain of events stranger than
those that led to the receipt of the report itself [37]. He was Hans Ferdi-
nand Mayer, who had been a physicist for Siemens and Halske and who
was imprisoned later in the war for incautious talk.
If Bomber Command was not disposed to see the December disaster
as evidence of German radar, it was disposed to see it as evidence for the
futility of daylight attacks, for this had been simply the worst of the handful
of raids made over the North Sea, and the others had entailed losses too
heavy to be sustained. At the same time Bomber Command had been
covering the German countryside at night with leaflets and experiencing
very light losses. The lesson seemed clear: strategic bombing will have to
be carried out at night. Whether this extra ration of toilet paper had landed
on the cities for which it was intended was, however, unknown [38].
A sortie of the German battle cruisers Scharnhorst and Gneisenau [39]
for a few days in late November, both of which had just received Seetakt
sets, was the way Admiral Raeder wanted to deal with British shipping.
Konteradmiral Dönitz had a different vision, but it was a view not shared by
many at the Kriegsmarine because the technical capabilities of the new U-
boats were not significantly improved over those of 1918, and the U-boats
had been roundly defeated in that year. There was much more confidence
104
First Clashes
in surface raiders, which had the additional advantage of sinking or tak-

ing their prizes according to accepted international rules. Hitler did not
want to bring the United States into the conflict as unrestricted submarine
warfare had done 22 years earlier.
The sortie succeeded in sinking the armed merchant cruiser
Rawalpindi with accurate initial fire, which should have been ranged by
radar, but the new equipment had failed shortly before the engagement
[40]. The two raiders evaded their pursuers, who were not radar equipped
[41]. For the two battle cruisers to have sunk a much weaker adversary was
a bit of historical inversion. In 1914 two cruisers of Vizeadmiral von Spee’s
squadron carried the names of the two modern vessels and had been sunk
off the Falkland Islands by much heavier British naval artillery.
The pocket battleship Graf Spee with attendant supply ships had put
to sea in late August and was well into the South Atlantic when war was
declared. This class of warship had much greater cruising range than
ships planned for bruising fleet actions in the North Sea, and she was to
demonstrate this characteristic in sinking nine merchantmen before being
brought to action off the River Plate by one heavy and two light cruisers
on 13 December 1939. In that action her fire was quite accurate throughout
the engagement and the attacking cruisers were badly damaged without
inflicting serious harm on their stronger adversary. The consistent accu-
racy of the Graf Spee’s fire indicates that the crew had mastered the arts of
maintaining their radar and had adapted it to the harsh environment of
a warship, a reasonable assumption given that she was the first warship
equipped with operational radar.
It had not been easy. On 3 October the set had failed and had resisted
attempts by the operator to repair it—a task not made easier by his very
brief training and the absence of technical manuals and circuit diagrams—
they were too secret to be allowed aboard ship! The importance this new
device had assumed is made clear by the captain’s order to his chief radio
mate to drop everything until the radar worked again, which succeeded
after about a week [42].
The Graf Spee took the temporary refuge allowed in the neutral harbor
of Montevideo where her captain decided from various sources that a much
greater force was assembling outside the harbor. An erroneous sighting of
a presumed heavy unit confirmed his fears and caused him to scuttle his
ship [43].
The Admiralty sent L H Bainbridge-Bell, one of the early Bawdsey
team, to examine the wreck; he noted the radar and made a full report to
the Admiralty, where its significance seems to have been lost on everyone
but Jones [44]. There is evidence that it was suppressed for fear distribution
might compromise Britain’s great secret [45].
The Phoney War, known to the other side as the Sitzkrieg, ended
on 9 April 1940 with the invasion of Denmark and Norway. Britain had
naturally been displeased that iron ore was passing into Germany from
105
Sweden through the Norwegan port of Narvik thence by a sea route that
followed the coast within Norway’s own territory. Churchill, at the time
First Lord of the Admiralty, had proposed planting mines in these waters
as early as September 1939, thereby forcing the German vessels out of neu-
tral water where they could be attacked. This had been suggested in 1918
but rejected on the grounds of violating the sovereignty of a friendly, neu-
tral state. By spring the Cabinet had been brought around to Churchill’s
view. It was assumed that this action would provoke German interven-
tion, and to forestall this British troops prepared to invade first. Admiral
Raeder had alerted Hitler in October of the advantages of having Nor-
wegian naval bases, but the dictator had preferred to leave Norway neu-
tral. Diplomatic and military activities during the ensuing months made
it clear that Swedish iron was going to bring on trouble, so German war
plans proceeded accordingly. Hitler did react—and rapidly—to the British
moves [46].
The invasion of Denmark and Norway furnishes the historian with
naval action in which German skill and audacity together with British
slowness and bungling brought Norway under Nazi domination in the
face of overwhelming naval superiority. The campaign has much of inter-
est, including the first airborne assault, but little in the way of radar. It
did include the air lift of three Freyas to southern Norway: at Stavanger,
Mandel (near Kristiansand) and Bergen [47].
Radar’s part was restricted to the early morning of 9 April. The
Gneisenau and Scharnhorst were covering a group of destroyers transporting
the troops that occupied Narvik. At 0450 hours the radar of the Gneisenau
reported a target to the southwest at a range of 25 km, but the commander
waited for optical sighting, as the Seetakt equipment could not provide
direction accurate enough for blind fire; he was also a little skeptical, as
there had been trouble with the set earlier. Visual contact clarified the
matter but also gave the advantage of light to HMS Renown, the target,
which fired the first, well ranged salvo despite the want of radar [48]. The
German vessels replied accurately, but lost the use of their Seetakt sets to
the first shock of their own artillery [49]. The landing was safely ashore, so
the Germans broke off action. Neither side suffered serious damage. The
two battle cruisers encountered the carrier HMS Glorious returning from
Narvik and sank her and two escorting destroyers quickly with accurate
initial fire at 25 km. Presumably Seetakt was functioning.
The Royal Navy had radar installed on three anti-aircraft (AA) cruis-
ers that were used to protect the British troops put ashore at Narvik from
air attack, the landing party having no AA units. They found the cliffs of
the fjords made their radar useless. They also found this a poor way to use
these specialized warships, one of which was sunk [50].
When the British Expeditionary Force went to France at the outbreak
of war, air defense had two components: one to function in the forward ar-
eas in support of the Army, the other to defend ports and industrial centers
106
First Clashes
in cooperation with the French. For the latter a method of control similar
to that of the Home Defense was sought, and to this end a left-flank group
of MRUs began their emplacement in November just behind the Belgian
border. The stations at Boulogne, Calais, Lille, Arras, Cambrai, Aresne
and Sedan together with the station at Dover reported to a filter center at
Arras. As equipment became available a right-flank group was extended
to Verdun, Mount Haut, Bas le Duc and Troyes reporting to Rheims.
To function effectively these stations had to be linked to the filter
centers that in turn had to be connected to Headquarters Fighter Command
and the War Room in Paris. This proved more difficult than setting up radar
stations. The telephone lines were inadequate both in number and quality;
they were noisy, overhead lines that were easily put of action or sabotaged.
It was a sorry contrast to the excellent system used in England.
The French radio screens and the ground observers of both nations
were tied into this system, but there was neither IFF nor the substitute
used in England that located friendly fighters by a grid of direction-finding
stations that triangulated them as they emitted the radio carrier for 14 s
once every minute. The mobile units gave very poor height data; their
design had evolved out of CH, which had had the vertical lobes determined
essentially by the ocean, conditions not well approximated by the local
terrain. The result of this was a radar contribution to stopping the Blitzkrieg
that was trivial, very likely zero [51].
3.2. THE BATTLE OF BRITAIN AND THE BLITZ

The only virtue of a battle is that it sometimes renders a decision, and it
is the nature and importance of this decision that elevates or degrades the
bravery and sacrifices of the warriors. The Battle of Britain was not, like
Waterloo, immediately recognized as decisive. Indeed, it took the passage
of weeks for people to realize that there had been an important victory; it
took months for them to realize that it ranked with Trafalgar in preventing
invasion by a foreign despot. It was like Trafalgar in being the result of
one man forging and exercising the required weapon during the preceding
years and then using it with skill. But whereas the leader at Trafalgar was
immediately raised to the pinnacle of national glory, Dowding, the leader of
the Battle of Britain, was relieved of his command, sent out of the country to
do he knew not what and, after being assigned a task of finding economies
in the Air Ministry, transferred to the retired list. He did not attain the
highest rank in the Air Force although he did receive a barony three years
after the Battle, a noble rank three degrees below that conferred on the
victor of Waterloo.
It was a unique battle. It is the only set-piece battle fought entirely in
the air. It is the only air struggle that resulted in an unequivocal outcome.
It was fought with an absence of hate in a war dominated by hate. The
age of chivalry had no battle that surpassed it in such ideals, and it was
107
fought by the knights in far greater proportion to the commons than any
medieval clash. Measured by the total number of combatants killed, it was
a small battle; measured in the fraction of the combatants who survived it,
it was a slaughter. Far more people watched from the ground, chancing
the stray bullet or shell splinter, than fought in the air, making it the most
observed battle of all time. Yet the ardent public rarely knew whether to
cheer or cry as they watched a plane burning its way to the ground. It was
a bizarre battle in which a parachuted escape from death was sometimes
followed minutes later by being guest of honor at a pub.
But out of this chivalric battle grew, as inevitably as the seasons, a
form of warfare as savage as any fought by the Iroquois. The opposing fliers
conducted until the end of the war an honorable if brutal fight between one
another, but the war was not to be won in the air but on the ground, and
the war that began—at least in Europe—with careful consideration of the
civil population slowly became one of attempted mutual extermination.
If Dowding was the Horatio Nelson of his time, Churchill was William
Pitt, the Younger, and having to face a much more desperate situation than
Pitt, Churchill emerged the greater historical figure. Defeat in France left
Britain in an impossible position, but Churchill placed the English language
in the front rank and so matched his words to the race that the defeat of
France was met with a ‘wave of lunatic relief’. The people faced the future
exhilarated! ‘Now we are on our own. Now we know where we stand.’
Churchill’s words earned the right to be remembered with those given by
Shakespeare to Henry V. That they imparted strategic nonsense troubled
the Islanders not at all.
On 16 May, less than a week after the start of the German offensive
against France, Dowding had comprehended the situation and Britain’s
position in it perfectly and had set forth his analysis in a letter to the Air
Ministry that ranks with any of the battlefield writings of Caesar or Grant
for incisiveness. He had seen that France was almost certainly defeated and
that Britain would quickly have to face Germany alone. Through his sum-
mer air exercises he had learned what his fliers with their electronic eyes
could do; he had determined how many squadrons he would have to have
and had repeatedly given the number as 46. The French Armée de l’Air had
collapsed as rapidly as had the Polish air arm eight months earlier, leaving
the Royal Air Force to face the Luftwaffe in France essentially by itself. The
Hurricane squadrons were unable to protect the outclassed Allied bombers
in doing the army support the Air Force had continually rejected for them
and were too few to prevent the Stukas from doing the army support for
which they had been designed. Fighters were the key to everything—in
blatant disregard of Air Force doctrine—and their losses were eating up
Dowding’s reserves. From Paris, where he learned the shocking state of
French defense on the same day as Dowding’s letter, Churchill called for
six more squadrons of Hurricanes to bolster the French, causing an imme-
diate confrontation and rescission of the order. The debacle in France cost
108
First Clashes
the Air Force a quarter of its fighter strength and irreplaceable pilots.
Britain’s defiance after France’s defeat left Hitler at a loss. The Soviet
Union had just expanded its borders to the west by annexing the Baltic
states and part of Rumania, steps going beyond their assignment to the
Soviet sphere of influence indicated in the secret protocols of the 1939
German–Soviet treaty, steps undertaken without consulting Berlin. In her
path to independence Finland had received substantial German aid in 1918,
and the two nations had maintained friendly relations subsequently. The
Soviet attack on Finland took place just weeks after the pact had been
signed and had struck Berlin decidedly the wrong way. Furthermore, the
attack’s objective was obviously to strengthen Soviet defense against Ger-
many. When all this was added to more than a decade of fulmination
against the bolsheviks it left Hitler disposed to see a much greater enemy
in the east, one which he thought should be dealt with quickly. He was,
therefore, willing to come to terms with Britain and spent the first weeks
after the French defeat pursuing that goal. In the meantime the Wehrmacht
gave home leaves generously. There was a general feeling that the war was
over. There was certainly no feeling of urgency in the Luftwaffe.
Hitler’s confidence in speedy victory or his lack of comprehension
of the nature of technical warfare (probably both) is reflected in the oft
encountered statements that he ordered in early 1940 military research not
completed for 1940 or shortly thereafter to be discontinued [1]. It did not,
of course, apply to research done by private corporations and was ignored
by GEMA, Telefunken and Lorenz.
When it became obvious that Britain could not be subdued by a de-
based diplomacy, two alternatives were open: invasion or strangulation.
The latter seemed too slow, so the former was selected—without war plans
or prior staff work—for the summer of 1940. The Kriegsmarine had suf-
fered losses in Norway that made it incapable of escorting the invasion
flotilla across the channel in the face of the Royal Navy, making the only
solution the countering of the Navy by the Luftwaffe. To do this meant that
the Royal Air Force would first have to be defeated, which Göring assured
Hitler could be done. Thus the battle was set, almost with the formality of
the Middle Ages after the last futile exchange of couriers.
Technically, CH was ready. The east and south coasts of England and
Scotland had adequate warning with long waves, and the most vulnerable
sections had warning against low fliers with meter waves. In the plotting
rooms markers on large map boards indicated formations of aircraft, hostile
or friendly, with the time of the last fix shown by a color to match that on
the five-minute sectors of the clocks that were standard throughout the
Air Force and Observer Corps. The women and men on duty were well
trained and ready.
Radar gave valuable information about the horizontal position and
speed of the bombers but was much less reliable about height and num-
bers. Once they were over land, information had to come from the ground
109
observers with all the infirmities that weather and lighting could bestow,
although some CHL (Chain home, low cover with 1.5 m) stations could
follow planes inland depending on station location and siting.
Particularly difficult for the observers was distinguishing friend from
foe, something of obvious importance for the controllers. Inland radar
coverage with electronic identification was how it should have been done,
but in this British radar was not ready, so a substitute was used that had
been devised years before for directing fighters. It employed a network of
direction-finding stations that tracked radio signals emitted by the fighters.
This system of triangulation had allowed fighter control to be developed.
The last modification, called Pip Squeak, had the aircraft radio automati-
cally emit the transmitter carrier for 14 s out of every minute with the pilot
able to override this when he needed the set for communication.
Tactically, Chain Home was also ready. Stations had blast proof build-
ings, and the large towers were not easily knocked down. Radar mechanics
were well trained in repair and had necessary spares. Mobile Radio Units
(MRUs) were available to be erected, were a station knocked out. The
crews expected and were ready to face the worst.
Colonel Martini’s men were examining (unimpressed) the wreckage
of British MRU and GL sets left at Dunkerque before the smoke of battle
had cleared. It soon became apparent to them from listening to radio
transmissions that the RAF was sending fighters aloft too soon to have
been the result of visual observation, so they set up directional antennas on
Cape Gris Nez and ascertained what the strange towers across the channel
at Dover emitted [2]. But a detailed Luftwaffe intelligence briefing in July
makes no mention of radar [3]. It was but one of many failings of Luftwaffe
intelligence that summer.
British intelligence did much better. By July 1940 virtually every ra-
dio transmission of the German forces within range was overheard, and
the information gleaned from it put to immediate use. The extremely se-
cret Ultra was beginning its cryptanalysis of German signals, but since the
Luftwaffe used land lines for much of its communication not directed to
and from fliers, Ultra’s part was not comparable to radar in conducting the
battle. In those cases where details of an attack were so obtained the infor-
mation was of small value to the fighter controllers, who needed to know
where the enemy was to be expected with positional and temporal accu-
racy not obtainable from a tactical order. Nevertheless, the information
Ultra provided Dowding was helpful, and Göring’s Order of the Day on 8
August initiating the battle with the command to ‘wipe the British Air Force
from the sky’ was in Dowding’s hand within an hour of its transmittal [4].
The attacks that began the main Battle on 13 August 1940 were pre-
ceded by fighter sweeps and raids on Channel shipping intended to wear
down Fighter Command. These were not strongly countered in order to
conserve strength for the serious fighting that lay ahead and were wrongly
interpreted by the Luftwaffe as weakness. They helped finish for the RAF
110
First Clashes
the painful lesson started in France that their fighter tactics based on tight
formations were no match for the loose ones used by their opponents in
Me-109s. They brought home to the Luftwaffe the weaknesses of the long-
range fighter, Me-110, which had been intended to accompany and protect
the bombers and of which so much had been expected but which proved
to be a liability in this kind of war. The Stuka dive bomber, Ju-87, which
had terrorized ground forces east and west and which was needed to sink
the Royal Navy, fared even worse. Both were incapable of holding their
own in a fight with Spitfires or Hurricanes.
When battle was joined on 13 August it came with a bruising crash,
and five days later produced the greatest number of casualties for a sin-
gle day [5]. The attackers concentrated primarily on air defense targets in
southern England with civilian targets avoided. Dowding’s system was
well exercised by then and its radar component proved of inordinate value
by giving the defenders frequent opportunities for ambush. Attacks on
five CH stations on the 12th knocked out Ventnor on the Isle of Wight and
damaged the others [6], but repairs succeeded quickly either by returning
them to service, setting up an MRU or at least by generating transmitter
emissions, leading the Luftwaffe signal troops to interpret the attacks as
failures and Göring to suspend them. Poling, near Portsmouth, was heav-
ily damaged on the 18th, requiring the placement of an MRU for a period.
This was probably a target of opportunity because Göring had suspended
further attacks directed at radar.
Radar was not without fault. Estimates of the number of aircraft in an
attacking formation had to be discerned from the magnitude of the radar
pulse returned and the degree with which it fluctuated, and unlike the skills
that attempted to extract height from the ratio of the signals received from
the dipole pairs at various heights, these skills had not been calibrated on
realistically sized formations numbering hundreds. Furthermore, accurate
knowledge of the numbers in an attacking formation were seldom known
at the time, so operators were unable to gain much from their experience
and the estimates were frequently off by a factor of two or more.
Göring’s important subordinate commanders were both Field Mar-
shals: Albert Kesselring for Luftflotte 2 and Hugo Sperrle for Luftflotte 3
with a boundary running north-north-west through the middle of England
roughly allotting the east to Luftflotte 2, the west to 3. Kesselring was to
prove a skillful opponent for the Allies until the last day, and it is probably
fortunate for Britain that Göring and Hitler made strategic decisions dur-
ing that summer of 1940—and later too, for that matter—rather than he,
although it is really incorrect to speak of a German strategic plan at all that
summer, for plans were made on a day-to-day basis.
Dowding’s important subordinate commanders were two Air Vice
Marshals: Keith Rodney Park for 11 Group and Trafford Leigh-Mallory for
12 Group, with 11 Group covering southeast England and the Thames Es-
tuary and with 12 to protect the Midlands and serve as a tactical reserve for
111
11 Group. Dowding’s strategic plan was the result of five years’ thought.
Park believed in it with heart and soul and executed it with consummate
skill. Leigh-Mallory had other ideas about the conduct of the defense, and
as the tension of August and September weighed upon these men, these
ideas took on forms that skirted perilously close to insubordination. That
Leigh-Mallory was Park’s senior yet Park’s command was the first line of
defense did nothing to lessen a dispute that eventually went to cabinet
level and possibly to Dowding’s replacement.
The few thousand young men that fought one another were remark-
ably similar. Both sides showed courage, skill and daring in extreme mea-
sure; both took great losses stoically; both went to the limits of exhaustion.
But whereas one leader cared for his men with an affection that earned
them the name of ‘Dowding’s chicks’, and the Prime Minister gave them
perpetual honor as ‘the Few’, Göring heaped abuse on his men, blaming
them for the Luftwaffe’s defeats and on occasion implying it was the result
of cowardice.
Radar’s part in the battle appears rather less than heroic at this point
because it functioned just as expected. It had its greatest moment on 15
August. On that day Luftflotten 2 and 3 carried out a maximum effort just
before an attack from Norway over the North Sea against Scotland and
the Midlands by Luftflotte 5, initiated after it was thought the first two
had tied up Fighter Command. These attackers from the north-east had
expected to meet little resistance, expectations that show clear evidence
both of how the importance of radar had not yet been grasped and how
faulty was the knowledge of the RAF order of battle. The distance was too
great for the bombers to be protected by the Me-109s and great enough to
allow the controllers an almost leisurely placement of their fighters. The
result was serious losses for the attack and few for the defense. Luftflotte
5 found other occupation for its skills during the Battle of Britain.
As the days passed, the German attacks on RAF bases began to tell
and, had they been continued, might have produced the desired result, but
a change altered the tactics. On 25 August German bombers had, in viola-
tion of Hitler’s orders, dropped bombs at night on central London through
navigational error, hardly surprising given the difficulties of that art even
without the distraction of defensive fire. Churchill seized on this as an
intentional attack on civilian targets, which he almost certainly knew was
not the case, and ‘suggested’ night attacks on Berlin by Bomber Command,
which followed the next night [7]. They had little effect on Germany’s war-
making potential and were, because of necessarily poor navigation, nearly
all on civilian targets. The inability of the Luftwaffe to do anything to stop
the attacks added to the embarrassment of Hitler and Göring, who allowed
such personal feelings to override whatever judgement they were bring-
ing to the matter. Beginning 7 September London was to feel the full brunt
of air attack as retribution. This decision gave the RAF bases the respite
they sorely needed and simplified the tasks of the air controllers by giving
112
First Clashes
them a single, obvious target to defend. It also allowed Leigh-Mallory’s

ideas to be fitted perfectly into the scheme of things. The result was an
epic struggle subsequently commemorated as Battle of Britain Day.
On 15 September the Luftwaffe put into its attacks on London ev-
erything that its exhausted crews could fly, who had been bolstered by
intelligence reports that Fighter Command was at the end of its strength.
As they headed for their target Park’s squadrons were onto them with their
usual fury, apparently unaware of the reports of their demise, but the real
shock came on reaching the city, for Leigh-Mallory had had time enough
to form his squadrons into the big wings he favored (over attack by in-
dividual squadrons) and to send them straight onto the already harried
bombers. The Germans could scarcely believe the number of fighters in
the air. The Royal Air Force was clearly not defeated, the invasion was
postponed indefinitely, and the British people began to experience a less
exhilarating phase of aerial warfare, the Blitz.
A division of the air war that enveloped Britain from July 1940
through May 1941 into the Battle of Britain and the Blitz is, of course,
arbitrary but is useful here because radar functioned differently for the
two parts. In the former it executed a well performed drill; in the latter
a desperate makeshift. There were attacks at night before 15 September
and in daylight afterwards with the last great daylight fight on 30 Septem-
ber, but the nature of the conflict took on a completely new form with
large attacks the rule at night and those in daylight usually as low-flying
fighter–bomber sweeps for which CHL became almost the only warning
device.
If the Germans lagged behind the British in realizing the importance
of radio techniques for defense, they were well in advance in understand-
ing the need for improved navigational aids for attacks at night or through
clouds. In the summer of 1937 Oberstleutnant Friedrick Aschenbrenner,
commander of the Luftnachrichten-Versuchs-Regiment Köthen, described
to various notables radio navigational equipment being developed since
1934 by Dr Hans Plendl of Deutsche Versuchsanstalt für Luftfahrt (Ger-
man Aeronautical Research Establishment) at Rechlin in cooperation with
Kühnhold of NVA [8]. It was based on the Lorenz Company’s blind ap-
proach techniques that were well known in aviation circles: invented by
Dr E Kramar, introduced for Lufthansa in 1934, and sold worldwide. In it a
pilot received a pure audio-frequency tone if he were flying where the left
and right lobes defining the radio beam had equal amplitude. If he strayed
to either side, the pure tone became dominated by either a Morse code dot,
indicating he was on one side, or a dash indicating the other; added to-
gether and heard sequentially the two codes, which were transmitted on
the same radio frequency carrier, formed the reassuring continuous pure
tone.
This system used a single dipole transmitting an audio-modulated
carrier on the 10 m band. At each side of this dipole was located a reflector
113
dipole with relay-operated switches in the center. Closing the switch of

one of them threw the radiation pattern to the other side. A motor-driven
cam alternately switched between the two, longer for the dash reflector
than for the dot [9]. Depending on the sensitivity of the detector, be it
the navigator’s ear or some electronic instrument, there was a range of
position for which the equipment would indicate that the flier was on the
prescribed course. This range was 8 km at a distance of 100 km but when
used to guide the aircraft towards the transmitter for landing it became
narrower as the distance decreased, so the relatively large lobes of this
simple antenna could be tolerated.
Guiding a bomber over a distant target where positional accuracies of
200 to 400 m were demanded at distances of as many km required narrow
and accurately directed lobes. Plendl followed a straightforward way of
accomplishing this. Using dipole arrays he could direct beams of sufficient
accuracy that were detectable at great distances, depending on the height
of the aircraft. The pilot would fly out on a beam, constrained in motions
left and right until he intersected a second beam directed from a station
positioned to give good triangulation. The accurate topographic maps, of
which the European states were justly proud, allowed a completely blind
bomb release. In winter 1938 a bomber formation, KGr (Kampfgruppe)
100, was formed out of the Köthen Regiment to apply the technique with
Aschenbrenner, by then a Colonel, in command.
The range allowed by this method was restricted by the curvature
of the Earth’s surface because meter waves travel very nearly—but not
exactly—line of sight. Substantially greater range than line of sight had
been reported in the open engineering literature in 1933 [10], and Plendl
had either read this paper or in all probability made the discovery himself,
as he was a propagation specialist. Ignorance of these deviations caused
early confusion for the British in realizing what was happening.
The first equipment along these lines was called X-Leitstrahlbake
(direction beacon) and was a straightforward enhancement of the blind-
landing method. This equipment showed good results in 1937 blind-
bombing tests but had a clumsy antenna. At about that time Kühnhold de-
signed an ingenious antenna that was much smaller yet generated equally
narrow beams—unfortunately, it generated 14 of them, all of equal inten-
sity [11]. This system, called both X-Gerät and Wotan 1, was adopted, and
by 1939 14 stations were in operation. The confusion the air navigators
might encounter owing to the multiple beams, especially from the cross
beams that determined bomb release, was to be countered by training.
Plendl expanded the X-Leitstrahlbake into a heavy, rigidly mounted
dipole array 30 m high and 90 m wide, bent in the middle with a dihedral
angle of 165◦ to form two principal lobes. The shape of the array caused
it to be called Knickebein (bent leg). The whole arrangement rotated on
an accurately positionable turntable and was easier for the fliers to use
because each array had but one prominent lobe. It was transportable with
114
First Clashes
difficulty [12] and used the Lorenz navigation wavelengths, double the
5 m wavelength used in X-Gerät. This allowed the bombers to use Lorenz
navigation receivers that had been given increased sensitivity.
The difficulties of moving Knickebein and the confusions in using
X-Gerät prompted the design of Y-Gerät, also called Wotan 2, which elimi-
nated the need of cross beams, an obvious operational complication of the
other two. In Y-Gerät the array of dipoles and reflectors was so fashioned
that a single lobe, larger than those of X-Gerät but still using lobe switch-
ing, predominated over the side lobes. A more sensitive detector was used
to locate the equilibrium point between the switched beams and thereby
overcome the inaccuracy inherent in the larger beam. The radio-frequency
carrier was modulated with an audio-frequency signal, which was used to
determine the distance of the aircraft from the transmitter. This modulat-
ing signal was retransmitted by the aircraft on another radio frequency and
received at the transmitter where the shift in phase of the audio signal was
measured. This phase shift gave a measure of the distance of the plane
from the transmitter, which allowed a signal for the bomb release to be
sent. It was significantly easier to use and did not require the cooperation
of two greatly distant stations [13]. The smaller dipole array of both X-
and Y-Geräte meant, of course, less precision in directional accuracy than
Knickebein.
X-Gerät was used in the war against Poland in bombing a munitions
factory in Warsaw, but the results could not be evaluated: the target had
been bombed by other units. The system was employed in Norway and
in the invasion of France, often in the daytime. One is forced to assume
that X-Gerät was not found to be particularly successful, because Knicke-
bein stations were quickly built in Holland and France [14]. X-Gerät was
first used against Britain in a successful night attack of 13/14 August on
a machine-tool factory at Birmingham, although not recognized by the
defenders as a radio navigation flight [15].
R V Jones, a young protégé of Lindemann at Balliol College, Oxford,
was sufficiently concerned about Britain’s defense that he chose in 1936 to
become a Scientific Officer in the Air Ministry rather than continue studies
at the Carnegie Institution’s Mount Wilson Observatory in California. He
became Deputy Director of Intelligence Research and can be said, without
stretching things too far, to have invented scientific military intelligence.
His first interest lay in evaluating what the enemy had by way of radar,
and he began to use the Oslo Report with ever increasing confidence as a
guide.
One of the first major problems he had to face was the bits and pieces
that pointed to the German’s having radio-beam navigation equipment.
The first clues came from conversations overheard from prisoners in which
X-Gerät had been mentioned. As information began to come from Ultra at
about the time of the fall of France other indications came to be seen with
Knickebein making its way into the reports.
115
Lindemann alerted Churchill to these concerns, and Jones found him-

self in a cabinet meeting with Churchill on 21 June 1940, a moment of great
national peril, which tells us the degree of high level concern about the
beams and certainly demonstrates the seriousness with which the Prime
Minister approached technical matters [16]. (That Air Marshal Arthur Har-
ris, soon to lead Bomber Command in its night attacks on Germany, was
of the opinion that celestial navigation sufficed well enough at night and
that the concern with the German beams was misplaced tells us about the
state of mind in some Air Force circles [17].)
There now began a fascinating struggle, the first electronic warfare,
at least of a form that current practitioners of that art would recognize.
Jones began putting together every possible bit of evidence that he could
gather: prisoner interrogation, decrypts from Ultra, searches for the beams
in aircraft or perched with receivers on top of the huge CH towers, statistical
analyses of bombing attacks, even studies of Norse mythology for clues in
the names given the systems by the Germans and in the background ever
the Oslo Report.
Two defenses offered themselves: destroying the stations and inter-
fering with the beams to make navigation either impossible of deceptively
wrong. Both were tried. Dr Robert Cockburn of TRE was assigned re-
sponsibility for inventing electronic countermeasures, a project he was to
carry far beyond the ‘Battle of the Beams’, and 80 Wing of the Air Force
was organized to put these measures into practice.
This initiated a long period of espionage having some of the thrilling
elements that go into that kind of activity, which Jones has recounted in
his extremely readable memoirs. It was a confusing story as the two ad-
versaries thrust about for one another in the dark. To be effective the
countermeasures had to be carried out on exactly the right frequencies—
both radio and audio—that had to be determined in advance of the attack
and for which Ultra was to be of great value. It was not made easier by
the poor calibration of the British receivers used at the beginning. In eval-
uating the success of the jamming it was difficult to distinguish results
caused by interference from those caused by the deficiencies of the Ger-
man equipment, deficiencies unknown to the British at the time and with
which KGr 100 struggled. Furthermore, the most accurately delivered at-
tack, one in which the jammers knew they had failed, took place on a clear
moonlit night [18]. How successful was the interference? It must remain
an unanswered question.
With the onset of the Blitz, Fighter Command and Anti-aircraft Com-
mand, their headquarters next to one another at Bentley Priory, faced the
problem that they had expected for years. While CH was in construction
Tizard had pronounced that it would be a success and that this success
would bring on night attacks requiring airborne radar for night fighters.
The predicted moment had arrived but the AI radar for night fighters had
not; it was in the form of unsatisfactory prototypes.
116
First Clashes
The maximum range of AI was blocked by the enormous ground

return at a range equal to the height of the fighter. Hanbury Brown saw
that the immediate solution to this, until microwave AI became operational,
was a night version of the daytime system in which ground controllers
guided the fighters into attack positions on the basis of ground radar signals
[19], data CH could not supply. A completely new radar system had to be
constructed based on CHL, which had demonstrated its ability to follow
planes inland from the coast when station siting permitted. Out of this
design grew Ground Controlled Interception (GCI) with prototypes having
to be tested in combat during the winter of 1940–1941.
The 1.5 m band was the obvious one to use until power was available
at shorter wavelength. The first sets used separate antennas for transmitter
and receiver, each having its own man-powered pointing gear with the
transmitter following the receiver. Height information was as important
as ever, so the receiver had two dipole arrays positioned one over the other
and from which the operator used the relative amplitudes in determining
the angle of elevation. The determination of horizontal position used the
new Plan Position Indicator (PPI) in which the oscilloscope trace began at
the center of the tube face and extended outward in whatever direction
the antennas were pointing with targets shown by a brightening of the
range trace [20]. This was the practical beginning of the most widely used
radar display but differed from what has become universal in that initially
the antennas were being pointed by the operators, not rotated steadily
so as to produce a maplike picture. A common-antenna version of GCI
quickly replaced the separate antennas, and operations rooms began to
be organized around the functions of this new craft [21]. The early GCI
suffered from an inability to deal with more than one interception at a time,
although skilled operators could conduct two simultaneous interceptions.
Later stations had multiple displays with a controller for each.
An entirely new technique of radar observation and fighter control
had to be developed and personnel trained for it—while fighting off the at-
tacker. The lesson of proper siting had to be learned first, as the designers of
Freya had learned a couple of years earlier. Placing the set in terrain having
the shape of a shallow bowl produced strong nearby ground returns but
had clear traces beyond its rim, which had to provide uncluttered ranges
for tracking bomber and fighter. The collecting of multiple observations
and filtering them for the controllers proved quite incapable of functioning
as it had during the daylight; the night controller had to deal more rapidly
and with much more accurately defined positions. CH and CHL alerted
the GCI stations to the incoming raiders, but the fighter controller now sat
at his PPI scope rather than looking down at a map table as he followed
the two adversaries and guided them to their deadly encounter by moving
his antenna back and forth from bomber to fighter while talking the fighter
pilot along his way until the AI radar picked up the prey. The first GCI
interception was in the night of 18 October 1940 by a Beaufighter equipped
117
with AI mark IV [22]. This was the beginning of a successful defense, but
one which required months of frantic effort to bring to fruition, as GCI and
AI sets were built and deployed, and fliers, radar operators and controllers
were trained in this new endeavor. A dear price was now paid for the time
lost through the moves from Bawdsey to Dundee to Swanage with side
trips to Perth and St Athan for the airborne work3 .
By the time US Army Air Corps officers visited Britain in mid-1941
GCI and AI worked well enough for them to insist that they be copied
by the Signal Corps, GCI becoming SCR-527 and AI mark IV SCR-540.
This went more slowly than hoped, in no small part because the British
design was not yet fixed, and the first of the 527s were not produced until
the spring of 1943 and the 540s were never deployed because microwave
equipment was ready first [23].
If Fighter Command had troubles during the Blitz, AA Command
faced the attackers with equipment quite incapable of even giving the
bombers a proper fright, the consequence of attitudes about AA artillery
dating from the First World War. Although the 3.7 inch gun was bal-
listically up to the task, the director was of poor quality and dependent
on optical sighting. This was difficult enough during daylight raids, but
meant at night that the attacker had first to be picked up by searchlights,
and putting a searchlight beam onto an aircraft at high altitude was a tall
order just in itself. The gun-laying radar sets, GL marks I and II, were of
more use than the acoustical horns in helping the searchlight operators,
and they could provide accurate range, providing there was not a confus-
ing number of planes within their huge beams, but General Sir Frederick
Pile, the chief of AA Command, reported that not a single aircraft had been
brought down with the help of radar by October 1940 [24].
Pile sought technical help desperately and found it in P M S Blackett,
who took as his first task securing personnel capable of instructing troops
in the operation and maintenance of the new equipment so that the sets
worked at all. This he accomplished by recruiting school teachers and bi-
ologists as radar officers—the pool of physicists and engineers had already
been taken up. They were given a brush-up course in electricity, a crash
course in radar and distributed to the batteries where they were quickly
appreciated for their competence [25]. The next problem was to make the
best use of the radar data acquired other than helping the searchlight op-
erators; the directors were not capable of incorporating the output of the
GL sets in their calculations, and he finally settled on plotting the positions
by hand so as to predict the optimum location for a gunfire barrage [26].
It was a wretched method but superior to firing barrages on no data at all.
Blackett left AA Command in April 1941 for Coastal Command where his
newly invented ‘operational research’ was to play a much more decisive
part.
3 For details of the move to Swanage see Chapter 4.1 (p 153).
118
First Clashes
As the Blitz proceeded the gunners shot away the rifling of their gun
tubes so that projectiles followed unpredictable trajectories and fell on the
towns supposedly being protected because the time fuzes were not acti-
vated, owing to the lack of spin. When, in order to save usable guns, Pile
restricted fire on courses where it could not be effective he received a per-
sonal call from Churchill demanding fire [27]. Few besides Pile looked on
the guns as having any value other than sound effects for the public. What
Pile needed were Würzburgs or SCR-268s. AA Command eventually be-
came a deadly arm, but not in time to affect the Blitz. Even as evidence
accumulated that excellent gun-laying radar could be made, Lord Cher-
well, Churchill’s scientific advisor, was opposed to investing much effort
on such things [28].
At Pile’s strong, early recommendation AA Command also had
women from the Auxiliary Territorial Service (ATS). They operated search-
lights, radar sets and optical sighting equipment but masculine delicacy
did not allow them to load and fire the guns. They lived in field conditions
distinguished by cold, mud, rain and snow, often endured in the absence of
heat. Whereas women in the Air Force (WAAF) and Women’s Royal Naval
Service (WRNS) received steady recognition, the AA Command ATSs fared
little better than the gunners, who were frequently reviled in the pubs after
nights without gunfire when Pile and Dowding tried using the searchlights
to illuminate the bombers for the fighters—not with outstanding success.
The Blitz set the moral course of the remainder of the air war. The
attack on Coventry during the night of 14/15 November 1940 resulted not
only in significant industrial damage but in severe damage or destruction
to some 50 000 houses and to St Michael’s Cathedral; more than 400 were
killed and double that number wounded. British propaganda broadcast to
the world the barbarity, and as if to emphasize the point, Dr Goebbels’s own
propaganda trumpeted it as a harbinger of things to come, if Britain per-
sisted in defying German war aims. He promised that other cities would
be ‘coventrisiert’ [29] and in so doing removed future cause for complaint
when the situation was reversed.
May 1941 generally marks the end of the Blitz, for in that month the
GCI-AI system began to work effectively with 102 bombers shot down
that month, some during the day [30] and leaving the Luftwaffe to try
their luck on the Eastern Front. The skies over England were not to be free
of bombing planes until near the end of the war, but the Luftwaffe never
returned again in force.
Lord Dowding observed final victory as a bitter observer. The mode
of his relief from command agitates his countrymen to this day. It was by
no means a simple matter, being composed of ample portions of anxious
concern for the conduct of the war and old animosities within the Air Min-
istry. The Blitz was going hard against Fighter Command in November
when Dowding departed, and his enthusiastic support for the develop-
ment and deployment of GCI-AI counted for little, as their success came
119
later. Dowding was one of the most scientifically minded military com-
manders of the Second World War. His analysis and conduct of the air
defense of Britain has stood the test of hindsight. His failing was as a com-
manding officer. Two of his lieutenants ended the Battle of Britain almost
incapable of cooperating with one another. Dowding either ignored the
situation or was unaware of it, which is difficult to imagine, but then Dowd-
ing seldom visited his subordinates. A dispute of the kind between Park
and Leigh-Mallory is not unusual among military commanders. War is a
trade that nurtures strong ambition and firm opinion, characteristics that
must inevitably lead to clashes, but it is the responsibility of a commander
to deal with that sort of thing. Dowding did not.
And radar? Did this miracle weapon turn the tide? Of course radar
did not win the Battle of Britain. Brave men and women won it. But could
they have won it without radar? That question has been posed many
times and has, more than for any other engagement of the War, received
a negative answer. It is difficult to imagine a satisfactory outcome for
Britain had the Air Force not had radar. Had the Germans proceeded with
a carefully thought out strategy instead of day-to-day makeshifts, there is
good reason to believe that they might have won in spite of radar. With
such a narrow margin of victory one is compelled to hand a victory medal
to radar, but it is a medal that the Prime Minister would not have awarded.
He gave radar but a single condescending sentence in the chapter of his
memoirs devoted to the Battle of Britain [31], in remarkable contrast to the
eight pages about the German beams [32]. He was an active partisan in the
Lindemann–Tizard fight, and Tizard left high council before Dowding.
Churchill’s failure to realize or give credit to radar’s part in this epic
battle slips into insignificance when viewed against his own contribution.
If it is difficult to imagine a positive outcome without radar, it is even more
difficult to imagine it without Churchill’s leadership, for it was he who
inspired a whole people. It was expressed simply in the response to his
visit to a poor London neighborhood that had been very hard hit during
the Blitz: ‘Good old Winnie! We knew he’d come’.
On 18 June 1941 Lord Beaverbrook, British Minister of Aircraft Pro-
duction, released in a radio broadcast that a ‘radiolocator’ was an impor-
tant part of the country’s defense and called for volunteers with radio skills
[33]. The United States revealed something about its own radio location
work at the same time and called for electronic volunteers [34].
3.3. THE ATLANTIC, 1941

The complete defeat of the U-boat in 1918 had left Admiral Erich Raeder
with little interest in using that arm for the commerce raiding he planned
for the expected war. He and most of the top ranks of the Kriegsmarine
were convinced that surface vessels could better take on this task. Only
120
First Clashes
Dönitz held back from this viewpoint and obtained limited support for his
submarines.
Surface raiders were of two classes, armed merchant ships and war-
ships, both the modern embodiment of naval tactics as old as ships.
Whereas the warships went about their work in a direct and straightfor-
ward manner, the armed merchant ships brought to their task disguise
and guile. For the disguised raiders there were sailings in exotic waters,
captured ships sent back under prize crews, supply ships met on lonely
stretches of ocean, mysterious communications with agents on shore, re-
pairs at island bases of the quasi-ally Japan, selected prisoners with whom
to enliven evening conversation and the excitement of being chased by
and evading the Royal Navy. They conducted it—with one notorious
exception—according to the rules of warfare, indeed in an almost gentle-
manly manner. They emulated the exploits of the Emden and the Seeadler
of the First World War and, like them, caused comparatively little loss of
life. Seven such vessels sank a total of 39 ships and removed 590 000 tons of
shipping from British usage. One, the Atlantis, known to the Admiralty as
Raider C, put to sea on 31 March 1940 and cruised until sunk in November
of the following year [1]. The story of these ships is an interesting one, all
the better for being generally devoid of the viciousness which the war by
U-boat became, but we must leave it—these ships used no radar.
The warship raiders were different. They were big ships, 8 inch gun
cruisers being the smallest. More often than not they were unable to cap-
ture ships but sank them with gunfire. Their crews certainly felt the ex-
citement of being chased by and evading the Royal Navy, but despite their
size and power they had to be timid in the face of serious force, for even
slight damage could seldom be repaired at sea, thereby forcing the cruise
to be terminated, so speed was essential. The captains of these ships rec-
ognized the value of radar from the start, for it allowed them to locate their
prey at night or in fog and to distinguish, usually by size and speed, the
escort vessels or any other ships that meant trouble. Radar proved to be
an exceptionally important aid to navigation, and to this the 80 cm Seetakt
was well adapted.
After the loss of the pocket battleship Admiral Graf Spee six warship
raiders followed that calling. The Admiral Scheer was a pocket battleship
that mounted 11 inch guns, as did the two battle cruisers, Scharnhorst and
Gneisenau. The Admiral Hipper was an 8 inch gun cruiser and sister ship
to the Prinz Eugen, which would enter upon surface raiding briefly but
eventfully as consort of the battleship Bismarck in May 1941. All were
equipped with radar.
Both classes of raider were dependent on world-wide rendezvous
with supply ships and tankers to remain at sea for long periods, and theirs
is a good story too but one without radar except as used by the warship
needing supply.
121
3.3.1. Admiral Hipper, 1–27 December 1940

Of the warship cruises this one by the Hipper was the least successful,
primarily the result of mechanical problems that forced curtailment, but
her radar worked perfectly. After a false start in September 1940, termi-
nated by engine troubles, she departed German waters at the beginning
of December. Her entry into the North Atlantic was through the Denmark
Strait, which lies between Greenland and Iceland and was a favorite point
of exit for warship sorties because of the dependably foul weather, and
her escape went unnoticed. Her orders were to attack convoys only, not
independently routed shipping.
On 19 December she easily found with Seetakt the tanker spotted
for her refueling despite a very uncertain position after days without an
astronomical fix. Following this she used radar to avoid the forbidden
encounters with single ships, finally observing a convoy at 1700 hours on
the 24th at a range of 15 km. Soon eight to ten ships were observed, one
as distant as 22 km—a maximum for Seetakt. The commander’s plan was
to track the convoy until morning and then attack by gunfire, but during
the night he thought it worthwhile to try a torpedo attack based on radar
data. At midnight Hipper closed on the convoy and launched a spread
of torpedoes at a range of 6 km. No hits. Investigation disclosed the
torpedo officer had ignored the radar data and launched at a shadow he
was convinced was the convoy.
When Hipper prepared to attack at 0630 hours on Christmas Day,
radar reported intruders that proved to be a heavy cruiser with other escort
vessels following. Fire was exchanged from which Hipper received little
damage, but mechanical problems again set in and she had to make for
Brest, the first large German ship to make use of the new French base [2].
3.3.2. Scharnhorst and Gneisenau, 22 January–22 March 1941

These two vessels left Kiel under the command of Vizeadmiral Günther
Lütjens. The plan was for the Hipper to sail out of Brest whenever the two
battle cruisers had been detected and had attracted the hounds, leaving
much of the Atlantic unprotected, and it worked out pretty much according
to plan.
The battle-cruiser pair followed the Norwegian coast, left it where it
turns north-east, and attempted to pass to the south of Iceland. During the
early morning of the 28th the Gneisenau’s radar failed, and during 90 min-
utes of frantic repair work the Scharnhorst reported blips that gave every
indication of being cruisers. Course was reversed and the two raced their
way through an unnerving series of radar sightings and course changes
until free. From there it was far to the north for a rendezvous with a tanker,
which was 30 km from the expected position and very likely could not have
been found during such long polar nights without radar.
With fuel tanks full they tried the Denmark Strait, a risky business
122
First Clashes
in such weather with no lights on Iceland’s shore nor stars to steer by.
The Seetakt sets transformed it all into routine off-shore navigation by
providing the range and direction to prominent features of the island, and
the passage was almost routine with a couple of twists and turns to put
distance between them and other vessels. By noon of 2 February they were
clear.
On breaking into the open sea they separated during the day to en-
hance the chance of finding ships but rejoined again at night. On the 8th the
Scharnhorst was sighted by the battleship Ramillies, which was escorting a
convoy. Lütjens broke off well to the northwest for more oil, whence he
began seeking convoys on the Halifax route and was rewarded by sinking
five unescorted ships on the 22nd. Their wanderings continued until they
lay off the coast of Africa, again sighted, then off to the north where their
luck improved and they sank 16 unescorted ships. Lütjens was ordered
to put into Brest, luring many of his pursuers after him and clearing the
sea for the return of the Hipper and the Scheer to Germany by the north-
ern route. The Scharnhorst and Gneisenau began preparing for expanded
raiding with the new battleship Bismarck planned for the following month.
This cruise raised the mariners’ opinion of radar to an unprecedented
high. The troublesome early failures at sea were studied on this trip by
an engineer from the Nachrichtenmittel-Versuchs-Kommando (NVK, the
newly renamed NVA) aboard the Gneisenau who was able to isolate causes
of many past and present troubles. They also found that the nearly iden-
tical wavelengths used by the two ships gave unmistakable indications
on their respective sets when they approached one another after daytime
separation, an unexpected sort of IFF that was much appreciated when
extremely hostile ships were about [3].
3.3.3. Admiral Hipper, 2–14 February, 16–31 March 1941

The February sortie was made in accordance with the plan, once the two
battle cruisers were observed in northern waters. During the night of
11/12th the Hipper observed a large convoy by radar, which they used to
maneuver into the best position for attack at dawn. The convoy consisted
of 18 unescorted vessels, seven of which were sunk by gunfire. The Hipper
then returned to Brest to prepare for the return trip to Germany where her
numerous mechanical problems were to be removed.
Thanks to Seetakt this return trip in March was uneventful, but even
with radar Hipper required eleven hours to find the tanker needed for
mid-trip refueling. In what had now become routine a heavy cruiser was
successfully dodged in the Denmark Strait [4].
3.3.4. Admiral Scheer, 27 October 1940–1 April 1941

Of these warship cruises this one began first and ended last, ranged over
North and South Atlantic and into the Indian Ocean, and accounted for
123
16 ships. The outbound passage of the Denmark Strait was typical of

what was to follow weeks later by the other raiders: secure navigation and
avoidance of unwelcome encounters. On 5 November she encountered
the homeward-bound convoy HX84, escorted by the armed merchant ship
Jervis Bay under Capt E S F Fegen, who drove straight at the raider while
his charges made smoke and scattered. The Jervis Bay was, of course,
destroyed but only five of the convoy were sunk. Capt Fegen received the
Victoria Cross posthumously.
The Scheer’s long trip supplied her radar personnel with a generous
supply of problems. They repaired an uncommon number of failures,
generally caused by the shock of their own gunfire, and on leaving the
Indian Ocean found themselves with only two spares of a critical tube.
Radio provided for replacement at a supply-ship rendezvous on 11 March,
only to have an oscillator crystal fail. Again replacement was sent, this
time by submarine U-124. The weeks without radar left the command
experiencing the sensations of the recently blinded.
They passed the Denmark Strait safely homeward bound, easily
avoiding a heavy cruiser—and then had the set go out again, the conse-
quence of a short circuit caused by condensed water that was removed the
following morning. The remainder of the voyage to Kiel was uneventful
both navigationally and electronically [5].
3.3.5. Bismarck and Prinz Eugen, 18–27 May 1941

Late in the evening of 18 May 1941 two new units of Raeder’s surface
fleet left their Baltic port after completing careful shakedown cruises and
training. They were the 15 inch battleship Bismarck and her consort, the
8 inch heavy cruiser Prinz Eugen. The former was named for the chancellor
whose foreign policies had made friendship with England a vital element,
attained by avoiding naval and colonial rivalry. The latter was named for
the comrade-in-arms of Winston Churchill’s ancestor, John Churchill, 1st
Duke of Marlborough, Eugen’s partner in a long and successful struggle
by the Germans and the British against Louis XIV’s attempts to subjugate
Europe. Both ships were the ultimate of naval architecture. Both were
equipped with Seetakt; both had special radar rooms as a part of the origi-
nal design. Their assignment was commerce raiding under the command
of Admiral Lütjens. More was expected of them than of previous surface
actions, for with their armor, speed and radar they would be difficult to
stop, an opinion shared in Berlin and London.
Previous surface raiding had found the Royal Navy radar poor and
the raiders making good use of their own. Now the balance was to swing
in the other direction with the Royal Navy, Coastal Command and the Fleet
Air Arm radar equipped to some extent. The exact time and route of the
pair were not known to the Admiralty, but the break-out was no surprise,
and a significant reception was prepared.
124
First Clashes
On the 21st the two were sighted at Bergen by air reconnaissance.

That observation had been followed up on the following day in near-
impossible flying weather, and the harbor was found to be empty. The two
had sailed for the Denmark Strait, at the time about two-thirds blocked by
ice and with most of the remainder the recent depository of 6100 mines.
Retreating ice had left a safe passage that Seetakt easily traced, allowing
them to avoid the floating bergs as well as the pack ice even in the deep
fog that kept British non-radar air patrols from sighting them.
The cruiser Suffolk had received one of the first two 7.5 m type 79Zs
in May 1939, later upgraded to type 279, and now was also equipped with
the 50 cm type 284 radar for directing the fire of her main armament. She
waited at her station at the exit of the mine field. The cruiser Norfolk, which
patrolled 80 km to the west, had only the 1.5 m fixed-antenna type 286M,
the one that required swinging ship for direction [6].
At 1920 hours on the 23rd the Suffolk and the Bismarck sighted one
another visually as the latter broke briefly from a fog bank. The type 284
transmitter tubes were pushed to the limit to gain the needed power at
such short wavelength; this normally allowed operation for only a couple
of hours at a time, not too restrictive for gun-laying but hardly suitable for
searching. The vertical lobe structure of the 7.5 m set precluded using it
for surface search except at very close range. It was the intermittent use
required to conserve the 284 that caused the British sighting to be visual.
Suffolk scurried for fog before 15 inch shells could be sent her way, got off
a sighting report and began tracking the big ship with the 50 cm type 284.
The Bismarck, whose two 80 cm sets were not restricted in duration of
operation, had located the Suffolk both with radar and underwater sound
before the visual sighting. Fortunately for the cruiser the Seetakt did not
incorporate lobe switching and thus could not direct blind fire, having
a directional accuracy of only 5◦ [7]. Because of iced insulators on the
radio antenna the Suffolk’s first sighting report was received only by the
Norfolk and the Prinz Eugen, where it was promptly decoded [8]. The
Norfolk soon had a glimpse of the battleship and narrowly escaped a salvo
of heavy shells. The shock of gunfire had the effect of knocking out the
forward Seetakt [9] to Lütjens’s great displeasure, so Prinz Eugen had to
lead, as both her radars still functioned. The Suffolk managed to keep her
quarry in optical or radar sight and hold the Norfolk close with radio. The
Admiralty soon learned of the chase and dispatched the new battleship
Prince of Wales and the flagship Hood to intercept. They met the enemy
early in the morning of the 24th, despite the Suffolk having lost contact a
few hours before. Vice Admiral L E Holland, commanding the squadron,
ordered complete radio silence for his ships, including radar, until the
German ships were sighted, his fear being that with their greater speed the
Germans could escape if alerted.
The Hood was the finest of that most unfortunate kind of warship, the
battle cruiser. As large as a battleship with guns as heavy, it sacrificed armor
125
to gain speed. It was a stylish idea in naval circles before the demonstration
that a 5 knot difference in speed did not matter to well aimed projectiles that
easily penetrated thin steel. Three ships of this type had disappeared in the
Battle of Jutland in catastrophic explosions. (The German battle cruisers
Scharnhorst and Gneisenau traded gun power for speed rather than armor
plate, having only 11 inch artillery, small for battleship-class vessels.)
The Hood had a type 279M air warning radar and a type 284 gun-
laying set, but radar did not protect her from the first salvos of the two
German ships, and she blew up in a mighty explosion, the presumed con-
sequence of a heavy shell penetrating her thin deck armor and detonating
the magazines. The German optical fire control was up to the same high
standards it had so startlingly demonstrated in action in the North Sea in
the previous war and the Bismarck’s defective radar was not missed.
The Prince of Wales had a 3.5 m type 281 air warning set and nine fire
control radars, but the ship was so new that civilian workmen were still
on board, as bad luck would have it, because of problems with the main
armament. She was also so new that the gunnery officers had not incor-
porated radar into their procedures. The radar officer reported accurate
ranges throughout the brief fight, but they were not used in calculating
gun orders, and it was only the sixth salvo that had the correct range [10].
So it came to pass that in the first encounter of big-gun ships equipped
with radar the use of the new technique is enveloped in fog: the forward
German set on which the First Gunnery Officer would have relied was
dead, and the British set was ignored [11]. What the Hood did will remain
unknown, but her first salvo was not on target.
The Prince of Wales developed serious malfunctions in her artillery
and sustained enough damage to cause her to withdraw behind a smoke
screen. The Bismarck had unintentionally begun replacing fuel oil with
seawater though retaining a speed of 28 knots. Why Lütjens did not pursue
and very likely sink the Prince of Wales is a puzzle few have understood.
At this point the Bismarck was sufficiently damaged that commerce raiding
without repair was not possible, and sinking the two most powerful ships
of the Royal Navy would have certainly justified the attempt. Lütjens
detached the Prinz Eugen to proceed independently to the south and began
a straight run for the safety in the Bay of Biscay.
Now the Bismarck was pursued by an ever growing assortment of
very heavy ships with the Suffolk again doggedly tracking, but on the 25th
she lost radar contact, the almost certain consequence of the intermittent
use required of the 284. Lütjens was so impressed with the ability of the
Suffolk to follow that he broke radio silence to inform his chief of the radar
capability of which he had not been informed and the range capability
of which he greatly overestimated. The overestimation probably resulted
from navigational errors of one or both ships, as Lütjens compared his
calculated position with the continual flow of messages that the Suffolk
was transmitting [12]. Lütjens’s message allowed British radio-direction
126
First Clashes
finders to get a rough idea of his position, but at the time he incorrectly
thought he was being held fast by British radar.
This incident is linked to reports [13] that the Bismarck had a passive
radar receiver and had monitored the tracking. If so, it must have been
an experimental set of which there is no other record [14], and the passive
receivers that first came into use more than a year later would not have
responded to 50 cm waves. It is plausible that the radar operators, presum-
ably briefed on British use of long waves, picked up on communications
receivers some of the abundant 7.5 m transmissions, which they would
have recognized as radar. Given the circumstances it is unlikely that they
would have realized that this equipment was incapable of observing them
at the ranges involved.
A sighting through the swirling clouds over a rough sea by a Catalina
flying boat equipped with ASV mark II [15] established the Bismarck’s po-
sition accurately enough for the cruiser Sheffield to be ordered to pick her
up with the type 79Y radar, if possible. At this point aircraft from the
carriers Victorious and Ark Royal were decisive. Both were equipped with
the famous Swordfish biplanes, slow but very tough and possessed of a
remarkably long range and a deceiving agility, if not encumbered with tor-
pedo or bomb. They probably sank more tonnage than any other torpedo
bomber during the war and were valuable participants until the very end.
We shall return to them when describing action in the Mediterranean, the
high point of the Swordfish’s service.
One of the Swordfish from each carrier was equipped with ASV mark
II, and green fliers from the Victorious, which had not had time to work
up her crews, even to allow them to practice take-off and landing from
the deck, found the target and got an ineffective hit on the armor belt.
The first attack by 14 planes from the much more experienced Ark Royal
went after the shadowing Sheffield instead, of whose presence they had
not been informed, but their torpedoes missed. Their next attack of 15
planes found the Bismarck with radar in conditions of ‘low rain cloud,
strong wind, stormy seas, fading daylight and intense and accurate enemy
gunfire’. One torpedo struck the armor belt, another jammed the steering
gear, and with that the great ship was doomed. The radar that found the
target also found the home ship, and all 15 aircraft returned, to be sure
with wounded crewmen, perforated fabric and three crash landings [16].
With the stricken ship no longer able to reach the protective cover of
land-based bombers, dawn came as a death sentence to be executed by the
battleships Rodney—ordered to the spot with a deck cargo for installation
in America and 300 passengers—and the King George V. Accurate fire, soon
delivered at close range, destroyed the ship that refused to surrender. There
are several accounts of this famous battle. The reader is advised to read
the one by the Bismarck’s Adjutant and Fourth Gunnery Officer [17] and
that of the under-water explorer who found the wreck in 1989 [18].
The sinking of the Bismarck put an end to German surface raiding
127
with large ships. Even without that dramatic climax it was becoming
increasingly obvious that it simply did not pay. The Scharnhorst cost as
much to build as 100 submarines, required a huge crew and elaborate
supply, and was not immune to sinking. There was an attempt by the
pocket battleship Lützow to renew raiding, but her sortie of 10 June 1941
was countered by a torpedo-plane attack that sent her back to Kiel for
months of repair. When Hitler attacked the Soviet Union, he required
many of his surface units for the Baltic. The disguised raiders continued
until the Royal Navy removed them, their tankers and supply ships from
the seas. Commerce raiding would be left to the U-boats of Konteradmiral
Karl Dönitz, and all traces of romance disappeared.
The use by the Kriegsmarine in 1939–1941 of Seetakt was a most im-
pressive consequence of the power of pure radar, the result of a naked
radar set mounted on a ship for which no thought had been given as to
what its exact tactical function was to be. The naval personnel received
little training, but the set was simply ideal for a commerce raider. It was the
kind of thing that every alert officer recognized when he first encountered
it—the torpedo officer of the Hipper a conspicuous exception. Application
came immediately and instinctively. There is no evidence of captains con-
sidering radar as just ‘an interesting device’; they regarded its malfunction
to be a major problem for which they demanded the delivery of spare parts
by special ship and submarine.
It had not been planned that way by Raeder. On first seeing a radar
demonstration he was impressed enough not to interfere but cautioned
Kühnhold that his primary research mission was under-water sound. It
was the line officers who recognized the new weapon for its value, and
their use of it in the few months of surface action was beyond criticism.
Except for a technically dull-witted command they could have had blind-
fire directed gunnery in 1938. German naval radar had a brilliant beginning
that led nowhere.
Typical of the want of understanding at the top was the vacancy of the
position of Chef der Abteilung Entwicklung der Nachrichtenmittel (Chief
for Development of Signals) from November 1939 until April 1943 [19]!
Moreover it was not until mid-1941 that the Marine-Nachrichtendienst
(Navy Signals Service) was formed and with it a naval career specialty for
radar, Seetaktischer Funkmessdienst (Tactical Radar Service) [20]. Progress
remained slow, and Dönitz was to find his U-boats completely outclassed
in either defensive or offensive radar techniques.
A comparison between the two navies offers instruction about their
respective use of radar 21 months into the war. The Germans had mounted
a prototype Seetakt in 1938, modified it in small ways, and haltingly made it
reliable aboard a warship, the obvious responses of competent engineers;
it was their only shipborne radar for months yet to come. Despite the
Navy’s introduction of the equally good air-warning Freya, it was never
taken to sea except on vessels in the North Sea as part of the country’s air
128
First Clashes
warning system, nor was the excellent gun-laying Würzburg used aboard
ship to improve AA fire, although GEMA soon adapted the Seetakt for
dual purpose. The British by contrast had by May 1941 almost a dozen
different kinds of shipborne radar installed [21], but it was not until the
10 cm type 271 appeared, with sea trials in March and April 1941, that they
had a surface-search set competitive with Seetakt [22]. In their hunt for
the Bismarck only one shipborne radar set of the entire pack of hounds was
effective, and its inability to maintain continuous search caused it to lose
the target vessel at a critical moment, saved by the splendid ASV mark II.
It remains a puzzle that a naval command that gave high priority to radar
placed so little importance on surface search equipment. The answer to
the puzzle probably lies in Britain’s approach to radar from the long-wave
side.
3.4. FRIEND, FOE OR HOME?

As soon as radar sets began to move from plans to reality designers began
to consider the matter of determining whether the blip observed on the
cathode-ray screen came from a hostile or friendly ship or plane. It was
a problem as old as warfare itself. Visual identification has never been
certain, and in the confusion of war a significant fraction of losses has
always come from ‘friendly fire’. In unconstrained naval battles fought at
long ranges in poor visibility it had become a pressing problem before the
complication or help of radar. For the relevant case of war in the air there
were many cases of attacks on the wrong aircraft through failure of visual
recognition, and the new kind of vision offered by radar was to bring the
same old problem back in a new form.
Watt discussed it in his original memorandum to the Tizard Commit-
tee [1]. The British name for it, IFF for identification friend or foe, never
had any serious competition in English. The first idea was the same ev-
erywhere: equip radar targets from your own side with some structure
or circuit that would resonate at the radar wavelength and that could be
modulated and re-radiated. This would give an enhanced return signal,
the amplitude of which would vary in a manner recognizable to the oper-
ator at the screen.
The people at Bawdsey elected a tuned dipole with a switch in the
middle that allowed keying a signal recognizable at the radar set. Arnold
Wilkins produced equipment of this kind for testing in the Home Defence
Exercise of 1938. The return signal proved difficult to distinguish from the
variation of intensity produced when the target was a flight of planes, so
he decided on an active rather than a passive unit [2]. Here passive means
that only a reflected wave is returned; active means energy is added to the
response through a vacuum tube circuit. The Naval Research Laboratory
tried similar tuned dipoles in testing XAF during the Caribbean exercise
of 1939 and also decided to use an active system [3]. The opening of hos-
129
tilities found neither side with operating IFF. Britain rightly expected to be
under air attack soon and felt dire need for the new device. This led to the
introduction of IFF mark I in time for the Battle of Britain.
Owing to the urgency of the times, the designers elected to have
mark I respond only to CH, a band from 10 to 15 m, but not to MRU, CHL,
GL or Navy wavelengths because of the complication of incorporating
these bands. It consisted of an airborne receiver whose tuning was swept
mechanically through the range in a few seconds. The receiver was a super-
regenerative circuit that functioned as the responding transmitter as well;
it was a receiver from the earliest days of vacuum-tube radio, but which
was seldom encountered in 1940.
This kind of circuit came out of the first experiments by De Forest and
Armstrong. A triode used for the amplification of radio waves had a tuned
circuit in its output, comprised of a coil in parallel with a capacitor. Both
investigators found that if one connected some fraction of the output to
the input, a process called feedback, the circuit would go into oscillation.
This was the first electronic oscillator, which transformed radio completely
and soon eliminated spark and arc equipment. They also found that with
feedback just below the threshold for oscillation the circuit would amplify
with much higher gain than without it, and the regenerative receiver was
born. One of the skills required of people listening to early broadcasts
was that of adjusting this critical element. There was an intermediate
region with even higher sensitivity in which the detector, once tickled by
an input signal of the frequency to which it was tuned, had its output grow
into sustained oscillation. This was clearly of no use for the reception of
voice, but could be adapted to Morse code if the oscillations could be
appropriately squelched.
IFF mark I functioned by being triggered by the CH signal when its
tuning matched the frequency of the interrogating radar, which meant that
for a given CH station the IFF would respond not for every radar pulse
but only those for which its tuning corresponded. On being triggered the
oscillation amplitude would grow until the detector became a small trans-
mitter. The radar operator would recognize a friendly blip by its increasing
in size in a repetitive manner determined by the speed at which the IFF
transponder tuning was being swept. This had both simplicity and econ-
omy. Unfortunately, it required in-flight adjustment of the feedback by a
flier whose mind was fixed on other matters and was not appreciated by
the pilots and radio operators who had to use it. An improper adjustment
resulted in no response or in radiating random pulses to the great annoy-
ance of the ground stations [4]. Installation of the first 100 sets, made by
Ferranti, began in November 1939 with 1000 eventually delivered. Oper-
ational experience was poor, as only about half of the interrogations with
properly adjusted equipment yielded an identifying response [5].
Inadequacies of mark I were apparent before it went into service,
development of mark II was under way in the spring of 1939 and even that
130
First Clashes
design was seen to be only a provisional solution. Mark II responded to

the 10–15 m band of CH, the 7 m band of the Mobile Radio Unit and the
Navy, the 3–5 m band of the Army sets and the 1.5 m band of CHL and the
Navy. It functioned the same way as mark I but had a ‘complicated system
of cams and cogs and Geneva mechanisms’ [6]. Unfortunately, it retained
the sensitive in-flight adjustments. Improved maintenance, supervision
and training brought operational success up from the 50% of mark I but
never to satisfaction.
After the disclosure of radar information to the United States by the
Tizard Mission in September 19404 mark II was also adopted by American
forces as SCR-535 and manufactured by Philco. By July 1942 the US Army
Air Forces had 18 000 on order [7]. IFF equipment soon dominated the
costs of radar.
When it became apparent that microwaves would have to be added to
the spectrum for which IFF must respond, the basic design of mark II was
recognized as unusable because of the disparity of electronic techniques
that separate 15 m waves from 10 cm. Furthermore, mark II would respond
just as well to German radar, if they placed equipment using Allied bands
in service. F C Williams began to concern himself early with this problem,
casting it in terms of a universal interrogating system, one independent of
the primary radar frequency. IFF sets were not to reply directly to the radar
signals but to a special interrogating transmitter. The interrogation took
place in the range 1.6 to 1.9 m, which the transponder swept, with reply on
the interrogation wavelength. It used the super-regenerative method but
had automatic gain stablization to maintain the receiver at optimum sen-
sitivity, eliminating in-flight adjustment. The reply was a pulse of coded
length with an extremely long pulse signifying distress. The basic de-
sign was ready in early 1941 with manufacture begun by Ferranti Limited.
American production was desired, and Hazeltine Corporation received
a contract to design the set for American production with an acceptable
prototype completed by mid-1942 [8].
The mark III became the Allied standard for the war but not with-
out a small serving of acrimony, quickly settled by America’s precipitate
entrance into the war. The Naval Research Laboratory and the General
Electric Company had already designed an IFF system not dependent on
interrogation by the radar pulse, and the Signal Corps Aircraft Radio Labo-
ratory had adopted it, the airborne component being SCR-515. The Ameri-
cans did not like mark III because they saw the sweeping of a 30 MHz band
an open invitation for enemy mischief. Their own system had passed pre-
liminary tests and was scheduled to begin final trials on USS Hornet on 8
December 1941. The British did not like the NRL-GE design because the
interrogation used a single frequency that overlapped the Würzburg band
[9] and that would be easier to jam [10]. In retrospect these seem a curious
4 See Chapter 4.2 (pp 159–66).
131
objections. The Würzburg would generally be encountered with IFF sets

turned off, and mark III responded to a range not far removed from the
Freya frequencies; jamming IFF never became an important part of coun-
termeasures. The NRL-GE system also used a single but different response
frequency.
That Britain would use mark III come what may presented the awk-
ward situation that would have required American planes operating in
Europe to have two transponders, impossible for fighters. This quickly
forced the decision in favor of mark III as the Allied standard. The Amer-
ican system, of which several hundred had been manufactured, was sub-
sequently referred to as mark IV and held in reserve in the event mark III
was seriously compromised [11]. It was compromised as early as 1943 and
used as range-enhancing secondary radar regularly by Germany, but these
radar struggles come later in the story. Mark III was retained.
As use increased IFF problems grew apace. When large formations
began taking to the air the interrogating equipment began to show masses
of signals, quickly earning the name of ‘IFF clutter’ and easily obscuring
the presence of enemy aircraft. Just as bad, the transponder of one air-
craft could trigger the response of another, the consequence of using the
same frequency for interrogation and response; the solution was to lower
transponder gains [12]. At sea the vertical lobes of the interrogating radia-
tion were much larger than those of the shorter-wavelength primary radar,
which when coupled with low transponder gain, led to vessels being sub-
ject to radar-directed gunfire from their own side [13].
Few things illustrate the organizational problems that plagued Ger-
man radar better than IFF. German electronics work was invariably marked
with high degrees of competence at engineering levels but confusion and
want of understanding at the upper levels of command with obtuseness
increasing with altitude. It is obvious that successful IFF requires agree-
ment by all branches of one’s own and allied forces. On the American
side a Joint Radio Committee had agreed on a national IFF standard early
enough for a system to be starting production in late 1941. Even before
America’s formal entrance into the war discussions had begun with Britain
over this matter, leading to a decision shortly after the Pearl Harbor attack.
The German contrast is stark.
The German work started soon enough. In November 1938 a meeting
took place about recognition. It was only then that the Naval Command
learned of the separate radar work for the Luftwaffe by Telefunken and
Lorenz. At that time GEMA offered an IFF, eventually to be called Erstling
(first born), for Freya. The need to have such equipment respond to 80, 62
and 53 cm as well as Freya’s 2.4 m was obvious, but no agreement came
from the discussions. Not only was there no agreement then, there was still
none by the end of the war. In the fall of 1939 after a GEMA demonstration
the Chief of Luftwaffe Signals ordered 2000 to 3000 units of Erstling, still
with response only to 2.4 m [14]. At about the same time Dr Hans Plendl
132
First Clashes
had designed an IFF for use with the Würzburg called Stichling (a prickly
fish). It gave identification only in direction, not range, at the time not
thought important for Flak. Martini’s demands that the functions of these
two devices be combined were ignored [15].
The Technical Bureau of the Reichsluftfahrtmisisterium (National Air
Ministry), the government agency resposible for aviation matters, brought
forth a modification of Stichling called Zwilling (twin) for which produc-
tion began in early 1941 with 10 000 eventually being installed, this despite
a report from the Air Research Station at Rechlin that the device was com-
pletely unusable. Martini was not able to stop production of this unit until
10 January 1942 [16]. GEMA cannibalized these units for components used
to make Erstling [17].
Zwilling was deficient in that it responded only to the Würzburg and
did not even do that satisfactorily. If a Würzburg sighted an aircraft for
which the identification was in question, it altered the pulse repetition rate
from 3750 Hz to 5000 Hz. The receiver, a diode detector sensitive to a
wide range of wavelengths, responded to this change of repetition rate by
transmitting on an auxiliary 1.9 m wavelength Morse code formed by a
notched rotating disc in an audible 800 Hz. The radar operators listened
on a receiver tuned to the auxiliary frequency. The deficiencies of Zwilling
quickly became apparent: there was no range information and the direction
of the reply signal could not be determined from the auxiliary transmission,
as only a simple receiver was used at the radar station. This meant if
there were more than one plane in the few degrees of the Würzburg’s
beam, one did not know which was responding. Further, when the radar
set interrogated a target its normal functions were completely disturbed
because of the need to alter the pulse repetition rate [18].
Dissatisfaction with Zwilling led to a field modification called Zwill-
ing J1, which dispensed with the change in pulse repetition rate and re-
placed the amplifier that had been sensitive to the 5000 Hz with an amplifier
for the radar pulses. These were then transmitted to the radar set on the
1.9 m auxiliary wavelength. The 1.9 m receiver at the radar set was altered
to broad band width in order to accept the pulses, and its output fed to the
radar oscilloscope screen [19].
On 9 July 1942 Erstling was introduced for the Würzburg as well as
Freya by simply equipping the radar set with an interrogation transmitter
on 2.4 m [20]. The interrogating system received the name Kuckuck—the
same cover name used early by the British Army for MRU.
German IFF problems grew seriously in 1943 as a result of their own
confusion, Allied jamming and the fear among their own pilots that IFF was
being used against them by the enemy [21]. Various designs, the most im-
portant being Neuling (newcomer), were never deployed. The Allies also
planned more advanced IFF equipment, Hazeltine and the Naval Research
Laboratory designing IFF mark V [22], also not deployed during the war.
For both sides IFF had become such an enormous endeavor that change
133
came to be looked on with horror, not only because of the magnitude but
because of the complications of a changeover that would necessarily extend
for an uncomfortably long period.
The problem of IFF between aircraft became critical for Germany as
the great struggle between Bomber Command and the Luftwaffe night
fighters became intense in 1942 and was a problem never satisfactorily
solved for the defenders—or the attackers either for that matter. Night
fighters on both sides were twin-engine aircraft because two-man crews
were required, necessary not only because of the complications of radar
operation but also because of the absolute requirement that the pilot retain
his night vision, which was ruined by looking at an oscilloscope. The
absence of German plane-to-plane IFF led to an order to pilots not to attack
any two-motor aircraft [23].
IFF did not prevent losses to friendly fire. An air crew sometimes for-
got to switch it on when returning from enemy territory, where it should
have been off; sometimes it had been damaged in action or it occasionally
malfunctioned as a result of postponed or deferred maintenance; it was,
after all, a piece of equipment that did not call for attention as would a
defective communications transmitter or receiver. There was always the
problem presented by hastily trained crews, who were not alert to a de-
vice that was supposed to be automatic. There were cases of an insufficient
number of sets available for an operation, with occasionally disastrous out-
come, the paratroop invasion of Sicily being one [24]. Ill use of IFF reached
such alarming levels for the US in 1944 that a 40 min joint Army–Navy
training film was issued for the benefit of front-line units [25]. A P Rowe,
Director of TRE, perhaps put it best: ‘. . . the problem of IFF, like the poor,
was always with us’ [26].
IFF was, of course, very secret yet it invariably fell into enemy hands
from aircraft wreckage, despite the explosive charges and thermite incor-
porated to destroy it, and furnished the eagerly sought information about
interrogating frequencies. The first certain evidence that General Martini
had about CH wavelengths was from a reconstructed IFF mark I in April
1940 [27].
Radar beacons are technically members of the IFF family. The early
application of ASV radar by Coastal Command in 1940 to this purpose
led to its use growing enormously on the Allied side. Beacons quickly
became important to British night fighters, who were just as desirous of
knowing where they were and how to get home as were their comrades
in Coastal Command. This enthusiasm led to much user individualism
so that matters became and remained rather complicated and local [28].
It grew beyond these uses. Portable beacons were set up behind enemy
lines to assist drops of agents, supplies or paratroops. For these the gen-
eral response of IFF mark III was not desired. One wanted the beacon to
respond only to a specific interrogating wavelength rather than a band of
wavelengths and to respond with a coded identification, which required
134
First Clashes
a special interrogating radar that the British called Rebecca and a beacon
called Eureka. A prime virtue of Eureka was in Watson-Watt’s words ‘to
speak only when spoken to’ [29].
3.5. THE JAPANESE REALIZE THEY ARE BEHIND

Japan’s first dealings with Nazi Germany aroused both esteem and sus-
picion. Hitler’s ejection of constitutional government coincided with the
Imperial Army’s own achievement and fostered a moderate amount of
political comfort. The Anti-Comintern Pact concluded between the two
in 1936 and expanded to include Italy the following year was much to
Tokyo’s liking, being directed toward the ominous power that threatened
the Manchurian border, where, in fact, two full-scale battles erupted in
1938 and 1939. But Japan was outraged when Germany concluded the
Non-aggression Pact with that same Soviet Union, considered an enemy
three years earlier, and renounced the 1936 treaty—not all of the outrage at
the German–Soviet pact was found among the democracies. A year later,
seduced by Germany’s astounding defeat of France, Japan was persuaded
to enter the Tripartite Pact, pledging unspecified help to Germany and Italy
for equally equivocal aid in return. The Berlin–Rome–Tokyo Axis was thus
formally presented to the world.
As a consequence of this renewed and presumably friendly relation-
ship, Japan sent two missions to Germany to learn what they could of
military value. An Imperial Army delegation of 20 left in December 1940,
but it was the Navy group that most actively sought information touching
on radar. This group of 22, the largest naval mission ever sent abroad,
had Commander Yoji Ito handling electronic matters with two others from
the Naval Technical Research Department (NTRD) accompanying him [1].
They decided against the Trans-Siberian Railway used by the Army and
chose a sea route through the Panama Canal. The separate travel arrange-
ments illustrate the serious rivalry of the two services. The naval delega-
tion left Yokohama on 16 January 1941 aboard the Asaka Maru, the same
day they learned that American armed guards would board all foreign flag
ships during passage of the Canal. By 6 February the Asaka Maru received
assurances from the United States that this would cause no affront to the
prestige of the Imperial Navy and dropped plans for a voyage around one
of the capes. Indeed the passage of the Canal proved to be quite friendly
with the senior officers accepting a dinner invitation ashore from the Com-
manding Officer of the Canal Zone. They departed Port of Cristobal 9
February and arrived in Berlin on the 24th by way of Portugal [2].
Ito and a few others immediately met with the military attachés in
Berlin and left to see the points of naval interest from the French campaign,
boarding 13 new Packard sedans at the Düsseldorf railway station, which
they noted was severely damaged by a recent visit of the Royal Air Force.
The first sight of radar that Ito gained was British, not German, the remains
135
of GL mark 1s and MRUs displayed in the wreckage at Dunkirk. At the

submarine base of Lorient Admiral Dönitz greeted the delegation, which
proceeded with their inspection after a reception. It was then that Ito saw
a Würzburg; it took no effort on his part to ascertain that it was a gun-
laying radar, and he was suitably impressed. To obtain a closer look at the
Würzburg took a bit of persuading and was limited to half an hour for a
few, including Ito and Rear Admiral Naosaburo Irifune, head of the Navy’s
Gunnery School. Attempts to obtain details came to nothing, as Göring had
declared the Würzburg the most secret of Germany’s weapons; not even
their formal ally, Italy, had been given such information [3]. Ito noticed a
cage-shaped antenna on a pole that he took to be for a very-high-frequency
directed beam. It was probably a Freya, as they were being installed along
the coast at that time, but his inquiries received no answer [4].
The delegation toured Germany for several weeks with numerous
meetings and exchanges of information, but this was not an ‘Axis Tizard
Mission’, the very open exchange of secrets between the British and Amer-
icans that took place in late 19405 . Ito did not learn many details of German
radar, but neither did he let out anything about his magnetron. Given the
continued bad performance of German torpedoes, one must assume that
they told them nothing about their own designs. One suspects that the
Germans did not expect to be taught anything by the Japanese, which may
have restrained their probing.
While the Naval delegation was in Germany, alarming intelligence
began to accumulate, both to the delegation and in Tokyo, about British and
American radar. This was more often than not inaccurate in the direction of
exaggerating the capabilities, but this only added to the rising concern that
Japan was being left behind in the knowledge of what was finally realized
to be an important new weapon.
While the delegation was in Germany the Axis suffered two serious
naval defeats at the hands of the Royal Navy, both resulting from British
radar superiority. During the night of 27/28 March 1941 Italy lost three
cruisers and two destroyers in the night action off Cape Matapan6 . Rumor
of this filtered to Tokyo from the Italian Naval Attaché in Washington in the
form of a report that British ships could deliver accurate blind fire at night
without searchlights. As a consequence of this report Commander Iwao
Arisaka, an ordnance officer then the Naval Inspector resident in New York
City, was ordered to investigate discretely any radar capability that the US
Navy might have. His report, based on what evidence is not clear, came
back with exaggeration that matched the Italian: American battleships,
carriers and cruisers all had antennas on their foremasts suspected of being
this new method of night vision. Inasmuch as the Japanese considered
themselves to be masters of naval night fighting, this was serious [5].

136
First Clashes
This was followed by news reports—attended by rumors of radar’s

involvement—of the sinking of the German battleship Bismarck on 27 May7 ,
which led to instructions to the Naval Attaché in London, Commander
Ryo Hamazaki, to look into the matter. He came up with precious little
about what radar had to do with the sinking of the Bismarck but did send
a description of what must have been a GL mark II that he saw set up with
an AA battery in Hyde Park [6].
Part of the delegation to Germany had planned to leave in June, leav-
ing Ito and the others to remain a little longer to extract what more they
could, but the news of the radar capabilities of Japan’s prospective ene-
mies brought a peremptory order on 19 June for all to return immediately.
If information gained about radar had been meager, the overall results of
the mission were considered to be extremely valuable for the details of
industrial processes, submarines, high-speed torpedo boats and methods
for manufacturing artificial rubber.
The delegation returned home in two different groups. The first,
which included Ito, departed Rome on 15 August in Italian aircraft for
Recife, Brazil by way of Villa Cisneros in western Sahara. German aircraft
took them from Recife to Rio de Janeiro, whence they boarded a Japanese
steamer for Yokohama. The remainder went through Switzerland and
Vichy France to Spain where passage on a Spanish ship to Rio was arranged,
thence by freighter to Japan [7]. Although the Army’s interest in radar did
not equal the Navy’s, one officer of its delegation, Lieutenant Colonel Kinji
Satake, remained in Germany and became well instructed about radar and
the Würzburg in particular after Japan entered the war.
By summer of 1941 the reports from abroad had begun to alarm key
members of the Naval General Staff (Kaigun Kansei Hombu) and caused
the issuance of an order on 2 August for the expenditure of ¥11 million
($4.4 million) on radar. Rear Admiral Kiyoyasu Sasaki, head of the Elec-
trical Engineering Research Department (Denki Kenkyu Bu), was quite
eager to get started and called a meeting of his entire staff. Commander
Chuji Hashimoto was placed temporarily in charge until Ito returned, and
he consulted closely with Ito’s staff. Industrial support was added from
NHK Japan Broadcasting Corporation’s Technical Research Laboratory,
specifically and importantly by Kenjiro Takayanagi, who had developed
Japan’s television, and NEC (Nihon Denki). Most of the design elements
needed for a meter-wave apparatus were at hand among these groups,
and assembling them into a lashed-up pulse-modulated 4.2 m set of 5 kW
proved remarkably easy.
A 3 m refined prototype was set up on the grounds of the Naval Mine
School on the Miura Peninsula by 8 September. At this time they had no
data on the reflectivity of aircraft nor any idea as to the best polarization
to use, but they were soon able to detect a medium-sized bomber at 97 km
137
and a flight of three at 145 km.

From this prototype the mark 1 model 1 (Ichi-go Ichi-gata) was con-
tracted to three firms for immediate production: NEC for transmitters,
Japan Victor (Nihon Victor) for receivers and Fuji Electrical Apparatus
Manufacturing Co. (Fuji Denki Seizo) for antennas. The first industrially
produced radar was placed in November 1941 at the Katsu-ura Lighthouse,
where it was used throughout the war [8].
The Army relied primarily upon NEC and Toshiba for its radar, the
key developmental elements of which were the NEC Ikuta Research Office
Branch (Ikuta Kenkyu Bunsho) and the Toshiba Research Institute (Toshiba
Kenkyusho) [9]. The Doppler interference between widely spaced trans-
mitters and receivers, such as had been observed at the US Naval Research
Laboratory, in the Watt’s Daventry experiment, by David in France and
by Oshchepkov in Russia, had not escaped Japanese observation, and the
Japanese Army put this equipment, designated as type A, operating on 4
to 7.5 m, to use in China in 1941. The longest such link was from Formosa
to Shanghai, a distance of 650 km [10]. In June 1942 NEC and Toshiba each
began developing 1.5 m searchlight radars, designated Tachi-1 and Tachi-2
respectively. Both proved too fragile for field service and only about 25 of
each were manufactured, although Tachi-2 proved a financial success to
the manufacturer at ¥200 000 ($80 000) each [11]. The Army’s air-warning
radar was the 4 m type B, soon designated as Tachi-6. It used a broad-
cast transmitter reminiscent of CH but with some degree of electronically
adjustable direction and had up to four separately directional receivers
[12].
Trustworthy information about Allied radar came with the first Japan-
ese conquests. A report by the military commander in Singapore described
what were thought to be electronic weapons captured from the British,
which led to a delegation that included Masatsugu Kobayashi of NEC
and Shigenori Hamada of Toshiba flying to inspect [13]. At Singapore
they obtained a GL mark II and a searchlight control radar (SLC), which
startled them in its use of Yagi antennas. This antenna, named for their own
illustrious high-frequency expert, had been used extensively in America,
Great Britain, Germany and Russia for experimental work and a few final
designs, but it had found very little use in Japan. Along with the SLC came
a nice extra. A Corporal Newmann had made—in violation of draconian
orders to the contrary—a complete set of notes that the Japanese had typed
in English and duplicated in a 22-page booklet [14]. In the Philippines they
obtained an SCR-270 and 268.
The British SLC had impressed Hamada with its compact simplicity
and its similarity to the design already employed in Tachi-1 and 2, and he
had its improvements copied into Tachi-4. It had a single Yagi for the trans-
mitter and four Yagis positioned about it for the receiver and connected
through a rotating capacitor that generated a conical scan, a technique
already used in Tachi-1 and 2. It also worked on 1.5 m and had trailer-
138
First Clashes
mounted antennas that could be pointed in azimuth and elevation as had

the two predecessors.
Kobayashi allowed Masanori Morita to make the improved copy of
the GL mark II called Tachi-3 that had dominated his thoughts since he
had learned of the British set [15]. Transmitter and receiver were mounted
separately about 30 m apart over underground shelters that rotated in
azimuth. The radiation pattern of the transmitter could be adjusted in
elevation by altering the phase between pairs of dipoles. The receiver had
five dipoles in a horizontal array, four forming a diamond that yielded lobe-
switching for azimuth and elevation with the fifth used for determining
range. Both Tachi-3 and 4 became widely used sets for searchlight and
somewhat limited gun-laying use.
The Navy successfully developed its own version of SCR-268 as mark
4 model 2 and placed it in operation at the important base of Rabaul in 1943.
It was widely used and one of the most produced Japanese radars, 2000
units having been manufactured. As the war progressed the nomenclature
of Japanese radar becomes more confused, although generally with regard
to equipment that was never deployed.
With Japan’s entrance into the war against two of Germany’s three
principal enemies a change in attitude prevailed in Berlin—specifically
Japan could have the Würzburg secrets. To obtain the details and an oper-
ating set the new and very large submarine I-30 was dispatched on a long,
adventurous voyage to France with 120 men on board and arrived at the
growing U-boat base of Lorient on 8 August 1942 to a rousing welcome.
I-30 departed on 22 August with all that could be desired for building
Japanese copies of the now famous gun-laying radar and arrived at the
Penang Naval Base on the Malay Peninsula on 10 October. After refueling
she continued this remarkable voyage only to strike a British mine seven
days out of Penang. The sample Würzburg was damaged beyond repair
and the drawings became a soggy mass. In summer 1943 Colonel Satake
and a Telefunken engineer, Heinrich Foders, made a harrowing voyage on
an Italian submarine to bring Würzburg capability to Japan; unfortunately,
important components and data were on an accompanying submarine that
was sunk [16]. The Navy re-engineered it as mark 2 model 3, and the Army
produced its own version independently as Tachi-24, illustrating the funda-
mental problem the nation had in apportioning scarce resources. Neither
advanced beyond prototypes completed in April 1945 nor affected the war
in any consequential manner [17].
GEMA also sent an engineer, Dr Emil Brinker, to Japan by a sub-
marine that landed in December 1943 with details about Freya, including
anti-jamming circuits. The Japanese considered their own air-warning
equipment adequate and were not interested in Freya, so Brinker spent his
time developing radar test equipment. Although an expert in IFF, he was
not allowed to enter Japanese radar research because his security clearance
was only for work on a Freya [18].
139
By the end of the war Japan’s radar had been completely outclassed
by the Allies. This came about from having resources unequal to the task
and a military divided from beginning to end as to radar’s relative im-
portance in the disputes about allocations. Until dramatic examples were
presented in the late war years of the great power of this new method
of waging war, this attitude held back Japanese radar. There were high
levels in the government that appreciated the military need for scientific
and technical research, so Prime Minister Fumimaro Konoe established the
Cabinet Board of Technology (Gijutsuin) to organize such efforts and ap-
pointed Viscount Kyoshiro Inoue, Professor at Tokyo Imperial University
and Minister of Railways, as president of the Board on 31 January 1942,
but the Board was severely hampered by contemptible annual budgets and
ignored by the feuding Army and Navy. Its effect on the course of the war
is difficult to find [19].
Shigeru Nakajima headed research at Japan Radio, which conducted
Japan’s research in magnetrons, and saw his staff shrink from 800 at the
beginning of the war to half that number at the end [20]. Technical special-
ists were simply drafted into the Army. It was a startling comparison to
the huge growths of similar American, British and, after an initial pause,
German groups.
Nevertheless, much was accomplished. Fielding an industrially pro-
duced 3 m air warning set in November 1941 in a program begun in August
is rivaled in speed only by the South African production of the JB and the
Australian of the LW/AW8 . As we shall see, the Japanese 10 cm equipment
was at sea only months behind the British and weeks behind the Ameri-
cans. Yet the value of their sets was limited by retaining only the A-scope,
the most primitive indicator, the display of signal size against range, some-
thing that immeasurably reduced the effectiveness of this equipment and
that so remained until the end. Only 100 IFF sets were manufactured [21].
140
First Clashes
PHOTOGRAPHS: AIR WARNING—PACIFIC
Australian–American cooperation. A 1.5 m LW/AW mark II antenna

(light-weight/air-warning radar) at Rose Hill Racecourse in Sydney with US
personnel in training. This set, which the outbreak of the Pacific War caused to
go from design to production in a few months, was light enough to be transported
in the ubiquitous DC-3 transport and was widely used by Allied forces in the
South Pacific during the early years when American equipment was both rare and
too heavy for some island operation. No landing was considered secure until an
LW/AW was on the air. The frame below the antenna supported a tent. Australian
knowledge of the jungle requirements of electronics proved valuable for the Allied
forces operating in the South Pacific. Worledge Collection, Radar Research and
Archive Collection, RAAF, Williamtown, NSW.
141
An American SCR-271 3 m air warning radar at New Caledonia in May 1943.

The development of this set resulted from the Army Air Corps eagerness for a
device to protect their bases from surprise attack. The 271 was a fixed-station
mount of the electronically similar SCR-270, which was portable. The cabin at
the left rear housed the engine-driven generator. The military unit was Company
A, 579th Signal Air Warning Battalion. National Archives photograph 111-SC
245874.
British Light Warning radar LW.

This 1.5 m set weighed only 1200
kg and was generally the first radar
on the beaches. It was copied by
the United States as the SCR-602.
Troops in the Pacific found its elec-
tronics superior to the Australian
LW/AW, which they used until
mid-1944, but preferred the latter’s
dipole-array antenna to the four
Yagis of the original design. Histor-
ical Radar Archives. Crown Copy-
right.
142
First Clashes
An American SCR-602 light-weight air-warning radar. This 1.5 m set was a

copy of the British LW and went through a number of modifications. Here the
original Yagi antennas have been superseded by a dipole array. It replaced the
Australian LW/AW for American units as the war progressed. The device shown
here was used by the 583rd Signal Air Warning Battalion on Tanahmerah Beach,
New Guinea, 23 April 1944. This battalion placed detachments with even smaller
radars, AN/TPS-2, among Filipino guerrillas to monitor Japanese air and sea
traffic before the invasion of the Philippines. National Archives photograph 111-SC
254238.
Japanese Navy 3 m fixed air-warning

radar mark 1 model 1. This set was
one of two captured in August 1942
on Guadalcanal to the great surprise
of American radar men. It had a peak
power of 5 kW, pulse lengths of 3 to
20 µs, repetition frequency of 750 to
1500 Hz and a range of about 150 km.
National Archives photograph 80-G
11293.
143
Japanese Army 4 m Tachi-6

air-warning radar transmit-
ter. This used a broadcast
transmission with a vertical
dipole array configured to pro-
duce radiation lobes cover-
ing the region from which at-
tack was expected. Three or
four directional receivers com-
pleted the station. This de-
sign was the Army’s stan-
dard air-warning radar. This
well camouflaged antenna was
captured on Noemfoor Island,
Dutch New Guinea in July
1944. National Archives pho-
tograph 111-SC 267148.
Japanese Army 4 m Tachi-6

air-warning radar receiver.
Rotatable horizontal dipoles
with reflectors on two levels
provided directional accuracy
of 5◦ by maximizing. The
receivers were located about
100 meters from the transmit-
ter. The system has elements
common to CH but was al-
most certainly designed with-
out knowledge of the British
equipment. This antenna was
captured on Noemfoor Island,
Dutch New Guinea in July
1944. National Archives pho-
tograph 111-SC 267161.
144
CHAPTER 4
NEW IDEAS
4.1. MICROWAVES
Most of the originators of radar wanted to use microwaves from the very
start. The main reason for this was the desire to construct a radio search-
light, and the mirror for such a searchlight had to be several wavelengths
in diameter. By forming a narrow beam one could also concentrate more
power onto the target, technically called increasing the antenna gain, some-
thing of no mean importance considering the low power of the generators
then available. A narrow beam also yielded much less reflected power
from nearby ground or water, making the operator’s task simpler. Not
appreciated at the time was the decreased susceptibility of narrow beams
to jamming or to the reflecting dipoles that attacking aircraft were to throw
into the radio eyes of the defenders. That shorter wavelengths allow one
to resolve smaller target structure did not figure into their thoughts either,
although the importance would soon be felt.
Preliminary experiments, such as done in the United States, France
and Germany, had shown clearly that microwaves yielded detectable re-
flections from ships, automobiles and aircraft and indicated that practical
equipment could be made, had there been a generator of sufficient power.
But there were arguments from theorists that microwaves would not be ef-
fective because of specular reflection, and an experiment in the summer of
1936 at the Nachrichtenmittel-Versuchs-Anstalt gave the opponents of mi-
crowaves some supportive data. A metal screen was suspended between
the masts of a small vessel from which the reflection of very short waves
was observed. Swinging the ship by 90◦ produced very large effects [1].
This experiment, which satisfied some investigators of the unsuitability of
microwaves, was about forty years ahead of its time and was supported by
inadequate theory and computing capability. An airframe whose surface
is made up entirely of flat surfaces—an aerodynamic horror—can be made
invisible to radar for any reasonable orientation of aircraft to radar beam.
This is the basis of the Stealth technique [2].
For the first few decades of microwave research the only design ele-
ment not common to radio in general was the parabolic mirror. This drew
145
on both optical and acoustical experience, the optical from astronomical

telescopes and searchlights, the acoustical from the whispering chambers
of renaissance architecture. The next truly microwave design element came
from ancient acoustical experience. Almost every child has learned for
himself or from his playmates that blowing across the mouth of a bottle
or jug produces a low-pitched tone. This phenomenon comes about from
the sound being restricted to certain discrete wavelengths because the air
molecules cannot vibrate freely when in contact with the walls of the jug.
Hermann Ludwig Ferdinand von Helmholtz, whose investigations prob-
ably encompassed more branches of learning than any other prominent
scientist, used these simple devices in explaining the physics of sound and
in formulating a physical theory of music. For our part he earned a place
of honor for having guided his favorite pupil, Heinrich Hertz, toward elec-
tromagnetic waves.
Stimulated by von Helmholtz’s ‘On the Sensation of Tone’, Lord
Rayleigh undertook a mathematical analysis that ended in the publica-
tion in 1877 of The Theory of Sound, a book still in print for its tutorial
as well as historical value. The analysis of acoustical resonators can be
transferred to electromagnetic waves in cavities by requiring that the tan-
gential component of the electric field be zero at the conducting surface of
the cavity, essentially the same requirement as saying that air molecules
cannot vibrate when touching a hard surface. This transfer was made by
W W Hansen of Stanford University in the mid-1930s [3]. Hansen’s analy-
sis went beyond solving the partial differential equations that describe the
waves with the boundary conditions imposed by the cavity. He reduced
the results to a form that allowed a designer to consider them as equiva-
lent circuit elements and described how one made the connection from a
two-element conductor to a cavity. He predicted and verified that energy
loss in such chambers was remarkably small.
Hansen’s original goal was to accelerate electrons to produce high-
energy x-rays for medical application. David Sloan had built an accelerator
based on this principle at Lawrence’s laboratory in 1934, but the frequencies
available from the oscillators then available restricted its use to ions much
heavier than the protons, deuterons and alpha particles that the nuclear
physicists wanted to use. Electrons required techniques of much shorter
wavelength applied over distances that precluded accelerating electrodes
connected conventionally [4].
Closely related to a resonant cavity and just as vital for microwave
work is the waveguide. Rayleigh had shown that electromagnetic waves
could be propagated between two parallel conducting sheets whose sep-
aration was comparable to the wavelength, and this was seen to be the
explanation of the worldwide propagation of very long wavelengths, the
earth’s surface and the ionosphere [5] forming the two conducting surfaces.
He also showed that waves could be propagated in tubes of circular or rect-
angular cross section [6]. George C Southworth’s thesis at Yale had left a
146
New Ideas
residue of ideas that grew into waveguides some years later at Bell Tele-
phone Laboratories [7]. At the same time at the University of Kiel an exper-
imental thesis by O Schriever reported the propagation of radio waves in
dielectric rods using the new Barkhausen–Kurz oscillator [8]. Southworth
read and appreciated the paper’s significance, but he remained unaware
of Rayleigh’s paper until years later when it was found during the research
for the waveguide patent.
Southworth continued work in high frequencies at Bell Labs and as
sources of 15 cm radiation became available took up again the dormant
idea of propagation through a pipe. In August 1933 he succeeded in build-
ing a circular waveguide of 12.5 cm diameter. He had had to hide the
early work as a form of ‘test’ rather than ‘research’ because of the restric-
tions on his duties. When the truth came out it attracted the attention of a
theorist, who almost wrecked the project through a calculation error. The
transmission of electric power without a return conductor was a bit too
startling and evoked skepticism, but success brought approval and theo-
retical help. Soon Bell was full behind the project. Southworth used the
term ‘wave-guide’ in an early memorandum [9].
At the Massachusetts Institute of Technology (MIT) Wilmer L Barrow
was making waveguides independently. After earning a degree in elec-
trical engineering at MIT, Barrow went to the Munich Technical Institute,
completing a dissertation there in acoustics. He also received instruction
in mathematical physics from Arnold Sommerfeld with important conse-
quences for work he was to do. He returned to MIT and began research
in very-high-frequency antennas and propagation. His approach to the
waveguide, unsuccessful at first, was as a method of picking up signals
reflected from a paraboloid mirror. By Christmas 1935 he understood the
critical elements and by the following March he had worked out the na-
ture of the vibration modes in a rectangular guide and derived the cutoff
wavelength beyond which the guide would not transmit [10].
The two investigators learned of one another’s work a few weeks
before both were scheduled to present papers to a combined meeting of
the American Physical Society and the Institute of Radio Engineers in May
1936. Patent problems were worked out amicably.
An open-ended waveguide squirts radiation much like water from a
hose. Barrow determined the optimum shape for the horns that terminate
the waveguide—the nozzle, as it were—so that a minimum of energy was
reflected back down the pipe. There was soon great activity at Bell Labs
and MIT on working out the details of waveguides and horns as circuit
elements.
Microwaves can be transmitted by coaxial cables but reach practical
limits beyond which waveguides have the advantage. As frequencies in-
crease the rapidly changing electric field is able to penetrate ever shallower
depths of conductors, the skin effect. For a coaxial cable this means that
less and less of the center conductor is used, which results in ever increas-
147
ing resistance. The same thing happens on the inner surface of the outer
shield, but it has more surface area. For a coaxial cable this means that with
microwaves only a small fraction of the center conductor is used and trans-
mission becomes increasingly attenuated. For high power the cable will
become hot, even with the low-loss insulator, polyethylene. In a waveg-
uide all the necessary conduction is in the large periphery, so attenuation
is much less.
Most of this waveguide research was carried out with the Barkhausen
tube, later replaced by something far better—the klystron.
The idea of using the electron’s relatively slow speed to effect the gen-
eration of very short wavelengths had been in the air since the invention of
the Barkhausen–Kurz tube and the split-anode magnetron, both of which
we have encountered. In 1935 a paper from Italy [11] presented a new idea,
the Heil tube. A beam of electrons passes through an electrode that is con-
nected to a resonant circuit; a gap precedes the electrode and another gap
follows it; when the resonant circuit oscillates the electrons are bunched in
the first gap (the buncher) and energy extracted from them at the second
(the catcher); the oscillations build up, if the electron velocity is properly
matched to the dimensions of the electrodes and the resonant frequency.
Electron bunching is brought about by reducing the speeds of the first of
a group of electrons and increasing those of later ones, thereby produc-
ing a grouping further down the beam. Some excellent Heil tubes were
made, but the dependence on external circuit elements greatly reduced
their suitability for microwaves [12].
A couple of years after the Heil tube Hansen built a microwave tube
that combined one of his resonant cavities with an Acorn triode and an elec-
tron beam in a design he called the ‘rhumbatron’ [13], but it was quickly
swept aside by the klystron. The klystron was the product of two broth-
ers, Russell H and Sigurd F Varian, working with Hansen in the Stanford
Physics Department. They dispensed with the Acorn and used an electron
beam much as in the Heil tube (of which they were unaware) but used
a resonant cavity, which Hansen designed specially for the tube [14], in
such a way that the standing waves of the cavity acted directly on the elec-
tron beam instead of imposing resonance with an external resonant circuit.
This proved to be a superb generator of microwave oscillations and went
through several development generations in short order. Its name is de-
rived from the Greek verb ‘klyzein’ that describes the breaking of waves
on a beach.
The team of Hansen and the Varian brothers combined three remark-
able men. Hansen was simply brilliant. He graduated from high school
at the age of 14 and went immediately to Stanford where he mastered the
disparate skills of master mathematician and machinist. Russell Varian,
the older of the two brothers, had such difficulty in learning to read that
he graduated from high school four years older than his classmates but
with a solid grasp of the things he learned and a good knowledge of sci-
148
New Ideas
ence. His perseverance earned him a BA in physics at Stanford. Sigurd

wanted nothing to do with college, purchased a war surplus Curtiss Jenny
and entered a career as pilot, eventually trying out new routes for Pan
American Airlines. His knowledge and interest in aerial navigation later
supplemented by concerns for air defense led him to work with Hansen
and his brother on microwaves, contributing his enormous practical and
organizational skills [15].
In frequency stability the klystron was all that the radar engineers
could have wanted, but it did not have adequate power. The power level
is proportional to the intensity of the electron beam, and as in the cathode
ray oscilloscope1 the electrons repel one another, so that as one attempts
to increase intensity the beam diverges and the desired effects are soon
dissipated. Unlike the transmitter triodes, no dramatic increase in power
could be obtained by pulsing the tube because the limit was space charge,
not the thermal characteristics of the anode. If not useful as a transmitter,
although there were attempts, it proved to be invaluable for microwave
radar in more delicate functions, such as amplification and as the local
oscillator for producing intermediate frequencies.
The Stanford work attracted the attention of Professor Edward L
Bowles of MIT and the Sperry Gyroscope Company, who were experi-
menting with methods for assisting pilots in making blind landings. As a
consequence of Bowles’s favorable recommendation Sperry gave financial
assistance to Stanford and opened a shop in nearby San Carlos for design-
ing a blind-landing system based on the klystron. The Varian brothers
went to the new Sperry shop, and Hansen remained at Stanford to con-
tinue klystron development. In the fall of 1940 Sperry transferred these
operations to Brooklyn [16].
Alfred Lee Loomis was perhaps the last of a long tradition of am-
ateur scientists, amateur only in not earning his livelihood from science;
his income was derived from investment banking, and his science done
on the side. He was also a successful lawyer and Army officer in World
War I. It was while serving as a major at Aberdeen Proving Grounds that
Loomis became friends with the Johns Hopkins physicist R W Wood. Af-
ter the war Wood helped Loomis set up a private laboratory at Tuxedo
Park, New York from which an impressive list of publications arose [17].
Conversations with Bowles convinced Loomis of the value of the new mi-
crowave equipment, and he agreed to set up a program to be conducted by
J A Stratton to study propagation [18]. They soon had a continuous-wave
radar measuring automobile speeds at Tuxedo Park.
The collapse of France in June 1940 altered entirely Americans’ view
of the war. It placed the heads of four leading scientific organizations in
much the same position that Tizard and his associates had found them-
selves five years earlier: Vannevar Bush of the Carnegie Institution, James
1 See Chapter 2.1 (p 33-5).
149
Bryant Conant of Harvard University, Karl T Compton of MIT and Frank

B Jewett of the National Academy of Sciences. They had discussed infor-
mally in the past the ways that the country might organize its scientific
and technical strengths for defense, but the defeat of France and Britain’s
extremity demanded immediate action. Bush incorporated their ideas into
a memorandum for President Roosevelt that he was able to present in per-
son in early June, receiving the famous ‘OK FDR’ initials. The result was
the creation of the National Defense Research Committee, NDRC.
Bush began to form sections for the varied forms of research to be pur-
sued and chose Loomis to head Section D-1, the Microwave Committee.
Loomis selected Bowles as Executive Secretary, included representatives
from Bell Labs and Sperry Gyroscope and called a meeting at Tuxedo Park
on 14 July 1940. They quickly added the heads of General Electric, West-
inghouse and RCA along with nuclear physicist Ernest O Lawrence. The
Committee took as its first business the survey of what was being done
by the armed forces in radar and within days visited the Naval Research
and the Signal Corps Labs. One of the impressions that they gained, and
which was to remain preserved in the memory of the NDRC people, was
the conviction that the two services kept their radar work so completely
covered in secrecy that neither knew what the other was doing [19]. This
bit of misinformation was soon expanded to include the belief that the
service laboratories had never consulted the electronic industry.
In surveying the work that had been done in microwaves the Com-
mittee learned of another very-high-frequency tube, a demountable, con-
tinuously pumped beam tetrode having mechanically tunable cavities; it
delivered 1 kW continuous and 5 kW peak power at 50 cm. Generally
referred to as the Sloan–Marshall tube or the Resnatron, it had been devel-
oped at Lawrence’s Berkeley Laboratory and reported at the June meeting
of the American Physical Society. Fortunately for the security that now be-
gan to cover microwaves, the abstract [20] disclosed no details in its record
briefness of two sentences and no publication followed. The Committee
wanted a shorter wavelength but nevertheless thought the work of value
and secured $20 560 to further it. This tube would be swept aside within
weeks, when the long-sought generator for 10 cm waves arrived, but was
to have a curious part to play by and by.
Britain had placed far and away more resources into the radar pro-
gram prior to the outbreak of war than any other nation but nothing into
microwaves, although the Royal Navy was developing decimeter equip-
ment. The need for very short waves became pressing as the shortcomings
in the equipment for airborne interception (AI) became apparent. This
concern was not really so serious as thought, as the horrendous losses that
Bomber Command were to suffer from German night fighters equipped
with 3 m radar were to demonstrate, but radar AI had not been used in
combat in early 1940, and the designers were rightly concerned with what
they perceived as a serious defect of their product.
150
New Ideas
Bowen had proposed very early pushing microwaves but found ‘no
interest within Bawdsey itself and anyone talking about centimetre waves
was thought of as some kind of crank’, although he did find interest with
Sir Charles Wright, Director of Scientific Research for the Admiralty [21].
Decimeter waves had already engaged the interest of the defense establish-
ment, and the Admiralty awarded a development contract to the General
Electric Company, Ltd (GEC) in November 1938. GEC entered into the
problem of AI a year later and had a complete 25 cm pulsed radar lashed
up at their Wembley laboratory in April 1940. It used high-power triodes
and attained an output of 2 kW. Of particular note was its use of crystal
diodes for the mixer. The set had, in fact, exceeded the goals set for it four
months before [22].
Two eminent nuclear physicists recently brought into radar, John D
Cockcroft and M L Oliphant, became concerned about the AI problem and
about the general absence of research in very short waves. Oliphant se-
cured a development contract for microwaves in September 1939 for the
Physics Department of the University of Birmingham from the Admiralty
Department of Scientific Research and Experiment [23]. Work began imme-
diately on constructing klystrons, and Oliphant soon had James Sayers de-
sign one yielding impressive continuous-wave power but of temperamen-
tal performance: ‘ten minutes once a fortnight’ some claimed [24]. Such
results were hardly encouraging, especially the inability to compress the
impressive power into pulses [25]. One of the Birmingham staff, J T Ran-
dall, known at that time for having been the inventor of the phosphor used
in fluorescent lamps, was assigned to microwave generation with a lecturer
in radiophysics, H A H Boot. They saw that a resonant component had to
be incorporated within the vacuum tube and that Hansen’s cavity offered
no solution beyond the klystron. This conclusion pointed them toward the
magnetron.
The demands for speed fortunately prevented them from making the
usual survey of the magnetron literature, much of it not in English, which
by that time had grown to impressive dimensions. Hoping to find a new
approach, Randall read a translation of Hertz’s work and recognized in
the loop-wire resonator, used in the very first radio experiments, the de-
sign element they sought. They thus dropped Hansen’s three-dimensional
cavity for Hertz’s two-dimensional loop [26].
The result was a success on the first trial that fixed the fundamentals of
this device. A plurality of loops was connected in a circular pattern around
the cathode to increase the output power; the magnetic field was oriented
perpendicular to the plane of the loops. This had two important advan-
tages over any of the magnetrons previously made: the Hertzian loops
were made out of copper as thick as practical and maintained at ground
potential. This allowed the anodes, which must take up the energy of the
electrons that had not been expended in radiation, to be much larger for a
given magnet size than the split-anode magnetron and to be water cooled,
151
as in the first model, although generally replaced by air cooling as a con-

sequence of the high efficiency of the device in converting direct-current
power into microwave power. The difficulties with previous magnetrons—
soon to disappear from the radio world—were the requirement of an ex-
ternal resonant circuit, the inability to dissipate the heat generated on the
anode except by thermal radiation, and an inherent frequency instability.
With the goal of generating 10 cm waves they made their first anode
block according to the formula for the Hertzian loop, which specified 1.2 cm
diameter; the thickness was arbitrarily chosen to be 4 cm. The first model
was pumped by a diffusion pump and sealed with wax in the tradition
of the golden age of physics. The first trial on 21 February 1940 was an
immediate success. The continuous wave output at 9.4 cm was 400 W and,
unlike the early klystrons, pulsed operation would yield greatly enhanced
power. The powerful continuous output allowed dramatic demonstrations
of glowing lamp bulbs and heated flesh. Worries about serious frequency
instability, a continuing problem with split-anode magnetrons, were soon
set to rest.
In April a contract was awarded GEC for the design of a manufac-
turable, sealed-off tube. E C S Megaw incorporated a large, cylindrical
oxide cathode, drawing on the experience of Henri Gutton of the Société
Française Radioélectrique, who had demonstrated the high currents attain-
able and the ability of such cathodes to work at high voltage [27]. Pulsed
operation followed quickly because of Blumlein’s recent contribution of the
high-voltage modulator. Despite the importance of the work, Randall and
Boot worked alone on the project until June, a measure of the shortage of
qualified personnel. Within a few months tubes delivered peak powers of
10 to 50 kW with 10 to 20% of the power supplied converted to microwaves
[28].
In June E G Bowen was relieved of his duties at St Athan in devel-
oping meter-wave AI and ASV, which was turned over to his assistant,
A G Touch, who continued in much more suitable quarters at the Royal
Aircraft Establishment at Farnborough. He and Brown soon completed
development of ASV mark II. Bowen was re-assigned to ill defined duties
at the principal Air Ministry Research Establishment (AMRE) laboratories,
which by then had left their unsuitable quarters in Dundee for Swanage,
located on the Channel coast 35 km due west of the Isle of Wight. It was
suitably placed to participate first hand in the invasion Hitler would soon
be planning. There Bowen learned about the capabilities of 10 cm radiation
and lectured informally on the general problems of airborne interception.
The resonant magnetron naturally changed the direction of the 25 cm
work being so successfully carried forward at GEC, and their research was
combined with the work being done at Swanage at a meeting held on 29
July 1940. It was not to prove a ‘happy marriage’, primarily because the
GEC engineers saw their new colleagues as being considerably behind in
radio engineering. Dr C C Patterson, Director of GEC’s Research, wanted
152
New Ideas
Plan and section views of the resonant magnetron E1189. E C S Megaw of GEC
designed this tube from the laboratory model of Randall and Boot. It used a cylin-
drical oxide cathode and produced in pulsed operation 12 kW at 9.5 cm wavelength.
Number 12 of this series went to America with the Tizard Mission. E B Callick,
Metres to Microwaves.
AI development done at their well equipped and well staffed laboratories,

whereas Rowe wanted the work done at AMRE under the supervision of
P I Dee, a nuclear physicist with little radio experience. That the working
conditions at Swanage were primitive added more to Patterson’s displea-
sure. Rowe won out [29].
The most effective receiver for microwaves would use the heterodyne
principle, by then almost universal for communication work. This mixes
the receiver input with a tunable local oscillator to form a much lower
and constant beat frequency, called the intermediate frequency (IF) that is
more conveniently amplified. To do this requires a tunable and stable local
oscillator of frequency comparable to the signal frequency. The klystron
came immediately to mind. It was stable enough but tuning was done by
mechanically altering the shape and hence volume of the resonant cavity,
a clumsy way. Robert W Sutton, a master tube designer at His Majesty’s
Signal School, was presented the problem and soon returned with a very
153
valuable microwave component based on principles outlined earlier by the

Stanford group—the reflex klystron [30].
As the reader now knows, the klystron generates microwaves by ex-
citing one of Hansen’s cavities to a resonance that matches the velocity of
the electron beam. If the electron velocity is not perfectly matched to the
cavity, oscillations will still take place but at a smaller amplitude; there is
a range of frequencies centered on the cavity resonance for which oscilla-
tions with amplitudes usable for purposes of a local oscillator will occur.
Sutton placed an electrode to reflect the electrons back into the tube. By
varying the voltage on the reflector, he could alter the frequency slightly,
which enabled the development of circuitry to perform that function. In
this way the intermediate frequency could be held constant, a requirement
for constant amplifier gain, when the magnetron frequency drifted slightly.
A protype tube operated in September 1940.
The mixing of frequencies for an IF amplifier requires a non-linear
circuit element. A diode rectifier, a tube or circuit element that permits
current to flow predominantly in one direction, is certainly non-linear, but
for wavelengths of 10 cm the vacuum tube diodes had too much inter-
electrode capacitance, the same thing that made trouble for some uses of
triodes at very high frequencies and prevented use of the usual hetero-
dyne mixers. The solution lay in resurrecting another circuit component
from early radio to join the super-regenerative receiver in doing modern
service—the crystal diode. Many a home wireless enthusiast in the 1920s
and as many children in the 1930s had delicately positioned a needle on
the surface of a galena crystal to make their crystal sets work. The delicacy
of this operation saw its replacement by vacuum tubes occur without a
great deal of nostalgia, but the crystal diode had for the radar engineer the
important property of having much lower capacitance than one formed by
the electrodes of a vacuum tube.
The utility of the crystal diode had not escaped the attention of early
experimenters with microwave devices. H E Hollmann described their
incorporation in regenerative receivers [31], and George Southworth re-
ported their routine use at Bell Labs as detectors [32].
Silicon crystals proved to have better properties than galena, and a
plentiful supply of metallic silicon was found in the Birmingham depart-
ment. H W B Skinner at TRE designed a wax-encapsulated diode suitable
for production but for a time was the sole manufacturer [33]. The main
problems in constructing a microwave receiver had been solved: high-
power generator, local oscillator, mixer.
Bernard Lovell went into radar during the summer of 1939, one of
the university physicists assigned by the expedient of professors putting
together duty rosters much as sergeants assign men to guard or kitchen
duty. After his apprenticeship at a CH station he began to work with Bowen
and Brown on airborne radar and by the summer of 1940 he found himself
making his first and fateful encounter with microwaves. He first set up
154
New Ideas
shop at Worth Matravers but soon moved to Leeson House nearby in the
collection of TRE laboratories [34].
When the first of the General Electric Company’s sealed magnetrons
came to TRE on 19 July, Lovell and his colleagues set about to make a crude
pulsed system as soon as possible and by 12 August they had observed re-
flections from aircraft [35]. Within a few weeks they had demonstrated
its utility for detecting ships, including a surfaced submarine. This was a
simpler design problem than AI, and the Admiralty type 271 was rushed to
production at HM Signal School—the first operational 10 cm radar [36]. It
used separate transmitting and receiving antennas, which were enclosed
in a lantern-like structure of teak and plastic at the top of a mast. Its
function was to scan about the ship, but the waveguide technology was
not far advanced nor was the technique of making simple rotating con-
nections. Also missing was the duplexing technique for microwaves that
would have allowed a common antenna to be used. Polyethylene coaxial
cables were available, and they allowed rotation of the antennas for two
revolutions, after which the cables had to be unwound. The plan position
indicator, which gives the operator a maplike presentation of the targets
as the antenna continually rotates, was not yet available, so this restriction
was not particularly onerous. The maximum range for detecting a surfaced
submarine was 5 to 7 km [37].
Having something good invariably seems to goad humans to want
something better. The stability both in amplitude and frequency of the cav-
ity magnetron was so much better than that of the split anode variety that
it was hailed as a triumph, but it was soon noted that there were different
oscillation modes for which the device showed somewhat random prefer-
ence and that required an adjustment of the local oscillator of the receiver
in order to return the target blips to the scope. This had been an incentive
for the invention of the reflex klystron, but Sayers sought a more direct
correction and determined that the problem lay in maintaining the same
phase between non-adjacent poles of the oscillating Hertzian loops. He
forced the phase with by-passing anode conductors, called straps, which
stabilized the frequency sufficiently that one had to wait another genera-
tion for new complaints to arise, and power miraculously grew in strapped
magnetrons [38].
There are no simple explanations of the cavity magnetron, only com-
plicated ones, none satisfying. The best approach is to describe the forces
acting on the electrons as they attempt to reach the anode. The cathode has
a potential some kilovolts negative relative to the anode, so electric forces
push the electrons toward the anode. Perpendicular to the plane of their
motion is a magnetic field that pushes them at right angles to the direction
of their velocities. The magnitude of this magnetic force is proportional
to the electron velocity but at right angles to its ever changing direction.
Left to itself this situation would admit analysis, but a further complica-
tion enters—a complication on which the whole matter depends. All of
155
these changes of electron velocities, both of their directions and speeds,

require accelerations, and Maxwell’s equations entail radiation whenever
there is acceleration, just the process that produces the microwaves. So
as the electrons dance their way to the anode, they lose energy and in so
doing produce the microwaves as well as introducing severe difficulties
into the equations of motion.
The design elements of the magnetron were known 20 years before
Randall and Boot put them to use: (1) vacuum good enough for electron
tubes, (2) the diode, (3) the law governing the motion of electrons in electric
and magnetic fields and (4) the Hertzian loop resonator. Thus it is not really
surprising to learn that it had already been invented, not once but three
times.
That one of these should have come from Japan can hardly be a sur-
prise. Yoji Ito, the Imperial Navy’s leading electronics expert, had been
greatly impressed by a magnetron based on the split-anode configuration
invented in 1927 by Professor Kinjiro Okabe of Tohoku University, but Ok-
abe left magnetron research in 1936. In 1937 Ito was rewarded by two mem-
bers of his group, Tsuneo Ito (unrelated to Yoji Ito) and Kanjiro Takahashi,
producing a tube with eight segments that excited an internal resonance
in the manner of a cavity magnetron and that produced centimeter waves.
The concept soon yielded a 6 cm magnetron that produced 30 W, a 3 cm
tube of 3 W and a 1.5 cm tube of 1 W. The value of industrial electron-tube
laboratories led to the decision in early 1939 to reveal the secret aspects of
the magnetron to Japan Radio where Shigeru Nakajima soon operated a
10 cm eight-sector tube that delivered 500 W of continuous power [39].
Some Japanese have subsequently credited Arthur L Samuel of the
Bell Telephone Laboratories for the invention of the cavity magnetron
[40]. This is an interesting credit because Samuel was never able to ex-
tract enough power to warrant describing the device in a publication, and
the Bell Labs—certainly interested in microwaves—did not carry the idea
further. An examination of Samuel’s patent [41] shows one of the three
electrode shapes to be highly suggestive of the cavity magnetron, and Ok-
abe was sufficiently impressed that he showed it in his 1937 book [42]. A
critical difference lies in the wide gap Samuel placed at the mouth of the
loops; the Japanese design differs primarily in tightening this parameter
and thereby forming a Hertzian resonator. The arrangement Samuel chose
for the electrodes forced the use of a coil without an iron core to produce
the magnetic field, which very likely kept operation below the minimum
field required for magnetron oscillation, were it possible. Both the Samuel
and the Japanese designs extracted microwave power through two bal-
anced leads attached to cavities of opposite polarity. Samuel had seen the
need to place the resonant elements within the tube but thought in terms
of discrete components having inductance and capacitance.
The 1939 Japanese tube was water cooled and had an anode 1.2 cm
thick. Continuing work naturally brought forth a variety of anode designs.
156
New Ideas
The delegation to Germany in 1941 had learned the advantages of pulse

modulation, which was applied immediately on their return to Japan and
put an end to continuous-wave methods.
These magnetrons had glass envelopes and used electromagnets
throughout the war, owing to a shortage of materials for permanent mag-
nets. The first sets used a super-regenerative receiver that was extremely
difficult to tune and that was replaced by super-heterodyne equipment.
Each type required a stable oscillator for the detector for which a very
low-power magnetron, the M-60, with the anode made from a thin molyb-
denum plate rather than massive copper was designed. It operated at
much lower voltages and currents than one normally encounters in cavity
magnetrons [43]. Mechanical deformation of the thin anode of the M-60 al-
tered the frequency slightly, an effect utilized for tuning [44]. The Japanese
did not use klystrons even though descriptions of them were in the open
literature and never learned of the reflex klystron as a microwave local
oscillator. The M-60 was paired with the M-312, which had the high emis-
sion needed for high-power pulse work; it also had cavities of alternating
size, which suppressed unwanted modes. The 10 cm prototype radar was
quickly transformed into the mark 2, model 2 Shipborne Surface Search
Radar and mounted on the battleship Hyuga just before the Battle of Mid-
way. Maximum ranges were 55 km for aircraft at 3 000 m and 20 km for
a battleship [45]. This radar had separate horns for transmission and re-
ception, a configuration that marked Japanese microwave sets throughout
the war.
The Japanese Navy excelled in night actions and wanted to add this
new technique to their ways of penetrating the darkness, although the
microwave project did not gain really strong support among line officers
until events in the Solomon Island later drove the point home. When
the Navy was first introduced to the possibilities inherent in microwave
techniques they saw it as an improved way of station keeping at night that
would prevent collisions, a use that came to other navies only after having
had experience with it at sea [46].
As already discussed, Soviet radar development suffered from lack of
interest in the high command, confusion as to its mission and the dispatch
of excellent radar engineers to the Gulag during the purges of the late
1930s. That anything at all came out is remarkable. Because of or in spite
of these extraordinary circumstances there occurred what must be one of
the most baffling incidents in the history of radar. In April 1940 when
the cavity magnetron was Britain’s most precious military secret, when it
traveled under armed guard, when its use was discussed at cabinet level,
when it was described as the most valuable cargo ever to arrive in America
[47], when the United States was preparing to open a special laboratory
just to exploit its properties, when all these circumstances applied, two
Soviet engineers published a complete description of it in the open scientific
literature [48].
157
During 1936 and 1937 N F Alekseev and D D Malairov produced

a series of cavity magnetrons as part of a project for building an anti-
aircraft (AA) gun-laying radar at Scientific Research Institute 9 (NII-9) from
proposals by its director, radio-eminence Mikhail Alexandrovich Bonch-
Bruyevich [49]. The magnetrons were discarded in favor of a pulsed trans-
mitter that used very-high-frequency triodes that worked on 64 cm, had
12 kW peak power, was called Zenith and was abandoned in 1940 [50].
One can presume—and little other evidence is at hand—that the lack
of success of magnetrons in this work, for whatever reason, taken together
with Professor Joffe’s long-standing opposition to microwaves for radar
allowed the publication of the paper. Irrespective of the reasons, the paper
is a complete disclosure of the elements of the cavity magnetron. One
does not even need to know Russian. It suffices to see the tables giving
wavelengths and powers and to think a bit about the drawings of the
characteristic electrode shapes. That the drawings showed water-cooled
anodes tells one a lot. It was all there. There is a report of the independent
invention of the klystron at NII-9 by N D Devyatkov during those same
years, but even less was made of it than the magnetron. It was quickly
followed by a reflex klystron [51].
At the Swiss electro-technical company of Brown, Boveri & Cie, F Lüdi
began experimenting with a magnetron for generating centimeter waves
in 1936. By 1939 he had developed an oscillator that would have delighted
radio engineers in Britain or America, a delight they could have enjoyed,
had they followed foreign technical publications and studied them with
a bit more attention than usual, for the basic elements were described in
an article published in 1937 [52]. Lüdi continued to work on the device,
eventually called the ‘Turbator’, but without incorporating it into a radar
set. It did not have as much power as the Randall–Boot magnetron but had
good frequency stability, certainly better than the un-strapped magnetron.
A high power level could have been attained, had that been a design goal,
but it was unnecessary for the microwave communication links that were
intended. With a large oxide cathode it attained 10 kW peak power in
pulsed operation.
Lüdi’s magnetron resembled the Randall–Boot device except that the
resonating cavity was in fact a three-dimensional toroidal cavity with a
rectangular cross section, which was maintained at positive potential rela-
tive to the axially located cathode by 0.5 to 2 kV. Alternating slits coupled
the cavity to the rotating electron stream that excited various modes of
oscillation when this ‘electron wheel’ was synchronized with one of the
cavity modes. The wheel’s rotation velocity depended on the anode volt-
age and the magnetic field. The anode cavity was supended in a glass
envelope and cooled by thermal radiation, thereby limiting power. Lüdi
published another paper in 1942 [53] that expanded on the first. It in-
cluded a photograph of the device that clarified ideas until then generally
obscured by pages of not particularly enlightening mathematics, although
158
New Ideas
still a little coy about the details of the resonator. Experimental results for
10 cm removed it yet another step from the abstract. The Turbator is best
described in a postwar paper [54].
The Swiss were keenly interested in devising modern weapons, both
for the defense of their homeland and for export, but neither Brown Boveri
nor the Swiss Army initiated a program to provide methods of radio loca-
tion. Army intelligence became aware of radar early in the war and set up
high-altitude monitoring stations using equipment of their own design and
from belligerent aircraft that landed in Switzerland [55], but for enigmatic
reasons, possibly related to the sensitive nature of Swiss neutrality or to a
negative evaluation of its tactical significance for the defense of the moun-
tain réduit, the Army did not tell Brown Boveri about their knowledge of
radar.
I leave it to the reader to ponder the strange meanings found in these
last two tales.
4.2. THE TIZARD MISSION

It was to Professor A V Hill that Wimperis had turned in 1934 when con-
sidering the formation of what became known as the Tizard Committee,
and it was to Hill that Tizard turned in November 1939 with the idea of
opening a scientific liaison with the United States. Tizard, ever thinking
well in advance of present problems, feared a long war for which Britain’s
technical capacities were inadequate and saw America as an important re-
source. Hill, whose scientific stature rested on medicine and physiology,
had made important contributions to air defense in the 1914–1918 war and
had become thoroughly acquainted with radar in serving on the Tizard
Committee. As a consequence of his reputation and position as Secretary
of the Royal Society he was well known to many American scientists. All
these reasons spoke for his visit to the United States and Canada in the
spring of 1940 to investigate the possibilities of scientific liaison and to test
the Americans’ attitude toward the war.
Although Americans were overwhelmingly pro-Allied they were also
isolationist, the consequence of two decades of disillusionment over the
causes and outcome of the previous war and from their interpretation of
the manner in which their entrance into it had been secured: a naive,
unsophisticated people deceived by cunning Europeans. The case made by
Walter Millis in Road to War [1] caused reflection by many a serious reader.
But Hill found American scientists so repelled by the Nazi movement that
these concerns had been swept aside. Some were openly interventionist
and many wanted to co-operate with their British counterparts. Hill noted
that Canadian scientists were prevented by the requirements of secrecy
from collaborating with their natural and willing American colleagues.
On returning to England Hill proposed that official exchanges of se-
cret scientific material be opened between Britain, Canada and the United
159
States. There was a meeting about the proposal on 2 May 1940 made up of
the Tizard Committee, considerably expanded and including Watson Watt,
who maintained ‘that the Americans could not teach us anything, and that
we should get much the worst of the bargain’, and Admiral Sir James Som-
merville, who believed ‘that anything told to the American Navy went
straight to Germany’ [2]. The German offensive a few days later brought a
high degree of urgency and reality to the deliberations and an agreement
by the top military and civilian leaders for a unilateral disclosure of sci-
entific and technical material to the United States. It was hoped for, but
not placed as a condition, that the Americans would respond in kind. The
most important gain would be access to the full resources of the American
radio industry, which necessarily required some disclosure. There were
opponents to the idea on the western side of the Atlantic as discussion of
this remarkable proposal spread, but it had the support of Lord Lothian,
Ambassador to the United States, and gained the full support of President
and Prime Minister. Churchill gave Tizard a free hand [3]. Arrangements
were embodied in a memorandum between President Roosevelt and Lord
Lothian dated 8 July 1940.
The mission, officially the British Technical and Scientific Mission,
consisted of Tizard as leader with Brigadier F C Wallace (Army), Captain
H W Faulkner (Navy), Group Captain F L Pearce (Air Force), Professor John
Cockcroft (Army Research), Dr E G Bowen (radar) and Mr A E Woodward-
Nutt as Secretary [4]. Tizard insisted that the military officers had to have
served recently in combat and have personal knowledge of the technical
equipment they described. He flew across in the miserable conditions that
often served even the mighty as they crossed the Atlantic in wartime; the
remainder came on the Duchess of Richmond, named by them the Drunken
Duchess for her behavior in the seaway encountered. They landed in Hali-
fax and arrived in Washington on 8 September with the famous black deed
box containing documents describing Britain’s technical secrets and one of
the 12 magnetrons that had been produced by General Electric Company.
Tizard and Lord Lothian set up a series of meetings with the radar
discussions arranged through Major General Joseph O Mauborgne, Chief
Signal Officer since 1937, and Rear Admiral Harold Bowen, Director of
the Naval Research Laboratory (NRL). Equally important openings to the
recently established National Defense Research Committee (NDRC) came
easily through Vannevar Bush, one of which led to the Microwave Com-
mittee.
Tizard had planned to demonstrate two of his mighty gifts: the mag-
netron and ASV mark I. The magnetron was ready for display, having come
over in the black box, but an ASV mark I set to be mounted on arrival on a
suitable airplane was not sent, someone in London having decided to send
mark II instead, consequently not demonstrated until after the meetings
were over. The demonstration was left for Bowen to make after Tizard left,
but the magnetron demonstration went as planned.
160
New Ideas
The meetings started on 12 September in the Army and Navy build-

ings located on the Mall, temporaries from 1918 and now mercifully dis-
mantled. First the British described their radar equipment but held back
for the moment the details of how they actually produced the prodigious
power for centimeter waves with which they were so pleased. They were
then given tours of NRL in Anacostia, the Signal Corps Lab at Fort Mon-
mouth and Loomis’s Lab at Tuxedo Park [5].
The Americans representing radar interests were of three general
groups: the first group were Signal Corps and NRL men, who had strug-
gled intensely with the practical problems of radar for half a decade, the
second were the microwave men, who knew more about that technology
than any other group in the world—except how to make a high-power
generator—but had yet to make a useful radar set and the third were the
Army Air Corps officers who knew little about the techniques of radar but
had already realized its value.
The first group held the view prevalent among the American military
establishment that, if left to fight alone, Britain’s position was hopeless
and that their duties called for them to prepare the United States for the
consequences of Britain’s defeat. They saw little hope that America would
enter of its own volition and more or less assumed the Soviet Union was
Germany’s ally. They were also acutely aware of the incredibly bad state
of the Army, one of the smallest in the world, and of its Air Corps. On
the naval side of things there were still traces of a long-standing rivalry
between the British and American fleets.
This group was favorably impressed by only two British sets: CD/
CHL, which had strong similarities to American equipment, and ASV of
which they had no counterpart, although they did have a good radio al-
timeter. They were unable to separate what the British had done from what
they were planning to do. The NRL people thought the Royal Navy’s
radar second rate. CH came across as crude, expensive and immobile,
and the extravagant claims made by Brigadier Wallace [6] for GL mark I
must have brought serious doubts to the Americans about the veracity
of their informants once they had examined the specifications of that set.
They feared that giving any secrets to Britain would bring them to German
hands through capture. They also found the British ‘snooty, crusty, scorn-
ful and antagonizing’ [7], but after their long series of disappointments
in microwaves these thoughts did not prevent a full appreciation of the
importance of the cavity magnetron.
The second group held the view dominant among American scien-
tists, deeply stirred by Churchill’s speeches and the ferocious air battle then
being fought, that Britain had set her face against a monstrous tyranny
and must be given every possible support. And this new magnetron was
the circuit element they had dreamed about. The excitement of learning
how radar was saving the island from invasion carried them to a high
degree of enthusiasm for the work they saw cut out for them. The mil-
161
itary’s cold analysis of Britain’s situation, if anything, heightened their

zeal. They wanted to work with their colleagues across the sea and start
immediately! In this they were to succeed in an unparalleled manner of
international wartime technical collaboration.
The third group, the US Army Air Corps, formed the opinion that
American radar was significantly inferior to the British [8]. This came
about in part because of the British mixing of plans and achievements
coupled with the Army and Navy’s slowness in telling what they had, the
result of personal reluctance on the part of those attending and a slowness
of higher commands to free them. From this unfortunate muddle came the
demand by the Air Corps that the United States begin producing British
sets for American forces, a demand bitterly resented and resisted by the
Signal Corps.
Signal Corps and Air Corps officers soon visited Britain, but each
group saw what it had been led to expect. The former found CH poorer
than advertized, especially in its height-finding capability; the latter found
British air defense superb and reasoned it came about because they had the
best radar. Both were right in part but were unable to reconcile their points
of view [9]. These biases stuck for most of the war. The microwave group,
soon to become the MIT Radiation Laboratory, retained a very fruitful and
even loving relationship with the British, eventually opening a branch in
England for closer liaison. E G Bowen found the working atmosphere so
congenial and his memories of Rowe so raw, especially after having been
relieved of his airborne development duties, that he arranged to remain at
Rad Lab.
The NDRC people formed a relationship to the Signal Corps that was
at least correct but with NRL it became destructively poor with Admiral
Bowen singled out by NDRC as the cause. He had been strongly opposed
to the exchange of information and at times boorish about it [10]. The at-
mosphere did not improve until Rear Admiral Van Keuren replaced Bowen
as Director of NRL. Individuals learned to cross these hindrances, but they
did not disappear.
E G Bowen’s demonstration of ASV mark II mounted on a PBY flying
boat in December did much to dispel lingering American doubts of British
competence and led to the Navy ordering 7000 from Philco [11]. The les-
son was reinforced by a quick, unsuccessful attempt to put the SCR-268
transmitter and receiver with small external dipoles in a B-18 bomber [12].
For their part the British ‘were agreeably surprised to find that both
the US Army and Navy had progressed a good deal further in radar than we
had been led to expect’ [13]. They found the reception by the Microwave
Committee exhilarating. Here were men that understood all and could
hardly wait to start. The enormous success attributed to the Tizard Mission
rests almost entirely on microwaves, which changed warfare during the
next few years more than any single weapon. The enmities engendered
between the British and the American services labs were sufficiently well
162
New Ideas
suppressed to prevent open hostility—until Watson Watt’s visit in early

1942, but the Alliance was strong enough by then to sustain even that.
An indirect result of the Tizard Mission was the Signal Corps Electron-
ics Training Group, the suggestion of James B Conant with implementation
through President Roosevelt. It was composed of men with sufficient tech-
nical training and education to be commissioned directly as reserve officers
and sent after a minimum of preliminary training for a radar apprentice-
ship with the British. The demand for persons with such qualifications
was high both by the armed services and industry, a circumstance that
only became worse with time, but on 12 September 1941 the first group of
35 left Fort Monmouth for their assignment. Seldom in the course of hu-
man events have soldiers served in such a strange organization. They were
publicized yet secret, observers yet combatants (two killed before America
was officially in the war), receiving and giving orders to British. The loca-
tions of many were lost to American authorities for months [14]. They were
remembered in Antiaircraft Command by its commander: ‘From America
we got a wonderful bunch of recruits. America was not yet in the War, so
these scientists operated on our gun-sites, in the guise of civilians’ [15].
The item brought in the black box was a naked magnetron, no mag-
net, no power supply, no modulator, so demonstration had to wait until
suitable auxiliary equipment was available. That this heightened the dra-
matic impact did not trouble Tizard and Bowen. The test came on Sunday,
6 October at Bell’s Whippany Laboratories where the necessary magnet
and a modulated power supply for 10 kV were present. Bowen’s fears that
something might have happened to the tube in its travels were quickly
forgotten as the air around the output terminal glowed for an inch as a
result of the radiation [16]. Power and wavelength were quickly measured
with the power exceeding the advertized amount and the wavelength was
right on, 9.8 cm. The Bell Lab’s tube department was to manufacture 30
for NDRC immediately, which were ready in a month.
Microwave equipment quickly moved from the United States to the
United Kingdom: klystrons, Acorn tubes, lighthouse tubes, waveguides
and horns, all essential elements in building microwave radar.
What to do now? On the weekend following the dramatic test of the
new tube, a meeting took place at Tuxedo Park. Included among others
were Loomis, Bowen, Edward Bowles from MIT, Carrol Wilson represent-
ing NDRC and Ernest Lawrence, Loomis’s friend and a figure of eminence
in American science. There was immediate agreement that a new labo-
ratory modeled on the Air Ministry Research Establishment, soon to be
re-named yet again as the Telecommunications Research Establishment
(TRE), had to be founded. They undertook to discuss locations for the
imagined laboratory without deciding where and to discuss its staffing.
The British example of drawing on university research groups, not just
physicists and engineers but from all disciplines, brought immediate agree-
ment, and Lawrence was ready to start recruiting. That out of the way, the
163
objectives of the laboratory were fixed: (1) a 10 cm airborne interceptor (AI)

set, (2) a 10 cm gun-laying (GL) set and (3) an as yet unspecified long-range
radio navigation system. It was an exhilarating weekend to say the least,
and it all came to pass [17].
On 18 October a meeting with Bush at the Carnegie Institution in
Washington saw the organization of a Laboratory Management Commit-
tee, and the MIT Radiation Laboratory (Rad Lab) was about to be born,
the location of which had only just been selected [18]. Bush saw to it that
orders were placed for magnetrons from Bell Labs, permanent magnets
for them from General Electric, modulated power supplies from Westing-
house and RCA, 12 inch cathode-ray tubes and IF amplifiers from RCA,
parabolic reflectors from Sperry and various other components [19].
A decision was made to have the service labs continue their meter-
wave work, leaving microwaves to Rad Lab and Bell. The who, how or
when of this decision remains unclear. It just seems to have happened.
It may have been a collective decision compelled by its obvious wisdom.
Meter-wave radar was now a proven and very valuable weapon just enter-
ing production in the United States. Indeed it was to be decisive during the
next two years, and there would be an abundance of tasks for its designers
during these months. Microwave radar showed great promise but some
risk and would require the invention of a whole new series of techniques.
How much better to leave it to a fresh crew, unencumbered with concerns
about production, mobilization and design modification.
For whatever reason and however justified, the decision to entrust
the magnetron to a new, as yet non-existent laboratory did nothing to
remove the vexation of the service lab men. After delivering the country
first-class radar prototypes ready for production, done with trivial support
but much personal effort during the preceding years, they were now con-
sidered country cousins. That these numerous frictions were not allowed
to develop into nasty jurisdictional fights with attendant administrative
obstructions is a credit to those involved.
While all these events were unfolding the engineers at Whippany
thought about the wonder they had just seen but from a more practical
point of view. They were preparing the production of the CXAS fire di-
rection set that they had demonstrated to both Army and Navy about a
year earlier. There had been two versions, one for 40 cm, the other for
60. Power and with it range had dropped sufficiently in going from 60
to 40 that the Navy had elected the longer wavelength equipment for its
FA or mark 1 fire direction radar. The prototypes were modular in design
with a self-contained transmitter unit, which used special high-frequency
triodes. The way to improve this set was almost immediately apparent.
Re-design the transmitter unit for a magnetron at 40 cm and obtain better
directional resolution and increased range, not to mention a longer life for
the output tube, as the special triodes were used up fast. The magnetron
scales linearly with wavelength and made a manageable package for 40 cm
164
New Ideas
operation. All the other modules of the set would remain the same. As FC
or mark 3 it would be the first American magnetron radar [20]. It quickly
evolved into FD or mark 4, which had vertical as well as horizontal lobe
switching that made it the Navy’s standard AA fire control radar for the
entire war. And thus it came to pass that the first American application
of the device capable of making the long sought 10 cm waves required its
modification to produce waves four times as long.
Canada was included in the Tizard Mission, but almost as an af-
terthought. During the early discussions about whether to share technical
secrets with the United States, Canada did not fare well. Reluctance came
about not from fear that secrets would be lost to Germany, as some had
expressed as a concern for America, but that it simply was not worth the
effort.
Dr John T Henderson, Chief of the Radio Section of the National
Research Council, visited Britain on invitation in early 1939 to be given
‘information respecting a most secret device which they have adopted for
the detection of aircraft’. Henderson’s background prepared him for the
principles of radar but not for the extent to which it had been developed
as a weapon—without previous hints to Canada. During the winter of
1939–1940 a group of Canadian radio scientists began to approach this new
technique, and by March their number had grown to 22. Design details
and prototype sets from Britain that would have been extremely useful
were not forthcoming, so the Canadians began building from their own
designs using commercially available components. By June 1940 a 1.5 m
set for ship location proved successful in trials at Halifax and was given
the name of Night Watchman.
The Tizard Mission entered through Canada on its way to Washington
but primarily as a courtesy. It was only after they were actively engaged in
discussions south of the border that they learned of the extent of Canadian
progress and decided to deal with them as intellectual equals and gave
them the same information imparted to the Americans. The Canadians set
about designing a 10 cm gun-laying radar as their first priority, a device
destined to become GL mark 3C. (Mark 3B would designate the British set
and mark 3A the American, better known as SCR-584 [21].)
The Tizard Mission opened up liaison with scientists in more fields
than just radar, but it will be remembered more for the radar than for
any of the other matters. It was an overwhelming success for both sides.
Britain gained the much needed additional electronic manufacturing ca-
pacity along with a big jump in the knowledge of microwave techniques,
and the United States gained the magnetron in time to make good use of
it. The discovery might have been made in America independently, just
as it was made in Russia, Japan and Switzerland, but at the time of the Ti-
zard Mission none of the American microwave people seem to have been
thinking in terms of magnetrons because the klystron was exerting a strong
influence on their thoughts. One might rather expect that they would have
165
added an axial magnetic field to that device to allow much larger electron
beams, but who knows?
It came as a somewhat rude surprise for both parties to learn that
someone else had operating radar. In this encounter the British represen-
tatives were better prepared mentally. They had at least come to terms with
the necessity of disclosing their secrets weeks earlier and had their stories
ready. The Americans had learned of this high-level deal only shortly be-
fore the meetings, did not have immediate clearance to tell all and were
personally reluctant to let things out, so they were psychologically unpre-
pared for what was to take place. The result was an unfortunate split of
the American service-lab people not only from the British but also from the
Air Corps and the future Radiation Laboratory. It also gave rise to a repu-
tation they did not deserve of having produced inferior equipment. From
the Administrative History of the Office of Scientific Research and Development:
‘Historians may differ as to the reasons why with all of its remarkable
scientific advances the United States lagged so dangerously in the devel-
opment of weapons, but none will deny the fact’ [22]. It is hoped that the
fallacy of this statement is apparent to readers who have come this far.
4.3. THE RADIATION LABORATORY

Within most physicists there lies an engineer eager to design something.
The two branches of learning have always had a thin boundary separat-
ing them, one frequently crossed or dismissed. In principle the distinction
is simple enough: physicists discover nature’s laws, engineers synthesize
these laws into apparatus for good or evil. But in the day to day perfor-
mance of their trades the two are at times indistinguishable. An experimen-
talist designing equipment with which he plans to measure some atomic
quantity looks for all the world like an engineer designing some part of
an engine or radio; a theorist striving to describe a nuclear scattering pro-
cess with quantum mechanics looks like an engineer analyzing the stress
patterns of a bridge design. Names often confuse more than enlighten.
Designers of lasers manipulate atomic properties so as to produce some
form of radiation, operations no different from an electrical engineer de-
signing yet another form of high-frequency oscillator, but the former seems
invariably referred to as a laser scientist whereas the latter is an engineer.
Beauty is a common attribute for which both strive. There is never
a more satisfying achievement for a physicist than the reduction of some
complicated phenomenon to description by a simple, closed mathematical
statement. It is one of the misfortunes of our civilization that the beauty of
an equation cannot be shared with so many who specialize in the appre-
ciation of beauty. It is in beauty that engineer and scientist differ: for the
scientist it is nature’s beauty, for the engineer it is the beauty of creation.
Engineering is a form of art and has filled the world with things of obvious
visual beauty but also with subtle forms, much like a theorist’s equations.
166
New Ideas
A poet may enjoy the grace of the Brooklyn Bridge but will never appreciate
the charm of a well designed electronic circuit, but the beauty is there for
all that, and an electronics man will recognize grace, symmetry and style
in a design without ever having had a course in electronic art appreciation.
If ever there was a place where the latent engineer in the scientist burst
out, it was at the MIT Radiation Laboratory.
Lee DuBridge was Lawrence’s choice for Director of the new labora-
tory even before the formation of the Laboratory Management Committee
in Bush’s office on 18 October, and Lawrence’s choice met no opposition.
DuBridge had built a cyclotron at Rochester University that had delivered
its first beam in 1938 and formed the basis of a very active group in nuclear
physics. Recognition of DuBridge’s scientific and administrative skills was
apparent in his being Chairman of the Physics Department and Dean of
the Faculty of Arts and Sciences. Loomis and Lawrence persuaded him
to accept the post. DuBridge quickly selected another unrelated Loomis,
F Wheeler Loomis of the University of Illinois as his executive officer. Al-
fred Loomis was never a member of the Rad Lab staff, but he was always
in the background and made significant contributions.
Physicists not in the Microwave Committee were quickly added to
the roster. Kenneth T Bainbridge of Harvard was probably the first, and
recruiting was substantially helped by a meeting on applied physics at MIT
during 28–31 October attended by 600. A luncheon meeting at the Algo-
nquin Club in Boston introduced I I Rabi of Columbia, Edward U Condon of
Westinghouse, John C Slater and John G Trump of MIT and others to the op-
portunities of working at the Radiation Laboratory, and acceptances came
quickly. DuBridge presided over an organizational staff meeting on 11
November, attended ex-officio by Lawrence and Alfred Loomis, in which
eight sections were created: (1) pulse modulators, (2) transmitter tubes,
(3) antennas, (4) receivers, (5) theory, (6) cathode-ray tubes, (7) klystrons,
(8) integration. They chose sections like children picking sides for ball
games. Bainbridge wanted modulators, Rabi the magnetron until all sec-
tions were taken [1]. The various design projects were to draw on these
sections for support.
Recruits began arriving from far and wide: Luis Alvarez from Berke-
ley, Ivan Getting from Harvard, Norman Ramsey from MIT. Most were
nuclear physicists. As Alvarez remembered: ‘The first weeks at the lab-
oratory were like a family reunion’ [2]. The atmosphere was distinctly
that of Bawdsey with Bowen transferring tradition and spirit. Real work
started at once.
Project I, airborne interception, held E G Bowen’s attention fast.
Britain was experiencing the Blitz, the night attacks to which the Luft-
waffe had been forced after defeat during the day, and the bombers were
delivering their loads, inaccurately but with great indiscriminate damage,
and departing with few losses either to fighters or guns. A 10 cm AI set
seemed the most vital contribution toward stopping this. By 16 December
167
Rad Lab was beginning to take on some semblance of being a laboratory, so

a schedule was drawn up with AI the first goal. On 6 January 1941 a 10 cm
set was to be operating on the roof; on 1 February it was to be installed in a
B-18, the bomber version of the famous DC-3 transport, and a month later
in an A-20 attack bomber, the most likely candidate on hand for a night
fighter. It was, as intended to be, a hard-driving program.
The crucial components that Vanevar Bush had ordered on the basis of
the Microwave Committee’s suggestions allowed the roof-top equipment
to transmit its first beam in late December. Duplexing, or TR (transmit–
receive) switching as it became known among the microwave men, was
not available for this first unit, so two antennas were used, and by 4 Jan-
uary there were reflections from buildings in Boston. But if aircraft were
to be tracked from another aircraft, it was imperative that a single antenna
be used. They found that using a klystron as a pre-amplifier for the diode
detector prevented the transmitter pulse from burning out the diode; an
overloaded klystron did not transmit a damaging signal and it recovered
fast enough to function for the reflected pulses. This temporary expedient
allowed a common antenna but introduced tube noise at the input. Finally
by pointing the parabolic dish visually the hard-pressed crew observed
reflections from an aircraft at 3 km on 7 February 1941, an event so impor-
tant that the information was telephoned to DuBridge and Lawrence at a
meeting in Washington [3].
The problems with the roof-top unit were just about everything. Fre-
quencies were misaligned because of the lack of adequate test equipment,
there were no polyethylene cables and mismatched waveguides were pro-
ducing standing waves. A new engineering art had to be mastered, and
the most important teachers of this new art proved to be the radio ama-
teurs, with James Lawson especially remembered. By 5 March the roof-top
unit was delivering a creditable performance and was transferred to a B-18
with a plastic nose transparent to the radiation. On 27 March Edwin M
McMillan, E G Bowen and Luis Alvarez took off with their creation and
were rewarded by observing Cape Cod and numerous ships on their screen
[4]. After this success Rad Lab could proceed, if not in a relaxed manner
at least with assurance, toward other projects; by then the rolls listed 140
employees, six of them Canadian. There were to be many more roof-top
units, as all new equipment started by examining the Boston skyline.
A solution to the microwave TR switch problem came from Oxford’s
Clarendon Laboratories in spring 1941, the invention of A H Cooke. One of
Sutton’s klystrons was filled with water vapor at a low pressure so that the
electrodes of the resonant cavity, which were close together, would pass
a low-level signal picked up by the antenna but would form an immedi-
ate plasma discharge from the high power of the transmitter, creating a
short circuit that reflected the high power back, thus preventing it from
destroying the detector diode. It was Page’s duplexer transferred from
two-conductor transmission lines to waveguides. The results were suffi-
168
New Ideas
ciently encouraging to warrant making a special tube, and several kinds

of gas and of electrode configuration were tried. One problem was to have
the discharge begin quickly enough, because the diode was unforgiving of
even the briefest application of high power. To do this a small number of
ions had to be present always. Radioactivity was tried, but a small ambient
current from a filament proved best. An acceptable design, the CV43, took
lots of painstaking trials, but other versions were to follow as transmitter
power levels kept pushing the TR switch designers to new models [5].
Project I then began to undergo changes of direction and emphasis.
The Air Corps doubted whether the A-20 would make a suitable night
fighter and preferred designing the equipment for the P-61, an aircraft
specially designed for such service. A set was flown to Britain in June
1941 for comparison with their 10 cm AI equipment. The American set
had greater transmitter power, the British a more sensitive receiver, and
the return flight brought Rad Lab the much needed TR tube. Western
Electric made a few of these sets, designated SCR-520, by the end of 1942
but concentrated on designing a lighter version, SCR-720, for the P-61 [6].
By June there was much less emphasis on airborne interception. Ger-
many had attacked the Soviet Union, greatly lessening the night attacks
on Britain, and the attacking planes had begun to take severe losses from
meter-wave AI during the month before the greater part of the Luftwaffe
had been sent east. On the other hand U-boats were proving increasingly
effective as Doenitz’s ideas began to be taken seriously. The SCR-520 was
modified for this use and designated SCR-517. This set had only forward
scanning, and it was obvious that ASV radar should scan a full circle and
present the observations on a PPI scope. The set to accomplish this, the
Navy ASG and called ‘George’ by its users, was installed in blimps in
June 1942 and shortly thereafter in airplanes. It was to become the Navy’s
favorite ASV equipment [7].
Project II, gun-laying, was not nearly so high on the British list as AI
but struck a resonance among the Americans. First, it followed a direction
that the Coast Artillery Corps had pushed for over a decade; second, it was
a seed planted in rich soil for the development of such equipment and was
a project that quickly became self-generating. Out of it came the finest fire-
control radar set of World War II—SCR-584. The engineering was done by
Ivan Getting and Lee Davenport.
Getting was born in New York City into a family with strong Slovak
ties, his father being deeply involved in the interests of his countrymen.
With the formation of Czechoslovakia in 1919 his father accepted a position
in the new government, and the family moved to Bratislava only to return
when the senior Getting received a position in the Czechoslovak Embassy.
After seeing the postwar turmoil of Central Europe the mother decided
that the children were to remain in America. It was for Ivan the beginning
of an ever expanding range of experience that led him to nuclear physics
at Harvard by way of a Rhodes Scholarship. A useful skill learned along
169
the way came with a reserve commission in the Coast Artillery, giving him
first-hand knowledge of AA artillery. As one of Bainbridge’s students he
was one of the first recruits to Rad Lab [8].
Davenport’s background was quite different from Getting’s. It was
typical American training for a career in physics in the 1930s, one shared
by many of Rad Lab’s younger staff. He was born in Schenectady, New
York and grew up with a self-generated interest in all matters technical
with encouragement from his father, who taught high-school mathematics.
On graduating from high school at the bottom of the depression he got a
Federal Youth Administration job drawing the illustrations for a revised
edition of Kimball’s College Physics and then entered the local Union College
to study physics, which allowed him to live at home. Summer work at
General Electric paid his way. A teaching assistantship at the University
of Pittsburgh had brought him almost to a PhD when he was called to the
Rad Lab, where he immediately found himself on Project II [9].
In the first discussions the novice designers decided almost imme-
diately that the new equipment would incorporate automatic tracking, a
course neither expected nor desired by the higher ups, certainly not by the
members of the Tizard Mission. Reasons for their decision are not hard to
find. In 1940 MIT led engineering in servomechanism design, and these
ideas had been incorporated into the microwave equipment tried out for
Bowles’s blind-landing experiments. Barrow had even built a microwave
horn system used to track students moving about the campus [10]. The
opposition thought this approach would be too time consuming and pre-
ferred human interpretation of the signals. Automatic tracking proved nei-
ther difficult nor time consuming and was a most fortunate choice, doubly
fortunate because the Ordnance Department had incorporated automatic
tracking in the 90 mm AA gun that was replacing the 3 inch and because
Bell Labs was designing an electronic analog computer for predicting AA
fire that was to be much better than the mechanical computer. These were
decisions for which London would later be thankful.
A roof-top unit was the first step with the dish mounted on a servo-
controlled machine gun turret of the kind being manufactured for the then
developing B-29. Automatic tracking required something akin to lobe
switching, allowing the amplitude of two signals to be compared. The
solution was the conical scan generated by a rotating feed as used on Tele-
funken’s Würzburg. The roof-top unit was tracking aircraft on 31 May
1941 with remarkable skill, the principal uncertainty being the defect in-
herent in lobe switching or conical scanning caused by the ever changing
aspect of the target that yielded reflections of varying amplitude, giving
the tracking a nervous twitch. This was quickly removed by circuits that
produced running averages of the signals [11]. It was time to show off the
results.
A visit by Alfred Loomis soon brought the by-then Brigadier General
Colton to see the replacement for his beloved SCR-268. The effect on Colton
170
New Ideas
was dramatic. He was to provide the group with all the support it needed
from the Signal Corps, and Getting was just the kind of man to work
effectively with Colton. The General was so impressed that he decided
immediately that the Army had to have such equipment, but he was not
certain that Rad Lab would be able to deliver, so he immediately placed
an order with Bell Labs to design something similar as a back-up [12].
The next radical design step was to put everything in a closed truck,
designated XT-1 for experimental truck 1, with the dish lifted to the top
when in action and carried within for transport. This provided not only
mobility but gave the operators comfortable working conditions. They
would not be exposed to the weather as on SCR-268 or forced to keep
their head pressed against a light shield in order to see the oscilloscopes
traces during the day. In the production model a trailer replaced the truck.
Transmitter power and receiver sensitivity made important advances, and
the maximum range kept increasing until it reached 90 km. This allowed
it to function as a search radar so a PPI scope was added. On 6 February
1942 XT-1 was tested at Fort Monroe with the Bell Labs T-10 director and
a 90 mm gun firing on towed sleeve targets. They ‘shot down targets with
as few as eight rounds, all without human intervention or visual contact
with the target’ [13]. The Coast Artillery Board recommended the XT-1 be
procured as the standard gun-laying set with an allocation of one to each
AA-gun battery. On 2 April the Chief Signal Officer ordered 1256 units
[14]. XT-1 became SCR-584 and the Bell Labs director became the M-9.
Industrial production had problems of its own that even had Getting
confronting the president of the Chrysler Corporation about fabrication
of dishes and their high-precision drives. Production did not begin until
a year after the first order, and it was not until early 1944 that the first
set reached combat, delays in part attributed to a misunderstanding of
material priorities in the War Department that had their ultimate origin in
secrecy [15].
Even before reaching the battlefields SCR-584 began to attract a lot
of attention. Luis Alvarez saw it as the solution to the problem of blind
landing. It was not the solution, as we shall see, but its superior properties
put Alvarez onto the path that led to it. Its ability to track shells accurately
uncovered errors in the firing tables for the 90 mm gun, which detracted
from accuracy and were corrected for M-9 director, and the ability was
extended to determining bomb trajectories. At the front uses were found
for it never dreamed of during design: location of enemy artillery and
mortar positions by tracking projectiles, surveying islands by tracking a
photographing aircraft, accurately directing bomb release for a plane above
the clouds on near-front targets, detecting the movement of enemy vehicles
at night. A plan for countering the V-2 rockets was based on the 584, but
the launch sites were taken before it could be tried [16]. It was a masterly
design [17].
AI and gun-laying were the two microwave designs for which the
171
laboratory had been established, but if one were to ask what Rad Lab set
had the greatest effect on the course of the war and certainly if one asked
what set had had the greatest effect on the immediate postwar world, the
answer would have to be one that was not discussed in the excited meetings
that followed the Tizard disclosures and the founding of the Laboratory.
This was the SG, the naval set that Samuel Eliot Morison, historian but no
stranger to the bridges of warships, called the ‘greatest boon of scientist to
sailorman since the chronometer’ [18]. The antenna of its modern descen-
dant can now be seen rotating on the mast of almost any vessel traveling
the great and small waters.
It grew directly out of the first roof-top set for the same reason that
the Admiralty Type 271 went forward faster at HM Signals School than
airborne interception at Telecommunications Research Establishment—it
seemed easy and sensible. The B-18 equipment lacked two vital compo-
nents necessary for the SG to be an improvement over the 271: (1) the
TR switch to allow, without the degradation caused by a klystron pre-
amplifier, a common, rotating antenna and (2) the PPI scope. When these
elements were added the resulting equipment had much the same elec-
trifying effect on seamen as the installation of the radar beacon for ASV
mark I had had on airmen of Coastal Command. It presented to the watch
officer a map of the area surrounding his ship showing coastlines, harbor
markers and other ships. It greatly simplified the difficult problem of keep-
ing station for convoys and fleets running in blackout and foul weather. It
became the indispensable navigation aid for landing troops in an invasion
where the approach had to be made under cover of darkness. It allowed
a surfaced submarine to be picked out in the clutter of a convoy far more
easily than with the type 271. It allowed vessels to run for the first time
(legally) at full speed in fog. In the tangled naval actions of the southwest
Pacific it removed the confusion the meter-wave sets had when operating
where ships and islands appeared much the same on their A-scopes. All
this became apparent when the SG was first used. It was never planned to
be such; this set was an afterthought.
The first model of what was to become SG was mounted in May 1941
on USS Semmes, a four-stacker destroyer working out of New London.
The first sea trials of the device on 5 June caught the vessel in fog on her
way back to port, but a sea-sick Ernest Pollard picked out buoys with his
radar and piloted for the skipper, Lieutenant Commander W L Pryor, who
made an excited telephone call to Washington on docking [19]. One of the
first things learned was that the dish needed to be stabilized because the
roll of the ship—and the four-stackers were unsurpassed in their ability
to roll—caused the relatively narrow radar beam to rise above the surface
or to examine the near ocean too carefully, depending on the orientation
of the mast. Stable verticals use servomechanisms to maintain a direction
established by a gyroscope, and this solution was tried but rejected in favor
of the cut paraboloid, which was wider than tall and formed a beam 5◦ wide
172
New Ideas
and 15◦ tall, a vertically extended fan, and so the design has remained [20].
The Semmes gained a quasi-permanent laboratory crew and cruised
about the eastern seaboard for much of the year as successive alterations
were made in the design. At the insistence of Commander Pryor the final
form of the set remained aboard the Semmes until replaced in March 1943 by
a production model of SF, the destroyer model of SG. The Naval Research
Laboratory made the design sea and battle worthy and Raytheon produced
it. The first production set of SG was installed aboard USS Augusta on 5
April 1942. It reached the Solomon Islands, where it became an instant
favorite, in October.
These three sets, ASG, SCR-584 and SG, and their descendants form a
solid basis for Rad Lab’s high regard. They were created during the heroic
period in ways similar to the heroic years at Bawdsey. Just as Bawdsey
grew into the large, well run engineering laboratory at Malvern, so Rad
Lab became a large, well run engineering laboratory at Cambridge. Its
appropriations for 1941 were about five times what the service laboratories
had spent on radar before 1940, and by 1945 it spent as much in a day as
the service labs had spent annually during those austere years. The total
cost for the war was $142 million. Personnel reached a peak only slightly
less than 4000, and building floor space grew accordingly.
An Army Air Corps detachment stationed at East Boston Airport
began flying for the lab in July 1941. They outgrew this base and a new
one was created from scratch at Bedford, opening in May 1944. By the end
of the war 95 aircraft made up Rad Lab’s private air force. In addition to
the shops for building experimental equipment there was the Model Shop
operated under contract by the Research Construction Company, Inc. that
made small production runs for immediate military needs that could not
wait on industrial contract. There were field stations far and wide, and Rad
Lab men gave expert help to those at the front. To strengthen the ties with
their colleagues overseas the British Branch of the Radiation Laboratory
was established in September 1943 at Malvern where the TRE had taken up
final residence, ensuring a steady flow of Americans to the places where
the equipment was being used to advise and learn.
The Radiation Laboratory was a big business within a bigger busi-
ness. By the end of the war the United States had spent $2819 million on
radar, 48% being for equipment of Rad Lab design. The number and kind
of radar that had appeared on the military scene by 1945 was staggering
to anyone who had known radar in 1940. Few of those interesting sets can
be treated in a book this restricted and many came too late to affect the
outcome of the war in any important degree [21].
DuBridge directed the laboratory in a relaxed manner, but it was a
relaxed dictatorship, for it was he who made the decisions. He formed a
Steering Committee for guidance, eventually made up of about 20 persons
he thought best capable of helping him. He relied heavily on them and
seldom introduced his own technical ideas. It generally met at monthly
173
intervals on Saturdays at 1 pm, just when the normal work week con-
cluded. It was in the Steering Committee where policy, jurisdictions and
contentions were hammered out and where long-range plans were made
[22]. Positions on the Committee were much sought after and filled by
competitive men. Pollard remembered the sessions as very rough after-
noons. ‘I was worn down more by the Saturday afternoons, which used to
go from one until six, than the whole rest of the week’ [23]. The Steering
Committee differed from Rowe’s Sunday Soviets in having as its objective
the making of decisions rather than the exchange of ideas. Only rarely
were outsiders present.
By spring of 1943 Robert Oppenheimer began approaching nuclear
physicists at Rad Lab to join his Los Alamos staff. The attraction of a new,
mysterious project having closer ties to their scientific lives coincided with
a feeling that the Laboratory had fulfilled its most important work and the
details could now be entrusted to other, respected members of the staff.
Bainbridge, Alvarez and many others thus experienced two astounding
engineering laboratories within five years. Few of their coworkers knew
what was going on, but they noticed that people started to disappear.
Unlike Los Alamos the Radiation Laboratory shut down soon after
the end of the war, officially terminated on 31 December 1945. As a final
task Rabi had insisted that they make a record of the technology they had
learned, in part out of concern about possible future Congressional inves-
tigations as to what the country had obtained for the money spent. Writing
began in fall 1944 under the editorship of Louis N Ridenour and produced
28 volumes [24] from 49 authors (not including Rabi, who did not write
books), most of them protesting strongly about this waste of time when
men were dying at the front [25]. But it was everything but a waste of
time. Volumes from this electronics encyclopedia would be found on the
bookshelves of almost every electronics engineer and experimental physi-
cist for more than a generation, indeed some are found there today and
not just as mementoes. Adding to the impact of this injection of knowl-
edge into the postwar world was the return of the staff to civilian tasks,
either the academic ones they left or the industrial ones many had learned
to like. The effect on America of this influx into physics departments and
electronics companies is difficult to judge, but it was extremely large.
4.4. THE PROXIMITY FUZE—THE SMALLEST RADAR2

By 1938 Britain’s air defense problem was acute. Almost all of her fighters
were biplanes that were slower than the monoplane bombers that Germany
was building, an impossible situation for the defense. Hurricanes and
Spitfires were entering production and CH was advancing rapidly, but air
defense cried out for new ideas. One such idea was the ‘bomb the bombers’
2 Based on ‘The Proximity Fuze’ by Louis Brown that appeared in IEEE Aerospace and Elec-
tronics Systems Magazine Vol 8, pp 3-10, 1993. Copyright IEEE.
174
New Ideas
scheme. A number of relatively light bombs dropped from above might

have hope of doing damage, if a direct hit were not required. A nearby
burst could damage the bomber much as an AA artillery shell might, but
time fuzes such as used in AA shells required rapid measurements and
calculations impossible for air crew members; a fuze was required that
sensed the presence of the bomber [1].
Professor P M S Blackett, a distinguished nuclear physicist from
Rutherford’s Laboratory and veteran of the Royal Navy in World War I,
proposed a fuze based on the photoelectric cell in a memorandum to the Ti-
zard Committee on 7 July 1937. The idea was discussed at a meeting of the
Royal Aircraft Establishment on 22 October in which the idea of acoustical
triggering was injected. By May of the next year ground test results were
encouraging, but premature detonations could be caused by the fuze look-
ing at the sun or clouds. Tests made by dropping fuzed bombs on balloons
in March 1939 were at best a modest success, and the Tizard Committee
recommended 500 be manufactured for service trials. Disagreements and
misunderstandings marked the next few months with service tests called
off at the outbreak of war, it being proposed to substitute trials against the
enemy. Matters continued without anyone insisting on a fixed goal. Use
of the fuze against ground and sea targets was also pushed; using rockets
instead of bombs was tried. By mid–1940 there was little hope for this
approach.
The Air Defense Experimental Establishment experimented with an
acoustical fuze and went through a similar series of tests. It had a Rochelle-
salt crystal microphone incorporated in specially shaped tail fins. Tests in
August 1939 showed directive response to sounds of frequency above 5
kHz, but only three out of 18 detonated correctly. Rockets were tried with
no mentionable success.
Into this stepped W A S Butement, designer of radar sets CD/CHL
and GL, with a proposal on 30 October 1939 for two kinds of radio fuze: (1)
a radar set would track the projectile, and the operator would transmit a
signal to a radio receiver in the fuze when the range, the difficult quantity
for the gunners to determine, was the same as that of the target and (2)
a fuze would emit high-frequency radio waves that would interact with
the target and produce, as a consequence of the high relative speed of
target and projectile, a Doppler-frequency signal sensed in the oscillator.
Discussions with E S Shire and A F H Thomson yielded a simple design of
a continuous-wave oscillator capable of responding as desired [2].
William Alan Stewart Butement came from a pioneer New Zealand
family. He was born in 1904, educated in Australia and England, graduated
from London University with a Bachelor of Science Degree, and joined the
Signals Experimental Establishment of the War Office. His early experi-
ments with P E Pollard in radio location, which resulted in a design having
all the elements of an elemental radar, are mentioned in Chapter 2.3. That
he did not continue the bent so clearly disclosed was the immediate result
175
of the lack of a transmitter of sufficient power at the 50 cm used and the

more enduring lack of any real interest at the War Office. One of many idle
speculations is what the history of radar might have been had an enlight-
ened attitude toward scientific research given Butement support [3].
One of the disclosures of the Oslo Report3 was the knowledge that
the Germans were working on a proximity fuze; a tiny vacuum tube from
the project was even packed in the envelope containing the report. The
design attempted to utilize the change in electric capacitance between a
nose electrode and the shell body when some object came near. The Ger-
man work did not proceed to any useful result but made a small ripple
in England. The author of the Oslo Report was a Siemens und Halske
technical expert, Hans Mayer, who was a friend of an English instrument
maker and entrepreneur, Cobden Turner, owner of Salford Electrical In-
strument Company. In the summer of 1939 he and a few of his engineers
visited Siemens und Halske on business and got a hint of the fuze work.
On returning home they designed a radio-influence fuze and even tried it
in some bombs [4].
Unfortunately, Britain had too many serious problems to deal with.
Butement was heavily involved in designing radar for the Army. There
was no time to be spent on a device that showed so little promise. The
proximity fuze needed development by forced march and was proceeding
at a stroll. Fortunately, by the summer of 1940 others in America, less
pressed, began considering the problem.
While working for the newly formed National Defense Research
Committee (NDRC) C C Lauritsen, a nuclear physicist from Caltech, noted
in July 1940 that the Western Electric Company and RCA were manufac-
turing 20 000 thyratron and photoelectric tubes for the British Army [5]. A
thyratron is a vacuum tube filled with low-pressure gas and has an elec-
trode configuration like a triode. Unlike a triode it conducts only negligible
current for low-level signals, but goes into a plasma discharge once the grid
voltage exceeds a certain threshold value, thereby passing a large anode
current. It is an electronic switch. The combination of the two types and
the specifications for them brought a quick guess that something in the
nature of a proximity fuze was being made.
This information soon became an item of discussion between Van-
nevar Bush, President of the Carnegie Institution and Chairman of the
NDRC, and Merle Tuve, a physicist at Carnegie’s Department of Terres-
trial Magnetism (DTM) located in Washington. Tuve was already well
known to people at the Naval Research Laboratory through his invention
with Gregory Breit of ionosphere sounding with radio waves. Since 1927
he had worked to build a particle accelerator for nuclear physics, suc-
ceeded in adapting the Van de Graaff generator to that end, and by 1940
had created one of the major centers of experimental nuclear physics in the
176
New Ideas
United States with three Van de Graaff accelerators operating and a 60 inch
cyclotron under construction. Tuve and his colleagues at DTM were very
concerned about the war and wanted to get into war work immediately.
Bush had taken a quick liking to Tuve and brought him into discussions
of defense matters from the start. After discussions with naval ordnance
Bush formed Section T (for Tuve) on 17 August to work on a proximity
fuze at DTM [6].
Thus in mid-August 1940 Tuve asked Richard Roberts whether he
thought a vacuum tube could stand an acceleration of 20 000 g, and received
a tentative answer of yes the next day. Roberts mounted an obsolete tube,
a number 38, on a lead brick that he suspended from the ceiling and then
fired a bullet at the brick, the oft repeated experiment that demonstrates the
conservation of momentum to students in introductory physics and that
had completely altered the scientific study of guns two centuries earlier.
The tube still worked, and calculation showed it had briefly sustained
an acceleration of 5000 g. The next day Roberts mounted a tube on a
hemisphere of lead and dropped it from the roof of a three-story building
onto a steel plate. The indentation of the lead allowed an estimate of the
acceleration, which was even higher than before, and the tube still worked.
The fuze project was under way [7].
In fact, Tuve and Roberts were already on a war project, for both were
on President Roosevelt’s Advisory Committee on Uranium. In January of
the year before, Roberts had demonstrated fission in a startlingly simple
experiment to Niels Bohr, Enrico Fermi, Edward Teller and Gregory Breit,
who were attending a scientific meeting (on low-temperature physics!) in
Washington at which the knowledge of this new nuclear process had got
out. Roberts continued to work on fission and subsequently discovered
delayed neutrons, which allow fission to be controlled in a reactor, but the
events of the spring and summer of 1940 brought the men at DTM to the
viewpoint that an atomic bomb would come too late to affect the outcome
of the war. One of the DTM staff, Norman Heydenburg, continued making
measurements for the uranium project until all such work was transferred
to Los Alamos and construction of the cyclotron continued, but most of
DTM went to work on the fuze. Other thoughts may have been in Tuve’s
mind. When asked about leaving the bomb project years later he said: ‘. . .
and I didn’t want to make an atomic bomb’ [8].
Tuve and Roberts made interesting contrasts. Tuve was the son of
Norwegian immigrant grandparents who had settled in a small town of
South Dakota. He and his childhood friend Ernest Lawrence had linked
their houses with a telegraph line, replaced with wireless sets when Ernie’s
family moved. Both went on to build pioneer nuclear physics laboratories.
Roberts traced his lineage to colonial roots, had financial independence
with origins in Pennsylvania oil and had gone to the best schools. The two
of them guided DTM for four decades with a scientific leadership that kept
them active laboratory partners of their colleagues. They were implacable
177
Circuit for the proximity fuze. This is a slight modification (prescription E) of

the design of W A S Butement. The dc component of the plate current of the free
running Hartley oscillator was altered by a change in radiation resistance when
a conducting object came within a few wavelengths, modulating it by the motion
of the projectile and forming a signal that passes through a low-pass filter into a
two-stage audio amplifier. When the output of the amplifier exceeds a fixed level,
the thyratron conducts, discharging the semi-cap. Archives of the Department of
Terrestrial Magnetism, Carnegie Institution.
enemies of big science.

Roberts’s first experiments obviously called for shooting vacuum
tubes out of a gun, so the machine shop made a small muzzle-loading
smooth bore, which was taken to a farm owned by a friend of Tuve’s in
what is now the Virginia suburb of Vienna. The gun was pointed straight
up, a projectile with a small tube potted in wax loaded, and the gun fired.
178
New Ideas
And failure! Although the glass envelope had survived, the electrodes col-
lapsed completely. Navy ordnance experts suggested that they try again
using smokeless instead of black powder, which explodes instead of burns
and gives much higher initial acceleration than smokeless. A 37 mm gun
of 1916 vintage was procured, and tubes began to survive. For the next few
months projectiles were fired, sometimes hundreds a day, testing tubes and
other components. Initial nervousness of the experimenters about where
the shots would land was soon replaced by confidence on learning that
they could predict the point of impact within less than 100 m.
While the tests to determine whether electronic components could be
fired were being conducted, the Tizard Mission arrived in Washington, and
on 14 September R H Fowler and John Cockcroft had dinner at Tuve’s home,
open exchanges of information about fuzes soon following. The Americans
had not settled on the method of influence yet and were examining the same
methods the British had. Lawrence Hafstad worked on a photoelectric
method, and G.K. Green on an acoustic. The electronic circuit designed
by Butement, Shire and Thomson for a radio proximity fuze was extracted
from Tizard’s famous ‘black box’, and Roberts, who brought the additional
skills of an electronics enthusiast as well as a reserve officer of Field Artillery
to the project, had the circuit working in the lab in a couple of days. The
basic circuit remained unchanged throughout the project [9]. The anode
resistor of a Hartley oscillator was connected through a low-pass filter to a
two-stage audio amplifier connected to a thyratron, which passed current
through a detonator when its grid voltage exceeded a given threshold. Just
four tubes!
The laboratory circuit, tuned to 100 MHz, worked beautifully. The
thyratron output responded sensitively to the motions of a half-wave
dipole anywhere in the room. Roberts’s brother, Walter, a radio engineer
who had helped design the oscillator for the DTM cyclotron, worked out
the theory of the thing and found that if the target came within a few
wavelengths of the oscillator, it altered the loading of the antenna, thereby
changing the direct component of the anode current, which varied at a rate
determined by the relative motion of target and projectile. This amplified
and filtered signal triggered the thyratron. Doppler was not really needed.
It was an elegant design.
With evidence that vacuum tubes could be fired from guns and that
a simple electronic circuit could be made to trigger the explosion it was
obvious that a greatly expanded project was needed. Vacuum tube manu-
facturers had to begin furnishing prototype rugged tubes while preparing
for mass production. Batteries presented particular problems. Circuit and
mechanical design had to proceed toward a usable device, and a greatly
expanded testing program undertaken. All this required an increased staff,
which quickly had over a hundred persons working in a building that had
housed only a dozen a few weeks earlier. Tuve put out a set of rules, the
first of which was: ‘I don’t want any damn fool in this laboratory to save
179
money. I only want him to save time’. For those who had experienced
Tuve’s frugality before or after the war this was a startling rule.
In October fuzes made of non-rugged components in non-miniature
circuits for both the radio and photoelectric fuzes detonated 100 lb bombs
dropped at the Naval Proving Ground, Dahlgren, Virginia. At about this
time Tuve decided that fuzes for non-rotating projectiles presented differ-
ent kinds of developmental problem and turned the work on bombs and
rockets over to the Bureau of Standards under the direction of Harry Dia-
mond, who continued to work on the photoelectric method but dropped
the acoustic as impractical. The DTM group dropped all methods except
the radio fuze.
In February 1941 tubes were fired in 5 inch star shells with the
parachute intended to lower the flare being used to bring down the com-
ponents tested. On 20 April 1941 an oscillator was shot from the 37 mm
and observed to function, and about two weeks later seven oscillators were
fired from a 5 inch gun at Dahlgren, four being heard in flight. An oscillator
with a modulator to calibrate microphonics generated in flight disclosed
no such problem. It was time to make complete fuzes.
The small size of the 37 did not allow the firing of complete fuzes, so
the vertical firing was transferred to a 57 mm at Dahlgren. This gun had
not only a larger shell but a higher muzzle velocity. The Dahlgren firings
were enlivened by the caretaker’s dog, who raced into the river with each
shot, expecting that such a powerful gun would bring down plenty of
ducks, and who needed weeks of duckless firing to learn that the hunters
were incomparably bad shots. Firing became routine for testing prototype
industrial tubes as well as production lots. A more exciting aspect of the
Dahlgren firings was a poorer ability to predict where the shots would land,
the consequence of them ascending to much higher altitudes through more
complicated wind patterns. One landed completely out of bounds, a mile
from the gun.
Numerous tube manufacturers entered the competition, but Sylvania
proved most successful. Its T-3 tube weighed less than three grams. One
must remember that small-sized electronic components so common today
were not so much admired in 1940. It is also worth noting that the entire
US production of vacuum tubes in the last peacetime year was 600 000 per
day. By 1945 the production of tubes for proximity fuzes was 400 000 per
day with 95% from Sylvania.
The first batteries were specially adapted dry cells furnished by Na-
tional Carbon, but they soon showed serious shelf-life problems and were
replaced by wet batteries that had indefinite life with the added advantage
of being activated only at the firing of the gun. A sealed glass ampule con-
taining acid was placed within a stack of annular discs. One side of each
disc was zinc, the other carbon. On firing the glass ampule shattered and
the acid was flung into the plates by centrifugal force. If its shelf life was
long, its active life was short, about two minutes, just long enough for the
180
New Ideas
flight of any proximity-fuzed shell.

A fuze was mounted in a 5 inch shell with a hole cut in the side for
an ammeter in the anode current circuit. With this the radiation pattern of
the little transmitter driving a dipole formed by a cone-shaped electrode at
the nose and the shell body, was measured. The next step was to fire pilot
production fuzes in 5 inch guns at Dahlgren, which took place in August
1941. On 29 January 1942 the success rate at Dahlgren exceeded 50%, and
full production started while the bugs were still being removed. Unfor-
tunately, removing bugs did not mean they would stay removed. One of
the greatest problems in producing fuzes proved to be quality control at
all levels. It was a never ending problem, and there was no let up. A dud
rate no greater than 5% was sought, but it was hard to attain.
The introduction of the high-explosive shell at the turn of the century
brought an awkward period during which guns exploded on firing from
time to time, owing to imperfections in the fuzes. Improvement in design
soon made the simple impact and time fuzes bore safe, but the proximity
fuze obviously had many more ways to fail. Tuve was determined that his
race against time was not going to result in dead gunners, so a major effort
went into safety devices. The explosion was initiated by a detonator that
was activated by some tens of milliamperes, so the first line of safety was
to keep it shorted until the projectile was clear of the muzzle. A clockwork
located in the base of the fuze and actuated by projectile spin removed
a short circuit and a mechanical gate in the powder train half a second
after firing; it was eliminated in later models. The wet-cell battery also
helped by requiring a tenth of second to come up to voltage. A mercury
switch functioned in two ways. Before firing the mercury resided at the
center of a porous cylinder located slightly off the projectile axis where
it effected a second short. On firing, centrifugal force spun the mercury
through the porous material thereby opening the short and closing a switch
that activated the electrical components with a delay determined by the
diffusion time through the diaphragm. The thyratron and the last stage
of the audio amplifier, which operated in the range from 30 to 300 Hz,
were initially biased to cutoff and became active only after a time delay
determined by a capacitor charging time. Finally, the presence of the gun
tube so loaded the antenna that the oscillator would be quenched while
within the gun. Thousands of rounds with only one operable safety and a
reduced charge of black powder were fired to evaluate each separately.
Firing at air frames suspended from balloons and from towers at
the New Mexico Proving Ground by H R Crane and David M Dennison
measured the burst patterns, which could then be adjusted with the only
available parameter, the sensitivity. If the sensitivity were too great, the
shell would burst too far from the target; if it were too small, the shell would
burst close enough to assure the target’s destruction but allow many possi-
bly damaging rounds to pass by. These tests were all made with explosive
charges just great enough to permit photography, otherwise target replace-
181
ment would have become a major waste of time. After these successes it
was time for the critical test: firing from a ship at radio controlled targets,
called drones, under routine service conditions. The tests were made on
the shake-down of the new cruiser USS Cleveland in the Chesapeake Bay
on 12 August 1942. Roberts was aboard and later recorded the event.
The next day all was ready off Tangier Island and a drone ap-
proached on a torpedo run. At about 5000 yards the ship opened
fire with all its 5 inch guns. Immediately there were two hits and
the drone plunged into the water. Commander Parsons called
for another drone and out it came on a run at about 10,000 ft
altitude. Once again it came down promptly. Parsons called for
another and then raised hell when the drone people said there
were no more ready for use. He enjoyed this very much as he
had been on the receiving end of a lot of comments by the drone
people in other firing trials. The drone operators had one back-
up drone ready in case of troubles but they never expected to
have one shot down. In fact the Navy photographic crew who
took pictures of all the firing trials of the fleet had never seen
a drone shot down before. The ship was ordered to the Pacific
with no stops, as the crew had seen too much [10].
As the Cleveland was not to dock on her outbound voyage the techni-
cal personnel were loaded into a launch to take them ashore. In a somewhat
humorous gesture the skipper gave his evaluation of them when he pre-
sented each a life preserver as they descended to the small boat, which
naturally had a normal supply of such articles.
It is ironic that this test, which showed that a warship could defend
itself very well against air attack, took place less than 100 miles from the
location where, some 20 years before, Mitchell thought he had proved that
surface ships were obsolete as a result of air power.
By mid-November 1942 about 5000 rounds were on the way to Pearl
Harbor of which 4500 were sent to the South Pacific on USS Wright. At
Noumea they were distributed by Vice Admiral Halsey to the ships con-
sidered most likely to see action. On 5 January 1943 USS Helena, on her
way back with two other cruisers and two destroyers from an attack on
an airstrip on New Georgia the day before, shot down a Japanese plane
with a shell equipped with an industrially produced fuze [11], less than
30 months after the first discussions at the newly formed NDRC about the
need for such a device.
The security surrounding the device was extreme. Early models were
called ‘T3G Device’ and all shipments were guarded by Marines and signed
for by a commanding officer. Afloat and ashore they were kept under lock
and key, and on arrival at port no one was allowed to leave the vessel until
the fuzes were accounted for [12]. In production it became ‘mark 32’, and
in the summer of 1943 called the ‘VT’ after British suggestions meaning
182
New Ideas
‘variable time’ or ‘velocity triggered’. The proximity fuze may have been
enveloped in extreme secrecy, but it was the subject of enough rumor by
the time of the Battle of the Eastern Solomons to be mentioned in the after-
battle reports, the same month as the Cleveland trials in the Chesapeake
Bay [13].
The supreme driving force behind fuze development was its use
against aircraft, but once this problem was solved thoughts naturally pro-
ceeded to an older problem of the artillerist: air bursts against ground tar-
gets. The first explosive artillery shells, which introduced the term ‘bomb
shell’ into the language, had used powder-train fuzes. If this fuze were cut
short, it led to ‘bombs bursting in air’. With better fuzes and more accurate
guns this had been refined by General Henry Shrapnel of the British Army
into a shell filled with lead balls and that burst in the air with devastating
effect on exposed infantry. After World War I shrapnel had been replaced
by the high-explosive shell that did its killing with jagged shell fragments
instead of lead bullets, but the time fuze remained. Up to 15 seconds flight
time could be obtained with a powder-train fuze, 25 with a clockwork fuze.
With flat trajectory guns at moderate ranges and observed fire these could
be effective. At long range, at night or in fog, or unobserved, time fire was
almost useless. Use of the proximity fuze was obvious.
The Field Artillery had gone over to howitzers to a large degree, and
they presented a few problems. They never had the high muzzle velocities
of the AA guns and even had a variety of velocities from which to choose,
determined by the amount of propelling charge loaded. Varying muzzle
velocities meant varying spins, and spin operated the safeties. Thus high
acceleration, that horrible problem in the summer of 1940, became a ne-
cessity. It was soon decided that only the top three powder charges for
howitzers would be considered. Fuzes were soon ready.
The army equivalent to the Cleveland firings was a demonstration to
the Field Artillery Board at Ft Bragg on 24 and 25 September 1943 with
Lieutenant General Leslie McNair, Chief of Army Ground Forces in atten-
dance. It was fouled up, yet a stunning success. The fuzes for different
caliber weapons were mixed up at the gun positions causing up to 30%
duds and bursts at the wrong heights. The Section T men were frantic, and
it showed. The Board was so startled to see air bursts at extreme ranges,
air bursts unobserved, air bursts with high-angle fire (shells descending
almost vertically), air bursts at night that its excitement was almost un-
controlled. When the fuze men went on about the performance, McNair
answered: ‘Gentlemen, you want all this and the moon too?’ [14].
The account of the story at this point does not convey a proper picture
of what had been going on. Tuve’s objective was a weapon to be placed in
the hands of the warriors—and soon! This meant that production had to
be brought in early, well before designs were final, and the entire project
grew at an incredible pace. The early fuze work had more than 40 industrial
and academic contractors, and Canadians helped with battery design. The
183
year 1940 was a good time to place orders because industrial mobilization
had just started and there was plenty of slack yet to be taken up. In April
1942 Section T had outgrown the space at the Carnegie department and
moved to a large building on Georgia Avenue in nearby Silver Spring,
Maryland. At that time the Carnegie Institution transferred administrative
control to Johns Hopkins University, and the newly established unit was
named the Applied Physics Laboratory. By the time of the Cleveland firings
production of fuzes was already beginning. Needless to say, this gamble
brought on no small number of emergencies. Strange infirmities would
appear, in a product that had a built-in bias against diagnosis, yet diagnosis
was demanded immediately. US and British forces had between them 40
different kinds of shell for which the fuze was required, and each had to
be individually fitted.
Secrecy had adverse effects in complicating procurement, and curious
ways were found to conceal the true function of various components. The
plastic noses were ordered through Johns Hopkins Medical School under
the name of ‘rectal spreaders’. Worse, because they were not told what they
were making, workers came to believe it was not important, and to keep
from arousing curiosity, fuze plants were never given the Army-Navy ‘E’
for excellence flag. In a product requiring high quality control, this was a
definite embarrassment [15].
By the end of the war 112 companies were engaged in production
work on fuzes and more than 22 000 000 had been manufactured with the
price eventually falling to $18. As a wartime project it was exceeded in
magnitude only by the bomb and what we might call ‘large-set’ radar. Yet
the entire project was directed to the end by Tuve, who controlled both
the technical and the business aspects and who before 1940 had never
supervised more than half a dozen persons.
The first wide-scale employment was in the Pacific, in part because it
was more the Navy’s weapon than anyone else’s but mostly because fleet
use gave the smallest probability of one being captured. Section T was
well aware that the first danger from a fuze falling into enemy hands was
jamming, and recovery of just one of the all too many duds could give
the whole thing away. Jamming really meant causing premature bursts
and could be effected by sweeping a high-frequency oscillator through
the frequency band of the fuzes. When the frequency of some electronic
device interfered with that of the Hartley oscillator of the fuze prematures
did occur. On Okinawa 105 mm howitzers using the fuze had to stop using
them because of bursts all along the trajectories, bringing severe protests
from the infantry. The cause was determined to have been the meter-
wave radars of nearby destroyers [16]. (This explanation is questionable.
There are no Navy reports of fuzes being set off at sea, where the radars in
question would have had excellent opportunity. If radar was the cause of
prematures, the source is more likely the SCR-270, which was present on
land, as its frequency band of 100 MHz was that of the fuze, and its 100 kW
184
New Ideas
peak power and pulse repetition rates of 200 to 400 Hz were well suited to
trigger the fuze.)
The effect on naval action was immediate. Each naval air engagement
saw the new weapon playing an ever greater role, culminating at the Battle
of the Philippine Sea on 19 June 1944 and in the defense against the suicide
pilots, principally at Okinawa. The first use in the European theater was
again naval, during the invasion of Sicily.
The most spectacular triumph of the fuze was in the defense against
the flying bombs, but it was a triumph shared with the gun-laying radar
SCR-584 and the electronic director M-9. It was crucial in the Battle of the
Bulge, where it was used to devastating effect against infantry advancing
in fog.
The German proximity fuze work continued in fits and starts.
Siemens und Halske dropped work on the original balanced-capacitance
fuze, but others took up the task later in the war when large AA rockets
were being designed at Peenemünde, the location of the V-2 rocket-bomb
development. A proximity fuze was necessary for these, but it was not
subject to the severe constraints of space and shock resistance imposed on
the fuze of an artillery shell. The work was directed from Peenemünde
West and was both of local and contracted origin. Four fuzes were under
simultaneous development: Kranich, Kakadu, Marabu and Fox.
Kranich was a purely acoustical fuze that had a resonant cavity di-
mensioned to the principal frequency of heavy-bomber motors. A wire
whip was fastened to a diaphragm that formed one wall of the cavity.
Vibrations set the whip in motion, causing it to touch a ring electrode
and close the firing circuit. Kakadu and Marabu were 50 and 70 cm
continuous-wave transmitter–receiver pairs with separate tuned-dipole
antennas. Kakadu made use of the Doppler shift in the reflected wave,
Marabu a frequency-modulation effect. Fox operated on 3 m and was sim-
ilar to the Butement design in using the alteration of antenna loading. It
was not small, having a dipole antenna.
These competing designs were tested at Peenemünde by fastening
them to long poles mounted on a wooden tower that held the necessary test
equipment. The individual designs caused lights to flash when actuated,
and their spatial relationship to the aircraft that flew over the poles was
recorded on film. The end of the war prevented the work from proceeding
further [17].
Britain also continued fuze work. By November 1940 GEC furnished
miniature pentodes that withstood the shock of firing, and in August 1941
a shell fired from a gun was detonated in the air by a radio pulse from the
ground. This approach continued, and in February 1942 yielded a test in
which 75% of the shells were burst by signals from the ground. By October
1942 a report to the Prime Minister placed the American efforts well ahead;
work continued nevertheless, as it was not clear that US fuze production
would suffice for Britain as well as for the extraordinarily hungry Pacific
185
fleet. When it became clear that Britain would receive an ample supply
of the new devices, work lagged and no satisfactory design emerged from
the war effort [18].
4.5. GREATER AND LESSER MICROWAVE SETS

By 1942 radar engineers had begun to think of their powerful new mi-
crowave generator as a standard element of design rather than as a sen-
sational invention. It was still unknown to the Germans, ignored by the
Soviets and faltering in application by the Japanese. It brought with it a
completely new kind of electronics, one that used waveguides—pipes!—
for transmitting power, a concept that had caused a small amount of con-
sternation when first demonstrated, as electricians had long thought con-
ductor pairs necessary for this function. It required technicians to master an
assemblage of new crafts for the workbench, and antenna designers found
themselves pulling down textbooks on physical optics. These first mi-
crowave years saw a truly astounding growth in the number of practition-
ers of the new art, many quite new to electronics, and to an overwhelming
degree they found themselves fairly taken with it. The predictable result
was a harvest of designs for equipment both to excel in the tasks that radar
had been doing and to enter unsuspected fields of usage.
In Britain radar was being developed by three service laboratories.
For the Air Force it was done by the Telecommunications Research Estab-
lishment (TRE), the lineal descendant of the Air Ministry Research Estab-
lishment (AMRE), in turn the descendant of Bawdsey Research Station. For
the Army it was done by a group that started as the Army cell at Bawdsey
and which had had three name changes, first the Air Defence Experimental
Establishment (ADEE), next the Radar Research and Development Estab-
lishment (RRDE) and finally the Air Defence Research and Development
Establishment (ADRDE). Naval radar began at His Majesty’s Signal School
(HMSS) but conformed to the prevailing administrative atmosphere by be-
coming the Admiralty Signal Establishment (ASE).
On the American side of the Atlantic four laboratories, all retaining
their original names, provided designs: the Naval Research Laboratory
(NRL), the Signal Corps Laboratory, the Bell Telephone Laboratories (some-
times BTL) and the new Radiation Laboratory (Rad Lab). The two service
laboratories had extensive commitments to improving existing sets and to
supervising the general supply of radar and communication equipment to
combat and training units. They also had the important responsibility of
insuring that the designs from Bell and Rad Lab were satisfactory for the
harsh conditions under which they had to function.
The constraints on the two American service labs applied even more
strenuously to the three British service labs, which were soon in the middle
of a radar war with the Luftwaffe that had designers and technicians be-
ing rushed from one emergency to another and that frequently had them
186
New Ideas
building special sets for combat one day from a prototype of the day before.
They also found working in a war zone disrupting of the calm conditions
generally sought for research. As a consequence the clever new designs
using microwaves came more often from Rad Lab and Bell.
P I Dee came from the heroic age of nuclear physics at Rutherford’s
Cambridge laboratory, as did others who occupied themselves with radar,
such as P M S Blackett, John Cockcroft, C W Gilbert, W B Lewis and
M L E Oliphant, all of whom acquired honors as a consequence of their
scientific and technical contributions. Dee arrived at Swanage in May
1940, the time when the cavity magnetron had reached the stage of an in-
dustrial prototype. His intense nature quickly led to his assuming control
of this new work. At the onset airborne interception was to be its obvious
purpose. Defense against night bombing during the Blitz was bringing
down few attackers and airborne radar was the only hope. Meter-wave AI
mark IV working with GCI finally became effective in early 1941 and the
shift of the Luftwaffe to the east took much of the pressure off obtaining
10 cm AI, or AIS as the project was called. The S in AIS was derived from
the code designation for the 10 cm band; X was soon added for 3 cm and
K for 1 cm. These designations have remained despite modern attempts
to rationalize the nomenclature of microwave bands.
Those were years during which new disasters were ever ready to
replace any whose virulence had receded, and the struggle with the U-boats
replaced the Blitz. Meter-wave ASV mark II proved to be one of the best
designs of the war and was able to go into significant production once
priority went to ASV.
Microwaves were better than meter waves for both AI and ASV, so
these projects continued, but there was quite a bit of uncertainty during the
summer of 1940 about how best to use microwaves. To the general surprise
of the designers, Air Chief Marshal Sir Philip Joubert de la Ferté, Assistant
Chief of Air Staff announced on 22 September that AI mark IV seemed
satisfactory and 10 cm gun-laying equipment for AA should be the goal
rather than AIS, which involved Dee in extensive discussions with those
of higher authority [1].
When the GCI-AI skills began to mature toward the end of 1941, the
Luftwaffe began changing their tactics. Noting that the aircraft assigned to
planting mines in British waterways were seldom intercepted, the bombers
became fond of low altitudes too, where the ground return on AI mark IV
swallowed up all but very near targets [2]. The situation was not des-
perate because the raids were much fewer since Hitler had attacked the
Soviet Union, but they required improvisation by the air crews and placed
renewed pressure on TRE to produce 10 cm AI.
Independently, Oliphant’s group at Birmingham was working with
the firm British Thompson–Houston on a 10 cm GL set using the pow-
erful klystrons they had designed. Oliphant had pushed to completion a
working, two-paraboloid set that he called for some unrecorded reason the
187
‘Dog’s Breakfast’ [3], and the Navy offered its own diversion, as we learned
earlier, in making the first operational 10 cm radar in its type 271. Anew, un-
suspected purpose was to grow out of these contradictory tasks—guiding
the night bombing of Germany—and it was soon to dominate TRE.
4.5.1. AI marks VII, VIII and IX

A critical element for AIS was the microwave TR switch that came from Ox-
ford’s Clarendon Laboratories in spring 1941 (the invention of A H Cooke)
because it removed the need for separate transmitting and receiving dishes.
Double dishes required too much space and for aircraft even a single dish
had to be made as small as possible. By March 1941 the first AIS flew suc-
cessfully and was compared to the Rad Lab’s AI-10, which they sent over
in June4 .
Accuracy of direction was attained for meter-wave equipment
through lobe switching, a method neither practical nor desirable for mi-
crowaves, which allowed the more elegant solution of changing the direc-
tion of the beam. This was done first for a 50 cm beam in the Würzburg
by off-setting the dipole at the focus of the antenna dish and rotating it,
a method invented independently at Rad Lab for SCR-584. The solution
at TRE was to move the dish rather than the feed, the ingenious design of
Alan Hodgkin working with the firm of Nash & Thompson. Dee pushed
the project through the General Electric Company, which required con-
vincing them that 10 cm was better than the 25 cm equipment they favored
[4].
In March 1942 100 sets for operational use were designated AI mark
VII, and installation in Beaufighters and Mosquitoes began. Within a few
months an improved version, mark VIII, that corrected a few faults and
included IFF, had replaced mark VII in production. Fifteen hundred sets
were ordered.
The next step was to have the AI set lock onto or track the target
and allow blind firing, which was a more difficult problem than simply
tracking the target. The project took a bad turn on 23 December 1942 when
A C Downing, who had taken Lovell’s place in the work, was shot down
and killed by a Spitfire that misidentified the Beaufighter in which he was
testing the only mark IX prototype for a Ju-88. The arrival of the successful
SCR-720 that Bell Labs had developed out of the original Rad Lab AI-10
effectively put an end to the mark IX [5].
4.5.2. H2S and ASV mark III

Hitler’s attack on the Soviet Union changed Britain’s strategic situation to
a remarkable degree and with it the tasks facing TRE. On one day defense
was paramount, defense against air attack and against the U-boat. On the
next day air defense was much less important—so long as Russia held out.
4 See Chapter 4.3 (p 168–9).
188
New Ideas
It was not in Churchill’s nature to use this relief to rest; he wanted to use the
resources freed to attack Germany, and the only way Britain could attack
was from the air with Bomber Command. As the night raids intensified it
became ever more apparent that the navigation of the bombers was wildly
inaccurate. Thus when Lord Cherwell, whose primary interest—at times it
seemed his only interest—had become the bombing of Germany, reported
in September 1941 a study proving that two-thirds of British bombs were
falling at distances greater than 8 km from the target, it was clear that a
first-class crisis was at hand.
At Rowe’s Sunday Soviet of 26 October Cherwell insisted that radar
was going to have to furnish the navigational aid to guide the bombers into
Germany beyond the 500 km that current radio-navigational methods were
planning to attain5 . All present knew that Cherwell spoke with the voice of
Churchill. No one had a solution, but Bowen had noted observing towns
in experiments with airborne radar as early as 1938 [6], so Dee had an AIS
set modified to aim downward at 10 degrees with some lateral scanning.
In flying toward Southampton echoes from the city were evident. On 29
December 1941 Rowe placed Bernard Lovell in charge of making a device
to be mounted in the four-engine bombers for guiding them deep into
Germany by looking at the ground with 10 cm radar.
Lovell quickly enjoyed the sensation of having high priority and soon
had good co-workers joining him along with cooperation from A D Blum-
lein at EMI. On 23 April 1942 they tested an experimental system mounted
in the belly of the microwave section’s own Halifax. It had a rotating an-
tenna that pointed slightly downward and presented the observations of a
map-like display on a PPI scope. It received the code name H2S, meaning
‘home sweet home’ or ‘it stinks’, depending on the raconteur.
The prospect of flying their most secret of secrets, the magnetron, over
Germany proved exceedingly troubling. All were certain that it would
soon be in German hands, and they were right. This led to a second Hali-
fax being equipped with a similar system using klystrons instead of mag-
netrons, but their pulsed power was 20 to 30 times less than magnetrons
produced, and the klystron project quickly died.
The project was delayed by two severe setbacks. First, in March 1942
TRE moved abruptly yet again and again with brief planning, but more
about that later. Second, and more tragic, Blumlein and two of his EMI
co-workers died when their Halifax crashed in an accident on 7 June 1942.
That evening Lovell recognized only the magnetron in the smoldering
wreckage of their single prototype.
On 3 July Churchill summoned the Secretary of State for Air, the
Minister of Aircraft Production, the leader of Bomber Command to meet
with Dee and Lovell. He demanded 200 sets of H2S by 15 October, although
not a single aircraft had flown with H2S since the crash that killed Blumlein.
5 See Chapter 6.2 (pp 301–4.
189
This unrealistic deadline was not met, but two squadrons were ready by
the end of December.
What had been built as a device for navigating—most imperfectly—
deep into Germany was an almost perfect ASV set, something not lost on
those seriously concerned about U-boats. Lovell’s group made the minor
modifications that created ASV mark III out of H2S and mounted three of
them in Coastal Command Wellingtons by the end of January 1943 [7].
Microwave antennas have the capability of having the beam shaped
for specific purposes by the forms given the reflector and the feed. This
came about first for the Royal Navy’s type 271 that produced a vertical
fan-shaped beam for 10 cm surface search radar that would yield a good
horizontal resolution and not be affected by the ship’s roll. With H2S a
more sophisticated approach began that has seen significant later evolu-
tion. In sweeping the ground from a bomber one does not need or even
want as much antenna gain for close-range points as for far range, but one
does want to observe close-in reflections that are nearly directly below.
The fulfillment of these requirements gave the PPI display a more maplike
appearance and resulted in a reflector for H2S that produced a reflected
signal at the receiver with an approximately constant amplitude indepen-
dent of range for the same size target. Its dependence on vertical angle
caused it to be called a ‘cosecant-squared’ beam pattern.
The maplike display for H2S had two inherent and troubling prob-
lems. The intensity of the display depended on the amount of reflected
signal, which depended on the composition of the target and the angle at
which it was irradiated. Of the various target materials, only a surface of
water could be relied on to give specular reflection with essentially noth-
ing returned to the receiver; this presented a dark surface that was easily
identified on the scope. But even here the edges of a body of water were
affected by landforms, vegetation or structures that lined the banks, pre-
venting an unambiguous definition of the shore, and the picture changed,
depending on the distance and direction of approach. Even worse was the
consequence of the radar necessarily measuring slant range from altitudes
comparable to the ground ranges, not only distorting the depiction but
having the distortion change as the aircraft moved [8].
4.5.3. Gun-laying radar

Joubert’s surprising announcement in September 1940 of the high priority
to be given the design of a 10 cm radar for AAartillery had been carried over
with the Tizard Mission as one of the three high-priority projects suggested
for the new Radiation Laboratory, the other two being airborne interception
and an unspecified long-range long-wave navigation system using ground
stations, which became Loran6 . The end of the Blitz changed all priorities,
and neither of the two microwave projects received the priority in Britain
190
New Ideas
that it did in the United States. When one combined the general lack of
belief in AA artillery in Britain with Cherwell’s disdain of it and favor for
any method of improving the means for bombing Germany, the result was
low priority indeed for GL mark III [9]. That this lessened zeal did not visit
the United States resulted from the happy combination of Ivan Getting’s
drive and the Coast Artillery’s long-standing objective of radar-controlled
AA fire.
Yet when SCR-584 made its spectacular appearance in early 1944 four
competing 10 cm GL sets were also in production, pointing either to a very
high priority that demanded insurance of success by duplicating design
efforts or to confusion in inter-Allied planning. As we have seen, Col
Colton saw the roof-top demonstration of automatic tracking at Rad Lab
as so important that he immediately covered his bets by ordering Bell Labs
to design a 10 cm GL set, which became SCR-545. This equipment added
a 1.5 m array to the 10 cm dish for increasing the field of view and had
automatic tracking for both wavelengths [10]. It was an excellent set but
was outclassed by the 584.
Although aware of the American work, ADRDE continued the design
that had begun as the ‘Dog’s Breakfast’ and that became GL mark 3, or more
correctly GL mark 3B, designating the British set. The obvious superiority
of SCR-584 led to British purchases for which the designation GL mark 3A
was applied. There was also a GL mark 3C, the Canadian design. Neither
3B nor 3C had automatic tracking, and both used separate transmitter and
receiver dishes [11]. They were not in the same class as the two American
sets and had deficiencies that became all too apparent when the robot
bombs headed for London in 1944, deficiencies that could be traced to
Lord Cherwell and the prevailing British attitude about AA artillery.
4.5.4. The Alvarez antenna

Of the 16 chapters of Luis Alvarez’s autobiography [12] only one is devoted
to radar, the record of two and a half years, for he went to Los Alamos in July
1943. In that brief period he initiated two remarkable radar sets, Eagle and
Ground Controlled Approach (GCA), and had an important influence on a
third, Microwave Early Warning (MEW). The brilliance of his style makes
an interesting contrast with that of Getting. Alvarez would perceive a com-
pletely new concept and follow its design through to the point where the
principles were safely established, and then rush on to the next idea. Get-
ting appreciated the dire need for directing AA fire and pushed through
the fundamental design, prototype, industrial production, battle-field de-
ployment and final modification based on experience to make SCR-584 the
finest radar set of the war. For Alvarez the basic idea was everything. For
Getting the final device was everything. Both were true to their styles in
their equally brilliant postwar careers.
The three radars that interested Alvarez all required higher resolu-
191
tion of targets than any previous sets. Resolution, the ability to distinguish
between adjacent targets, comes with narrow beams, and narrow beams
come with increased antenna size for a given wavelength. To obtain the
beams he wanted with reflecting dishes required impossible dimensions.
The lobe needed to be narrow in only one dimension, which allowed the
antenna width to be a linear array of dipole elements. Alvarez questioned
whether a waveguide with slots, a ‘leaky pipe’, could not be configured
so that each slot radiated as a dipole. This proved possible but invari-
ably yielded unwanted side lobes, which allow targets several degrees off
the beam direction to be observed as if they were in the beam. The next
approach reduced side lobes to tolerable levels by replacing the slots with
dipoles of alternating polarity connected to the waveguide [13], mimicking
the technique of meter-wave arrays.
In the first application of this antenna it was not possible to scan by
rotating the structure, so Alvarez devised an electrical technique capable of
scanning 30◦ to the right and left. The wavelength in a waveguide is not the
same as in free space and depends on the size of the guide. The direction of
the emitted beam is altered if the phase of the wave differs by some amount
from dipole to dipole, and phase shift is just what happens if one changes
wavelength. The wavelength in the waveguide can be altered by changing
a resonant dimension mechanically, specifically its width, and this became
the basis for the electrical scanning method, frequently referred to as the
Eagle system.
4.5.5. Eagle (AN/APQ-7)

The US Army Air Force faced the world conflict in 1939 with a confidence
in their ability to do accurate high-level bombing based on their regard for
the Norden bombsight, a mechanical analogue computer that incorporated
data entered by the bombardier for altitude, ground speed and wind. It
was so secret that the United States refused to share it with Great Britain,
even though the enemy would inevitably gain samples soon after American
bombers appeared over their territory. The RAF had, in fact, a very good,
very practical bombsight, and the accuracy of both air forces was affected
in a trivial way by the shortcomings of these instruments.
There were two serious problems with its application: (1) accurate
bombing required a carefully adjusted approach at constant altitude, devi-
ating from a straight line only to the degree necessary to follow the bom-
bardier’s corrections, which made it ideal for a defending AA gunner; (2)
it depended on being able to see the target for nearly all of the entire run,
something that was routine in the American deserts but often a rarity in the
skies of Europe and, of course, it depended on doing the bombing during
normal business hours.
Alvarez became interested in this bombsight and was permitted to
visit Carl L Norden at his Brooklyn shop where he personally made all the
192
New Ideas
drawings. Alvarez recognized the application weaknesses but saw that if

radar replaced optical vision, darkness and clouds became advantageous
for the attacker rather than the defender, as optically controlled gun fire
would be much less dangerous and with proper jamming, then actively
being planned, even radar-controlled fire would be nullified.
Thus Alvarez approached the same problem of radar navigation and
blind bombing as Lovell but found an entirely different solution than H2S.
First, he insisted on having much finer resolution on the ground. To ob-
tain this he used the 3 cm radiation that Rad Lab was quickly perfecting
and a wide antenna capable of projecting a much narrower vertical fan
than the 10 cm with a small dish. Second, he saw no reason to scan the
entire horizon, just a 60◦ swath ahead of the plane, which he could do with
electrical scanning. The wide antenna, a linear waveguide array of 250
dipoles, was first intended for the leading edge of a wing, although this
idea was dropped in favor of a separate airfoil vane mounted beneath the
plane. These elements became Eagle or AN/APQ-7 as it was entered in
the catalogues [14].
The novel antenna awoke a great deal of skepticism among those not
directly involved with it, and the Eagle group received low priorities and
numerous challenges to have it replaced by some other system. On becom-
ing project engineer E A Luebke had to defend it against cancellation and
to decide between five competing computers: Norden mark 15, General
Electric, Librascope, Bell Labs’s BTO and UBS, some weighing half a ton
[15]. A design emerged that was independent of computer choice [16], but
a Norden design was used. Eagle had a moderately successful prototype
flight test on 16 June 1943, and a very successful production model from
Western Electric was tested on 16 May 1944. It was capable of picking up
cities at ranges of 250 km and had a beam width of 0.4◦ [17].
Because of the delays resulting from its poor support Eagle did not
reach the European theater in time for useful deployment, although it was
used effectively against Japan. Its importance lies in the new concepts of
antenna design that it introduced, concepts that were to have a profound
effect on later radar. It was renamed Eagle by DuBridge, who found the
acronym in use by the group, EHIB, too flippant for transmission to out-
siders. It stood for Every House in Berlin [18].
4.5.6. Ground-controlled approach (GCA, AN/MPN-1)

It was the flier in Luis Alvarez who saw in Getting’s XT-1, the prototype
of SCR-584 a method for landing an airplane under conditions of near
blindness. There had been ‘blind-landing’ methods since the early 1930s,
the Lorenz system being the most widely used [19]. The early microwave
work of Edward Bowles at MIT was aimed at this problem, but all these
schemes only got the pilot to the point where the landing field or its lights
could be seen. If things were really ‘souped in’, they were not much help,
193
and all required that the aircraft carry special equipment that the pilot had
to master. Blind landing was no small affair during the war. Alarge fraction
of aircraft losses were through accident, and poor landing conditions led
the list of causes.
The XT-1 could locate an aircraft accurately in three coordinates and
had been observed a couple of times to follow the plane all the way to the
ground during a landing. Alvarez saw this as a way of ‘talking’ a flier
down simply by comparing his measured location with a desired glide
path. He secured the use of XT-1 in April 1942 [20] with Lee Davenport
testing the idea by following a number of aircraft onto the runway. The
result was a failure.
For reasons within the electronic soul of XT-1 it would sometimes and
without warning break away from the line of sight and locate the plane
below the runway! The radar beam saw the reflection of the target on
the ground. No amount of adjustment or operator skill altered this un-
comfortable fact. When Alvarez finally conceded that the problem lay in
deficiencies of the method rather than in the immediately assumed inad-
equacies of Davenport’s operator skill, he consulted with Alfred Loomis.
The two of them worked out the solution [21] that was incorporated in
mark I GCA, for which L H Johnston became project engineer in July
1942 with the construction of ten units contracted to the Gilfillan Broth-
ers Company, an organization that contributed much to the final design
[22]. The design used three separate radars: a 10 cm search set, which
scanned the horizon and fed its information to a PPI, and two 3 cm sets,
one having a narrow horizontal fan beam for elevation, the other a narrow
vertical fan beam for direction. The antennas used the electrically scanned
dipole arrays from Eagle. The vertical positioning antenna scanned up
and down, the horizontal from side to side, thereby fixing the three coordi-
nates of the unseeing pilot. The shorter wavelength of the two fan-shaped
beams provided accurate positioning through beam narrowness, which
eliminated the possibility of a target being observed through its reflection
on the ground. An ideal glide path was measured and reproduced as an
electrical analog signal. Deviations of the aircraft’s path from the ideal
were presented to the controller by instrument deflections by which the
controller could tell the pilot the corrections necessary to position himself
about 40 m above the runway, although on occasion completely blind land-
ings were made. The search radar with PPI allowed planes to be stacked
while awaiting landing [23].
A single mark I, which had mechanical scanning, was tried under
field conditions in England in 1943. The demonstration was so successful
that demands for the equipment began to come from all theaters. Five
models of mark II, which had advanced to electromechanical scanning
(Eagle), were quickly produced, and by the time mark III appeared there
had been more than 2000 successful blind landings by crews new to the
equipment [24]. Many bombers and their crews returning from Germany
194
New Ideas
were saved by these sets, but its greatest triumph was to come during the
Berlin blockade in 1948–1949, as the weather would have never permitted
the continual flights needed to supply the city without GCA.
Curiously, GCA was not applied to commercial aviation after the
war, and airports continued to cease operation when obscured by fog. The
reason for this was the refusal of civilian pilots to relinquish responsibil-
ity for the aircraft during such critical moments. During blind landing
the GCA controller effectively flew the plane. A few civilian airports had
GCA sets for military aircraft, and there were times when civilian airliners
were brought home with them. The very high installation, operation and
maintenance costs did not appeal to commercial aviation.
4.5.7. Microwave early warning (MEW, AN/CPS-1)

The great problem of air warning with radar using wavelengths of meters
and tens of meters was the vertical lobe structure that allowed aircraft to
approach undetected at low altitudes. This was recognized for CH and
resulted in CHL. The difference was the use of wavelengths one-tenth as
long for CHL as for CH. But 1.5 m sets still could not detect low fliers until
they were closer than fighter controllers would like to have things. A fur-
ther shift to microwaves was the obvious solution. Poor target resolution
resulting from the large beam widths was also a problem that severely re-
duced traffic handling capacity and microwaves would help there equally
well. The gaps formed by the vertical lobes were used for height determi-
nation, but they had as a consequence gaps in coverage at moments when
things became hectic and when no one had time for height determination;
microwaves would not eliminate the gaps but formed so many that the
transitions were indistinguishable from normal target fluctuations. There
was, of course, no height finding capability.
Morton H Kanner of Rad Lab went to England in January 1942 to
study air warning requirements with the defense of the American west
coast uppermost in his mind. On returning he discussed the problem with
Alvarez, whose use of the ‘leaky pipe’ feed then being considered for Eagle
might be used for producing the desired narrow vertical beam. Kanner
became project engineer in June 1942.
Rough calculations indicated an S-band transmitter of the order of a
megawatt would be required, a power level comparable to those of the big
meter-wave sets. The reduction in size would require insulators to hold
much higher voltages than for the same power at longer wavelengths. The
polyethylene coaxial cables that had so pleased designers only months
before for their splendid electrical characteristics were inadequate. All
radio-frequency power would have to go through waveguides. The work
encountered arcing of magnetrons and waveguides in forms never before
encountered in the long two and half year history of microwave design
and met equally exasperating problems with the transmit–receive (TR)
195
junctions that allowed crystal mixers to burn out. All had to be solved [25].
The antenna design that would secure good low-level coverage at
long ranges would not be tall enough vertically for near-in targets that
were high, so a second antenna was planned for high coverage. Beam
widths of 0.8◦ required a horizontal parabolic-cylinder reflector 7.6 m wide;
the low-coverage reflector was 2.4 m high and the high coverage 1.5 m.
Each antenna was fed through a linear array of 106 dipoles mounted on a
waveguide similar in concept to that used in Eagle but the scanning was
by rotating the antenna, as 360◦ coverage was required. These dimensions
and the component sizes needed for the peak high power, eventually to
be 700 kW, began to give microwave early warning, MEW as it was called
in the field and lab, and AN/CPS-1 in the office, a tremendous size. The
two reflectors were mounted back to back. The large number of targets
that the system could track required five 30 cm faced oscilloscopes with
assorted auxiliaries, all of which went into a small house. When it was all
put together MEW weighed 66 tons and consumed 23 kW from an engine-
driven generator. Transport required eight trucks [26].
Only a few MEWs were manufactured, all hand crafted at Rad Lab.
Set number one operated in England in January 1944 and quickly estab-
lished its value to both air forces, and two were ready in time to help control
air traffic on D-Day and locate the V-1s, something for which the absence
of height finding capability was of no consequence [27]. For fighter control
over France it was necessary to add separately a set with a very narrow,
horizontal, fan-shaped beam, a modification of a Bell Labs design for the
Navy, much like similar equipment that had come into use to determine
heights better at British GCI stations. A veteran British fighter control offi-
cer gave MEW the highest marks for its ability to deal with large numbers
of planes [28].
In the Pacific where the density of aircraft was not so great as in Eu-
rope and a general satisfaction with meter-wave equipment for air warning
prevailed, the result of island siting, MEW was not initially greeted with
enthusiasm—66 tons to be moved under the worst transport conditions
and no height-finding capability! MEW Number 4 arrived at Saipan, the
base for the B-29 attacks on Japan, on 21 September 1944. On 27 Novem-
ber it was still not in operation, and the base was surprised by fighters
from Iwo Jima that came in just above the ocean and left burning and dam-
aged bombers behind them. Repeat performances brought telegrams from
Washington ordering immediate installation of MEW. After a sizeable engi-
neering effort the set was established atop Mount Tapochau by New Year’s
eve. A raid on 3 January was picked up at 200 km and intercepted. This
naturally changed attitudes, and fondness for the monster grew when it
began to give remarkably accurate positions for downed air crews in need
of rescue [29].
196
New Ideas
4.5.8. Airborne early warning, Project Cadillac

By the end of the war one-fifth of all the Rad Lab personnel together with
160 officers and men from the Navy were working on a single project, one
that became responsible for 12% of all outside purchases for the entire five
years of the Laboratory’s existence. It was Project Cadillac, named for the
mountain in Maine where sunlight first touches the United States each
morning, not for the luxury automobile with the same name, not that that
would have been inappropriate. It was a project that was not finished in
time to have any effect on the outcome of the struggle, but it pointed to
the future for radar and weapons systems in a manner unlike anything
previously designed.
Cadillac was to overcome some of the restrictions placed on radar by
the Earth’s curvature. It was a 10 cm search radar mounted in an aircraft
flying so high as to extend the range, especially for low-flying planes and
ships well beyond what a ship’s search radar could detect. The need had
been recognized early. An inter-service committee had recommended in
June 1942 such a system—and here ‘system’ begins to take on a modern
meaning—and requested Rad Lab to investigate. In August RCA loaned a
television link with which an oscilloscope trace was transmitted between
an airplane and a ground station. The amplitude-modulated equipment
showed serious problems as the distance separating transmitter and re-
ceiver continually changed, but Zenith Radio had developed a frequency-
modulated television of the kind that came into use after the war, and a
PPI display was satisfactorily transmitted in May 1943. An air–ship link
distance of 160 km was attained.
This part of the task was, in fact, the easy part. The amount of infor-
mation about the location of ships and aircraft, with the perennial problem
of IFF, would be simply enormous for the area observed and over which
a fleet might be maneuvering. To be of any value this information had
to be communicated rapidly to all of the fleet’s CICs. It also required for
command communication many separate channels of very high-frequency
radio that had to be established through the aircraft in order to avoid the
same line-of-sight restrictions as had the radar. Methods had to be incorpo-
rated using radar beacons for the accurate determination of the observing
aircraft’s position.
J B Wiesner as project engineer presided over a true crash program.
There were five divisions of labor: shipboard system, airborne system,
relay radar, relay radar transmitter, and beacons and IFF. Two complete
systems were to be delivered from a complex assortment of suppliers by
February 1945, and one did become available in March. USS Ranger en-
tered tests in April 1945 that extended two months and established Cadil-
lac’s value beyond doubt, but the end of hostilities removed the dreaded
need for it in the invasion of Japan [30].
197
PHOTOGRAPHS: LAND AND NAVAL FIGHTER CONTROL
British AMES type 16 fighter direction station. This 50 cm equipment had a

pencil beam that allowed ground controlled interception to be made with a single
radar. Its normal mode was a panoramic scan at about 2 per minute as it oscillated
vertically between 0 and 10◦ . Historical Radar Archives, RAF photograph. Crown
Copyright.
British AMES type 13 height

finder. This 10 cm set generated
with its truncated paraboloid an-
tenna a fanlike beam narrow in
vertical extent but wide in hori-
zontal, which allowed the height
of a target to be determined. It
worked in ground-controlled in-
terception, generally with the
AMES type 15 1.5 m CGI set,
which could accurately deter-
mine horizontal but not vertical
coordinates. It became the con-
genial companion of the Rad Lab
MEW (microwave early warn-
ing) radar. The design was the
basis of the American SCR-615.
Historical Radar Archives, RAF
photograph. Crown Copyright.
198
New Ideas
Fighter control with radar: Britain. The indicator of a CHL radar. This 1.5 m
equipment could observe low-flying attackers for which the large vertical lobe
structure of CH allowed evasion until too late. The coordinate grid on the left
oscilloscope indicates it was an A-scope display yielding range; the right scope had
a maplike PPI. Historical Radar Archives, RAF photograph, Crown Copyright.
German Würzburg-Riese (FuMG 39T-R). This radar used the same wavelength
and electronics as the small Würzburg but attained greater range by increasing
the size of the reflecting paraboloid dish from 3.0 to 7.4 m in diameter. Its function
was to guide night fighters close enough to the attacking bomber for the pilot to see
it visually or with his airborne Lichtenstein radar. Photograph courtesy of Bernd
Röde.
199
A Würzburg-Riese used as radio telescope. The excellent structural integrity and

availability of these dishes led to them to be used to observe the 21 cm radiation
of atomic hydrogen in The Netherlands, England and America. One is shown
here attached to an equatorial mount at the Department of Terrestrial Magnetism,
Carnegie Institution in April 1953. DTM Archives, photograph 18167.
Fighter control with radar Germany: a Jagdschloss (FuMG 404) panoramic set
based on the electronics of the Freya. It had 18 dipoles arranged on a 20 m
supporting structure to produce a narrow vertical fan-shaped beam. Jagdschloss
incorporated a PPI display, which was transmitted from the station to command
headquarters by high-frequency cable or the widely used 50 cm directed-beam
communication links. It had a broad-band antenna surmounted by an IFF antenna
that allowed modest changes of frequency and specifically that allowed a change to
excite the fighters’ IFF set, which then served as secondary radars that enhanced
the positions of the German aircraft at the closing of a switch. The scaffolding to
the left is not part of the antenna. National Archives photograph 111-SC 269084.
200
New Ideas
Air defense center for Berlin. This huge concrete building located in the Tier-
garten (Zoo) coordinated fighter and AA defense for the region around the city.
Mounted on the top are a small and a giant Würzburg. Similar towers were used
for gun batteries to allow unrestricted fields of fire in cities. The military unit
was Turmflak-abteilung 123 (tower flak battalion). Photograph courtesy of Fritz
Trenkle and Werner Müller.
Fighter control with radar:

Germany. The women shown
here received filtered infor-
mation from radar stations
and projected it in coded form
onto a large map so that
the squadron-control officers
could make use of it. This
photograph presents the un-
usual circumstance of hav-
ing officers of both the Royal
Air Force and the Luftwaffe
observing the tracking of
British aircraft during Oper-
ation Post Mortem. Imperial
War Museum photograph CL
3317. Crown Copyright.
201
American microwave early warning (MEW, AN/CPS-1) on Okinawa. This 10 cm

set removed two serious problems of meter-wave air-warning sets: their poor an-
gular resolution of targets and their blindness for low-flying aircraft. In providing
excellent 0.8◦ horizontal location it relinquished any knowledge of height, which
proved to be no disadvantage in guiding fighters and alerting gun crews against
the uniformly low-flying V-1 flying bombs. A very narrow vertical fan-shaped
beam was attained through a linear array of 106 dipoles cut into a waveguide across
the focus of a cylindrical parabolic reflector. A reflector that would secure good
low-level coverage at long ranges would not form a beam tall enough vertically for
near-in targets that were high, so there was a second antenna for high coverage, the
two mounted back to back. Only six hand-crafted sets were made, none of which
were available at the outbreak of the Korean War, leading to the use of SCR-270
for air warning for Japan. National Archives photograph 111-SC 238090.
202
New Ideas
Fighter control with radar: Germany. Filtered data from radar stations were
projected onto the large map located in front of the desks of the squadron-control
officers. Officers of both the Royal Air Force and the Luftwaffe are shown dis-
cussing the tactics of their recent encounters during Operation Post Mortem.
Imperial War Museum photograph CL 3316. Crown Copyright.
American Little Abner height finder (AN/TPS-10) on Okinawa. This 3.3 cm set
was used to make good the inability of MEW to determine heights. The very thin
horizontal fan-shaped, or beaver-tail beam, allowed it to observe close to the ground
without excessive ground returns. The design was essentially a copy of the US
Navy mark 22. National Archives photograph 111-SC 238093.
203
American MEW Operations Room on Okinawa. The combination of an MEW

and a Little Abner could provide extraordinary amounts of data that had to be
converted rapidly into information useful combat units. There were five 30 cm
PPI scopes and 5 18 cm B-scopes (plotting range against azimuth in rectangular
coordinates) for the MEW alone. Two men are seen plotting on the reverse side of
the plastic map display. National Archives photograph 111-SC 238089.
204
CHAPTER 5
YEARS OF ALLIED DESPAIR AND

HOPE
5.1. THE MEDITERRANEAN, 1940–1942

Neither Germany nor Britain had wanted war in the Mediterranean. Ge-
ography should have given Italy a dominant role, and pre-war estimates of
the strength of Mussolini’s navy and air force suggested that he might have
been able to assert power successfully once France was out of the way, if
her fleet were not to come into British hands. With the vital passageway to
Middle East oil, India and the Far East passing through Suez Britain would
have certainly preferred quiet in the waters north of Africa. But Mussolini,
Il Duce, did not want to miss ‘the march of history’.
In entering the war against the Allies when France was near surrender
he gave the first of many demonstrations of policy determined by delusion.
Mussolini saw Italy’s entrance into the war as heroic; others, even the new
German allies, saw it as despicable. Furthermore, for all the show-window
displays in the 1930s of the power of Fascism, for all the military might
exhibited in Ethiopia and Spain, Italy was woefully unprepared even for
a minor war, something nearly everyone at high level in Italy seemed to
know but Mussolini. The German command saw the Italian army as third
rate in equipment, logistics, training and spirit, and acting on their and his
own evaluations Hitler insisted that Italy not begin land actions that could
require German assistance. Harass the British in the air and on the sea but
do not try to take on even the meager British forces in Egypt.
The German high command did have one immediate military request
for Mussolini: seize Malta at once! World War II was to rage over the
Mediterranean from Gibraltar to Suez, on every shoreline except those of
Turkey and Spain, and everything would turn on Malta.
It required only a map and certainly no great training in naval strat-
egy to see that this island—the unsinkable aircraft carrier—was the key to
the Mediterranean. The German request was not only strategically sound,
it was at the time of Italy’s entrance into the war tactically achievable. The
defenses of the island were extremely weak. Fear of air attack had caused
205
the fleet to depart for the safer harbor in Alexandria, and the garrison was
but a handful with air defense no better. In the Mediterranean the Italian
navy greatly outnumbered the British in every class except aircraft carrier,
a deficiency countered by the large number of planes based on Sicily. Mus-
solini disregarded the German request while Churchill put every possible
reinforcement into the fortress that could be sent from a homeland expect-
ing invasion, slowly and painfully correcting the island’s defenselessness.
The need to take Malta never left the minds of Axis strategists, but when
another opportunity presented itself Hitler faltered.
The issue in the Mediterranean was supply, supplies for two armies
contending for the restricted space between an ocean of sand and an ocean
of water. These supplies had to reach them across the sea—generally on,
but sometimes beneath or above the surface. Supplies were dispatched or
intercepted with whatever cunning man could devise, precipitating some
of the most vicious convoy battles of the war. The struggle assumed the
aspect of a double siege: Malta and a few North African ports.
For two years navies struggled continually, augmenting the tradi-
tional fire of guns and the half-century old torpedo with the attack of air-
planes in a completely new kind of warfare, a kind of warfare soon to take
on increased dimension in the inappropriately named Pacific. Naval and
merchant seamen were pushed to exhaustion and death in a struggle that
wavered from month to month, favoring first one, then the other. Britain,
stretched to the limit, quickly proved capable of dealing with Italy alone.
Germany’s contributions would be momentarily decisive but not consis-
tent, as they were provided by a command riveted to the great struggle in
the east where first opportunity and then danger dominated the thoughts
of the dictator.
In the first months of the war the Royal Navy sent four radar-
equipped vessels to the Mediterranean: the battleship Valiant, the cruisers
Orion and Ajax and the carrier Illustrious. All except the Orion had type 279,
a 7.5 m set with separate, rotatable antennas for transmission and recep-
tion, the small antenna size and long wavelength resulting in a very broad
beam. It was capable of detecting planes 3 km high at a range of 80 km.
Owing to a large vertical lobe pattern it did much worse with low-flying
aircraft or ships. The Orion had type 286, the 1.5 m set intended for small
ships that was an adaptation of ASV mark II. Despite the shorter wave-
length the early models did not have rotatable antennas, requiring that
even very rough directional information be obtained by swinging ship.
The shorter wavelength and smaller vertical lobes allowed it to do a better
job in detecting ships and low-flying planes [1].
Because she was commissioned in April 1940 HMS Illustrious was
the first Royal Navy carrier to be radar equipped, as the older carriers
were considered to be too busy for the installation, but the initial absence
of radar aboard flagships points to some high-rank coolness toward this
new method. Early tactical doctrine also displayed a little aversion. In-
206
Years of Allied Despair and Hope
formation as to the whereabouts of enemy aircraft obtained by radar in

capital ships was to be shared with carriers by flag signals. The Navy’s
strict rules for radio silence, following a lesson learned from the ill effects
of indiscriminate German use of wireless in 1916, were applied to radar as
well—one radar sweep every five minutes, which was at least an improve-
ment on the earlier restriction of one minute every hour. Serving officers
quickly learned the power of their new weapon and disposed, at first un-
officially, later officially, of such restrictions. Aboard the Illustrious fighter
direction evolved out of combat at the same time American sailor–aviators
were developing their combat information centers in fleet exercises. Of-
ficers aboard non-radar carriers were quick in picking up the new tricks.
Lieutenant Commander Charles Coke of the Ark Royal gained renown in
the technique despite the clumsy communication imposed [2].
As experience accumulated with long-wave radar the operators en-
countered properties that diminished the initial euphoria for the miracu-
lous device. When operating near land the A-scope indicators presented
confusing echoes from the extended land masses illuminated by wide
beams. There was a clear need for panoramic scanning and screens with
plan position indicators, the manner of presentation that draws the opera-
tor a map of his surroundings and is now the form universally associated
with radar. In principal the height of aircraft could be determined us-
ing a knowledge of the vertical lobes, but in practice this was difficult or
impossible in a rolling ship or if multiple targets had to be tracked. The
lobe structure yielded such weak reflections for low-flying aircraft that they
were often detected visually first. A related electronics problem caused the
fighter control officers as much grief as the deficiencies of the radar. The
intelligence gathered by radar was difficult to communicate rapidly to the
fliers, both because of restrictions on radio transmissions and deficiencies
in the frequency band used—a deficiency removed for Fighter Command
in England with very-high-frequency (VHF) sets. The HF bands then in
use were noisy and allowed relatively few channels. The US Navy was to
encounter many of these problems a year later in the Pacific.
Two weeks after Italy entered the war her bombers attacked Malta,
for which the air defenses were a few RAF Gladiators, biplane fighters of
a type that saw wide use during the first year in the Mediterranean and
throughout Africa. Coupled with the 7.5 m Mobile Radio Unit (MRU)
radar installed on the island at Fort Dingli in April 1939 [3] they proved
a match for the tri-motor Savoia bombers and their protecting fighters,
but by the end of July Malta had only three Gladiators left, the legendary
‘Faith’, ‘Hope’ and ‘Charity’, and five Hurricanes. Air raids continued and
battles over convoys to the island began, the first ending in a fleet action
off Calabria fought without radar [4]. In September Mussolini blithely
ordered his forces in Libya to invade Egypt and followed a month later
with an invasion of Greece from Albania, specifically the kinds of action
that his German allies did not want. He soon needed German support.
207
The Italian navy outnumbered the British to such an extent in the

early stages that Admiral Sir Andrew Cunningham knew he must do some-
thing to improve the odds and to gain the moral upper hand. He did this
on 11 November 1940, and radar helped. On the 7th a force left Alexan-
dria escorting convoys bound for Malta and Greece but with other things
planned than the safe conduct of merchant vessels. HMS Illustrious and es-
corts headed for Taranto where important elements of the Italian fleet were
based. The harbor, alerted to the movements of the British, was guarded by
reconnaissance planes, but the Illustrious located them with her radar and
vectored ship-board fighters to eliminate them [5]. Twenty-one Swordfish
torpedo bombers from the carrier removed three battleships and various
lesser vessels from the enemy lists but did not gain this through the hoped-
for surprise, for the entire harbor erupted in a ‘volcano’ of anti-aircraft (AA)
fire well in advance of their arrival. Instead they gained it through their
famous skills in night action [6]. The success of this venture was not lost
on the Japanese military attaches.
This daring attack, made with a realized expectation of a strong de-
fense, was carried out in machines that were to gain a mythical reputa-
tion. The Swordfish was a sturdy, open-cockpit biplane of 1933 design,
affectionately known as the ‘Stringbag’, and not fast even by World War I
standards. It was the Royal Navy’s air striking arm from land and carrier
through most of the war [7]. It also became in July 1941 the first aircraft
in the Mediterranean to carry 1.5 m ASV radar with a transmitter dipole
on the upper wing and receiver Yagis fixed to the wing struts. If Luftwaffe
planes protected Axis convoys during the day, the Swordfish could find
and destroy them at night. An aircraft with ASV would locate the ships
and drop flares to illuminate them for the planes with torpedoes. In late
summer 1941 a few ASV-equipped Wellingtons began performing a similar
function for Malta’s Force K of cruisers and destroyers by guiding them
at night towards a convoy and illuminating them with flares to permit
sinking by gunfire. In both cases arrangements were made for surviving
vessels to be met by submarines during the day [8].
Four months later Cunningham dealt another crushing blow to the
Italian fleet and this time radar was decisive. British supplies to the Greeks
forced the Italians to attempt a surface interdiction that led to action off
Greece’s Cape Matapan, the southernmost tip of mainland Greece. Dur-
ing daylight of 27 March 1941 a Swordfish had disabled the cruiser Pola,
and Admiral Angelo Iachino decided to send back two cruisers and four
destroyers to aid the stricken ship under cover of darkness. An alert radar
operator on the cruiser Orion insisted that an insignificant blip on his os-
cilloscope was a ship and soon had the squadron heading toward it. Two
other ships had radar, the battleship Valiant and the cruiser Ajax, and their
sets soon showed a number of Italian ships. Approaching unseen and
unsuspected they closed to comfortable and accurately known range and
illuminated their targets with searchlights. The result was destruction from
208
which only two of the destroyers survived [9].

The Italian navy had relied on the ‘fleet-in-being’ for much of its
effect. That is, its mere existence should tie up enemy units as a guard
against sorties. It was a strategy attributable to Mahan that contrasted
notably with the British doctrine of ‘bring the enemy to action’. Whatever
value Italy’s naval force had ever had as a ‘fleet-in-being’ was lost after
Matapan.
From the disastrous outcome of this battle the Italians realized that
the suspicion of Britain having radar must be true and re-initiated the pro-
gram they had abandoned. By the end of 1941 they were able to install a
set aboard the battleship Littoria, although without significant effect. Mus-
solini had consistently opposed requests by his navy for aircraft carriers,
arguing that land-based planes could satisfy the need. With the power of
British carriers demonstrated beyond doubt, he also gave orders to con-
vert two liners into carriers, not completed before Italy left the war [10].
The Italian Navy suffered from having no aircraft under its own control,
being dependent on requests of Regia Aeronautica and the Luftwaffe, both
considering requests as just that, not orders.
South African forces had joined the British in stopping the Italian
drive against British and French territories and then in clearing them out
of Ethiopia, the first liberation of a nation conquered by the Axis. Moreover
they had brought their own, home-made radar with them, called JB (for
Johannesburg) and provided air warning for Mombasa and Nairobi. After
the Italians had been defeated they were assigned positions in the Sinai
desert for the defense of the RAF installations covering the Suez Canal.
There they had the satisfaction of seeing their equipment frequently out-
perform the huge factory-produced MRUs. When passage of the Mediter-
ranean became blocked, greatly increased amounts of shipping, vital to
Allied forces in Egypt, had to pass the Cape, so the South African radar
personnel were withdrawn to take up the duties originally intended for
them and their JBs: protection and navigation of ships off the storm-lashed
southern coast. The COL sets (an overseas version of CD/CHL) were slow
in coming, but twelve JB-3s together with a few ASV sets modified for
ground use were erected in the vicinity of Cape Town, Port Elizabeth, East
London and Durban.
The story of the JBs is another example of radar being a straightfor-
ward, not particularly mysterious device. Dr Basil Schonland, Director of
the Bernard Price Institute, had learned about it from Dr Ernest Marsden
of New Zealand, who imparted what he could of his recent instruction in
England while his ship made the short passage from Cape Town to Dur-
ban, weeks after the outbreak of war. Schonland learned enough to start
radar development at his institute with a small group that kept a copy of
The Radio Amateur’s Handbook nearby. They recorded their first echo on 16
December 1939; by March 1940 they had detected aircraft at ranges of 80
km. It worked on 3.5 m, radiated 5 kW peak power from a steerable array
209
of dipoles and was built from components obtained locally [11].

Radar was not the only secret weapon Britain had in the field. The
Mediterranean was the training ground for Ultra as an important, later the
most important segment of military intelligence. Its use in the Battles of
Britain and of the Atlantic was relatively simple, but operations in North
Africa necessitated heavy traffic by radio, furnishing ample feed stuff for
Bletchley Park but little of it simple. Just as practice and experience had
been needed to transform radar equipment into a weapon, practice and
experience were needed to turn the English translations of Enigma signals
into useful information. The first problem was speed. The Axis transmis-
sions had to be recorded and transmitted to England. Decryption there was
not necessarily immediate, depending on the key used and the care with
which the messages had been encrypted. The vital intelligence had then to
be made available to the commanders without disclosing the source; the
secrecy surrounding Ultra was necessarily severe—and successful. This
presented numerous problems that had to be worked out with intelligence
personnel, the greatest being the general distrust of information thought to
be obtained from agents, which bitter experience had long taught generals
to suspect.
Information from Ultra about Axis convoy sailings became reliable
in June 1941 but had to be used with care for fear of compromising the
source. The policy imposed required that vessels be spotted by aircraft
apparently on normal patrol before being attacked. As sinkings increased
the Germans suspected Italian treason, which created a few side benefits
for the British [12]. Information from Ultra about strategic and tactical
affairs was much more confused than details of ship sailings, as it described
only part of a complicated situation. When Ultra was integrated into a
comprehensive intelligence organization that gave each part thoughtful
evaluation, significant benefits accrued. Ultra’s part in stopping Rommel’s
advance at El Alamein demonstrated the maturity of such application, but
this maturity had been reached in a tough school that had begun with
the confusion Rommel caused by disobeying orders in starting his first
offensive in April 1941 [13]. After El Alamein Ultra became thoroughly
integrated in command thought.
The war in the Mediterranean was brutally continuous. Behind the
complex military struggle between Italy, Germany and Britain was an even
more complex political struggle, often lapsing into violence, involving the
three principals with various combinations of Arabs, Jews, Iraqis, Iranians,
Turks, Vichy French, de Gaulle French, Greeks, Spanish, Balkans and minor
groups. To describe it in detail here would serve no useful purpose, for
the story is well told in many books. Let it suffice to describe the principal
acts of this drama, pointing out the parts that radar played.
In early 1941 the British army with strong Commonwealth compo-
nents under the inspired leadership of Major General R N O’Connor drove
the Italians out of Egypt and back into Cyrenaica. This predictably brought
210
German aid, which took the form of Generalmajor Erwin Rommel with
the Afrika Korps and Fliegerkorps X, an air unit to protect Axis and at-
tack British convoys. Fliegerkorps X had specialized in operation against
shipping when based in Norway. This initiated the ‘First Malta Blitz’, of-
ten called the ‘Illustrious Blitz’ because of the attractive presence of the
damaged carrier in the harbor following a convoy action.
The air defense was stout but technically deficient. By the end of
January 1941 an additional MRU was operating at Fort Dingli, the two
working alternately to ensure continuous coverage, and low cover had
been secured by COLs at Forts Madalena, Ta Silch and Dingli. (Overseas
versions of CH and CHL were designated CO and COL and had minor tech-
nical differences.) Probably because of bad siting and the lack of enough
aircraft flying time for the extensive calibration required, the Malta radar
gave very poor indication of altitude; it also had blind spots, and the small
size of the island precluded effective use of ground observers. The intense
activity can be appreciated by noting that there were 20 000 plots in the
Filter Room in 34 days [14]. When coupled with poor communication this
made the basic tactic one of sending up fighters to wait near the island
rather than meeting the enemy far out [15]. Fliegerkorps X caused near
strangulation of the island, and soon Axis convoys were reaching their
destinations with few losses.
With their supply line secure German and Italian forces attacked
British ground forces in Libya on 31 March 1941 and scored substantial
advances because British troops had been sent to help Greece and because
Rommel proved to be a master of desert warfare. Fliegerkorps X shifted
much of its strength to support the Afrika Korps, giving the island a slight
breather but adding to the British Army’s troubles. It was a difficult spring
for Britain, to say the least, for in April the Germans invaded Greece and
Yugoslavia, subsequently driving the British from the mainland. A con-
tingent remaining to defend Crete, which had some of the characteristics
of an eastern Malta, was defeated by paratroop assault. The British radar
deployed in Greece and Crete had little effect on the melancholy outcome
[16].
The Crete victory was extremely costly for Göring’s airborne troops,
something not forgotten in Berlin during the ensuing months. The losses
to the Royal Navy in its generally successful withdrawal of the Army were
so great that surface forces could no longer be based at Malta. The Navy’s
losses came about by operating beyond the range of land-based air cover
and because their AA fire was inadequate to protect them alone. It was
a lesson that they had, in fact, already learned; the decision to proceed
was made in cold blood as the Army had to be supported. Rommel soon
stretched his supply line to the limit and Fliegerkorps X was sent to help
support the invasion of the Soviet Union, which began on 22 June 1941.
A year of desert warfare had caused the RAF to forget their pre-
war reluctance for army support duties. When General Wavell attempted
211
to dislodge Rommel from the frontier of Egypt they demonstrated them-

selves increasingly skillful under Air Marshal Tedder in attacking ground
targets. These skills made little use of help by ground radar to keep fliers
informed of the approach of enemy planes by GCI stations, although they
were pleased enough to learn the location of a downed aircraft for rescue.
Secrecy may have played a part, as pilots did not know the basis for the in-
structions they received over the radio. A ground radar officer of the time
did not remember having had discussions with the Air Officer Command-
ing’s staff as to how radar would fit into air defense [17]. A suggestion of
the reason can be found in Tedder’s memoirs, which mention radar only
three times in 417 pages. It was a problem that did not improve much with
time so far as army support went. A ground radar officer in the Normandy
invasion found the attitude of the fliers toward radar control to be ‘do not
interfere’ [18]. The unyielding belief in the absolute necessity of prevent-
ing radar equipment from being captured, coupled with its poor mobility,
kept sets sufficiently far from the front to reduce their effectiveness. The
best use of radar in the desert campaign was in the long defense of Tobruk
by the Australians.
Although outnumbered in men and material the Italian and German
forces held. With O’Connor a prisoner of war the deficiencies in Army
leadership, so obvious in Belgium the year before, were there for all to see.
The Germans introduced their first radar into North Africa, a Freya
to Tripoli, in January 1942, with no small degree of apprehension. This was
still the time that everyone thought himself master of a very secret thing,
but tactics won out as they built an effective air warning system—with local
sharp-eyed Arabs added as irregular observers. The first Würzburg made
its appearance at Bomba Bay in April, and by June there were three Freyas
and two Würzburgs in service [19]. The Germans made good use of their
radar in air defense, probably because the Luftnachrichtentrupp had been
so closely integrated with the flying units. An attempt by commandos to
take a Freya on 14 September failed [20].
With Hitler’s attention fixed in the east, Malta quickly recovered and
was soon dealing harshly with the Italian merchantmen bound for North
Africa. The combination of Britain being able to read relevant Axis signals
and searching with airborne radar at night resulted in 63% of the ves-
sels dispatched in November 1941 being sunk, the same month the British
opened a successful land offensive and drove back into Libya. In an exam-
ple of desperation the Italians loaded two cruisers with drum-loaded fuel
for transport. They were discovered by airborne radar and transformed
into gigantic conflagrations by destroyers. By the end of the year Axis
forces were reeling.
The successful attacks on Axis convoys caused Hitler to order many
of his U-boats to the Mediterranean in August 1941 against the strong ob-
jections of Admiral Dönitz. The damage they did there was more than
balanced for Britain by the damage they did not do in the Atlantic. Fur-
212
thermore, the Mediterranean was a submarine trap because the currents

at Gibraltar made entrance relatively easy, exit nearly impossible, but they
long remained a problem there, sinking carriers Ark Royal and Eagle, bat-
tleship Barham and 95 merchantmen at a cost of 68 of their own number
[21]. Coastal Command four-engine Sunderland flying boats equipped
with ASV began seeking them in the eastern reaches, where radar confu-
sion of the numerous fishing boats with submarine conning towers gave
the chase its own flavor [22].
With the eastern front in winter stalemate Hitler sent Field Marshal
Albert Kesselring as Commander in Chief South with a contingent of his
Luftflotte 2 to turn matters around. Malta had to endure the ‘Second Malta
Blitz’, and it proved to be worse than the first with starvation seriously
threatening garrison and population. The nadir of the island’s fortunes
was marked by the failure in February 1942 of a convoy to deliver even
a single shipload of cargo. Other convoys that had attempted to run the
gauntlet had been able to bring at least a token cargo through. Malta’s air
defenses were reinforced by US and British carriers ferrying Spitfires to
within flying distance of the island, a technique developed during the first
Blitz.
During this Blitz the Germans set up radar on Sicily to monitor British
air activity and used it to observe the transfers of fighters from carrier to
land, attacking them on the ground shortly after their arrival [23], which
motivated extreme measures to get them airborne again as quickly as pos-
sible. They also started electronic countermeasures. Martini supervised
the installation of jammers for the Malta radars, which effectively put them
out of action. This was reported to R V Jones in London, who told them
to continue operation as if nothing had happened [24]. Those on the scene
took more direct action. An Air Force officer struggling with an MRU and
an Army engineer sent out to rescue the GL sets independently discovered
that the jamming, either by accident or design, was modulated at the 50 Hz
power frequency. They synchronized with the peaks of the jamming and
operated where the disturbing noise was least, restoring much capability
[25]. For whatever cause German electronic intelligence concluded the
jamming was ineffective, and it was stopped. When Martini met Jones af-
ter the war his first question was how the jamming had been circumvented.
Whether he or Jones ever learned what happened is not clear.
For some reason Kesselring did not attempt to destroy Malta’s radar
sets, although his photo intelligence had picked 14 positions that they
marked as radar stations in June 1942 [26]. This came as a pleasant surprise
to the operators, who grimly expected a full-scale attack [27].
Despite clever radar tactics by the defenders Malta was neutralized
and Axis shipping again went through. If the Malta Blitz succeeded in
opening the supply route to Rommel again, it was not able to subdue
the air defense of the island completely and came to be like the labors of
Sisyphus to the attacking fliers, who took heavy losses [28]. Kesselring
213
and Rommel insisted that Malta must be invaded and began planning an
airborne invasion, measures that found little enthusiasm in Berlin because
of the losses in taking Crete. By May Axis supplies on the desert had
reached a level sufficient to sustain an offensive, and British strength had
shrunk as a result of urgent demands for help from Burma and India.
To no-one’s surprise Rommel attacked but with an effect that surprised
both sides—the heavily fortified port of Tobruk fell. This disaster saved
Malta, curiously enough, because within the Tobruk fortress were immense
quantities of supplies that fell to the attackers, Rommel thought enough
to sustain him to Suez, if the air forces attacking Malta were transferred
to him. The invasion of Malta was forgotten, to the relief of Hitler and
Göring. But the gamble failed; the British Eighth Army, its left flank secure
on the Qattara Depression and its right on the sea, stopped Rommel at El
Alamein in early July 1942, and he was never to go beyond it. The air and
naval forces of Malta soon began again to take their toll of Axis shipping.
On 10 August a large, heavily escorted convoy departed Gibraltar
and soon found itself the center of a bitterly fought convoy action, attacks
coming from aircraft, surface ships and submarines. The type 279 radars
were unable to sort out the many targets, resulting in poor fighter direction
[29]. Several ships were lost, including the Mediterranean veteran carrier
Eagle, but the sacrifices were not in vain, as a significant cargo was landed,
effectively ending the siege. On 30 August Rommel opened a desperate
attack on the British positions with inadequate fuel and ammunition. Its
failure was a direct result of the decision not to take Malta. On 23 Novem-
ber the British forces opened a long-prepared attack on Rommel’s lines
and succeeded in forcing him to retreat after ten days of heavy fighting.
Less than a week later American and British forces landed in Morocco and
Algeria. The nature of the struggle altered significantly after that as did the
employment of radar, which concluded a phase one might well call ‘heroic’.
It entered a new phase in which it was one of many weapons. Unnoticed
by nearly all of the participants, high and low, was the capture of German
radar equipment at El Alamein, in a variety of smashed conditions, to be
sure, but Allied electronic intelligence soon had an operating Würzburg
[30]. Those German units that retained their radars did so with determi-
nation, holding their tractors at gun point from others during the retreat
[31]. Of course the capture of radar did not work only in one direction.
The Afrika Korps captured a Wellington equipped with ASV mark II [32].
The question now arises. What effect did radar have on this part
of the War? Did it turn the outcome in favor of the Allies? The radar
enthusiast will be quick to point to Matapan and sometimes to Taranto as
important victories in which radar was a major contributor. He will also
point to the steady use of the technique in interdicting Axis and protecting
British convoys and in defending Malta. That it was an important weapon
for Britain—it also served Germany but more as a technician than as a
warrior—is beyond question, but did it prevent Axis victory?
214
For the Axis only one victory counted: to seize Alexandria and the
canal. The consequences for Britain would have been severe. Their loss
would have almost certainly meant the loss of the vital oil of Iraq and Iran
and its gain for Germany. Lost too would have been the reliable supply line
to the Soviet Union, and all this when Singapore had surrendered, Burma
had been given up and India was threatened. There were limits on what
America’s growing power could do. They were terrifying times.
Would Rommel have reached Suez, had Britain not had radar? Is it
really possible to decide? Indeed one can just as well pose the question in
terms of Ultra, and will be no closer to an answer for having done so. There
must be a dozen instances in which the outcome seems to have turned on
a single engagement. But there were for both sides defeats that were made
good, and much of what took place resulted from the variation of Hitler’s
attention. Both Axis failure and Axis involvement in the Mediterranean
had a common origin: the near absence of strategic planning in Berlin and
Rome. It is possible that had some of Britain’s early successes, obtained
with the help of radar and Ultra, not been made Germany might have
sent a smaller force to aid Italy with a correspondingly smaller threat to
Suez. Action in the Mediterranean until the end of 1942 was unquestion-
ably the most tangled of the War. Change any part and one is lost in a
morass of conjecture. One thing is clear. Britain needed every source of
strength available just to stay in the war, and in such circumstances any
new weapon, like any new ally, must be remembered as vital to victory.
5.2. WAR IN THE PACIFIC

Harold Zahl’s first thoughts when he learned of the Japanese attack on
Pearl Harbor were about SCR-270 [1]. He knew that this early warning set
into which he and many others at the Signal Corps Laboratory had put so
much labor and ingenuity was deployed in Hawaii. Had it failed? It was a
question that stood out above all others as the engineers at Fort Monmouth
discussed the news, especially as the extent of the disaster became known
to them. For a decade their professional lives had been spent in devising
the means to prevent just this, and now it had all come to nought.
The defeat suffered by American arms on 7 December 1941 was
remarkably similar to the one suffered by the British a year and a half
earlier on the fields of Belgium and France and had the same roots—
commanding officers lacking comprehension of the importance of air
power compounded by their inadequate imaginations and poor communi-
cation with their subordinates. For the British it was accepted as impossible
that an armored force could come through the Ardennes; for the Americans
it was accepted that Hawaii would not be attacked by carriers at such a
great distance from their home base. Both should have been in high states
of readiness: the British were at war, and the Americans had received a
war warning on 27 November, but both had made the ancient military
215
blunder of basing action, more properly inaction, on what was expected

of the enemy.
Like many catastrophes Pearl Harbor has compelled extensive stud-
ies to determine its causes and to extract the lessons that it has to teach,
and the literature matches the magnitude of the event, demonstrated by
the publication of a bibliography of 1514 relevant publications and docu-
ments [2] and by the 110 cm of shelf space occupied by the Congressional
and various service investigational reports. Not surprisingly it has gen-
erated controversies, two of which remain. One is whether a misguided
foreign policy of President Roosevelt and Secretary of State Cordell Hull
unnecessarily forced Japan to such desperate action. This clearly lies be-
yond the limits of this book. The other is whether Roosevelt and Army
Chief of Staff George Marshall withheld specific knowledge from the lo-
cal commander of the impending attack because they wished to bring the
United States into the war on the side of Britain. This stretches the limits of
common sense, being founded on the incredible assumption that an attack
resolutely defended would have left America at peace!
But what about radar? Had it failed? Here the answer is as unam-
biguous as are the consequences of the attack. No, the instrument had
functioned perfectly but to no avail. The responsibility for failure to act
on the warning that radar provided can be fixed uncluttered by issues that
complicate determining overall responsibility. The object of the Japanese
attack was the destruction or severe injury of the Pacific Fleet commanded
by Admiral Husband E Kimmel. When in port it was the responsibility of
the Hawaiian Department of the Army under Lieutenant General Walter C
Short to protect it against air or ground attack. The key to air defense was
early warning, something that the Air Ministry in Britain had understood
from the first and that had given the proposals for radio location such
power. Similar thoughts had guided the prewar work at the Naval Re-
search and Signal Corps Laboratories. What was lacking in the Hawaiian
Department was the realization that radar gave warnings of only 20 to 50
minutes and that the utilization of this precious intelligence required an or-
ganization whose complexity equaled that of a radar set. It was Dowding’s
clear grasp of this elemental fact that was decisive in 1940.
Of the officers who reported for duty with the Pacific Fleet in 1941
few had had the varied service of Lieutenant Commander William E G
Taylor. Commissioned as a reserve officer in 1927 he had transferred to
Marine Aviation after a year as a carrier pilot aboard USS Lexington. Or-
dered to inactive duty during the depression he had worked a variety of
flying jobs until going to England in the summer of 1939 in anticipation of
war. The Royal Navy accepted him as a Sub-Lieutenant, and he soon had
duties as a fighter pilot both from carriers and land. When the RAF began
forming squadrons of Americans he resigned from the Navy and became
the Squadron Leader of 71 Eagle Squadron. Intimate contact made him ex-
tremely interested in British fighter direction with radar, and he secured a
216
rather extensive knowledge of it. At the instigation of the American Naval

Attache he was offered, with British good will, a commission in the Amer-
ican Navy that he accepted and returned to the United States in August
1941 where he lectured on fighter control [3].
On arriving in Hawaii he was sent to work with Major Kenneth P
Bergquist of the Hawaiian Department Air Force [4], who was building an
air warning system based on radar equipment that began arriving dur-
ing the summer of 1941. Taylor liked what Bergquist was doing, the
two worked well together and both were very impatient with progress.
Bergquist had been in the first class of ten officers and 40 enlisted men who
attended the new Air Defense School at Mitchel Field (named for John P
Mitchel, former Mayor of New York City) in March and April to learn the
new technique. His teachers, Air Forces [5] and Signal Corps officers just
returned from inspection trips to Britain, had quickly put together demon-
strations of radar air warning and fighter control, and Bergquist returned
to Hawaii filled with evangelical zeal.
On return to his duty station he was assigned to his old interceptor
unit until the end of May, when he was told to establish an air warning
system, something he had been pushing since his return. But official con-
sent was not official drive, and equipment and personnel were difficult to
obtain.
The Signal Corps began setting up the SCR-270s as they arrived and
training operators and maintenance men. That went well, the main de-
ficiency being that the sets required siting where no electric power was
available, and no amount of talking to Chiefs of Staff could bring in elec-
tric lines, making it necessary to run with engine-driven generators, which
were having enough breakdowns that practice was severely restricted. Fi-
nally they were able to build a small, wooden information center at Fort
Shafter with tactical and administrative telephone lines to the five radars
in operation. By December the radars functioned and the plotters moved
the markers about the big map board during the few hours a day that
they could practice, but in the balcony sat neither fighter control nor li-
aison officers. This left the ‘Information Center’ devoid of information
about their own interceptors, bombers, naval air and AA guns on ship or
ashore. Equipment might be acquired in devious ways but officers were
in much too short supply to be obtained for assignment to this work. They
could never be spared; there was too much to do. Majors have difficulty
instructing flag ranks in their duties.
Thus by 7 December Bergquist and Taylor’s little center was able to
track what the SCR-270s reported but had nothing to help identify the
various plots; IFF lay months in the future for Hawaii. If through some
occult process they decided a particular radar target was hostile, they could
do nothing more than call some commanding officer at his office. There
were no tactical lines out of the Information Center to any fighting unit
capable of making use of the information. There were no serviced fighters
217
ready for take-off, no rooms filled with pilots waiting on call [6].
On the fateful Sunday morning two privates were operating an SCR-
270 alone on Opana Mountain, the northernmost radar on the island of
Oahu. They were restricted by General Short to operate between 4 and
7 AM and had reported distances and directions of the planes that made
up a quiet morning’s traffic. Joseph Lockard was experienced with the set
and was teaching George E Elliott its use. A truck was to arrive at 7 to take
them to breakfast and leave replacement guards, but the truck was late
and Elliott asked whether they could continue with his practice, to which
Lockard agreed, although in fact a violation of orders. At 7:02 Elliott saw
a blip on the cathode-ray tube at a range of 220 km, which Lockard said
was the biggest reflection he had ever seen at such a range. It was a large
formation of planes nearly due north and flying straight at them.
They decided to call the Information Center but were unable to obtain
an answer on the tactical line, as the exercise was over. It seemed important
enough, although they were not sure why, to try again by calling on the
administrative line where they eventually spoke to First Lieutenant Kermit
A Tyler, who had been ordered to observe the morning exercise to become
familiar with the procedures as the first step in becoming a fighter con-
troller. He had no other operational function and had no way of knowing
what the formation of planes was. He did know informally from a friend
in the bomber command that flights of B-17s were passing through on way
to the Philippines but knew nothing officially. He had also been told the
flights used a broadcast station playing Hawaiian music uninterruptedly
all night as a radio homing beacon and had noted that the station had been
playing just that as he drove to his 4 AM duty station. He reasoned that
this was what the Opana station was seeing and told the two not to give it
more thought. There was indeed a flight of B-17s bound for the island, but
that was not what made the large reflections, they were made by Japanese
carrier planes [7], who listened to the music too.
After the attack there were plenty of investigations into the causes,
all neatly collected in the congressional publication, to keep people busy
for decades. Some of the recorded interviews wander in the direction of a
disordered intellect: two privates and a lieutenant being interrogated with
an underlying implication that they might have saved the naval base.
Needless to say Hawaii’s air warning system received ample support
in equipment and personnel during the days following the attack. There
was a valiant but generally ineffective effort to make use of it during the
hours of and after the attack, but it was unable to prevent aircraft of USS
Enterprise from being shot down by Navy guns as they flew in from the car-
rier that evening. The Opana station had followed the departing Japanese
aircraft back to the north, but this information was lost in the confusion
of the day and searches for the Japanese fleet went south, all considered
probably just as well. Radar was now the favored instrument for which an
almost insatiable appetite grew. The CXAM from the seriously damaged
218
USS California found its duty station on land.

The garrison in the Philippines knew of the outbreak of war hours
before any Japanese action against them, but General MacArthur used this
advantage to little effect. As in Hawaii the failure lay at the top, not at
the bottom, and radar had brief but honorable service. A Signal Corps
Aircraft Warning Company had arrived in the islands in August 1941, and
their first SCR-270 followed on 1 October. They set about unpacking and
assembling the device and learning to use it. It worked well from the start,
which was particularly helpful as no test equipment came with it.
Lt C J Wimer and a detachment of 30 men positioned this first set at
Iba on the coast about 150 km northwest of Clark Field, the base for the ever
increasing number of B-17s, the Flying Fortresses that were the hope for
defending the islands, as the Air Force was confident that these bombers
were the match for any invading fleet. The Iba station was operational
by the end of October. More equipment arrived, and by the outbreak of
war there were additional sets positioned, though hardly fully operational,
on the coast of the Philippine Sea about 200 km southeast of Manila, at
Tagaytay Ridge 70 km south of Manila, and at Burgos Point at the extreme
northern tip of Luzon. In late November the Marines called the Signal
Corps to report the arrival of an SCR-268 for which no one was trained,
and a detachment was sent to get them started.
In the early morning hours of 8 December (7 December on the other
side of the International Date Line) and before they had learned of war
the Iba station picked up intruders at 180 km and a squadron of fighters
was scrambled to intercept; they failed, probably because of an inaccurate
height datum. From then on Japanese planes made many attacks that kept
the radar station busy. It reported a large formation of bombers at 11:45
AM headed for Clark Field. The responsibility for allowing the destruction
on the ground of half of the B-17s from this attack, hours after the start of
war, has been a matter of continuing dispute, but the transmission of the
Iba station report by teletype to Clark Field is not contested, although its
reception is. The Japanese bombed the Iba airstrip and destroyed the radar
at 12:20 PM.
Radar played no further role in the defense of the Philippines. What
reports came from the remaining sets went to combat air units unable to
utilize them [8].
The Philippine radar was not caught off guard, sending fighters to
investigate intruders even before they knew of the outbreak of war. Their
disposition of the initial attacks was as good as their equipment allowed.
The United States was not alone in experiencing disasters. The
Japanese struck south not only toward the Philippines but toward Singa-
pore, approaching through landings on the Malay Peninsula to the north
and proceeding through the jungle, a path that the defenders had ruled
out and against which they had prepared no fortifications. To put an end
to this the Royal Navy sent the new battleship HMS Prince of Wales and
219
the old battle cruiser HMS Repulse to destroy the landing parties. They
moved without air cover, and were sunk by land-based torpedo planes on
10 December 1941 in the Gulf of Siam.
Radar may have played an unfortunate part in this. Admiral Sir Tom
Phillips, who commanded and died with the ill fated squadron, had from
his first day on board the Prince of Wales shown a keen interest in her radar,
to the point of discussing in detail the use of each set with the operators.
He was aware that this vessel was the best equipped in the world with this
new technique. It had a type 281 3.5 m air warning set, a type 284 50 cm
main-armament gun-laying set, four type 285 50 cm AA gun-laying sets,
four type 282 50 cm close-fire AA gun-laying sets and one type 273 10 cm
close-surface search set. Types 284 and 285 both employed lobe switching.
It was an impressive suite of equipment to say the least. The Repulse had
one type 286P 1.5 m air-warning set, much improved over the original 286
in having a rotating antenna, and one type 284 [9].
Phillips had held strongly for the invulnerability of modern capital
ships against air attack at a meeting of the Joint Planning Staff just before
outbreak of war [10] but presumably understood by late 1941 the dangers so
recently emphasized at Pearl Harbor. He had no carriers but could call on
land-based planes from Singapore. One is led to suspect that he may have
relied too heavily on radar’s ability to protect his ships. Two elements may
have sealed the fate of the two great ships: first, maintenance at Singapore
was interrupted by the outbreak of war, leaving inoperable three of the
four type 282s, needed for the very important 40 mm AA fire; second, he
delayed breaking radio silence to call for air support until well after the
first radar sighting of impending trouble. The ships fought well but were
overwhelmed. Air support arrived but too late.
Beyond the Philippines and Singapore lay Australia, clearly in the
path of Japanese aggression. As loyal members of the British Com-
monwealth, Australia and New Zealand had declared war on Germany
promptly and both had sent sizeable contingents to the deserts of North
Africa, contributing mightily to General R N O’Connor’s devastating dis-
posal of the Italian army’s invasion of Egypt.
Dangers of Japanese attack, especially from the air, were never far
from mind and had led to the organization of the Radiophysics Laboratory
on 29 November 1939, a name intended to mask its radar function. Sir John
Madsen, Professor of Electrical Engineering at Sydney, went to England for
discussions with Tizard and Watt among others. On his return the Radio-
physics Advisory Board proceeded on a course of radar training, research
and construction. In September 1940 Wing Commander A G Pither of the
Royal Australian Air Force (RAAF) left for training in England, whence he
returned in May 1941 to take over Section 7 of the Directorate of Signals
with responsibility for radar. On 15 September 1941 training began with
6-month courses for officers in radiophysics at Sydney University and for
mechanics in maintenance at Melbourne Technical College. Instruction of
220
women members of the Air Force to become operators commenced in June

1942.
In October 1941 Australia had three radar sets, one of their own exper-
imental construction and two British. Additional imports would certainly
be far too few for the wide expanse of the continent’s northern coast and
island responsibilities. Concern about air defense became acute after Pearl
Harbor and brought forth from the Radiophysics Lab an air warning set,
designated AW, working on 1.5 m that was successfully placed in operation
for Sydney’s defense only five days later.
Australia had no large electronic manufacturers, but this proved to
be no hindrance in making either the AW or its successor, a light-weight air
warning set, referred to only by the initials LW/AW, and surely one of the
most remarkable pieces of radar equipment to emerge from World War II.
Its critical light-weight antenna—making it the first radar set transportable
in a single airplane—was designed and built by the New South Wales
Government Railways in conjunction with the RAAF. The electronics for
both were designed by the Radiophysics Lab, who also made the first six
production sets during the early weeks of 1942, after which production was
by HMV Gramophone. There is nothing in radar history to compare with
this feat for speed linking development to full production and then into
action. Quite obviously the circumstances had ‘concentrated their minds
wonderfully’. LW/AW, which came off the production lines in the last
half of 1942, proved to be perfect for the fighting that soon enveloped the
islands of the southwest Pacific and equipped numerous American units
during the early years. The Australians also coined (of unknown origin)
the name ‘doover’ that held its own against ‘radar’.
Components for a prototype AW set were flown to Darwin on 5 Febru-
ary 1941; it was not operational in time to help counter the first destructive
raid of 19 February but did detect the arrival of the fifth, becoming the first
RAAF station to detect an enemy raid [11]. After the first radar-detected
raid on Darwin the Commander of the American fighter Group asked that
someone from the RAAF be stationed at Cape Fourcroy on Bathurst Island
to extend the range of air warning. The first radio report by Corporal Bill
Woodnutt gave an extra 20 minutes warning that enabled the P-40s to gain
height and intercept with excellent results. One observer and one radar
blunted further Japanese raids [12].
In February 1942 the first American troops, intended originally to re-
inforce Philippine or Dutch East Indies garrisons but unable to reach them,
arrived in Australia. They brought with them a few SCR-268s, which they
turned over to their hosts, as no AA units came with them. There being
no directors to link them with Australian AA guns, they were modified for
air warning and called Modified Air Warning Devices (MAWD). Obtain-
ing ranges as great as 160 km, MAWDs became highly esteemed doovers,
particularly in the Darwin area.
New Zealand was no less interested in radar than Australia and sent
221
Dr E Marsden to the same conference in Britain as attended by Madsen

in 1939. Both Auckland and Canterbury Universities trained scientists
in radiophysics, whom the Department of Scientific and Industrial Re-
search employed to design some experimental air warning equipment.
New Zealand elected not to produce their own equipment but to import
British equipment, something their less vulnerable strategic position al-
lowed. The radar program yielded a large number of trained personnel
who, in addition to serving with their own Royal New Zealand Air Force
in the Solomons and beyond, could be found wherever the RAF went, from
Greenland to Burma [13].
The attack on Pearl Harbor, followed a couple of days later by Hitler’s
declaration of war, brought the United States, both people and government,
to a state of high consternation and replaced feelings of security with those
of apprehension. AA guns appeared in New York’s Central Park and the
Greenpoint section of Brooklyn, and of course, there were false alarms.
Such events took place more frequently on the west coast with radio broad-
casting abruptly stopped and an impromptu blackout imposed on the San
Francisco region, accented by a submarine putting a few shells on the sa-
cred soil of California. In February the air over Los Angeles was filled with
bursting AA shells attempting to destroy an unidentified balloon.
No branch of the Army could have been more harassed during those
first few months following the outbreak of war than the Signal Corps. It
was deeply involved in expanding, purchasing equipment for the combat
arms as well as itself, instituting training for its extremely technical func-
tions, not to forget manning the electronic defenses needed to protect the
continental nation and its outlying territories. In such trying times it is
considered wise to call in experts to help officers and officials to do what
they have been wanting to do anyway, and such support was found for
those working in radar in the person of Robert Watson Watt.
Immediately after the Pearl Harbor attack the US military mission
in London suggested that Watt visit the United States to advise on radar
and associated matters. He accepted with alacrity and was on his way to
Lisbon for the first leg of his flight to the United States by 9 December.
On arrival in Washington he had discussions with Secretary of War Henry
Stimson followed by meetings with lesser dignitaries [14]. Stimson had
seen the great value of radar quite early, was naturally distressed about
the Hawaiian turn of events and wanted a report from Watt [15]. Secretary
of the Navy Frank Knox made no similar request.
After a visit to the Radiation Laboratory and sometimes in the com-
pany of Alfred Loomis (Stimson’s cousin and an accomplished radarman
in his own right), Watt examined the air defenses on the west coast, at
Panama and in the Caribbean. He found nothing but praise for what he
found at the Rad Lab and nothing to praise in the Signal Corps. Much of
what he criticized was the result of finding inadequately trained (or even
untrained) personnel in an army mobilizing at an unprecedented rate, a
222
problem about which the Corps considered itself adequately informed, but
he also demonstrated a narrow point of view that had been hardened by
four years away from the laboratory in his capacity of Director of Com-
munications Development. He dismissed all Signal Corps radar as nearly
worthless and pronounced that long-range early-warning radar must work
in the 10 m band and hence could not be truly mobile, the obvious success
of the Opana set notwithstanding. Consequently, he insisted that Chain
Home stations be erected at many crucial locations, relying on the pur-
chase of 100 MRUs temporarily until the required CH towers could be
erected. The deficiencies he found at Panama, primarily the inability of
following aircraft inland as a result of the mountainous terrain and of de-
tecting low-flying attack, were inherently those of meter-wave equipment,
but he ascribed them to the infamous design of SCR-270 [16]. To cover the
dangers of low-flying attack the Corps modified the 1.5 m SCR-268 to serve
for air warning as the Australians did in making MAWD, designating it
SCR-516 and incorporating a PPI display.
The Army Air Forces had sought earlier the procurement of Ground
Controlled Interception (GCI) equipment, which became SCR-527, but
none was available at this critical time. Watt insisted that the west coast had
to have GCI, but here reality took over. GCI was inherently a British, in fact
an English set. Its function required siting in a shallow bowl with a mini-
mum of fixed targets visible outside the bowl. This requirement is easily
satisfied by England’s gently rolling countryside but not on the American
west coast, where reconnaissance had failed to find even an approxima-
tion of the right surface conditions in the Pacific northwest with things not
much better to the south [17]. Everyone agreed there was a crying need
for ASV mark II for patrolling both coasts, but there were none.
Watt’s report to the Secretary of War was received with satisfaction
in Washington, especially by Colonel Gordon P Saville, Director of Air De-
fense, whose use of it did not improve relations between Air and Signals
[18]. Watt was thanked profusely and awarded the US Medal of Merit.
He left Signal Corps officers writing detailed rebuttals and striving to get
orders for CH and MRU equipment canceled. Within weeks the visit had
been forgotten, swallowed by extraordinary amounts of work. For what-
ever value Watt’s report had to the Signal Corps it left the unmistakable
understanding in the War Department that Stimson considered radar im-
portant. On 1 April 1942 he added Edward Bowles of MIT as ‘his special
consultant for the purpose of getting radar upon a thoroughly sound and
competent basis’ [19], allowing him to apply not too subtle incentives when
he and Stimson thought it expedient, something which worked strongly
to the Corps’s benefit.
223
5.3. THE CHANNEL, 1942

5.3.1. The dash of the warships
Students of comparative naval theory had assembled for them toward the
end of 1941 at Brest a conundrum worthy of their best analytical minds,
and because the solution was more than just an exercise for the faculty of
some war college it engaged an increasingly large circle of thinkers, who
in turn activated an ever increasing array of military force. In March the
battle cruisers Scharnhorst and Gneisenau had ended a successful cruise of
commerce raiding and went to Brest to prepare for their next sortie. Initially
Raeder planned to have them join with the Bismarck and Prinz Eugen in a
progressive advancement of the technique, but problems with their boilers
could not be resolved in time for the action, and a change in the fortunes
of war had brought the Prinz Eugen fleeing to Brest without having sunk
a single merchantman and having left the Bismarck resting on the ocean
floor.
Thus one naval theory, the value of warships for commerce raiding,
had just lost a degree of credibility, although its proponents were by no
means inclined to admit a general failure of doctrine. The presence of
three large-gun warships in Brest brought into play another naval theory,
one guarded by the sacred writings of Alfred Thayer Mahan: the value of
a fleet-in-being. So would the matter have remained a generation earlier,
but by 1942 there was another theory with which the contending admirals
had to deal, one that came from the pen of Billy Mitchell, a prophet as
highly honored by airmen as Mahan by seamen. Succinctly put, it stated
that surface ships, however well armed and armored, could be destroyed
with comparative ease by land-based heavy bombers. As so oft in affairs
of abstract thought, the situation at Brest by the end of 1941 was concluded
with little regard for theory.
A fleet-in-being is one harbored safely behind coast defense guns,
mine fields, breakwaters, nets, booms and torpedo boats, and by 1941 it
was one also guarded by components of air power. By its mere existence
it is a threat to an enemy who wishes to travel the adjacent waters. This
requires that forces must be at sea or in the air to prevent it from departing
unexpectedly. The presence of these ships at Brest meant that convoys
would have to have battleship escorts as well as the lighter vessels for
defense against submarines; three heavy ships at Brest skewed completely
the dispositions of the entire Royal Navy.
Brest lay within easy striking range of British heavy bombers, and
although the RAF had never placed the abolition of warships as part of
their mission, as had their sister service across the ocean, they were as-
sumed to be capable of sinking ships with bombs. Almost automatically
Bomber Command reluctantly found itself expected to destroy the ships at
Brest—reluctantly, because its avowed mission was to win the war alone
by destroying Germany, and every digression from this was resisted. This
in turn required the Luftwaffe to defend the ships with large numbers of
224
fighters and AA guns at a time when these resources were much needed
on the Russian front and for defending the Reich against ever growing air
attacks. So the three ships became a distracting evil for the Kriegsmarine,
the Royal Navy, Bomber Command and the Luftwaffe.
Into this tangle of conflicting theories stepped Hitler. He had decided
that the British would soon invade Norway in order to secure the northern
route to Murmansk and Archangel and to rid themselves of some trouble-
some U-boat bases, which he rightly saw as a greater threat to his enemy
than surface ships. This meant that these three ships had to be posted to
Norway to aid in its defense, and if the experts were of the opinion that it
was impossible—because of radar among other things—to return to Ger-
many by the old northern route, then they must return through the English
Channel. This brought on a naval chorus crying that it could not be done
and that another Atlantic sortie was the better use of the ships. Hitler
disregarded all this and ordered them to procede through the Channel,
threatening to remove the heavy guns for coast defense and pay off the
ships, if the admirals could not bring themselves to the task. This certainly
fixed the attention of the naval staffs, who, on examining the matter in
detail, began to realize that ocean raiding was probably finished and that
a dash up the channel might not be as foolhardy as their first reactions had
given voice [1].
On the English side of the Channel, naval thought began coming to
similar conclusions, for although little damage had been done by the thou-
sands of bombs dropped where the ships were thought to be, eventually
they would be destroyed or at least damaged beyond the repair capabili-
ties of Brest. A dash up the Channel was the only way of saving the ships.
That they were worth saving was a thought as natural to British minds
as to those of the Kriegsmarine. Only Hitler seemed to have realized that
events had reduced them to three nuisances.
Thus two sides began to plan for the dash up the Channel that be-
gan in the evening darkness of 11 February 1942, ten months after the two
battle cruisers had taken refuge at Brest. The Germans produced an ex-
tremely detailed plan that provided for scheduling the ships’ movement
to the minute, for sweeping mines from the route, for using radar for nav-
igation, for jamming the British radar, for preparing a vast cover of fighter
planes and for central control of all these activities. The British plan con-
sisted of alerting high-level commanders that the ships were expected to
attempt the passage. There were separate commanders for the Fleet, Fleet
Air Arm, Army Coast Defense, Coastal Command and Bomber Command
units, all of whom had responsibilities in combating the vessels, and no
direct communication links between them. Secrecy allowed no planning
at the command levels where the details of a response would have to be
carried out, and the news of the breakout caught them by surprise. It was
assumed that the Germans would elect to pass the Straits of Dover at night,
which meant they would have to depart Brest during daylight, which the
225
Germans inconveniently and prudently decided not to do.

Radar featured prominently for both sides in this action. Gen Martini
personally supervised this aspect on the German side. The Kriegsmarine
insisted on complete radio silence to include the ships’ radar. This pre-
vented them from using their Seetakt sets to obtain their positions from
prominent land features or radar beacons. Given that the antenna arrange-
ments of the Seetakt allowed almost no back-leakage of radiation and that
their direct beams could hardly have been received on the English side
until near Dover, this was probably an unnecessary and not very helpful
restriction, but the squadron was able to navigate successfully neverthe-
less. As a substitute the shore stations were to locate the ships, which had
IFF, and radio their positions to them [2]. In practice this proved diffi-
cult, owing to the inherently poor directional accuracy of the shore-based
Seetakt sets and an unreliable IFF performance that made distinguishing
between the ships difficult [3]. On top of this, communication failures at
times failed to get the information assembled and transferred [4], so the
navigation officers had to pay close attention to their fathometers and the
depths indicated on the charts. There were also a series of marker boats at
various points.
The British radar of which Martini was aware was the 1.5 m CD/CHL
sets all along the coast. No shore-based radar would be able to observe the
squadron except in the vicinity of the Dover Straits, so Martini set up 1.5 m
jamming devices there, starting with very little interference but increasing
it slightly from day to day until the level was thought adequate to mask
reflections from the ships. This tactic did not fool Lieutenant Colonel B E
Wallace, who quickly called attention to it but was ignored. In desperation
he went to R V Jones, Head of Scientific Intelligence, and implored to be
taken seriously. Jones had a top officer from the Telecommunications Re-
search Establishment on the spot the next day—just in time for the passage
of the ships [5].
Unknown to Martini, however, there were new 10 cm NT 271 coast
defense sets (also called CD mark IV) installed on the southern coast. These
were essentially the same as the Navy type 271 but with a 2 m paraboloid.
Four that were within range of the squadron’s path were located at Ventnor,
Beachy Head, Fairlight and Dover. A fire had put the Ventnor station out of
operation the night before, and it may have been out of range anyway, but
the others worked just as intended. During mid-morning of 12 February
1942 the Beachy Head and Fairlight stations reported a force of large vessels
at the far side of the Strait moving much faster than a normal convoy. These
two reports then began winding their way to the Dover Coast Defence
Operations Room. At the same time aircraft observed the ships visually
but made no report until after landing, as strict radio silence prevailed,
and their report also began its way toward Dover. Well over an hour later,
Dover observed the ships with their own NT 271. It was their first news
of the affair, as word of the other sightings had still not arrived [6].
226
The most important radar sighting was the one that was not made.
Patrols had been set up for Coastal Command in three regions in the vicin-
ity of Brest by Hudson bombers equipped with ASV mark II. These planes
had been watching the harbor for seven months and their procedures had
become slack. The patrols did not overlap and equipment failures were
not allowed to disturb routine. ASV mark II was capable of spotting sur-
faced submarines, and a capital ship would have been the largest target the
operators had ever encountered, but they were not there. In a maddening
series of equipment failures—both in aircraft and in radar—the German
ships passed the patrol regions when they were not covered [7].
Martini made no provision for jamming these sets, which is strange
given the demand that there must be no discovery during the dark part
of the passage [8]. It is especially puzzling because British ASV capability
had been known since the preceding May [9], yet the detailed plans saw no
problem in the possibility that air surveillance might have radar. One can
only conclude that this was another of the many examples of something
kept secret from those who needed to know.
The feeble attempts to sink the escaping ships concern this account
only peripherally. All attacks were made piecemeal. Some were marked
by great courage enhanced by knowledge of the small chance of either suc-
cess or survival. The heavy guns at Dover, which initiated British action,
seemed impressive until one inquired about their rates of fire and ability
to follow moving targets; none of their 33 rounds hit. Motor torpedo boats
failed to penetrate the protective cover of the German destroyers and E-
boats to a range that would have allowed hope of a successful launch. Next
came the pathetic attack of six Swordfish torpedo bombers, all of which
were shot down. Had they been able to attack as planned at night, the
specialty in which they excelled, the result might have been otherwise, but
they attacked during the day with one-tenth the fighter cover intended and
suffered the fate of all torpedo bombers that had to face an overwhelming
fighter and AAdefense. Aflotilla of six 20-year-old destroyers went straight
after the big ships accompanied by Beaufort torpedo bombers properly es-
corted by Spitfires. In the waning daylight and bad weather the mixture
of destroyers and aircraft from both sides gave generous examples of mis-
taken identity with fights between enemies at times appearing to be the
exception, because visual IFF had problems as severe as radar IFF. Bomber
Command’s high-level attacks at the end of the day seldom found the
target, let alone hit it.
As night closed the squadron had to thread its way—not too success-
fully—through mine fields. This would have been an excellent time to use
Seetakt with radar beacons on shore to navigate, but strict radio silence
was still enforced. The two battle cruisers were damaged by encounters
with three mines. The Gneisenau received a blast followed by subsequent
bombing from which she was never to recover; the Scharnhorst went on to
meet a futile but heroic end in the waters off northern Norway; the Prinz
227
Eugen served in the Baltic and ended her career as a test ship for an atomic
bomb at Bikini Atoll.
The removal of the fleet-in-being from Brest greatly simplified the
Royal Navy’s arduous Atlantic duties, so Britain emerged from this in a
better position strategically but having suffered a major defeat psycholog-
ically. It happened at the time of the sinking of the Prince of Wales and the
Repulse and the loss of Singapore. The reaction of the public and the lower
levels of military command was one of fury, as it was seen to be incom-
petence of the kind demonstrated at Pearl Harbor. The Prime Minister’s
investigation found everything in order, and no senior officers were disci-
plined or replaced. Disparaging remarks in the report about the capabilities
of radar triggered an extensive response in Watson-Watt’s memoirs [10].
5.3.2. The Channel Convoys

Despite its proximity to the enemy, the Channel was an important supply
line for both sides. Britain’s western harbors could not handle all of the
island’s cargo, and distributing it inland strained an already overloaded
rail network. This meant shipping had to face the dangers of attacks by
German torpedo boats, long-range guns and aircraft. The coin had two
sides, of course; transport shortages required the Germans to make use of
their own convoys bound to and from the French and Lowland ports. Each
side kept its ships close to a friendly shore, but neither shore was friendly
around Dover and Calais.
The Germans had set up long-range guns shortly after occupying
that part of the coast, and on 11 November 1940 scientists at His Majesty’s
Signal School were informed that a convoy had been subjected to accurate
gunfire at night. Suspicion of radar brought N E Davis, an experienced
Marconi television engineer, to the scene with a receiver capable of a wide
range of wavelength. (Unknown to the Navy, technicians from TRE had
observed the Seetakt radiation the month before and identified its origin
as radar [11].) Davis quickly determined the radiative characteristics of
Seetakt, providing the certain evidence of the existence of German radar.
He then set out to provide a jammer.
Using a high-power decimeter-wave triode Davis built an oscillator
tunable with Lecher wires over the Seetakt frequency band. Restrictions
on the form of jamming modulation came from fear that an effective noise
might be used against CH, forcing him to use a sinusoidal form. By Febru-
ary 1941 he had six experimental jammers in the Dover–Folkstone area,
which greatly reduced the accuracy of the gunfire. It was the beginning
the ‘radio war’ at Dover.
The experimental sets were replaced by engineered versions, the
Navy type 91, which had large waveguides fitted to flared apertures fed
with wide-band dipoles. The presence of Würzburgs, which could be used
for coast watching, was known by then, so the frequency range was ex-
228
tended to encompass them as well. The Germans countered by altering

frequency, and so it went. As the radio war expanded an inter-service or-
ganization was set up to collect relevant intelligence. Secrecy deemed it be
called the Noise Investigation Bureau [12].
5.3.3. The Bruneval Raid

With the defeat of France the English Channel became the no-man’s land
of the Second World War. As on the shell-scarred ground between the
trenches of 1914–1918, patrols felt out the enemy but by boat or aircraft
rather than by crawling in the mud. Lives disappeared in clashes forgotten
except in the most detailed of official histories and in the memories of loved
ones and survivors. Both shores saw the placement of the tangible elements
of the earlier war of attrition—barbed wire, mines, machine-gun posts,
trenches and boredom. And as before, on both sides radio technicians
set up receivers to record any transmission the enemy might make, while
others set about to decode them, after which staff officers tried to fit the
information so obtained into a coherent picture of the enemy activities or
plans.
A new component of this electronic warfare was the signals intelli-
gence units that specialized in radar. The Germans had begun this be-
fore the Battle of Britain and had quickly identified the CH and CHL/CD
chains. There remained for some months afterwards the general British be-
lief, unquestioned at upper levels, that the Germans had no radar. It was a
belief not shared by TRE, where radar had lost its mystique, nor the Royal
Navy, nor by R V Jones, where the combined results of the Oslo Report,
the interrogation of prisoners and a number of other pieces of intelligence
pointed strongly toward German radar [13]. As noted earlier, both H M
Signal School and TRE had observed Seetakt emission in late 1940. The
same techniques had seen signals from Freya but had confused them with
emission from British equipment of similar frequencies, and their certain
identification as radar had to wait until February 1941 [14].
The conclusive piece of evidence for Jones had to be a photograph,
and for this he was in luck because the Air Force had had by the end of
1940 the remarkable Photographic Reconnaissance Unit under Wing Com-
mander Geoffrey W Tuttle. This group of pilots flying unarmed and very
smooth-surfaced Spitfires could sneak into and out of the most heavily
guarded region to take pictures, which they then turned over to equally
skillful interpreters. This capability came into being at the beginning of
the war through the irrepressible efforts of F Sidney Cotton, wealthy Aus-
tralian business tycoon, flier with experience in the previous war and pho-
tographic expert of the highest order. With private money he began doing
his own freelance photo-reconnaissance of German military positions in
January 1939. His annoying methods—the worst being that he did a much
better job than the old RAF photographic units—alienated nearly all in
229
the Air Force except Dowding, whose keen technical understanding con-
vinced him to give Cotton two of his precious Spitfires at a time when
Fighter Command needed them desperately [15].
Once the identification was certain that the emissions from Freya were
radar, Jones quickly succeeded in obtaining a close-up view of one of the
stations [16], which finally convinced the top levels of British command on
24 February 1941. The success of electronic intelligence by ground stations
naturally called for equipping aircraft with suitable receivers, and a flight
of 109 Squadron was outfitted with receivers in Wellington bombers, which
gained the name of Ferret for this kind of aircraft, and began searching for
emissions over a wide spectrum. By October electronic intelligence had
located 27 Freya stations [17]. On 7 May they made the acquaintance of
the Würzburg, observed as pulsed 50 cm transmissions from nine locations
[18], but no suggestive antennas showed up on the kind of photographic
coverage that had given the first hints of Freya. The information gained
from the Ferrets indicated that the Würzburg had a much narrower beam
than the Freya, and Jones suspected that this was superior equipment al-
ready widely deployed, possibly the cause of the unnerving speed with
which searchlights were being brought to bear on bombers. The character-
istics of Freya were pretty easily understood from the emissions, but this
new set was surprising, and Jones wanted details. The need for a picture
was obvious.
In late 1941 a careful study of a Freya station revealed an object close
by and apparently associated with it, small but curious enough for Jones
to insist on a close-up shot. This required two flights for success, but the
second gave Jones the picture he wanted. This first view of a Würzburg
disclosed two intriguing elements: the device was indeed small and lo-
cated close to the beach. Could it or some of its important components be
taken in a commando raid?
Jones was reluctant to recommend action that put the lives of many
men at risk to secure this information, but a consultation with W B Lewis,
by then Deputy Superintendent of TRE, bolstered his belief in the value of
learning the details of the Würzburg. A raid also fitted with Churchill’s
wishes to agitate the German shore defenses as much as possible, so the
request went to Lord Louis Mountbatten, Chief of Combined Operations,
who favored the operation enthusiastically and ordered preparations. A
parachute company underwent the detailed training for a night landing
near the village of Bruneval where the radar station was located, and crews
of the landing craft that were to remove the raiders and their precious loot
from the beach practiced on a similar coastline.
The raid took place on the night of 27/28 February 1942 and was
an unqualified success. Its professionalism, greatly admired by the Ger-
mans, made up for the bewilderment that had marked British response to
the battleship dash of two weeks earlier. The disposition of the garrison
was known to the attackers from aerial photographs and the reports of
230
French resistance agents, and it had been taken completely by surprise.

The entire Würzburg was too large to be taken, but transmitter and re-
ceiver modules were easily removed and the antenna feed, which had the
dipole, was sawed off. The radar man made understanding sketches and
took a couple of photos until the camera’s flash drew fire. They loaded
the captured equipment onto a cart brought for that purpose and led a
captured Würzburg operator with it onto the landing craft on which they
departed. Loss of life on both sides amounted to two British and five
German [19].
Four important pieces of information resulted from what was brought
back. First and most important, the Würzburg was a fixed frequency set
with a very narrow band over which it could be tuned. Second, it had
no circuitry designed to deal with countermeasures. Third, it was an ex-
tremely well engineered piece of equipment, having modular design that
made isolating faults and repairing them extremely easy. Fourth, the op-
erator prisoner, though extremely cooperative, had poor technical compe-
tence. These last two characteristics were inversions of British procedure,
which was to move prototype designs to production as fast as possible and
have them operated by personnel trained almost to the level of electrical
engineers.
Another benefit accrued from the raid. German radar stations became
heavily entrenched, making them easy to spot from the air [20]—but also
difficult to take. For whatever it was worth, the raid, combined with strict
secrecy, gave rise to extravagant rumors within the Wehrmacht.
5.3.4. The move of TRE to Malvern

If the raid had convinced the Germans to fortify their radar stations, it
also made the British apprehensive about the exposed location they had
selected for TRE at Swanage. The move there in the summer of 1940 from
the admittedly unsuitable quarters at Dundee must remain a puzzle to an
outside observer. Fear of German air attack had caused the rapid departure
from Bawdsey to Dundee, but Swanage was a terrible location after the
defeat of France, yet the laboratory grew during the weeks when invasion
was by no means a trifling danger. After the Bruneval raid, high command
became concerned about a possible retaliation, and the knowledge that a
German parachute company had been stationed near Cherbourg began to
make a number of people uneasy.
Superintendent A P Rowe, who had been greatly alarmed by the posi-
tion at Bawdsey, had become quite comfortable at Swanage. The laboratory
had spread about the region and numbered 1000 employees. The method
used to procure his cooperation was to provide rumors of German activ-
ity. The parachute company quickly grew in the telling to be ‘seventeen
train-loads’, and the region around the laboratory was fortified. Rowe was
convinced the Prime Minister had ordered the evacuation. For whatever
231
reason, a new site was soon found in the buildings of a college at Malvern
overlooking the Severn Valley in Worcestershire. With the emphasis on
microwaves there was much less need for a location near the ocean. The
spa town of 15 000 had to take up the addition of the 1000—soon to grow to
3000—employees and their families, and found this decidedly less pleas-
ant than furnishing hotel space for the annual festivals of plays by Bernard
Shaw that had marked the pre-war decade.
So the last and final move of TRE began on 25 May 1942 [21]. It was
not the final name, however. It remains in Malvern today, although not
in the college. Five more name changes have left it the Defense Research
Agency.
5.3.5. The Dieppe Raid

At the first light of day on 19 August 1942 landing craft touched the beach
at the channel port of Dieppe at about the time of similar landings to the
east and to the west of the town. It was an action similar to others on the
channel during preceding months, but this one was conducted at division
strength, almost entirely by the Canadian 2nd Division with some British
Commandos and a token force of American Rangers. The interrogation of
an Allied officer prisoner posed a question: ‘What was it? It was too big
for a raid and too small for an invasion’. The answer speaks across the
decades: ‘When you find out, tell me’.
The first wave was met by infantry in their combat stations who de-
livered a withering fire, and when tanks followed—the first landed in a
raid—they either fouled on the obstacles that the sappers had been unable
to remove or were knocked out by antitank guns already in position. A
western flank landing did better, but not even a dramatic success there
could have altered the enormous measure of the disaster of the center, a
disaster for Canada—of the Canadians engaged 68% became casualties—
about as great in proportion to the nation’s population as the Viet Nam
War was for the United States and one that left as great a psychological
scar. Controversy has naturally followed Dieppe with books and maga-
zine articles in abundance. As the years pass an uncountable number of
relevant documents have been located as German files have been sorted
out and Allied files released from classification. Many have been lost or
were destroyed, and what remains is a great mix—thoroughly stirred by
the Official Secrets Act.
For those alert to the events of the time, Dieppe is well remembered,
in part as something for which rumor augmented the news reports and
led to their distrust. For those who came later the name does not have
the same impact, or has none at all. The work by Campbell in the general
references is an invaluable aid in sorting out the confusion that lies behind
the written record but is not suitable as an introduction. Although written
too soon for access to important material, Robertson’s book [22] presents
232
the important facts in a gripping way.

The radar story of Dieppe is a curious one. Succinctly put, one might
say that nothing happened, seemingly belying the extent of what follows,
but there are occasions where ‘nothing’ requires a bit of explanation. There
are three parts to the radar component of Dieppe: (1) the operational use
of radar by British and German forces, (2) the technical intelligence gained
by the raiders and (3) the effect on subsequent use of radar, specifically for
the invasion.
German coast watching was the responsibility of the Kriegsmarine,
which began to set up Seetakt equipment on the French coast as it became
available, the first going to the batteries of heavy guns at Cap Gris Nez and
Calais, and the first detected by British signals intelligence. A Seetakt had
been placed at Pointe d’Ally, a few kilometers west of Dieppe, quite capable
of observing the attacking flotilla. Its existence was unknown to the British,
but it was removed a week before the landing, bringing British intelligence
up to date. The Kriegsmarine made the change without troubling to tell
the Army, telling something about interservice cooperation.
The British did know of a Luftwaffe Freya–Würzburg pair at Pourville
to the west of Dieppe, the location of the Green Beach landing. Although
a 2.4 m air-warning set, this Freya was capable of observing large surface
craft at 30 km as a consequence of its splendid cliff location [23], more or
less the capability of the Würzburg, but the Allied operations order had
no provision for any kind of radar countermeasures. Air Vice-Marshal
Trafford Leigh-Mallory, commander of air support, had discussed various
means of jamming or deception, but the Pourville Freya was beyond the
horizon of ground jammers on the English coast, so only deception was
considered, such as flying aircraft to distract the operators or having motor
boats tow balloons carrying dipole reflectors. Neither was done [24].
This becomes a particularly puzzling attitude because complete sur-
prise was crucial to the operation, as there was to be no preliminary bom-
bardment. The Royal Navy refused to commit any vessel larger than a
destroyer to the operation for fear of losing it to dive bombers, and Bomber
Command initially refused, owing to a policy of not bombing French cities
at night. When the policy was relaxed by Churchill for the occasion, the
bombardment was rejected by the planners as unnecessary [25].
What would seem to have been an excellent capability of radar warn-
ing for the defenders was spoiled by poor interservice cooperation. The
Freya stations reported air and the Seetakt stations reported surface move-
ments, although after the Bruneval Raid local arrangements were made to
have Luftwaffe stations report maritime activity to the Kreigsmarine plot-
ting centers [26]. The Freya station, under the command of Oberleutnant
Willi Weber, sighted the flotilla at 0232 hours and decided by 0330 that it
was a raid of substantial size. His report to the Navy plotting center was
brushed aside because they thought it was of a coastal convoy headed for
Dieppe from Boulogne. Deduced from the nature of the movements We-
233
ber was sure it was not the convoy and called the 302nd Infantry Division,
where he thought the warning caused the troops to go into a condition
of enhanced alert. The outcome of the landing naturally reinforced in his
mind the opinion that he had given the alarm [27], but the enhanced alert
had had other origins. Since February Hitler had transferred his inva-
sion fears from Norway to France with the result that all lower levels of
command were thoroughly exercised. On top of that, British propaganda
had been trumpeting a second front for 1942, amplified by all communists
or communist sympathizers in the west. Troops went into alert positions
whenever moon, tide and weather were suitable, as they were that night.
The operation did have one radar objective in putting a technician,
Flight Sergeant Jack Nissenthal, ashore on the Green Beach to examine and,
if possible, remove components of the Freya. Unlike the Bruneval Raid,
Jones had not requested this as he considered his knowledge of Freya ade-
quate; the task had been tacked on because the planners of every raid now
wanted something about radar. Nissenthal went ashore with Company A
of the South Saskatchewan Regiment, who were unable to take the well
fortified radar station, but he was able to observe the movements of the an-
tenna, which told him the set had lobe switching [28]. Some extravagant
claims have been made by Nissenthal [29] and others about the accom-
plishments of this exploit. A big point was made and has been disputed
of his having cut telephone lines out of the Freya station [30], forcing them
to use radio to transmit their findings, but these stations had already used
radio for this purpose to the extent that there was little more intelligence
to be gleaned from listening to such transmissions [31]. However that may
be, his adventures on the beach had a distinction denied any of the others,
for he had had a personal guard assigned to protect him or kill him (he
learned later) rather than allow his capture. This resulted from an order
originally intended to apply to a TRE scientist and not altered when the
assignment was given to a technician [32].
The German convoy bound for Dieppe was an unforeseen compli-
cation for the attacking fleet, as the two collided in the darkness with a
resulting exchange of gunfire. This was interpreted by the Kriegsmarine
to be British torpedo boats attacking the ships. British shore radar had
tracked the convoy with their NT 271 equipment since 2140 hours of the
18th, even achieving remarkable long-range plots as a result of anomalous
propagation [33], but this information did not reach Captain John Hughes-
Hallett, the Naval Force Commander because of a communications failure
[34]. Two destroyers in the force had 10 cm type 272 (a modification of the
271) radar but failed to detect the convoy [35]. The reason for this prob-
ably lies in the absence of a PPI indicator in this set [36], which made it
difficult for the operators to untangle the confusion of so many ships. The
resulting melée scarcely helped get things off to the right start for landings
dependent on surprise, although the short fight actually raised no alarm.
Subsequent studies show that part of the confusion arose because some
234
officers thought plots beyond normal radar range, the result of anomalous
propagation, were not trustworthy data [37].
Thus the extensive radar activities add up to very little, if somewhat
more than nothing. German radar contributed nothing to warning the
garrison despite the alert radar commander’s report and insistence. No
jamming or deception was attempted by the British despite the impor-
tance of surprise. Information about the German convoy learned from
shore stations was not passed on to the responsible commander, and his
own ships did not spot it. The one piece of radar intelligence that Nis-
senthal learned, that the Freya had lobe switching, was not deemed worth
preparing a special report [38].
The small thought given to radar in preparing for Dieppe was not
repeated in planning for the invasion of Normandy, somewhat less than
two years later. The invasion had a detailed plan for radar at every level
and used jamming and deception in the most advanced forms as well as
radar in its primary mission. The debacle at Dieppe had many lessons to
teach for the Normandy and Mediterranean invasions. That they had to
result from such a bloody defeat does not necessarily follow.
5.4. CARRIER WARFARE DEFINED

In one day the entire foundation of naval thought had been turned upside
down. On 6 December 1941 the battleship was the principal weapon, by
11 December it was the aircraft carrier. The supremacy of the former had
been accepted doctrine by all navies, but it was replaced in one day’s action,
with the lesson hammered home three days later in the Gulf of Siam for
any who still needed it. The air power demonstrated by six fleet carriers at
Pearl Harbor was simply enormous. There remained the concern that the
fragile nature of carriers, the result of thin decks covering large quantities
of gasoline and explosives, would make their careers short, if exciting, in
the initial phases of a great sea battle, but the dive bomber and the torpedo
plane had demonstrated that the battleship had pronounced weaknesses
too. The coming year was to see the rapid evolution of a new kind of
maritime warfare, one increasingly controlled by radar, something that
had been unknown when most carriers had been laid down and absent
from tactical thought during the design period of all then in existence.
Vice Admiral Chuichi Nagumo, who led the Japanese fleet that had
attacked Hawaii, was concerned for the vulnerability of his carriers and
had opposed the mission as too risky. He was to use these carriers for
the next four months in one of the most spectacular naval exploits of all
time, sinking five battleships, one carrier, two cruisers, seven destroyers
and large numbers of lesser craft without losing a single ship. He was
to drive the British Eastern Fleet away from the water it had to defend, a
humiliation roughly equal to that suffered earlier by the US Pacific Fleet.
The Pearl Harbor Striking Force returned directly to Japan, though
235
two carriers were dispatched temporarily to support the second and suc-
cessful invasion attempt of Wake Island on 21–23 December. Nagumo then
supported the invasions in the southwest Pacific where lay the oil fields
and various minerals and produce of Borneo and Java, vital to sustain
Japan’s war effort. The hastily organized ABDA Command united the dis-
parate American, British, Dutch and Australian forces but had neither the
material nor the essential common training to deal with Japan.
Darwin, on the northwest coast of Australia, was the supply port
for the defense of the Malay Barrier, and Nagumo sent a powerful carrier
attack against it on 19 February 1942 that reduced its ability to function and
destroyed many of the vessels needed. A week later Vice Admiral Takeo
Takagi disposed of the ABDA fleet under the Dutch Rear Admiral K W
F M Doorman without carriers. The newly formed allies fought bravely
and desperately but to no avail, and Doorman went down with his ship.
USS Langley, the Navy’s first carrier, was lost in one of the actions fought
around Java, while serving as an aircraft ferry.
Nagumo’s next target was Ceylon and control of the Bay of Bengal to
secure the naval flank for the invasion of Burma and, for all anyone on the
Allied side knew, to open the way to India and a link with the Germans
in the Middle East. To counter this the Admiralty scraped together a force
that seemed strong enough on paper; three carriers, five battleships and
assorted cruisers; but the carriers had too few aircraft and those inferior to
the Japanese, and the battleships were old and slow. Action ended when
the Japanese left with other business in mind after sinking a carrier and
two heavy cruisers. Nagumo had lost no ships but had to note attrition of
his superb air crews that would prove much more difficult to replace than
the aircraft.
In none of these actions was radar used.
From the safe historical perspective that time allows, one can say that
Japan began the Pacific War unfavorably. The success Japan achieved at
the beginning masked far more enduring errors. Their concentration on
battleships at Pearl Harbor was in part a necessary consequence of absent
US carriers, but their neglect of fuel tank farms and repair yards resulted
from failure to appreciate fully the logistics of modern naval warfare. In
fact they left the American Navy with most of its fighting ability, although
this was hardly appreciated at the time. That the US Navy was outnum-
bered in important units did not result from the losses at Pearl Harbor,
and for whatever losses in ships and territory America suffered, it gained
a public united to a degree that would have been thought impossible the
day before the attack.
Battleships would be important, better said useful in the coming
struggle, but the encounters long planned by all naval tacticians of great
masses of capital ships were not to be. In their place came task forces made
up of one or more carriers as the principal striking element with protection
and support from cruisers, destroyers, oilers, supply ships and—if they
236
were available and fast enough to keep up—battleships. The enemies of

the task force were aircraft and submarines. The new battleships, such
as USS Washington, would be highly valued not for their main batteries
but for the enormous number of 5 inch dual-purpose guns they carried,
especially after the Bell Labs FD fire control radar and the proximity fuze
became standard in 1943. It was a different world from the time of HMS
Dreadnought, which had had initially no secondary batteries at all.
American naval officers had recognized the value of the radar demon-
strated to them in the Caribbean exercises of 1939 and were quick to invent
the Combat Information Center (CIC) to evaluate and act on radar intelli-
gence. At first put into whatever cramped cabin could be found on carriers,
not known for unused space, the CIC evolved over the months and years
into an elaborate battle station for the commander.
When the Pacific War began all American carriers had CXAM air
warning equipment, the RCA production model of the XAF tested in 1939.
On active duty it was to do all things expected of it, if not all things de-
sired. As an early warning set it was as valued in 1945 in its updated
version, the SK, as in 1939. (Later, when radar seemed a province only for
microwaves, the meter-wave air-warning set remained as early-warning
champion, when it was observed that the reflectivity of a jet, unencum-
bered with propellers, radial engines or radiators, was substantially less
for 10 cm than for 1.5 m.)
Use of the vertical lobes for height determination was not considered
in the design of the CXAM as it had been in British sets, but by late 1941 it
was well understood by the Navy radarmen, who could make a fair deter-
mination on a single aircraft approaching at great distance provided that
the vessel experienced very little roll. A gyroscopically controlled stable-
vertical mounting might have helped, but in practice height determination
was and continues to be a difficult art, making the complications of a sta-
ble vertical for such a heavy antenna of questionable value. Good height
determination also requires following the target for a substantial part of
its path with frequent referrals to the fade charts, which is not the kind
of thing one did well when the scope became crowded with targets and
action was imminent.
Perhaps the worst aspects of CXAM were its dependence on the A-
scope display (the plot of blip size against range) and on the operator
having to point the antenna. If there were many planes in the air at various
directions, the radar officer became a very harassed man. Each sighting
had to be hand plotted. The SK, which began to see service in early 1943,
had a rotating antenna and a PPI display [1]. Without computers good use
of these data relied on almost intuitive skills of the radarmen, and careless
preoccupation with one part of the sky could mean being taken by surprise
from another quarter. Small wonder the results were less than perfect in
the confusion of battle. One must bear in mind that fighter control for a
carrier was a different thing than for southern England during the Battle
237
of Britain, as all activity concentrated within a few km of the ship rather

than over hundreds of km of land; reaction times were correspondingly
short and the resolution of target blips more easily confused.
One thing radar was not able to do was replace or even greatly assist
the all-important task of reconnaissance. Early location of enemy ships
was the key to victory in this new kind of battle, and it needed to be done
at ranges greater than the striking range of the bombers. The maximum
range for the detection of surface ships by radar was generally determined
by the curvature of the earth, although atmospheric conditions occasion-
ally produced anomalously long ranges by forming propagation ducts. (In
some parts of the Indian Ocean anomalous propagation is often the rule.)
Moreover whereas aircraft might be detected 200 km distant, if suitably
high, surface ships could be difficult to observe at 15 km with meter-wave
equipment, the consequence of the vertical lobe structure and the curva-
ture of the earth. All carrier planes had much greater ranges than radar.
Reconnaissance was accomplished by the American PBY Catalina and the
Japanese Kawanishi 97 flying boats from island stations, by carrier planes
with fuel substituted for bombs and torpedoes, and by catapulted float
planes. The difficulties surrounding naval reconnaissance exceeded those
of radar. The observing aircraft often did not know their own position nor
consequently that of the ships sighted; for understandable reasons they
were shy about approaching closely, which caused serious misidentifica-
tions, frequently transforming whatever they saw into capital ships.
5.4.1. The raids

The immediate concern of Admiral Chester Nimitz, newly appointed com-
mander of the Pacific Fleet or CinCPac as the Navy chose to call him, was
the protection of his carriers and the initiation of action against the en-
emy. During the first months of 1942 he did this skillfully by task-force
raids on Japanese island bases and finally on the home islands. At first
glance these appear to have been merely a method of striking back at the
hated foe, who was then inflicting slow defeat on American and Philippine
armies, but they yielded a variety of benefits. They kept the carriers at sea
at unknown locations—until, of course, a task force struck—and offered
excellent training and experience against the Japanese before having to
meet them in a full-fledged carrier battle; they did much for fleet morale
and kept the enemy off balance.
On 1 February 1942 two task forces, one under Vice Admiral William
F Halsey with USS Enterprise and the other under Rear Admiral Frank
Jack Fletcher with USS Yorktown, raided the Marshall and Gilbert Islands
respectively. Japan had seized the Marshalls from Germany in 1914 and
had retained them as a League of Nations mandate and had taken the
Gilberts, a British Crown Colony, at the outbreak of war. On these raids
American radar began its apprenticeship in war, and Americans began to
238
add some exotic names to the list that had Shiloh and the Argonne.
Both forces caught the enemy by surprise. The Enterprise’s radar
opened this chapter of naval history by spotting a snooper, which was de-
termined to be on a course that would not allow it to see the ships, so it was
not molested [2]. The absence of IFF made itself acutely felt once enemy
planes were in the air, leading to less than desirable fighter control, but in
alerting the task forces to enemy reconnaissance aircraft, to be disposed
of or not as circumstances directed, the CXAM was superb. The absence
of the FD (mark 4) fire-control radar resulted in poor, sometimes wild AA
fire from the 5 inch guns. The Marshall–Gilbert raids did not do a lot of
damage, but American losses were light.
These raids were followed a few days later by a third under Vice
Admiral Wilson Brown with the Lexington. The objective was the recently
captured harbor of Rabaul in New Britain, a port quickly becoming a major
Japanese base. A Kawanishi discovered the force uncomfortably early but
radar made Brown aware of having been seen [3]; two of the big flying
boats were downed by radar vectoring. The discovery quite predictably
brought an attack by land-based bombers on the morning of 20 February,
but radar gave alert for a timely interception with a repeat performance
for a second attack in the afternoon. The defending fighters broke up the
attacking formations before they could reach the fleet, which prevented
damage to the ships and allowed few of the bombers to escape. Because
of the extreme range the bombers had had to fly, the defenders did not
have to contend with a fighter escort and as a result suffered few losses [4].
Brown called off the attack on Rabaul as not worth the risk to his carrier,
but his presence troubled the Japanese and delayed the advance on Port
Moresby.
At the end of February the Enterprise raided Wake Island and added
two more firsts for its radar. It tracked the attacking planes and corrected
their course through the YE homing radio. The YE was a meter-wave
directional beam used to guide planes back to the ship. It normally rotated
automatically, sending a coded signal that gave the direction to the ship.
Pilots listened as its beam swung by them and thereby learned the course
they were to take to return. Its directionality made it a relatively safe way
of communicating with the attackers. After the successful raid a plane with
a non-functioning homing device was located by radar and guided back
by risking a short normal radio communication [5].
The next raid had significant strategic impact. Brown led the Lexing-
ton and the Yorktown through the Coral Sea to attack from the south Lae
and Salamaua on the north side of New Guinea. Surprise was complete
and the raid gave Japan the greatest losses of the war to date, and their
advance on Port Moresby was delayed yet again. Radar was by then just
part of the varied techniques of a technical force.
The last raid of this preliminary period appeared to be one for the
public, the Tokyo Raid, but it so unnerved the Japanese that they made
239
overly hasty plans for what was to lead to the Battle of Midway. Radar’s
observation of picket boats [6] caused Halsey to launch the Army B-25s
earlier on 18 April than planned, but the raid was a spectacular success
and no ships were lost; the bombers were lost, but most of the crews were
saved by the Chinese amongst whom they landed. Japan exacted a terrible
retribution on the populations of the regions that had helped them.
5.4.2. The Battle of the Coral Sea

Japan’s strategic planning was marked from the beginning by disagree-
ments between the Army and the Navy. The overwhelming strength of
the Army was tied up in China and Manchuria where the intentions and
strength of the Soviet Union were matters of great and varying uncertainty.
They refused to provide the troops necessary to invade Ceylon and India,
and their refusal put an end to that Navy plan.
The Tokyo Raid brought these disputes to a head. Admiral Isoroku
Yamamoto realized the American Navy had to be dealt a crucial blow
before the great industrial power behind it could make it too strong. This he
planned to do by initiating a decisive battle. This fitted the Japanese warrior
tradition and had produced the desired effect in their famous defeat of the
Russian Baltic Fleet in the Strait of Tsushima in 1905. Yamamoto elected
to bring the Americans to battle by attacking Hawaii with the seizure of
Midway as the first step, reasoning correctly that this would make the
Pacific Fleet fight. Preparations for this grand battle began immediately
after the Tokyo Raid. There were to be no more such insults.
A faction in the Imperial Naval General Staff had been favoring the
isolation of Australia from American help with the drive on Port Moresby,
and the weakness of Australia’s defense made them disinclined to drop it.
The General Staff decided that Port Moresby was to be taken, as it could be
accomplished with relatively small naval and military forces. The recently
evacuated Australian base on Tulagi Island in the Solomons was to be oc-
cupied and an air field built on the larger, neighboring island, Guadalcanal,
while another force would take Port Moresby, on the south side of New
Guinea. These two bases would permit a thrust to the south that would
control the approaches to Australia. Yamamoto’s Hawaiian strategy was
not immediately apparent, but this move was; following President Roo-
sevelt’s order that Australia and New Zealand were to be defended, Nimitz
moved to counter the threat by sending two carriers under Fletcher to the
Coral Sea to stop the invasion of Port Moresby.
The Battle of the Coral Sea was the first important naval battle fought
during which the surface vessels never saw one another. It had everything
needed to try the nerves of the participants: confused reconnaissance, gar-
bled and failed radio communications, complicated refueling of ships and
aircraft, weather manifestations to drive everyone mad, difficulties in coor-
dinating air strike units that were straining to hit the fragile but dangerous
240
enemy first and worries about whether aircraft would be able to find their
way home in the dark—and whether home would be waiting for them.
The degree to which radar had taken hold of the naval airmen was il-
lustrated in an order by the commander of the Yorktown assigning his most
experienced squadron commander as Fighter Direction Officer, which re-
quired him to take his duty in the radar room, not in the cockpit of a fighter
[7]. Given the responsibilities and the fallibility of his equipment it was
not an enviable job and certainly not the one sought by an accomplished
flier. For the aircraft of both carriers there were only six IFF sets, and two
were lost with their Wildcat fighters in a strike against Tulagi that opened
the battle.
Radar picked up a snooper on the morning of 5 May with the now
well exercised CXAM, but it being too dangerous to guide the pilots to the
encounter by radio they had to head out on a compass bearing and depend
on skill and luck, which prevailed [8].
On 7 May the carriers Shokaku and Zuikaku under Rear Admiral
Chuichi Hara (the whole Port Moresby force was commanded by Vice
Admiral Takeo Takagi) launched an early strike force that destroyed a fleet
oiler and a destroyer, misidentified as a carrier and a cruiser. At about
the same time an American reconnaissance plane reported surface vessels
that were taken to be carriers through a coding error, and they launched
their ‘knock out blow’, not at the Shokaku and the Zuikaku but at the cov-
ering force for the invasion, which did have the light carrier Shoho. The
Shoho and her consorts were caught completely by surprise and the Amer-
icans showed the good effects of their on-the-job training. Their attack on
the carrier was nearly perfect—except that American torpedoes were slow
and often did not run true or explode—and demonstrated the fragility of
carriers by quickly converting the vessel into an exploding inferno.
The remainder of the day was spent by each side unsuccessfully seek-
ing to find the other. A Japanese strike went ineffectually after another
group of the wrong Allied ships, which were bombed equally ineffectu-
ally by US Army Air Forces planes from Australia. Some of the returning
Japanese aircraft were ambushed by Fletcher’s fighters on the basis of a
radar sighting [9], some were lost trying to find their way home in the
dark, all significant losses of excellent flight crews. That night found both
Fletcher and Hara intently examining every piece of what was by then
a huge amount of confusing and often contradictory evidence as to the
location of the other.
At first light of 8 May reconnaissance planes from both task forces
sought and found their adversaries with strikes launched immediately.
The American attack, which obviously came as no surprise, did not have
the sting as the one on the Shoho but damaged the Shokaku seriously enough
to put her in repair yards for the coming decisive Battle of Midway. Zuikaku
hid in a convenient rain squall. Both Lexington and Yorktown were dam-
aged, only the former seriously, but inexperience in damage control re-
241
sulted in gasoline fumes accumulating unnoticed in an electrical room

where a spark set off a tremendous explosion followed by uncontrolled
fires that destroyed the Lexington.
It certainly had not been the Lexington’s day. Her radar had picked up
the approach of the attackers, but the height determination was sufficiently
in error to cause half the interceptors to miss entirely and the rest to engage
too near the ship for good defense [10].
As the battle ended, warning signs began coming from a very secret
source. The US was reading the Japanese wireless signals and piecing
together their strategy from an abundance of clues. The picture that came
from the decoding rooms at times bordered on the incoherent, obtained
as it was from the messages that could not go by wire and underwater
cable, but Nimitz had compared the cryptographers’ earlier handiwork
with subsequent events and had come to trust it. A recent change in code
had made life difficult for the decoding personnel, but by March 1942 they
had broken the new version of JM25. Yamamoto’s plans for a great decisive
battle became, if not clear, strongly suggested.
Fletcher was ordered back to Hawaii where maintenance crews in a
record three days of hectic work were able to patch up the Yorktown well
enough to fight again. American losses of flight crews were much smaller
than the Japanese. The invasion of Port Moresby was called off, and the
two Japanese carriers would not be available for Midway. The Zuikaku had
lost too many of her flight crews; in the Japanese Navy fliers were part of
the ship’s crew and there was not enough time to work up replacements.
In the American Navy air squadrons were assigned to carriers, which were
simply their current floating air fields, and replacement squadrons could
relieve the battle weary. It was an administrative difference but an impor-
tant one.
In preparing for Midway Nimitz added a ruse that radar allowed.
Halsey with Enterprise and Hornet was ordered back to Hawaii from a can-
celed assignment but only after he was certain his force had been sighted.
A snooper, for whom the whole task force steamed, was duly picked up by
CXAM and determined to be on a course that assured he had seen them.
The snooper was allowed to depart unharmed at which moment the task
force made a rapid change of course [11].
5.4.3. The Battle of Midway

In American historical memory two battles will stand on equal footing:
Gettysburg and Midway. Both were fierce, dramatic and decisive, even
though the Civil War would continue nearly two years and the Pacific
War more than three; both left the defeated fighting a hopeless cause. For
either battle to have gone the other way would have had consequences
one does not like to consider. A victory by Lee in 1863 might well have
brought about the severing of the Union; a victory by Yamamoto in 1942
242
would have almost certainly resulted in the United States abandoning the
‘Germany first’ policy, a somber thought when Stalingrad and El Alamein
lay six months into the future.
Midway’s epic nature has naturally resulted in careful study and the
retelling of its story. Here only the barest essentials needed to link the
events with radar will be recounted. The reader who does not know it
would do well to start with Morison [12], a description almost fresh from
the scene. For accounts drawing on a greater number of sources Lord [13]
and Prange [14] are excellent. For those interested in minute details about
the air battles there is Lundstrom [15].
Yamamoto’s plan, which came into Nimitz’s hands from the Pearl
Harbor decoding rooms of Lieutenant Commander Joseph Rochefort, had
three parts: the main carrier force under veteran Nagumo approaching
Midway from the northwest, a diversionary attack on the Aleutians, and
the Midway occupation force approaching the island from the west. Ya-
mamoto committed six carriers against which Nimitz could oppose only
three. The recently repaired Saratoga was on the West Coast, but had nei-
ther escort cruisers and destroyers nor fully trained flight crews to man
her. Yamamoto conveniently attached two carriers to the Alaskan diver-
sion that he could have well used in the main force and that accomplished
nothing in the northern waters; this left the ratio four to three. Yamamoto
followed his carriers with a mighty fleet of battleships; Nimitz had six but
left them at San Francisco, considering them too slow and not yet converted
into the floating AA batteries that would really help. The Americans had
more IFF sets than a month earlier in the Coral Sea, but many fliers had
none.
Early on 3 June a Catalina spotted the Midway occupation force,
and nine Army B-17s dropped bombs moderately near it. Nimitz was not
deceived by this or the Aleutian task force and placed the two task forces
that made up his strength north of Midway. Halsey was in the hospital and
his force, made up of the Enterprise and the Hornet, was entrusted to Rear
Admiral Raymond Spruance, subordinate to Fletcher, who commanded
the Yorktown force.
Unaware of the location, or even the presence for that matter, of
Nimitz’s carriers Nagumo launched a very strong strike at Midway on
the morning of 4 June. This attack was sighted by a Catalina and shortly
thereafter by the island’s SCR-270s [16]. Strike planes immediately left
the island to hit the carriers, including the B-17s that were to drop bombs
all over the ocean in three different sorties that day; Marine fighters in
obsolete Buffalos and a few Wildcats knocked down a few bombers but
were badly mauled by the escorting Zeroes. These attacks on Nagumo’s
carriers did no harm but did contribute to a hectic command atmosphere,
which was not helped by the leader of the Midway strike force calling for
a repeat attack and by the first faltering evidence from a reconnaissance
plane of the position of American carriers to the northeast. The difficulty
243
of CXAM in dealing with the many targets provided by the task force’s
patrolling aircraft, which were incompletely equipped with IFF, delayed
the identification of this observer for the better part of an hour, allowing
him to report the presence of a carrier.
Fletcher and Spruance launched their attack at the earliest possible
moment and provided one of the most dramatic episodes in naval history.
The first to sight and attack were the ineptly named Devastator torpedo
planes. Given the circumstances of their attack it is doubtful whether the
use of the new and much improved Avengers would have made any dif-
ference for they were set upon by fighters with no more protection than
their rear cockpit machine guns. Of the 41 that attacked only six survived
and not a single torpedo went home. Their sacrifice very likely provided
victory, for while the Japanese combat air patrol and AA guns were de-
stroying the Devastators and their crews, dive bombers caught Nagumo’s
carriers by complete surprise.
Radar’s only part in this was in giving certain warning to the Mid-
way garrison, important but hardly decisive. More important was radar’s
absence. Nagumo had no idea that dive bombers were so near and conse-
quently did not make the critical preparations for receiving an attack, and
the decks were full of armed aircraft with filled tanks and gasoline hoses
lying about. Akagi, Soryu and Kaga found the consequences of this in a
matter of minutes. Yamamoto had radar but it was on the battleships Ise
and Hyuga, some hundreds of kilometers behind where it served no func-
tion [17]. Had the US Navy been using electronic intelligence receivers
they could have received a shock. The Ise had a 1.5 m air-warning radar of
mark 2 model 1, and the Hyuga had a 10 cm surface-search mark 2 model 2,
which was given credit for preventing a collision in the bad weather fol-
lowing the battle [18]. When the one remaining carrier Hiryu launched a
strike later in the day on the Yorktown, it was seen by the CXAM in plenty
of time to make preparations that prevented her from being turned into
an inferno [19]. It was to no avail, as two torpedoes from a submarine
(the Japanese variety worked consistently and well) caused her eventual
loss. Hiryu found the same fate as the other three carriers later in the
day.
An evaluation of radar for the early stages of the Pacific War has many
similarities with the evaluation of it for the Mediterranean. America’s
great disadvantage initially was in being outnumbered in carriers and in
having inferior aircraft and less experienced air crews. After Midway the
inequality began to be reversed, but it was a change that had had to be
earned, fate had not foreordained it. The advantages that the US Navy
had during the time before the country’s industrial might came to the fore
were (1) a well trained navy, if not at the peak that Japan had in December
1941, (2) an excellent understanding of the logistic support that a modern
navy requires, (3) the ability to read some enemy signals and (4) radar. The
question turns on whether the Navy would have emerged victorious had
244
any one of these elements been absent; more specifically, could it have won
without radar?
The prelude to the victory at Midway was the carrier raids. Given the
extreme value of an individual carrier, these raids were risky affairs, but
could be undertaken with moderate assurance that the task forces could
not be surprised from the air, and this proved to be the case. The Japanese
examples during those years demonstrated the fragility of these vessels.
Without radar a carrier was like a battleship without armor plate. In effect
radar gave carriers armor. One can easily imagine—and here one enters
the great swamp of conjecture—the aborted Rabaul raid taking a bad end
for the US had the land-based planes not been intercepted before reaching
the Lexington.
The confidence radar built up among its users was obvious by the
Battle of the Coral Sea. Just its use in ambushing the returning strike force
planes on the evening of 7 May may have spelled the difference the next
day by adding to Hara’s serious attrition of air crews and planes. The
presence of those lost planes might have led to the sinking of the Yorktown
too. But this is all idle. There is no end of possible outcomes. There is
no denial that radar was highly valued by those whose lives depended on
it. It was not a perfect instrument but it was something no one wanted
to sail without. CXAM had been designed as an air warning set and had
functioned as planned. Its performance drove all users to make their CICs
into good fighter direction devices. What was sorely needed was IFF, PPI,
very-high-frequency radio and CICs properly set up and manned. Radar’s
value at the tactical level was comparable to the ability to read Japanese
signals at the strategic level.
One tenet of air power doctrine was quietly laid aside as a result of the
experiences of these few months. The Army’s four-engine heavy bombers
proved useless against surface ships. It was one thing to put a bomb into
a pickle barrel from 6 km high, but something else again to put it onto a
fast moving ship. Army B-17s from Midway made a total of four squadron
sorties without hitting a single ship, although they certainly frightened a
destroyer on their last attempt. Such gross inaccuracy was welcomed when
their comrades in Australia had mistakenly targeted American ships in the
Coral Sea but not when they missed the Shoho, which they had attacked
before the carrier planes arrived to show how these things were done. The
Army had obtained four-engine bombers on the basis of their ability to
protect the nation’s coasts; they had been the great hope for protecting
the Philippines. Billy Mitchell’s thesis had agitated defense circles for two
decades and should have been forgotten, but airpower advocates trans-
ferred with an adroit inversion of the scriptures the prediction of carrier
power to their prophet and the myth endures.
245
5.5. THE SOUTH PACIFIC, 1942

Admiral King decided after the Midway victory that the Allies must take
the strategic initiative even though their resources were not adequate, espe-
cially in view of the Germany-first decision. This put him at cross purposes
with the Army General Staff, which was entertaining fantasies of an early
invasion across the Channel and was far more concerned with the slaugh-
ter of Allied merchant shipping in the Atlantic, something for which it and
the British saw the US Navy as showing no interest. There was certainly an
element of truth in the accusation that the Navy was only interested in the
Pacific War. It was the war they had been preparing for, thinking about,
and training for for two decades. It had started for them as a humiliation
and they wanted desperately to redeem themselves. Despite their half-
year apprenticeship as de facto belligerent on the North Atlantic, keeping
U-boats off the convoys by their mere presence, they were ill prepared for
the submarine enemy, ill prepared in equipment, planning, organization
and training, and it was going to take time for those elements to be ready.
In the meantime they were prepared for action against Japan.
It was apparent to the Allies that the Japanese drive toward Port
Moresby, which Tokyo had not written off, had to be contained. The
discovery on 4 July 1942 that Japan was building an airfield on Guadal-
canal provided King with support for the strategic initiative he wanted
and insured that the Army would agree, for such an airfield was almost as
great a threat to Australia as one at Port Moresby, which justified seizing
it, but to do so required attack, not defense. Furthermore the attack had
better be made before the airfield became operational. The only ground
troops available were the 1st Marine Division, formed of units that had un-
dergone multiple fissions to provide cadre for new units being organized.
It not only was filled to the brim with officers and enlisted men having
less than one year’s service but had had no unit training beyond battalion.
They had expected six months training on arriving in New Zealand but
only saw New Zealand briefly on their way to ‘that stinking island’. King
named Vice Admiral Robert L Ghormley Commander South Pacific and
as such, leader of the combined forces Operation Watchtower, laconically
renamed Operation Shoestring by those selected to execute it. The Allies
assembled a major naval force with three carriers to invade an island of
no possible significance except for war, which must have given Ghormley
long thoughts about the lessons of Midway. The Japanese intelligence had
not discerned the plan, making it importantly different from Midway.
The two powers had their logistic lines stretched to the limit, lead-
ing to major ground actions being fought by companies and attempts to
bring reinforcements of regimental or even smaller size precipitating great
naval battles. By the time the issue had been settled five major surface
engagements had been fought—without the meddling interference of air
power—along with two carrier battles in which the men of the surface
vessels never saw their adversaries. In between major actions there were
246
almost daily fights between ships and ships, ships and shore batteries and
ships and aircraft. A restricted part of the ocean north of Guadalcanal be-
came the resting place of more naval tonnage than any comparable place
in the world and was named Ironbottom Sound by the survivors. The air
field on the island, the possession of which hungry, often starving ground
troops contended, was continually subjected to air attack and defended by
gunner and pilot.
Radar assumed an ever increasing importance, becoming by the end
of 1942 a significant weapon, but the infantry noted it not. Their war had
few refinements: rifles, pistols, bayonets, machine guns, explosives, mud,
jungle, insects, disease, boredom, terror.
The reader who does not know the details of this campaign should
correct this fault. The best source for completeness, accuracy and readabil-
ity is Richard Frank.
At the beginning of 1942 the American Navy had CXAM radar
mounted only on a few capital ships. As the year progressed Bell Labs
FC (later mark 3) sets began to be mounted on battleships and cruisers for
main-battery fire direction, and by March FD (later mark 4) sets for dual-
purpose batteries began to appear on all capital ships with some destroyers
outfitted toward the end of the year. These radars operated at 40 cm and
used lobe switching for accurate direction, FC only for horizontal, FD for
horizontal and vertical. Originally designed for the very-high-frequency
triode invented at Bell Labs, their design had been altered immediately af-
ter the disclosure of the magnetron, and they became the first American sets
deployed using this invention. The basic FD design remained the Navy’s
AA fire-direction radar until the end of the war with a 33 cm automatic
tracking version, mark 12, introduced in late 1943 [1].
The FD had, of course, larger lobes than 10 cm equipment, but skilled
crews made good use of the size by tracking one air target while keeping
others on their scopes, ready to be picked up when the first departed the
place of honor. As it was, the beam of the FD was already so narrow that
picking up the target initially required good teamwork between search
and fire-direction radars. On the other hand the large lobes made it nearly
impossible to track very-low-flying planes, something not long kept secret
from Japanese fliers.
For surface targets the FC and FD were less satisfactory, for here the
large lobes did not have the usefulness apparent in AA fire direction. Sea
clutter, the reflection from surface waves, was a nuisance but could be
worked through in firing on single isolated targets, but multiple targets,
especially with similar range, caused the operators attempting to get cor-
rect bearing with lobe switching to aim at the center of gravity [2]. This
defect led to Bell Labs making the mark 8 fire control radar, about which
more in a later chapter. At about the same time the FCs and FDs were being
distributed, cruisers and destroyers began to receive the type SC radar, a
1.5 m set similar to the CXAM but with a smaller antenna that generated
247
larger lobes, of course.

More important, however, than any of these was the SG, the 10 cm set
that resulted from the first Rad Lab experiments that had been begun with
other purposes in mind. After its success on the Semmes, the prototype was
rushed to the Naval Research Lab for alteration that made it agreeable to the
harsh conditions of a warship; it then went to the Raytheon Corporation for
production, starting that company in a field where it would long remain. It
was mounted on the cruiser Augusta (which does not figure in these events)
in April 1942 and on the San Juan and the carrier Saratoga shortly thereafter
[3]. USS Helena had one when she came to the Solomons in August [4],
and when fighting began in the waters among the islands the SG was the
only radar that distinguished clearly ship from shore. What was confusion
on the A-scopes of the meter-wave sets became clarity on the PPIs of SG.
It became a navigation instrument mariners did not want to forego. By
the beginning of 1943 cruisers and destroyers began to carry routinely an
SG for surface search, an SC for air warning and one or more FDs for fire
control. It was their basic radar equipment for the remainder of the war.
The US Navy was well served by Nimitz’s comprehension of radar’s
strengths and limitations. He actively supported the establishment at Pearl
Harbor of the Radar Center comprised of schools for maintenance, oper-
ation, fighter direction and tactics, established in that order. The last, the
Radar Tactical School, gave instruction to commanding and flag officers
and was to prove its worth in overcoming two misconceptions among that
naval population: (1) radar was not used in the Battle of Jutland and was
therefore unimportant; (2) radar was a magic box on which one need only
press the button and the battle was won. Both of these onerous attitudes
were prevalent during 1942 [5].
American Marines had to fight briefly for the tiny island of Tulagi, but
they went ashore unopposed across the channel onto Guadalcanal after a
short bombardment on 7 August 1942, the garrison having been caught
most puzzlingly by surprise and prevailed upon to take up quarters in
the jungle. Even more surprising than the absence of resistance was the
discovery of two radar sets [6].
During the first months of the Pacific War no thought had been given
to the possibility that Japan might have mastered this technique. This was
partly the result of an affliction common to the British and Germans at the
beginning—not believing the enemy was smart enough to duplicate, let
alone exceed one’s own achievements—and partly because the first few
months of the war were occupied with thoughts about what the Japanese
obviously did have rather than what they might have, so no special re-
ceivers were available to search for the tell-tale signals that would have
disclosed radar. Had such Ferret aircraft been in operation earlier they
might have located a similar radar that had begun operation at Rabaul in
the spring.
248
The sets were Navy type mark 1 model 1 for air warning. One set
with its boxlike array of dipoles was restored to service shortly after its
arrival at the Naval Research Laboratory. It operated on the 3 m band with
a peak power of 5 kW, with pulse lengths of 10 to 30 µs [7]. It seemed a
poor thing compared with SCR-270, but that judgement was pronounced
without appreciating the short time that had elapsed between perceived
need and production. Furthermore, the Allies were soon to make good use
of the Australian LW/AW radars of similar power. The greater range of
SCR-270 had to be bought at great price in power and weight; the latter
quantity was at times uncomfortable in the islands of the southwest Pacific.
An SCR-270 was on one of the transports unloading Marines and
their supplies but did not get ashore before the ships were pulled back
to Nouméa; a night surface fight had suddenly left them with little naval
protection, and they departed. In desperation the landed radar technicians
tried to put the Japanese set into operation [8].
From the time of the American landings until the last Japanese with-
drew from Guadalcanal on 8 February 1943 almost continuous fighting on
land, sea and air ensued. The goal of the antagonists was quite simple:
control of the airfield. The methods of attainment would prove to be ex-
tremely varied. For the Americans the object of the ground fighting was
to hold the airfield, named Henderson Field in honor of one of the Marine
aviators who died at Midway. That done they had to expel the Japanese
infantry from the island to insure a more secure operation of the air base.
Naval forces on both sides had to bring reinforcements and prevent
the enemy from doing the same. A key element in this was control of
the air during the day—lack of sophisticated radar prevented any serious
night air activities—but control of the air turned on whether Henderson
Field could function. For the first two days carriers provided the airplanes
overhead, but Fletcher withdrew them as too precious to risk long in such
a tough neighborhood. Marine fighters did not arrive at Henderson Field
until 20 August when the engineers with their single bulldozer had made
it capable of getting planes aloft. Guadalcanal’s air defense came to be
called Cactus after the island’s radio identification.
Attacking aircraft came from Rabaul on New Britain and from Buka
and Buin on Bougainville near the north end of islands so arranged that the
path to Guadalcanal gained quite naturally the name of ‘the Slot’. Early
warning was crucial and came from two sources: radio messages from the
coast watchers, Australians with native assistance positioned on the islands
past which the raiders flew and Cactus radar. The first SCR-270 reached
Guadalcanal on 20 September, quickly followed by two more 270s and
two 268s. A few days later Lieutenant (USNR) Lewis C Mattison and three
other officers arrived from the Fleet Fighter Director School at Pearl Harbor.
Radar established itself as so important that Lieutenant Colonel Walter L
Bayler, an experienced flier, took over fighter direction. The Japanese were
unable to locate and eliminate the coast watchers, so the pilots began taking
249
routes that avoided observation, but by then radar was able to give reliable
warnings.
An effective air defense grew out of the strengths and weaknesses of
the equipment. The SCR-270s detected the raiders, depending on altitude,
at ranges as great as 200 km. The operators became skilled at determin-
ing both number and type of the attacking aircraft but had little success
in determining altitude. The warning time so received sufficed for the
defending fighters to spiral directly over Henderson Field to an altitude
that gave them advantage. Their radio equipment was too weak to direct
an interception in advance of the island, and the pilots did not want to
fly without control, something that produced excellent radio discipline.
When the attackers came within range of the 268s a reliable determination
of their altitude could be obtained and transmitted to the fighters over-
head. An extremely valuable aspect of the radar warning was that after its
installation no air patrols were needed. This conserved fuel, aircraft and
pilots, all in short supply and pushed to the limit [9].
So long as Henderson Field functioned, American surface ships were
relatively secure during the day, but when night blinded the fliers Japanese
destroyers ventured into those waters working as transports, attacking any
American ships they found, and putting a few shells into the American
camps. The rapid arrival and equally rapid departure of these ships caused
them to acquire the name Tokyo Express. This pattern of Japanese supply
was maintained throughout the struggle. The larger naval surface actions,
which were fought almost entirely by gunfire and torpedo, had the same
form, just heavier ships and bigger fights.
Five major surface actions were fought over Ironbottom Sound, and
two carrier battles were fought to the north and east of the islands. At
other times the carriers moved nervously but not idly about the Solomons.
They were fundamentally aggressive units, yet fearful because of their few
numbers and strategic importance. For details of their activities between
and during major battles read John Lundstrom.
5.5.1. The Battle of Savo Island

In 1898 the US Navy fought two very one-sided surface battles with the
outclassed Spanish Navy. Except for the encounter of a few of its ships as
part of the ABDA fleet in the Battle of Java Sea it had fought no gunfire ac-
tions since that time: the main engagements of the war it was then fighting
had employed its air arm. During the night of 8/9 August 1941 it came to
know how the Spanish felt.
Landing the Marines on the 7th had necessarily concentrated a num-
ber of transport and cargo ships near the beach, offering a target of which
the Japanese were aware and to which they responded with alacrity. Amer-
ican carrier planes made an attack by surface ships during the day much
too risky, but the Japanese Navy excelled in night action and sent a sizeable
250
force of cruisers and a destroyer to destroy the supply ships at the beach.
To intercept any raiding force were five American and three Aus-
tralian cruisers with about as many destroyers under Rear Admiral Rich-
mond Turner. Two American destroyers with SC radars, Blue and Ralph
Talbot, were placed as pickets beyond Savo Island. Shortly after 0100 hour
Rear Admiral Gunichi Mikawa led a somewhat smaller but much better
controlled and undivided force from Rabaul out of the Slot around the
south of Savo Island. It should have been detected by Blue but was not.
Japanese lookouts—unaided by radar—saw Blue, but Blue saw nothing
with either optical or radio eyes, and Mikawa left her to continue her pa-
trol. The reason for this serious failure seems to have been a mixture of
poor equipment performance, inadequate training and the confusion for
meter-wave sets caused by the presence of nearby land. The SC displayed
its observations on an A-scope, target amplitude against range with di-
rection selected by the operator. Large nearby land masses, even if many
degrees off the antenna axis, could form echoes capable of hiding targets at
greater range, very much as reflections from the ground obscured targets
for meter-wave AI equipment. The SC, like nearly all meter-wave sets,
also leaked radiation out the back to a small degree, and when this was
reflected off a land mass it gave the operator the appearance of something
small to the front. PPI and microwaves were the only sure way in such
circumstances. Nevertheless, the SC should have disclosed the Japanese
squadron and the reason for failure is not clear.
The cruisers San Juan and Quincy and the destroyer Patterson had
radar that should have been useful, but the San Juan, which had the only
SG, was placed such that she was never engaged. The Quincy used her
SC to note but disregard a Japanese float plane [10], the Patterson her FC
to fire briefly on the Chicago [11]. The Japanese squadron retired after
receiving only minor hurts but in fear of the planes that dawn would bring
and without executing the orders to put the beach and supply ships under
fire. The Allies had four cruisers sunk, including the Australian Canberra,
and one badly damaged, a thorough whipping. The beachhead, however,
remained.
Naval tacticians have examined this battle in great detail, as the reader
can well imagine. Such study lies outside our purpose, but it is obvious
that radar had been more a handicap than an advantage because of the
false confidence it gave, reducing the emphasis on alert lookouts. The
reasons for radar’s failure were equipment unsuitable for close-in surface
action (the SG was not committed), inexperienced operating personnel, the
absence of IFF on ships and a want of understanding by commanders of
radar’s limitations as well as its capabilities. Fortunately, the Americans
recognized these reasons as the cause of grief with the means of correction
evident; there was no loss of confidence in radar.
The Japanese immediately followed this battle with attempts to bring
in reinforcements by destroyers that raced down the Slot under cover of
251
darkness packed with men but little heavy equipment. Not wanting to rely
on this tactic, Admiral Yamamoto set much larger forces in motion, and for
two weeks the two opposing navies prepared for a major encounter.
5.5.2. The Battle of the Eastern Solomons

This classic carrier battle resulted from the attempt to land 1500 men on
Guadalcanal. To accomplish this Yamamoto sent two carriers, two battle-
ships, nine cruisers and many destroyers under veteran Nagumo south
from Truk to counter any American fleet opposition, and a lesser invasion
force under Rear Admiral Raizo Tanaka destined for Guadalcanal. The
Americans were not well served by the intelligence service that had stood
them so well at Midway, but it was not particularly difficult to know that
something of the sort was afoot. Fletcher had three carriers, a battleship
and seven cruisers to counter Nagumo but had sent the carrier Wasp to the
rear for refueling.
The use of the air warning radar was very effective in picking up
snoopers and attacking squadrons. There was some improvement in de-
termining target altitude but just as much confusion for the air controllers
when aircraft began to tangle close in. IFF mark II gained few admirers.
Worst of all, however, was the over-crowded high-frequency radio channel
used between pilots and controllers. The excitement of combat overcame
communications discipline and filled the air with unnecessary talk, greatly
hampering controllers in their duties. Furthermore, that frequency band
was noisy and many transmissions simply did not get through, especially
distressing when sightings of the enemy failed to be received or were gar-
bled. The new, very-high-frequency frequency-modulated radios were
needed almost as much as radar.
The Japanese carrier Shokaku and the battleship Kirishima received air-
warning sets, mark 2 model 1, just before sailing [12], and the operators
and technicians had just begun learning how to use them. On the carrier
they noted the approach of bombers from the Enterprise, but the report
was lost to the ship’s command; it was alarm from the lookouts that set off
defensive action [13]. Allied commanders were not the only ones who had
to learn this new way of a ship.
Action began near noon of 24 August with an attack on the island
launched by the Ryujo. The flight was sighted by the radar of the Saratoga
at a range of 150 km headed for Guadalcanal. Poor communications kept
the warning from being received on the island, but this sighting indicated
where the home carrier could be found, and it soon became a blazing
inferno. Guadalcanal did not yet have operating radar, but a patrol plane
alerted their newly arrived Marine fighter group, who managed to keep
damage and losses to a minimum and from the outcome to lose their fear
of the Zero [14].
The troop convoy for the invasion was effectively attacked with the
252
largest vessel set afire and sunk and the other two forced back. It was
here that the US Army Air Force finally sank a ship with a B-17: a hove-to
destroyer, so unconcerned at the approach of the heavy bombers that she
did not even get under way when the bombers appeared. Attacks by them
earlier on all three carriers were as ineffective as usual.
The Japanese lost a light carrier and a destroyer. The Americans lost
no ships but had to send the Enterprise to Pearl Harbor for repair. It was by
any measure an American victory. The CXAM radar of the Enterprise gave
ample warning of the attack that damaged her, allowing a strong defense
and the prompt clearing of combustibles and explosives, but it failed for
most fighter control, although as much from communication breakdown
as saturation of the radar scopes.
In the interlude between this and the next important naval battle
something happened of illustrative importance to radar. On 15 September
during relatively quiet and routine operation the Japanese submarine I-19
passed undetected through the destroyer screen surrounding the carrier
Wasp and hit her a fatal blast. Earlier that morning radar had vectored the
combat air patrol to dispose of a snooper, but the destroyers’ sonar had not
been up to the same standard. Without a warning Wasp was fragile indeed
and became after three major carrier battles the first American carrier to
become an inferno, the fate that had by then befallen six Japanese carriers,
all from surprise air attack. No amount of damage control skills sufficed
when plentiful quantities of gasoline were about. It left the fleet in a critical
condition because submarine I-26 had sent Saratoga off for three months
at repair yards two weeks earlier in a similar attack. Both submarines
escaped. The Navy’s anti-submarine ability was consistently the same in
both Atlantic and Pacific Oceans in 1942—poor.
5.5.3. The Battle of Cape Esperance

The piecemeal reinforcement of Guadalcanal by both Japan and the US
began to favor the Japanese, leading to a decision by Ghormley to send
the 164th Infantry Regiment to the island escorted by a strong naval force
much as Yamamoto had attempted. Reconnaissance aircraft let both sides
know what was under way, and Japanese cruisers raced down the Slot to
be met by an American covering force under Rear Admiral Norman Scott
at Savo Island near midnight of 11 October 1942.
Scott was not well informed about radar and specifically made the
heavy cruiser San Francisco, which did not have SG, his flagship. He was
also erroneously informed that the Japanese had radar-detecting receivers
for SC so he ordered them shut down, although this probably was not of
consequence, aircraft proving to have had no importance and the meter-
wave sets not having shown themselves particularly useful around Savo
Island. All cruisers had fire-control radar and the light cruisers Boise and
Helena had the vital SG search equipment that functioned throughout the
253
fight and were to prove crucial to the victory that emerged. Captain Ed-
ward J Moran, commander of the Boise, was well informed about radar
and saw the struggle clearly on his SG when Scott was essentially blind.
Moran applied a bit of judicious disobedience of orders to make Scott’s
simple and effective plan—crossing Rear Admiral Arimoto Goto’s T—a
success; however, he failed to communicate clearly what he saw to Scott,
with substantial confusion arising. Moran probably did not realize the
extent of Scott’s blindness.
When the cruisers opened fire in darkness their first rounds were hits,
and the Japanese squadron was caught completely by surprise and severely
damaged in the first few minutes [15]. The Japanese failed to execute their
mission and suffered the loss of a heavy cruiser, three destroyers and the
life of Admiral Goto; the Americans lost one destroyer. The Boise was
seriously damaged as a result of fire received when she began to illuminate
the enemy with searchlights rather than maintaining reliance on radar,
which became useless with so many ships and shell splashes about. In
one last contribution the Boise’s SG prevented her from grounding while
withdrawing [16]. The accurate opening fire produced effects far away in
Japan where elements of the Navy had resisted further development of
microwave equipment; the Boise’s salvos decided the matter: microwave
radar was to go into all kinds of warship [17].
The next night two Japanese battleships pounded Henderson Field
and vicinity in what was thereafter referred to as ‘The Bombardment’. It
left the air defense a shambles and gave the 164th Infantry more than a
taste of how life on the island was going to be. It also served as a cover
for the arrival—at no small cost—of the ‘Fast Convoy’ that unloaded sev-
eral thousand men who, when added to those already ashore, formed the
Japanese 17th Army. Thus, despite the victory of the cruiser action and
the American reinforcement, matters at Guadalcanal stood at their lowest.
The garrison on Guadalcanal was particularly displeased with the Navy’s
inability to cut off the steady flow of nightly reinforcements down the Slot.
Nimitz chose to replace Ghormley with Halsey, and whether justified by
events or not, it raised the spirits of the command measurably.
The 17th Army was to take Henderson Field and force the Marines
and soldiers to the beach; with the air base safely out of the way Ya-
mamoto’s ships and planes would destroy the American fleet that would
predictably come to the rescue. Acting on cryptanalysis that revealed the
key elements of the plan [18] Halsey moved his carriers from their usual
position south of Guadalcanal, where their function was defensive, to the
north to make a spoiling attack, and the Battle of Santa Cruz Islands began.
5.5.4. The Battle of Santa Cruz Islands

This battle was Japan’s last chance to regain naval superiority. Yamamoto
sent four carriers against Nimitz’s two—but for an engine room fire it
would have been five; he still had a small core of his excellent fliers, and
254
a large fraction of the Americans had only just completed air training.
Japan’s shipyards and training schools were being completely outclassed,
so it was now or never, although not fully appreciated at the time. For
the Americans the odds were not as unfavorable as the carrier ratio indi-
cated because Halsey had a new battleship, South Dakota, to help guard the
Enterprise. The basis for this apparently incongruous statement was the
transformation of this vessel into an floating AA regiment. It had an amaz-
ing number of 5 inch semi-automatic dual-purpose guns directed with FD
radar and had just been equipped, as had the Enterprise, with the new
40 mm Bofors automatics to replace the weaker and jam-prone 1.1 inchers.
Not only was the 40 a better gun for close-in defense, it had the excellent
mark 14 optical computing sight designed by Stark Draper of MIT. The
cruisers were not yet equipped with 40s, although all had the new 20 mm
machine guns. They did have FD radars, but heavy AA fire despite radar
control accounted for a much smaller fraction of kills than did the close-in
defense by 40s, 1.1s and 20s [19]. It was the beginning of a new phase of
carrier warfare, one that would change even more dramatically during the
subsequent year when the 5 inch shells would carry proximity fuzes.
Radar added another new element in the form of ASV mark II sets
mounted in the long-range Catalina observation planes, which spotted the
enemy shortly after midnight on 26 October 1942, formally initiating the
battle. They even attacked these ships with bombs and torpedoes after
reporting the sightings [20].
The carrier battle had the usual confusions in communications and
reconnaissance. It began with what had now become a tradition: B-17s
dropping bombs into the ocean. Both sides used air-warning radar with
attendant strengths and difficulties. The Shokaku’s radar returned the best
range of the day with 155 km, but American radar had an especially bad
day. On assuming command Halsey took the well proved Fighter Direc-
tion Officer of the Enterprise to be his Communications Officer at Nouméa
to clean up problems that had plagued that command’s signals, and un-
fortunately some excellent radar men from the carrier’s radar plot went
with him. The new Fighter Direction Officer was hardly able to find the
china-marking pencils before the battle began [21], and by coincidence the
operation also began with the maximum ranges of the CXAMs of both
Enterprise and Hornet down to about half. Other vessels were not so hand-
icapped but did not transfer the information to where it was needed [22].
Fighter pilots may have once resented control by a voice on the radio, but
by October 1942 they depended on it and were vocal in their criticism of
its failures at Santa Cruz.
The air combat can best be described as wild and confusing. Some
Japanese pilots, either wounded or with damaged aircraft crashed or at-
tempted to crash their targets, a morbid hint of things to come. American
fliers severely damaged carriers Zuiho and Shokaku; their enemy counter-
parts crippled the Hornet, which the defending airmen blamed on the poor
255
fighter control. Japan lost 99 of the 203 aircraft engaged, 48 of them to the
vicious new AA fire, especially in the Enterprise’s task force; the US lost 80
of the 175 engaged [23].
The Japanese Army failed to take Henderson Field as a result of the
difficulty of maneuvering in the jungle, the tenacity of the defenders and
the use of infantry tactics that went out of style for other armies in 1914.
Severe losses of air crews forced the Japanese to withdraw even though
two carriers were still operational. American ships withdrew to the south
leaving the stricken Hornet to be sunk by Japanese destroyers. The Japanese
did not choose to pursue, having injuries of their own, especially the loss
of their best aviators, and no wish to tangle with island-based bombers.
The Battle of Santa Cruz Islands was very likely unique in the balanced
importance of struggles on land, sea and in the air.
5.5.5. The Naval Battles of Guadalcanal

Just after midnight on Friday 13 November 1942 Rear Admiral Daniel J
Callaghan led a column of five cruisers and eight destroyers northwest
from Guadalcanal to protect the transports and supply ships he had just
left and collided with Rear Admiral Hiroaki Abe leading two battleships,
a cruiser and 11 destroyers intent on giving Henderson Field another taste
of 14 inch shells. Callaghan had been Ghormley’s Chief of Staff and had
displaced Scott as Commander owing to 15 days’ seniority. Scott remained
as second in command that night.
This First Naval Battle of Guadalcanal is the most difficult to follow
and the most contentious; much criticism has been directed at Callaghan,
generally centered on his use or misuse of radar. Callaghan had little
experience with radar and had not exercised command in a night action.
Of the five cruisers three had SG radars; Callaghan took the 8 inch gun San
Francisco, which had none, as his flagship and had not used his first two
weeks of command to alter that arrangement. The SG-equipped cruisers
were all 6 inch gun vessels; in expecting a rough fight a commander prefers
the most powerful ship for his flag. Scott, with the experience of Cape
Esperance behind him, elected to place his flag in the Atlanta, the only
remaining cruiser without an SG. Two destroyers had SGs, the fourth in
column and the last. Abe had an air warning radar on Kirishima.
That the value of radar had become common knowledge to American
seamen is attested in a bit of doggerel that circulated after the Battle of Cape
Esperance:
Yes, we’re heading for hell in column,
Scott is as proud as can be.
Only one thing he is lacking,
A brand new, working, SG! [24].
The SG was unquestionably the best sea-surface radar of the time but it
was not without faults, some of which negated its usefulness. The PPI
256
indicator was mounted vertically as part of the control panel. With time
a PPI repeater would become one of the vital instruments on the bridge,
but such was not the case in 1942, and a crowded radar room was hardly
the place for a commander. The intelligence gained by the SG was trans-
mitted by telephone to the bridge whence it went by the single voice-radio
communications channel (Talk Between Ships) to the two admirals. Un-
der the best conditions this was a terrible way to transmit the complicated,
rapidly varying information that appeared on the SG screens, something
overlooked by the radar poet just cited.
The battle quickly became one of complete confusion—‘a barroom
brawl after the lights had been shot out’. Fire direction was generally
optical because of ranges so close that machine-gun fire was exchanged.
There were a lot of star shells, searchlights, fires and gun flashes [25] to
aim with and at. Once this close-range battle began it is questionable
whether central command could have been exercised regardless of the use
of SG. Whether the battle might have developed more satisfactorily had
Callaghan watched the PPI in the approach phases is clearly beyond the
ability of later judgement to decide.
Callaghan and Scott died within minutes of the crash; both were
awarded posthumous Medals of Honor. Henderson Field and the trans-
ports were saved from bombardment at the cost of two cruisers and four
destroyers, but the battleship Hiei was so crippled from the many 8 inch
shells fired at close range that she could not escape the air attacks that
began the next morning. It took direct hits by four 1000 and one 500 lb
bombs and 11 torpedoes (some of which actually exploded) to cause her
to sink the following night, abandoned and unobserved [26]; she was a
tribute to her 1910 builders and designer, Sir George Thurston [27]; she
was also Japan’s first battleship loss and as such was the first of two terrific
psychological shocks to the Imperial Japanese Navy. Two of her accom-
panying destroyers ended with the accumulating collection of wrecks in
Ironbottom Sound.
During the night of 13/14 November cruisers, destroyers and de-
stroyer transports of the Tokyo Express landed reinforcements and bom-
barded Henderson Field with 8 inch shells, but most of a transport convoy
was blown apart by aircraft next morning. Halsey did not want to submit
his one remaining carrier, Enterprise, although he had her air group operate
out of Henderson Field and ordered his two battleships to stop the next
attempt at bombardment. Naval gunfire was very much more damaging
both to material and to the spirit than any of the air attacks Guadalcanal
suffered during that campaign.
Thus in the late hours of the 14th began round two. A battleship, four
cruisers, 18 destroyers and four transports under Vice Admiral Nobutake
Kondo and Tanaka came down the Slot to be met by two battleships and
four destroyers under Rear Admiral Wilis A Lee. Lee had a flagship with
an SG and an understanding of what it could do. None of his destroyers
257
had an SG and only one had an FD fire direction set. These ships had not
worked together before; the destroyers were selected at the last minute
according to how full their fuel tanks were.
Lee kept control throughout and sank a destroyer and the battleship
Kirishima, which ended in defiance of all hydrodynamic knowledge bottom
up in the mud [28]. The transports were run aground to discharge the
troops, but few escaped the morning air attack. Lee emerged with his
flagship, USS Washington, unscathed and at the end fighting alone but
with the loss of three destroyers and quite a bit of damage to the other
battleship, South Dakota. He attributed his success to radar in turning back
this large force and protecting the troops on Guadalcanal from another
version of ‘The Bombardment’ [29].
5.5.6. The Battle of Tassafaronga

This was the last of the major naval actions that determined the fate of
Guadalcanal and took place during the night of 30 November/1 December
1942. It was fought primarily with long-range torpedoes, which almost
guaranteed defeat for the US Navy.
The difficulty of landing food and ammunition for the Japanese gar-
rison, which by this time had been reduced to a condition of near starva-
tion, had brought the Imperial Navy to loading provisions in steel drums,
transporting them as deck cargo aboard destroyers, and pushing them into
off-shore waters to be retrieved with small boats by troops from land. It
was a desperate method of supply, but the Americans were determined to
cut it off and put an end to the terrible bombardments and jungle fighting.
American supplies to Guadalcanal came from Nouméa and Espiritu
Santo over a stretch of ocean beyond the range of bombers from Rabaul, and
unloading was hampered primarily by relatively infrequent naval bom-
bardments and air attacks. Japanese supplies to the island had to follow
paths that were within range of Henderson Field, resulting in slow con-
voys being attacked in daylight, which forced the use of destroyers capable
of moving supplies much faster at night. The result was a serious loss of
ships, whether the slow or fast option was taken. This lack of symme-
try caused Japan to lose despite winning most of the naval engagements.
Whoever held Henderson Field controlled supply. The object of the battle
was also the key.
Rear Admiral Thomas Kinkaid, Halsey’s immediate subordinate at
the Battle of Santa Cruz, took command of the cruiser squadron being
assembled to counter the next Tokyo Express and quickly formulated a
plan that incorporated lessons learned in the water about Savo Island with
special emphasis on the use of radar. Two days after he had assumed com-
mand he was ordered by Admiral King to duty in the north Pacific, and the
cruisers were given to Rear Admiral Carlton H Wright, who the next night
led five cruisers and six destroyers to meet Tanaka with eight destroyers.
258
Wright’s flagship, USS Minneapolis, had SG radar and led the column
of cruisers with four destroyers ahead but none far enough to serve as
pickets. At 2306 the flagship’s radar picked up the first evidence of intrud-
ers, who were seen by the radars of the van destroyers shortly thereafter.
Radarless Japanese observers made out the presence of the Americans six
minutes later. Both sides launched torpedo attacks at long range. Wright’s
force delayed a few minutes and missed the optimum position; it also had
inferior torpedoes. Tanaka launched at optimum position and had supe-
rior torpedoes. Wright lost the Northampton and had three other cruisers
sent to repair yards for many months. Tanaka lost a destroyer to cruiser
gunfire. It was another Japanese naval victory that went unnoticed by their
troops on Guadalcanal for it gained them no vital supplies.
Ground action pressed the Japanese ever harder which, combined
with the inability to supply these men, led to a decision on 25 December to
evacuate, completed 8 February. It was a decision to fall back, not abandon
the whole region, and they began construction of an air base further up
the Slot—about a third of the way from Guadalcanal to Rabaul—at Munda
on the Island of New Georgia. It was on the return from a bombardment
of this construction during the night of 4/5 February 1943 that the cruiser
Helena shot down a dive bomber using an FD and proximity fuzes. In 1943
there were nine more night battles in the Solomons, one of which claimed
the Helena, but the story of Guadalcanal makes a closed chapter, one rich
with lessons about radar.
Radar’s contribution to this campaign is mixed. Taken all together
this was one of the great naval actions of all time, but radar was seldom
decisive, although it certainly contributed to the American victories at the
Battle of Cape Esperance, the Battle of the Eastern Solomons and the sec-
ond part of the Naval Battle of Guadalcanal. Its worst failure was at Tassa-
faronga in which the Americans had a strong force with the best surface-
search and fire-direction radars; the Japanese had a weaker force with no
radar.
The naval radar failures can generally be attributed to the ever chang-
ing upper levels of command not understanding that this new technique
had to be studied and exercised. The aviation commanders had begun de-
signing their operations around radar as soon as they had encountered it
and had employed embryo combat information centers to direct their first
engagements. The surface commanders did nothing comparable, which
was understandable so long as they were constrained by the deficiencies
of meter-wave equipment for locating ships, but they showed inattention
when the extreme advantages of the microwave equipment became known,
advantages strangled by the means through which the information had to
reach commanders. As the problems of fighter direction illustrated all too
clearly, one had to practice seeing with the new eyes.
But the same can be said about learning to fight at night. Every en-
gagement had had a different American commander, frequently scarcely
259
able to orient himself before action began. Retrospective criticism is cheap

and often unjust, but one wonders why a single commander was not cho-
sen, at least for the waters around Guadalcanal, and allowed to develop
his skills at night fighting with the excellent equipment the inventors had
rushed to completion. As the verses composed after the Battle of Cape
Esperance witness, the power of radar was understood in the lower decks.
Failure of command was not restricted to the American side. The
Imperial Navy was skillful but timid. They allowed the Americans to
extract strategic victories from tactical defeats. Fear of seeing their ships
sink prevented them from disposing of the beachhead immediately after
the Battle of Savo Island, and this pattern repeated itself. For whatever
faults the American commanders had, they protected the troops on the
island, cost what it might. The Navy’s casualties were three times those of
the men on Guadalcanal.
There is instruction to be gained by comparing the use of radar in
naval surface action by the Americans in the Solomon Islands in 1942 and
the Germans in the Atlantic the two years before. In every one of the five
night actions fought around Savo Island the Americans had the benefit of
the superb 10 cm surface radar, SG, as well as the FC and FD gunnery sets.
One may disregard the 1.5 m SCs as being unsuitable for surface work
in such tight quarters, and there is little evidence of their utility in this
function. The German surface raiders had only the 80 cm Seetakt, but it
proved to be a good surface-search and ranging set, although not in the
same class as the SG. It was also never put to the test of multiple targets
in confined waters. The American use of this new technique was on the
average poor; the German use was excellent.
Modest reflection explains the difference and provides a lesson in how
to introduce radically new weapons, if time allows. The upper command
levels of the Kriegsmarine were much less interested in radar than the
American. Furthermore, they introduced the new equipment with much
secrecy and little technical support aboard ship. The German officers were
given little or no training before embarking but had ample time during
their month-long cruises to encounter and learn about radar. Duty aboard
a surface raider was perhaps the most relaxed of any kind of wartime
service. There was time for examining and trying out Seetakt. A few times
in which an enemy cruiser was effortlessly avoided at night was enough
to make true believers of the most technically recalcitrant. By the time of
the Bismarck actions in May Admiral Lütjens had come to depend on radar
superiority—real or imagined—to an extreme, perhaps crippling degree.
By comparison the American commanders at ship or squadron level
in the Solomons were not given the luxury of studying this new equip-
ment in the detail that builds understanding and confidence. Each day
brought a new crisis to which they had to react. When not in action, they
were deeply involved in planning and in the myriad details of a rapidly
changing tactical situation. They were briefed about the new radar sets,
260
probably even shown them, but this is no comparison to devoting days

to learning what the equipment could do. One may object that the com-
parison between German and American surface actions is an unfair one
because the battle conditions were so different and that it would be better
to compare German surface action with the American carrier raids of early
1942. Here the fighter controllers had been allowed to master the new
methods in much the same way as the Germans and with an equally pro-
fessional result. The Solomon Islands were the US Navy’s surface-action
radar school. The teacher was strict, the radar conditions the worst, but
the lessons were learned.
The story of the ground radar on Guadalcanal itself is not so check-
ered. Without it the Cactus Air Force would have simply been worn down.
The reason why ground radar functioned well and the Navy surface radar
did not is obvious at this stage: the Cactus radar men worked their job con-
tinuously and became proficient. The significance is stated without quali-
fication by Richard Frank: ‘Without these warnings, Henderson Field, and
ultimately Guadalcanal, could not have been defended’ [30].
In 1942 there were two threats to the Australian line of communi-
cation with the United States, and the descriptions of the fighting in the
Solomon Islands tend to obscure the one farther west. Where the strug-
gle for the control of Guadalcanal brought on air and naval actions that
excite one’s sense of drama and used the latest methods of technical war-
fare, the defense of Port Moresby and the eventual ejection of the Japanese
ground forces from Papua offer the reader the grim story of emaciated,
rag-clad soldiers of both sides disputing a pass through the Owen Stanley
range, but it ended just as effectively the threat to Australia and was the
beginning of the Australian–American push north under General Douglas
MacArthur and Major General Robert L Eichelberger. After occupation of
Port Moresby from the sea had been thwarted by the Battle of the Coral
Sea the Japanese elected to push across the mountains from Buna and to
land at Milne Bay, the eastern-most part of Papua.
In preparation for invasion an Australian radar unit set up a CHL at
Milne Bay. The size and weight of this equipment, completely out of place
in this kind of warfare, drew heavily on the ingenuity of 37 Radio Station,
which was furnished with irregular transport to say the least. The jungle
was deleterious to man and English electronics, but the station was on the
air by 8 August and greatly assisted in fending off air attacks preceding
and during the invasion attempt of 25 August to 5 September 1942. In this
they were aided by Japanese pilots uninstructed in radar who flew perfect
paths for detection [31].
The Papua offensive began about the same time as the American
invasion of Guadalcanal. It was contained by the Allies by the middle of
September, and the Japanese had been forced back to the beachhead at
Buna by late November [32].
Radar played a small part in all this compared with its use in the
261
Solomons. It did not yet count as equipment useful to the infantry in its
inventory, and the naval action was one of supply conducted with fleets
of trawlers and landing barges, whose attack and defense were neither
the stuff to delight naval historians nor to make use of the radar of the
time. But the Milne Bay example was continuously repeated—although
not with CHL—on all the beachheads of the South Pacific. No landing was
considered secure until the LW/AW was on the air.
In 1944 the American-made SCR-602, a copy of the British 1.5 m
lightweight set, began arriving in the Pacific Theater. Later it was fol-
lowed by the highly admired SCR-602 type 8. This 50 cm set made use
of a high-frequency triode, VT-158, designed by Harold Zahl at the Signal
Corps Laboratory and manufactured by Eitel-McCullough, the source of
the fabled Eimac tubes [33]. It incorporated lines of thought followed inde-
pendently by the makers of the Bell Labs door knob, the British micropup
and the German LS180, TS6 and RD12Tf.
As American ground troops and air units began to assemble in north-
ern Queensland in preparation for the coming push north, the Army Air
Forces set up five SCR-270s in the vicinity of Townsville to reinforce the
Australians, who had by July deployed ten of their Air Warning (AW) sets
and eight of the converted SCR-268s (MAWDs); by the end of the year they
had added another 27 pieces, two of them special rigs set up at Milne Bay
for sea search made out of ASV mark II equipment for which the originally
intended aircraft had been destroyed [34]. This was the beginning of a
growing Australian–American radar effort that was to see a mixing of per-
sonnel and equipment as the Allies fought their way across New Guinea,
the Halmaheras and Borneo.
5.6. THE EASTERN FRONT

The paths of civilization are marked by milestones set in place by wars.
By the strict meaning of the word these wars are countless, having varied
in size from clan feuds to strifes that envelop entire continents. They have
been fought in every portion of the globe to which man has had access ex-
cept the polar regions. They have invariably seen the use of all the technical
aids to killing and destruction that the levels of culture the warring groups
had attained. They have ranged from gentlemanly encounters defined by
strict ritual in which prisoners were treated as guests to clashes of mutual
extermination in which enemy individuals counted as nothing. The Sec-
ond World War offers extremes of every one of these manifestations, but
the Eastern Front had only the loathsome ones—in magnitude of suffering
and bestiality to the helpless. Its nature was foreordained. Nazi Germany
and Communist Russia were both ruled by criminal bands who condoned,
in fact relished the murder of opponents as acceptable elements of their po-
litical systems. Their clash in June 1941 secured for the 20th century the
distinction of being history’s most bloody century, bloodier perhaps than
262
all previous centuries combined.

Many people of eastern Europe, wearied by interminable suffering
in war, revolution, civil war, collectivization and purges, and divided by
hatreds extending beyond memory, had recollections of the disciplined
armies of the Kaiser and looked on the Germans as liberators and wel-
comed them. With the arrival of the Wehrmacht their hopes remained
alive, but with the passing of the first wave came the SS and SD, and the
true face of the enemy was quickly revealed. The struggle soon became
one of unfailing cruelty. Russian prisoners were so maltreated that even
the Poles pitied them, and those few who returned to their homeland were
received as traitors and sent to the Gulag. German prisoners received all
the harshness of which Siberia was capable, and those who returned to
their homeland years after the end of hostilities often came as broken men.
One horror was spared the people of the East—there was no signifi-
cant strategic bombing.
When the Luftwaffe opened the Great Patriotic War by destroying a
substantial fraction of Stalin’s air force, it was unhindered by Soviet radar.
There was only one kind of radar in use when the conflict began: RUS-2,
the pulse-type air-warning equipment working on 4 m. That statement dis-
counts completely the few units of the radio screen, RUS-1, the production
version of Oshchepkov’s Rapid of 1934, for which the war found no use; yet
despite its complete failure during the Finnish War of 1939–1940, 13 RUS-1
sets were manufactured in 1941. RUS-2 proved of value as an air-warning
set but suffered from the need to have transmitter and receiver separated by
about a kilometer, the antennas of which had to move synchronously. There
were six sets in existence when war broke out, but they had no effect on
events. The Scientific Research Institute of the Radio Industry (SRI) had de-
vised how to use a common antenna before the war began and had incorpo-
rated it into the modification, RUS-2S, but production had not yet begun [1].
The radar groups in Leningrad (LFTI, NII-9) and Kharkov (UFTI)
soon found their principal problem was evacuation to the east, both of
development laboratories and production plants, a process that removed
five months of any useful activity. The death in March 1940 of Professor
M A Bonch-Bruyevich, who had taken over the leadership of NII-9 after
the purges added to the turmoil with which that group had had to deal
[2]. The production of only 53 RUS-2S sets during 1942 tells the story more
eloquently with numbers than is possible with words [3].
The Soviet dismissal of radar at the beginning of the war was not
reflected in their other attitudes concerning AA defense. Large cities
had hundreds of guns, although their accuracy was poor [4]; the fighter
squadrons were based at all-weather fields, much superior to the usual So-
viet bases. Moscow was the best defended city in the world and, despite
its proximity to the ground fighting, did not suffer serious damage from
bombing. Besieged Leningrad suffered in every possible way, but it too put
up a very strong air defense [5]. Not surprisingly, the first effective Soviet
263
use of radar was in augmenting the defenses of Moscow and Leningrad.

An experimental station at Toksovo near Leningrad, used before the
war by the Physico-technical Institute (LIPT), assumed immediate tactical
functions and was manned by members of its technical staff. Its equip-
ment was RUS-2 but with more power for greater range. Transmitter and
receiver were mounted on separate 20 m steel towers; antenna movement
allowed a 270◦ sector of observation. Operation was turned over to military
personnel once they had been trained.
The Research Institute of the Red Army (NIIIS KA), which had overall
responsibility for radar, built in the first months of the war a large station
for the Moscow air defense, which also used the RUS-2 principle. Specifi-
cations differ enough from those of other air-warning sets to be of interest:
pulse duration 50–60 µs, which allowed a receiver pass-band of only 40
kHz and a repetition rate 50 Hz. It mimicked CH in more than pulse rate,
for it too used special demountable vacuum-pumped transmitter tubes;
they were designated type IG-8 and made by the Svetlana tube plant [6].
Leningrad received numerous air attacks, generally by formations
of about 100 aircraft. In 1942 there were 38 such bombings, all of which
were stoutly resisted. Radar’s performance opened the eyes of theretofore
uninterested military leaders, as 20 000 targets were picked up that year.
Attacking squadrons showed up on oscilloscope screens in plenty of time
to alarm the city and scramble fighters [7]. On a small scale the air defense
of Leningrad was similar to the Battle of Britain, and by the end of 1942 the
radar men did not have to beg for attention even though they had to beg
for production. RUS-2 and RUS-2S gained reputations as simple, reliable
pieces of equipment—if only they could have given height information.
Finland made common cause with Germany and attacked the Soviet
Union in 1941 to regain the territory lost the year before in a defense that
had amazed and thrilled the world and did succeed in keeping Stalin’s
armies out of most of the land. Their capital, Helsinki, and the nearby
city of Kotka came under air attack during both wars. As the Russian air
force began to recover from its disastrous 1941 defeat it renewed raids on
Finland, causing Germany to furnish the Finns with two Freyas and four
Würzburgs in the spring of 1943. Helsinki had four batteries of 88 mm guns
with schoolboy volunteers operating the radar. The data from the Freyas
were connected by telemetry to a control room that directed fighters and
alerted the gunners. As in 1939 and 1940, defense was stout and attacks in-
dolent. The fliers generally elected to drop their bombs on locations other
than the targets, the ocean being found most convenient. Helsinki suffered
less than any city that was subjected to major air attacks [8].
The air defense of Helsinki had an unusual element. A German
freighter, Togo, had been made into a night-fighter-control ship through
an uncommon cooperation between the Luftwaffe and the Kriegsmarine.
Equipped with a Freya, a Würzburg Giant and a Y-Gerät she had a capabil-
264
ity similar to one of Kammhuber’s Himmelbett cells1 . During the months

of March, April and May of 1944 she was stationed at Tallinn (Estonian
SSR) and would take position with the onset of darkness in the path taken
by Soviet bombers headed for the Finnish capital. When the Finns began
serious negotiations with Stalin in June 1944 to leave the war, the vessel
was withdrawn to Liepaja (Latvian SSR) and provided similar protection
for German capital ships [9].
The poor showing of early Soviet gun-laying radars did not eliminate
this type from the minds of designers, and NII-9 organized an experimen-
tal battalion-sized AA unit in October 1941, employed in the defense of
Moscow while trying out its new equipment. Initially the battalion had
four 75 mm, six 105 mm (German guns obtained during the time of the
non-aggression pact) and six 37 mm automatic guns. A team of engineers
headed by M L Sliozberg worked directly with the unit. They introduced
some experimental sets, Sleep, B-2 and B-3 that worked on 15 cm using
cavity magnetrons [10]. The possession of the cavity magnetron, viewed
in Britain and America as the ultimate microwave transmitter and the basis
for uncounted radar successes in the coming years, seemed to hold no ad-
vantages for the Soviets. They were unable to produce a transmitter or local
oscillator with sufficiently stable frequency to allow the construction of a
heterodyne receiver, which was presumably attempted without a crystal-
diode mixer. These gun-laying sets were failures and soon disappeared
from the experimental battalion’s gun positions.
The arrival of British GL mark IIs produced much more interest than
the experimental microwave sets [11]. It was not much of a gun-laying
set, to be sure, but it was a robust, reliable and well engineered piece that
found use for searchlights, fighter direction and even air warning. The
British technicians who had been sent to instruct the Russians, and who
had been led to believe that radar was unknown to the recipients, encoun-
tered personnel who mastered the equipment rapidly despite a significant
language barrier [12]. Sliozberg’s people soon made a copy of it, called
SON-2 [13]. It proved the favorite Soviet radar, but British imports of GL
mark II (generally called SON-2) overwhelmed native production, which
produced only 124 during the entire war [14].
Later Britain sent 44 microwave GL mark IIIs and America sent 25
SCR-268s, 15 SCR-545s and 49 of the superb SCR-584. A copy of GL mark III
appeared as Neptune, and the 584 was copied after the war as SON-4 [15].
In the summer and fall of 1941 Stalin’s gigantic army and air force
suffered a defeat coupled with losses of men and material of magnitude
unparalleled in history, but as Hitler’s forces stood before Moscow, every-
thing changed in an almost miraculous manner: (1) Japan was suddenly
found to be completely occupied with America and Britain, thereby freeing
many fresh Siberian divisions to board trains headed west; (2) the Russian
265
266

Radar plots of Soviet bombers in an attack on Helsinki during the night of 26/27 February 1944. The large number of plots came about
because the Soviets did not attack in formations but in a long series of sorties, allowing the paths of individual aircraft to be determined.
The city is shown by the shaded area at the left center. As can be seen, the attack was not pressed home and few bombs fell on the city.
The scale can be ascertained by noting that the bottom edge is approximately 35 km. Archives of the Finnish Anti-Aircraft Museum.
people, who may have been originally indifferent to the downfall of the
communist state, had come to realize that the war was against them, not
just Stalin; (3) Hitler had also conveniently declared war on the United
States, which was to prove a serious distraction for the Nazi state; (4) a
Wehrmacht without winter clothing or equipment had been assailed by a
winter as deadly as the enemy. In January 1942 Germany found herself in
total war, a discomfort theretofore left to her adversaries. Only then did
German total mobilization begin.
During the intoxicating summer of 1941 radar had been, if anything,
even less important to the Germans than the Russians. The new weapon,
so important in the west, was ignored in the east. The Luftwaffe dominated
the air and found little need for equipment in short supply and required for
the defense of the Reich against Bomber Command. There had been use
of Freya sets before the surprise attack of 22 June to insure that no Soviet
observation planes discovered the large assembly of forces [16], but few of
the clumsy Freyas followed the Blitzkrieg.
The Soviet air force had to make its recovery in the face of German
air superiority, but its slow progress called for correspondingly increased
vigilance by the Luftwaffe. A measure of Russian progress can be found in
the extent of German radar deployment. As Leningrad became besieged,
the air struggle there became more advanced, and Luftwaffe Signals set up
Freyas on the islands of Hiiumaa and Saaremann, located to the west of Es-
tonia, to protect German shipping from air raids [17]. A Freya unit covering
the south approaches of Leningrad found movement of the set in the terri-
ble winter retreat of January 1942 so difficult that it had to be destroyed [18].
In 1941 radar was closely associated with strategic bombing, and its
use with and against tactical ground forces lay a few years in the future.
The steady growth of Soviet power came from factories beyond the range
of German bombers, and the railway network continued to distribute sup-
plies and troops, hindered but not brought to collapse by air raids. The
four-engine bomber that General Walter Wever had favored to attack these
resources was absent and not going to appear.
An early German use of radar came from an unexpected quarter—
partisan warfare. When it became clear that Germany was the enemy of all
those Soviet peoples that did not have some ethnic status that made them
acceptable to the Nazis, partisan groups began to make no small amount of
trouble behind the German lines. Made up of soldiers cut off but not taken
prisoner and civilians escaping and fighting SS terror, these groups were
organized and maintained by the Soviet command. Night flying served
as the means of bringing vital supplies and officers to these units and the
carrying out of information and wounded. Soon an elaborate air transport
was established at night. Combating these infiltrating flights proved dif-
ficult, in great part because of the primitive Russian equipment used. The
most important aircraft was a biplane, paradoxically designated the U-2.
Flying slow and low and necessarily observing strict radio silence it was
267
difficult to detect. Radar was obviously called for, and it came as railway
radar trains, but an effective counter to the U-2 was never found [19].
Russia’s notoriously muddy roads made movement by rail essential
for heavy equipment—a Freya required 28 horses for movement by typical
road [20]—and radar trains were the obvious answer, first placed in service
in October 1942. They were portable fighter control units that consisted
of a Freya for early warning and two Würzburg giants, one to track the
enemy and the other to track the interceptor so the controller could bring
the two together [21]. Some trains made good use of searchlights. It was
the system called Himmelbett in the west2 . As the air situation deterio-
rated for the Luftwaffe, the radar trains became more numerous and more
important. In 1943 a radar train in the Orel-Bryansk sector took credit for
bringing down about 30 planes [22].
(The first radar trains may have been placed in service somewhat ear-
lier in France during the summer of 1942. By that time the activities of the
underground were beginning to be troublesome, and light aircraft trans-
ported agents and supplies between the continent and England. Finding
the resistance personnel was more important than bringing down the air-
planes, so railway-mounted equipment that could be moved to suspected
places of operation in order to observe where they landed was an obvious
answer. It is reported to have led to several arrests [23].)
Growing Soviet air power began forcing the Luftwaffe to bomb at
night, and their efforts had grown to such an extent that the Soviets began
organizing night fighter units in late 1943. These units were not particu-
larly effective because they lacked both airborne and ground radar capable
of bringing about interception [24].
The absence of strategic bombing in the east meant there was no cen-
tralized air defense, so radar use on both sides tended to take on local
character and ingenuity. A German bomber group at Shitomir (near Kiev)
used two Freyas for night bombing Russian concentrations at locations
beyond artillery range. One Freya directed a bomber by radio so as to
follow an arc of constant radius while the second controlled the release of
bombs. The attacks were not only complete surprises but remarkably ac-
curate [25]. The reader will encounter a similar but more elaborate method
of blind bombing, called Oboe, used against Germany in a later chapter3 .
By the time of the great tank–air battle at Kursk during 5–11 July 1943
the Soviet air force was something that had to be dealt with, and the Ger-
mans assigned five of the nine then existent radar trains to the sector. The
Wehrmacht lost decisively. The wreckage of hundreds of aircraft and tanks
littered the field, but one Freya was credited with saving Fliegerkorps VIII
from complete destruction [26].
Any Soviet use of radar at Kursk has escaped mention in the sources

268
available. Indeed, Soviet use of radar in general was hardly noticed by

the Luftwaffe until 1944 and never reached the stage where countermea-
sures were employed. They seem to have thought all of the enemy’s radar
was of British or American manufacture and been unaware that any of the
Russian equipment was of indigenous manufacture [27].
German radar found ever wider use on the Eastern Front as ever
more equipment became available and the pressure of Soviet air power
increased—and not just Soviet. The oil fields of Rumania received a gen-
erous allotment of AA and fighter units and with them came Freyas and
Würzburgs for Flak and fighter control. Their effectiveness is attested by
the heavy losses of the American bombers that attacked Ploesti. The sav-
ing or at least preventing the capture of the extensive radar deployment in
Rumania became a matter of serious concern when Russian forces secured
that nation in August and September 1944 [28].
Such was radar in the east. Compared with the use in the west and
at sea it was small indeed, being a mere perturbation on the cataclysmic
battles that were fought there. Germany’s deployment was, until near the
end, trivial when measured against the air defense system facing the Allies.
Russia used it first only in defense of her two largest cities, to what effect it
is difficult to say. In the east huge ground forces struggled with air power
restricted to army support. It was not until the appearance of remarkably
accurate 10 cm equipment, such as SCR-584, that radar showed real value
for this kind of warfare. In the hands of ingenious officers the equipment
could be of benefit, especially to local fighter squadrons, but these contri-
butions were never decisive. The actions in the deserts of North Africa are
apt illustrations of this. Given this tactical background it is difficult to fault
the Soviet command for not giving radar a greater priority. Were it not for
their demonstrated capacity for confused and self-destructive administra-
tion, one might be tempted to attribute wisdom to the Soviet leaders for
the low priority given radar. But whether from wisdom or folly, there is
little reason to fault the result. The critical industrial strengths required
for the manufacture and operation of radar were put to better purpose in
communication equipment vital to mobile ground warfare.
The quality of Soviet radar development before and during the war
must be evaluated by what was accomplished against what was attempted.
Here is a bewildering confusion of competence at its highest and lowest.
Soviet engineers invented the cavity magnetron, a device for which praise
in Britain and America exceeds that for any comparable device. That not
being enough they invented the klystron independently of the Varians and
Hansen. But their attempts at putting them to use failed, owing to an in-
ability to master the lesser arts of microwaves, and resulted in an especially
bad gun-laying set that was never produced. The klystron does not seem
to have entered a serious Soviet radar design. In meter-wave equipment
the advantage of an early entrance was lost. Postwar design started from
Allied and captured German sets.
269
PHOTOGRAPHS: NAVAL RADAR

German Seetakt (FuMG 40) sur-
face-search radar captured at
Toulon during the invasion of
southern France where it was
used with a coast-artillery bat-
tery. When used in this ser-
vice it was referred to by British
intelligence as ‘Coast Watcher’.
The same basic 80 cm set, known
originally as DeTe-I, was used by
the Kriegsmarine on ship and on
shore. The pocket battleship Ad-
miral Graf Spee received a See-
takt (60 cm version) in January
1938, the first warship to have a
tactical radar. The Kriegsmarine
rejected lobe switching that the
manufacturer, GEMA, offered
and thereby lost the capability
of blind fire. National Archives
photograph 111-SC 246248.
US Navy FC, later mark 3, fire-direction radar mounted on the forward main
battery director of USS New Mexico in December 1941. Optical range finders
can be seen ahead and behind the FC antenna. The mark 33 director could use
data from either source. Its blind-fire capability gave a severe shock to the Japanese
Navy at the Battle of Cape Esperance in October 1942. US Naval Historical
Center photograph NH 84811.
270
US Navy FD, later mark 4, 40 cm fire-direction radar mounted on a mark 37

director of the destroyer USS Nicholas in January 1944. This set was developed
by Bell Laboratories from a prototype demonstrated in July 1939 as the CXAS
and used horizontal and vertical lobe switching that allowed it to direct 5 inch
dual-purpose guns at surface or air targets; it was a three-dimensional version
of the FC. Originally designed for the Bell Labs high-frequency triode 316A, its
modular construction allowed the transmitter unit to be replaced by a 40 cm
cavity magnetron and as such became the first American production unit to use
the British invention. The similarity of the antenna to Hertz’s original is striking.
It was the standard fire-direction radar for most of the Navy throughout the war.
US Naval Historical Center photograph NH 84804.
271
Radar of the carrier USS Yorktown on 5 April 1945. This replacement for the
earlier Yorktown was commissioned 15 April 1943 and carried a full complement
of the latest radar. At the left and right are mark 12 fire-direction radars with
IFF antennas in their centers and mark 22 height finders at their right. The
mark 12 was an improved version of the FD; the mark 22 was an elevation-only
3 cm set used to thwart the attacks of very-low-flying aircraft. Second from the
right and insignificant in appearance but mighty in operation is the SG, the most
valuable aid to mariners since the invention of the chronometer. Next is the
flying bedspring of the SK, the improved version of the 1.5 m air-warning XAF
and CXAM surmounted by its IFF antenna. Next and in the background is an
SC, used as a back-up set for the SK. One step farther to the left finds in close
proximity another SG and a YE aircraft homing system; the latter was not a radar
but a very-high-frequency directed beam with a signal that gave the direction of
propagation, thereby allowing fliers to set a homeward-bound course. Finally just
abaft the forward mark 12 is an SM, a 10 cm radar with a parabolic reflector that is
mounted on a gyroscopically stabilized platform; its function was fighter control
close to the ship where the poor resolution and lack of height data made the SK
nearly useless. National Archives photograph 80-G-376152.
272
The radar on an escort car-

rier. These ships were intended
primarily for operations against
submarines and air support dur-
ing landings and, as a conse-
quence of the restricted space,
carried a smaller amount of
radar. The radar-like antenna at
the top is the YE homing sys-
tem, a very-high-frequency di-
rected beam with a signal that
gave the direction of propaga-
tion, thereby allowing fliers to
set a homeward course; below
is the 1.5 m SC air warning
and the 10 cm SG surface search
radar. National Archives photo-
graph 80-G 214980.
American microwave
phased-array radar, the
Bell Labs FH (later mark 8)
fire-direction set. Early use
of the FC showed that the
40 cm wavelength and the
dependence on an A-scope
display resulted in confusion
in a complicated surface
action. The 10 cm mark 8
with its improved resolution
and maplike display allowed
multiple targets to be kept in
a ±15◦ sector view without
confusing the direction of fire
on the one selected as target.
US Naval Historical Center
photograph NH 84813.
273
Fighter control with radar:

an American Essex-class car-
rier. The Combat Informa-
tion Centers of large carriers
had elaborate means of display-
ing information. Here is a
plastic plotting screen with a
seaman plotting from behind.
National Archives photograph
80-G-326751.
Fighter control with radar: USS Santee. Fighter control in the US Navy evolved
the Combat Information Center (CIC) that grew into large specialized rooms that
became the nerve center of the vessel in combat. The CIC shown is on an escort
carrier in which space is at a premium, as these ships were intended primarily
for operations against submarines and air support during landings. The plas-
tic-covered table at the right serves as the plotting board from which the fighter
directors extract information to vector onto the enemy. The carrier depended on
the SC radar for data. National Archives photograph 80-G-342577.
274
Front view of Royal Navy type 271, 10 cm surface-search radar. This was the first
operational microwave radar. Sea trials of a production model were made on 25
March 1941 aboard the Flower-class corvette Orchis. Transmitter and receiver had
separate identical antennas, the transmitter at the top. A dipole backed by a rod
reflector is located at the focus of a cylindrical parabolic mirror. The top and bottom
plates earned this style the name of ‘cheese’. Aboard ship this antenna was enclosed
in a plastic cylinder for protection. By September 32 corvettes mounted type 271s.
Churchill Archives Centre, Royal Navy photograph. Crown Copyright.
Rear view of Royal Navy type

273, 10 cm surface-search radar.
This was the microwave radar in-
tended for capital ships. It dif-
fered from the type 271 in using
90 cm paraboloid reflectors that
provided higher antenna gain than
the cheeses. The pencil beam re-
quired a stable vertical axis, the
first radar so mounted. Sea tri-
als were made on the cruiser Nige-
ria in August 1941. Churchill
Archives Centre, Royal Navy pho-
tograph. Crown Copyright.
275
Fighter control with radar: fleet carrier HMS Venerable. The Aircraft Direction
Room of British ships differed from the Combat Information Centers of US carriers
in little other than name. Here a projection system, called the Skiatron, allowed
control officers to survey the air situation. HMS Dryad archives, Royal Navy
photograph. Crown Copyright.
Japanese 1.5 m air-warning radar mounted on the aircraft carrier Junyo. This
mark 2 model 1 entered service in time for Midway. Had a carrier had it there
instead of a battleship, which was far from the action, the battle might well have
ended differently. National Archives photograph 80-G 264924.
276
Japanese 10 cm surface-search
radar. This mark 2 model 2
also entered service in time to
be used at Midway, but only
on a battleship. It was used
effectively throughout the Pa-
cific War and gave Japanese
submariners a weapon de-
nied to their German al-
lies. Japanese microwave
equipment always used horns
and circular waveguides. Na-
tional Archives photograph
111- SC 290054.
Japanese submarine RO-58

with the mark 2 model 2 radar
easily recognized on the con-
ning tower. This 10 cm radar
was widely used by Japanese
submarines. It lacked a PPI
display and could not be used
with the vessel submerged,
as could American boats, but
was a technique that Ger-
man U-boats lacked and could
have used effectively. Pho-
tographed at Yokosuka Naval
Base, 7 September 1945. Na-
tional Archives photograph
80-G 339842.
277
Japanese Navy 10 cm surface fire-direction radar, mark 3 model 2. This set used
lobe switching by a receiver having two horns below the single transmitter horn
on top. By the time it had been developed the Japanese Navy was no longer capable
of fighting surface actions. The set shown here is mounted for coast artillery but
was never used for that either. National Archives photograph 111-SC 290052.
278
CHAPTER 6
THE GREAT RADAR WAR
6.1. THE DESTRUCTION OF GERMAN CITIES INITIATED

When Hitler turned his magnificent war machine toward the Soviet Union
in June 1941 England found the air attacks they had borne for a year de-
creased to a much smaller intensity. But the war was expanded, not over,
and Churchill, ancient enemy of communism, embraced his new ally with
startling alacrity. Moreover Britain was wounded, not dead, and the nature
of her race was to strike back at Germany with all she had, and Bomber
Command was the only arm capable of hitting directly. The doctrines of
the Royal Air Force during the between-war years had seen in bombing
squadrons the vehicles for future victory, but the first 20 months of combat
had left these doctrines devoid of reality. Somehow everything had got
turned upside down. No ‘knock-out blow’ had been placed by either side,
and the mission of army support, so decried before, was the Air Force’s
only effective activity against the enemy. Victory through air power was
proving very elusive.
That the object of Britain’s air war against Germany became the de-
struction of her cities did not come about as the result of a carefully thought
out plan; it came about in the same way that the Battle of Britain and later
the Blitz had been forced on the Luftwaffe. Both came as the result of an
unforeseen military and technical situation. Without long-range fighters—
and the delay in bringing them into the picture is its own story—massive
daylight attacks could not be sustained and the precision of aim that had
been assumed would prevail during the day was never reached. With-
out navigational techniques that permitted blind bombing with accuracies
comparable to the size of a factory only area bombing could be done at
night. Translated into practical terms this meant that the targets would
have to be of city size. By mid-1941 this simple fact had put an end to
the initial strategy of selectively destroying Germany’s synthetic oil pro-
duction [1]. Thus in phase one of the Great Radar War defensive radar
had eliminated by its mere existence the possible selective destruction of
German industrial targets, leaving cities the only thing the Air Force could
hit—and all too often the bombers even missed them. Strategic doctrine
279
was made to conform with reality in a directive of 14 February 1942, made

in consultation with the Ministry of Economic Warfare, that designated
the primary objective as ‘the morale of the enemy civil population and in
particular, of the industrial worker’ [2].
While Bomber Command was trying to learn what it was to do and
how it was to do it, the Luftwaffe began evaluating its unexpected role
in defense. By the middle of 1940 it was obvious to them that Bomber
Command was going to attack at night and that the raids were already
showing the defense to be ill prepared. It was all well and good to say that
the British were not capable of doing any serious military damage; when
bombs fell on cities action was required, irrespective of the military signif-
icance. Night air defense had three parts: early warning, fighter direction
and Flak (AA). The engineers built a new generation of air warning radar,
the fliers organized to make best use of it, and Flak learned the value of
the Würzburg.
Freya had demonstrated the ability to detect aircraft at distances of
100 km, given objects at sufficiently high altitude. The Freya-Fahrstuhl,
a set wherein the antenna could be raised 20 m to alter the vertical lobe
pattern, had emerged out of Diehl’s experimentation at Wangerooge and
had brought home the importance of controlling the vertical lobe structure
in air warning equipment. Freya pointed the direction for improving early
warning, and the resulting GEMA designs yielded the best equipment of
that kind to be used until the appearance of MEW, the Rad Lab’s microwave
early warning in 1944. For the tasks at hand it was better than MEW, as it
had good height-finding capability.
The deficiencies of Freya were a range shorter than the curvature of
the earth imposed and a poor resolution of targets both in direction and
in height, very poor for the latter. Both of these defects could be reduced
by increasing the antenna size while retaining the basic 2.4 m wavelength
and the circuitry of Freya. Increasing the number of dipoles caused a nar-
rowing of the beam, which led to obvious improvement in the angular
resolution but also concentrated the power, thereby leading to an increase
in range. Directional accuracy was obtained through lobe switching (in
German, AN-Verfahren or Leitlinienpeilung [3]), which Erbslöh and von
Willisen had introduced in their early equipment only to see it dismissed
as too complicated. Out of these considerations two complementary de-
signs, Wassermann and Mammut, materialized from GEMA during the
summer of 1940 at a test site at Jüterborg, about 60 km southwest of Berlin
[4]. Both designs required little new in electronics because the improved
performance came about from adding dipoles. It economized by relying
on well tested basic circuitry. More powerful transmitters were to be added
in time.
As time passed a number of different models appeared on the coasts
of Europe, the details of which need not concern us here. Wassermann S
(schwer or heavy) had an array of 188 dipoles in a vertical pattern mounted
280
The Great Radar War
on a 4 m diameter steel pipe 60 m high, causing British intelligence people

to name it Chimney. (Only Wassermann S used the pipe; others were
constructed of structural steel braced with guy wires but retained both
the German and British names.) This immense array of dipoles formed a
thin, horizontal fan-shaped beam that could be moved up and down by
altering the phase shifts in the transmission lines feeding the dipoles. This
allowed a direct and rapid way of determining elevation to an accuracy of
0.75◦ for aircraft lying between 3 and 8◦ ; lobe switching gave an azimuthal
accuracy of 0.25◦ . At 100 km this localized the target in a box-shaped
region 1200 m high, 300 m long and 435 m wide for planes above 5200 m.
Aircraft 8000 m high could be detected at 210 km, and very high fliers at
300 km. Mechanical rotation about the vertical axis allowed a 360◦ field of
observation [5]. For ease of repair a crane surmounted the tower except
for Wassermann L.
The exceptional long-range accuracy of Wassermann was used to de-
termine the positions of individual formations of attackers, a process that
took minutes of valuable time. The designers assumed that they would
have to contend with multiple formations, and Mammut allowed these to
be tracked rapidly and accurately in the horizontal plane with no height
information. Mammut had 192 dipoles arranged in a fixed array 16 m
high and 30 m wide mounted on four vertical steel beams. Its appearance
secured for it the British intelligence name of Hoarding (in American, Bill-
board). Its wide array produced a narrow, vertical fan-shaped beam that
could be pointed in a horizontal direction by altering the phase shifts in
the transmission lines feeding the dipoles, just as done for Wassermann in
the other direction. This allowed 100◦ of coverage; lobe switching yielded
a directional accuracy of 0.5◦ [6].
These two basic designs formed most of the Luftwaffe’s early warn-
ing throughout the war. Together they proved capable of dissecting the
immense numbers of bombers that were to attack Germany as the air war
proceeded, numbers that would have saturated CH. Although these radars
operated on a wavelength one-fifth that of CH, they shared its unsuspected
characteristic of being difficult to put out of action. The tall, spindly towers
proved resistant to explosives, and the electronics and operation room was
in a bunker [7]. Later they proved vulnerable to rocket-firing aircraft. Like
CH they were rushed to completion by a few forward-looking individu-
als in anticipation of the terrible ordeal that still lay in the future and was
unsuspected by most.
Wassermann and Mammut are the first examples of what has become
common in modern radar: securing beam direction through a phased array.
The Bell Telephone Laboratory’s 10 cm FH (mark 8) main-battery naval
fire-control radar [8] and Alvarez’s ‘leaky pipe’ were developed at about
the same time, close enough and independent enough to make priority
unimportant [9].
The air-warning difficulty for night attacks was straightforward. The
281
basic ideas came to mind almost as soon as they were posed. Destroying
the attackers was something else again. Hermann Diehl’s invention of
daylight fighter direction at Wangerooge in 1939 was an obvious beginning,
but difficulties were easy to list. Freya was excellent for picking up an
attacker 75 to 100 km away but gave no height data useful for night work.
If a fighter pilot was to fire on a bomber, he had to be guided to within a
couple of hundred meters of his adversary in order for it to become visible
even to dark-adapted eyes.
Hauptmann Wolfgang Falck had commanded the Me-110s that Diehl
had vectored so successfully to the attacking bombers on 18 December 1939.
This had impressed him so much that he began practicing the technique
using the new lobe switching for Freya with which Diehl had modified his
equipment with help from GEMA. They began using it over the Zuider Zee
in poor visibility and eventually in darkness, where local tactical conditions
of predictable altitude allowed them to down several aircraft. Thus when
Colonel Josef Kammhuber, just released from a French prison, was assigned
the task of organizing night fighter defense on 17 July 1940, Diehl and Falck
were incorporated into his new organization [10].
Telefunken had demonstrated in July 1939 the ability of the Würzburg
to determine the three-dimensional coordinates of the target, and in Octo-
ber it was decided to use this device as an adjunct to Freya. Owing to the
Würzburg’s small size and portability, the first plan had been to sprinkle
them liberally over the countryside in order to follow the incoming aircraft,
making up for the short range of 25 to 30 km by fielding a large number.
To this end 4000 were ordered, even though the design was not yet final.
In the final acceptance tests on 9 April 1940 at the Erprobungsstelle (test
station) der Luftwaffe Rechlin the performance was spectacular, and the
sets that were soon to be in production were taken away from air warning
and given to Flak, as they were obviously able to point guns or searchlights
better than any other technique [11].
Flak invented an ingenious test firing for the new equipment at shoots
in Kühlungsborn on the Baltic Sea, which they called ‘mirror shooting’
and which allowed trials on targets of higher performance than towed
sleeves. For an aircraft passing to the west of the battery, the bearing
angle calculated from the radar data was given a negative sign, resulting
in the guns pointing to the east. A vertical glass plate lying in the north-
south plane at the battery allowed the reflection of the target to be seen
superimposed on the sight of the shell bursts. Theodolite data of target
and bursts allowed numerical analysis [12]. In September 1940 a battery
in the vicinity of Essen brought down the first bomber using gun-laying
radar [13], and the future of the Würzburg was settled. It was also to be
the heart of night-fighter control.
The night-fighter control evolved through a number of stages, the first
two of which concern us now. Helle Nachtjagd (illuminated night fighting)
came first. In it a Freya picked up the enemy at extreme range and provided
282
The Great Radar War
a companion Würzburg with an accurate azimuth, which sufficed for the

Würzburg to find it and point searchlights for almost immediate illumi-
nation. The Flak batteries also used radar to direct searchlights, and the
speed with which the light caught a plane remained an unnerving experi-
ence for the fliers. The limited range of the little radar and the searchlight
combined with the speed of the target gave the night fighter only a few
minutes to intercept. Naturally this method failed if there was cloud cover.
Dunkle Nachtjagd (dark night fighting) was the next stage and re-
quired the cooperation of a Freya and two Würzburgs, one to track the
bomber, the other the fighter. The fighter director would note the relative
positions of the two aircraft in a three-dimensional display called the See-
burg Plotting Table and guide the fighter to an interception close enough
for the pilot to see the bomber. The Würzburg was sufficiently accurate
to guide the fighter successfully often enough to make the method useful,
although it needed airborne radar to become truly effective.
Generaloberst Ernst Udet, a distinguished pilot from World War I,
greeted this idea with skepticism but found that it worked after trying it
personally [14]. He has been credited with remarking that radar ‘took all
the fun out of flying’, which disclosed an attitude toward technical matters
that brings into serious question his fitness as a Chief of Air Armament. It
can explain much of the backwardness that marked the end of the Luft-
waffe. His suicide on 17 November 1941 certainly had part of its origin in
his sense of failure [15].
Kammhuber recognized immediately that the range of the Würzburg,
although adequate for Flak, was too short for night interception. At
Telefunken Leo Brandt, who had taken over further development of the
Würzburg, suggested that the quickest way to extend the range was to in-
crease the antenna from 3 m to 7.5 m diameter just as GEMAhad boosted the
range of Freya by increasing the number of dipoles to make Wassermann
and Mammut; therewith the Würzburg-Riese (giant) enters the story. It
proved just right for night-fighter direction and went into immediate pro-
duction [16]. The Zeppelin Company had made Würzburg dishes from the
beginning, but it was with the Riese that the unique structures noted in air-
ships became evident. Kammhuber organized a belt of Freya–Würzburg–
searchlight stations 50 to 75 km back from the coast where the early warning
stations were located. British intelligence determined who the comman-
der was and dubbed it the Kammhuber Line. The line had no anti-aircraft
(AA) guns, the cities no fighter defense, which allowed each weapon to
function without concern about the other. As stations began using Dunkle
Nachtjagd their searchlights were moved farther back to take over with
Helle Nachtjagd once the bomber left the dark zone [17]. The combina-
tion of a Freya with lobe switching, two Würzburgs and a fighter formed
a defense cell 45 km long and 22 km wide and code named Himmelbett
[18], which translates as four-poster bed. Bombers at night did not fly in
tight formations because of the danger of collision, and they passed the
283
Kammhuber line almost as individuals, just right for Himmelbett. They

also had to re-cross the line on their return home.
Crucial to the effectiveness of Himmelbett was the eternal problem
of IFF. Without it the controllers could—and at times did—send one night
fighter to shoot down another. The IFF Erstling was still in trial and was not
making converts in the Luftwaffe. They preferred the Y-Gerät used dur-
ing the Blitz for beam navigation. The strict—and not too well fulfilled—
requirements for blind bombing were not required for following the fighter.
The Würzburg would hold its position accurately and the Y-Gerät, in its
new form called Y-Verfahren, would be close enough to identify the fighter
to the controller. It had a singular advantage over IFF of yielding range and
effectively solved the problem of getting the fighter back to a landing field
in the dark. Two changes in procedure were required: the ground anten-
nas that formed the guidance beam tracked the return signal as direction
finders; the returned audio signal that determined range by comparing the
outgoing with the incoming phase was transmitted 20 seconds out of every
minute on the radio telephone on which the pilot obtained instructions, in
open imitation of the British Pipsqueak [19]. The system was much pre-
ferred by the fliers because it was reliable IFF and could get them back
home. It was also used for day-fighter control. Attempts were made to
dispense with what seemed a redundant Würzburg, but Y-Verfahren was
too inaccurate [20].
German air defense had a component called Y-Dienst (Y-Service) that
must not be confused with Y-Gerät and Y-Verfahren. It was a system of
directional antennas that picked up the bomber’s radar, radio-telephone,
IFF and jamming emissions and located the approaching formations by
triangulations. With the cooperation of Bomber Command in providing
a generous supply of assorted radiation it proved to be an important part
of German early warning. It had for the defenders the singular virtue of
being un-jammable. The RAF also had a Y-Service, although probably not
named with the mischievous intent of confusing future readers of history,
whose duties encompassed monitoring with skilled use of special receivers
all German radio emissions [21].
An un-jammable early-warning radar that covered the English coast
very well was Klein-Heidelberg. This was a series of passive stations in the
lowlands that observed the arrival times of the direct pulses from a CH sta-
tion and those reflected from a bomber formation. The time delay between
the two established the position of the aircraft on an ellipse. There were
two receiver antennas: a small synchronizer for the direct CH pulse and a
very large steerable dipole array mounted on a Wassermann-S tower. The
array gave a directional accuracy of 3◦ for the echo signal; range resulted
from the intersection of this line and the ellipse [22]. This was the price
paid in CH using floodlight rather than searchlight transmission.
The Germans believed that a spy penetrated a Klein-Heidelberg sta-
tion in the form of a mysterious Leutnant Kunkel. No one knew him,
284
The Great Radar War
whence he came or where he went—hardly a glowing commentary on

security—but after that the CH stations applied some kind of time jitter
to their pulses. This proved to be no problem but convinced everyone of
Kunkel’s villainous nature. R V Jones says that Kunkel was not one of
his men and that knowledge of Klein-Heidelberg’s function came from the
analysis of communications traffic [23].
German air defense required a means for the control to follow the
bombing formations as they proceeded across the Reich. The Himmelbett
stations were too short ranged and much too involved with their own prob-
lems to furnish reliable information to higher command levels. In 1940 and
1941 GEMA designed a prototype set built by Siemens und Halske, called
Panorama and placed on a large tower near Berlin to sweep the region.
It had 18 dipoles arranged on a 20 m supporting beam to produce a nar-
row vertical fan-shaped beam at the Freya wavelength band. In addition
to overcoming design problems this set taught the lessons of how best to
site a meter-wave set on land without being swamped with local echoes.
From these experiments Siemens built Jagdschloss (hunting lodge) on the
same basic pattern but with two alterations: (1) Jagdschloss incorporated
a PPI display, which was transmitted from the station to command head-
quarters by high-frequency cable or the widely used 50 cm directed-beam
communication links; (2) a broad-band antenna allowed modest changes
of frequency and specifically allowed a change to excite the IFF set Erstling,
which then served as a secondary radar that enhanced the positions of the
German aircraft at the closing of a switch. Jagdschloss began entering
service toward the end of 1943 [24].
In September 1940 the Luftwaffe initiated a successful technique of
attacking the bombers as they took off or landed at their bases in England
or in the early or late stages of their journey. One of many decisions made
by Hitler that caused some to question his stature as a great, universal
military leader was to transfer the Fernnachtjagdgruppe (long-range night-
fighter wing), which had built up these careful and destructive techniques,
to the Mediterranean in October 1941 and to forbid such methods in the
future [25].
Flak also began preparing for the coming onslaught. The original
Würzburg had not been intended for gun laying, but when this important
capability was recognized Telefunken’s engineers began to work on a ver-
sion that improved accuracy for direction and range by factors of three.
Other virtues were a smoother flow of data to the gun director, assisted-
manual tracking and automatic tracking in a few sets. This new design
was named the Mannheim and used the same Würzburg wavelength [26].
The tongue had so accustomed itself to the name Würzburg that it was
generally applied to Mannheim, which looked similar.
The commando raid on Bruneval had been recognized as the prelim-
inary to some form of countermeasure, and both the Würzburg and the
Mannheim systems were adapted to the expected active jamming. This
285
was done by providing both with a front-end adapter called Wismar that
allowed a 15% change in wavelength that could be accomplished in about
half a minute. The active jamming came in the form of Carpet, which was
dealt with reasonably well with Wismar, but the excessive secrecy placed
on Düppel prevented any advance action for the coming passive jamming
by Window, and that was another story [27].
Perched at the top of a narrow and steep-sided peak in Swabia about
10 km south of Reutlingen sits the tiny Castle of Lichtenstein. Its small
size and high location may have inspired Telefunken’s engineers to name
their radar series for airborne interception after it, although this clue does
not seem to have been picked up by British intelligence as the names for
Freya and Wotan had been. Kammhuber had recognized the need for such
equipment shortly after assuming command of night air defense and by
early 1941 had presented his requirements to Runge at Telefunken. Runge
proposed a 20 cm design but was told to utilize the company’s experi-
ence with the 50 cm techniques to which they had now devoted nearly a
decade, as there was no time for experimentation [28]. By summer 1941
he had a prototype flying [29] and by February 1942 the first night fighters
were so equipped [30]. It made its entrance as a contract development
by Telefunken’s engineers, moving from drawing board to prototype to
production in a normal manner.
The Lichtenstein had a rectangular array of four dipoles with reflec-
tors mounted at the front of a twin-engine plane, the Me-110 and Ju-88
being initial favorites. This minimum array gave a wide beam, which pro-
vided a wide angular field for searching but also a short range, 3 to 4 km.
Lobe switching in two directions would have provided enough directional
accuracy for the radio operator to guide the pilot to the target; however,
Runge and Hans Muth found a more elegant solution, one that imitated the
Würzburg’s rotating dipole by using a rotating phase-shifter in the trans-
mission lines to the dipoles that produced the same twirling beam [31].
The vital minimum range achieved was 200 m, which fit perfectly with the
Freya–Würzburg system of fighter direction and greatly improved Dunkle
Nachtjagd.
A number of different Lichtenstein models were put into the field as
criticism of the first ones returned to Telefunken. The Lichtenstein B/C
went into active service in September 1942, but production was slow, pri-
marily caused by a bottleneck in manufacturing the special vacuum tubes,
and production was not generally of high quality, causing many sets to be
returned. The biggest complaints were in the range, the angular region
searched and indicators that were too complicated for combat. Several
modifications corrected some of these faults, but to increase the range it
was necessary to go to a 3.75 m band, the Lichtenstein SN-2, which allowed
more power than possible at 50 cm [32]. The long waves were to prove an
adroit choice for reasons yet to come.
The intitial response of the pilots was remarkably cold. The dipole
286
The Great Radar War
array detracted not only from the graceful appearance of the aircraft but
also reduced its speed quite a bit, and initially fighters so equipped were
not flown. A technically minded captain, Ludwig Becker, who had worked
with Diehl and Falck in their Zuider Zee operations [33], changed this
attitude by using the new device effectively [34].
Lichtenstein was Runge’s last important contribution to radar. On
graduating from the Darmstadt Technische Hochschule in 1923 he had
selected Telefunken in part to avoid working with Dr Karl Rottgardt at
another company at which he had been interviewed, as his dislike for
Rottgardt was immediate and enduring. When the object of his loathing
became Director of Telefunken years later, mutual animosity found fertile
ground, and by April 1942 Runge no longer had real responsibility. He left
Telefunken on 1 November 1944 [35].
An episode in German radar worth noting took place in October 1940
when Manfred von Ardenne and Hans Hollmann, each the owner of neigh-
boring private laboratories, proposed the plan of a panoramic radar based
on microwaves and the plan position indicator. They first approached
Göring, then a representative of Admiral Dönitz and finally Hans Plendl,
the Plenipotentiary for High Frequency Research. All rejected the proposal
[36]. There is reason to believe that the project as presented was not well
founded technically. They apparently had no suitable microwave gener-
ator, and a drawing that accompanied the proposal could easily have left
Plendl unimpressed, as it suggested that the two had lost contact with
radar research since the mid-1930s—the obvious result of the tight secrecy
imposed—but these were defects that two such men would have quickly
made good. The incident is revealing about both the Nazi state and the
two engineers. Hollmann had withdrawn from work on radar with GEMA
because of misgivings about the direction of the regime, and it is unlikely
that this had been ignored at top levels of the government. Von Ardenne,
a shrewd business man, had come near to joining the NSDAP in 1933 but
rejected the offer, as had his father. He had also come under the anti-Nazi
influence of Max Planck, Max von Laue and Graf Georg Arco, founder and
long director of Telefunken [37]. Neither of the two Lichterfelde labora-
tories was engaged in war work; Hollmann was doing medical research
for which von Ardenne was developing the use of tracer isotopes. Holl-
mann had, nevertheless, continued research in microwaves with emphasis
on magnetrons, and it is not certain what Germany’s—and until recently
Europe’s—leading microwave specialist had in mind. The regime thus
rejected the help of two of the nation’s outstanding electronic experts for
political reasons. What impelled the two to propose something of military
use was their concern about the air attacks they feared were coming [38].
There is an adage: ‘When war comes, you go with your people’.
While Germany was preparing the weapons for defending the Reich
from night attack TRE set out to furnish Bomber Command the means
for navigation; they had by mid-1941 finally come to realize this was an
287
absolute necessity. Beam systems had two faults. For the pilot to know
his position he had to be on the line that the beam defined, which was a
restriction of no small consequence, as it is difficult to conduct a raid, even
a peaceful flight, that requires the aircraft keep to an aerial railway. The
flier needs to know his position wherever he may stray. Second, a beam
is simple for the enemy to interdict, either electronically or by having the
defending fighters use it to locate the attackers.
R J Dippy approached A P Rowe, Director of TRE, shortly after that
organization had reached Swanage in the summer of 1940 with an idea for
a radio navigation system, which came to be called Gee, a curious non-
abbreviation of G, which in turn stood for grid. Dippy had discussed this
idea while at Bawdsey, but there had been no push to follow it then [39].
Gee was fundamentally simple but had complicating details. A chain
of stations consisted of a master and two or three slaves, all located at dis-
tances of about 150 km from one another. The master broadcast radar-like
pulses on wavelengths chosen from 3.5 to 15 m. The slaves had receivers
for the master pulses, which they then re-broadcast after predetermined
and accurately timed delays. Master and slaves used the same wavelength,
so the navigators received all relevant signals on the same receiver, limited,
of course, by line-of-sight requirements.
On receiving signals from any master–slave pair the operator mea-
sured the delay between them. A given time delay did not determine
position but did locate the receiver on a hyperbola. Repeating the process
for another pair located the receiver on a second curve. The navigator was
provided with a map on which these curious grid systems were superim-
posed and located his position from the intersection of the two curves he
had determined. For identification, the master and slave pulse rates dif-
fered. The master had a repetition rate of 500, the first two slaves had rates
of 250 transmitted alternately and the third transmitted double pulses ei-
ther at 500 or 500/3. Complicated as it would appear, it presented an
easily recognizable display on the receiver cathode ray tube. A fix could
generally be accomplished by any two station pairs; the third pair resolved
ambiguities and extended the range when its slave was the closest to the
aircraft. Accuracy decreased with distance, primarily because the inter-
secting curves became more nearly parallel; the uncertainty varied from
one to a few kilometers [40]. There was much enemy territory not covered
by the new system because of the line-of-sight restriction, but even for
those raids it furnished a secure starting point. It also reached out to guide
the tired, often wounded fliers home, a function of value comparable to
sending them toward their targets.
Small-scale tests using low-power stations in fall 1940 showed the
idea to be practical. By August 1941 three full-power stations tried opera-
tion with 12 aircraft that had prototype receivers, and the results were an
outstanding success [41]. Unfortunately, something disturbed the eupho-
ria. The tests were extended to a raid on the Ruhr, and one Gee-equipped
288
The Great Radar War
bomber failed to return, leaving the distinct possibility that a receiver was
being examined by Martini’s engineers, who would be onto the secret in
short order with jamming techniques running through their minds. This,
of course, was to be expected in normal operations, but these sets were
laboratory-built prototypes and production was months off, despite all
pressures to expedite it—months during which countermeasures could be
devised.
The matter was turned over to R V Jones as an inversion of his normal
duties. Now he must do what he could to obfuscate any intelligence the
enemy might have gained. Here Jones, a notorious practical joker, was
in his element. He had the Gee transmitters continue broadcasting but
with unsynchronized pulses; he had them disguised to look like radar
stations; he had some navigation beams set up, which the Air Force found
quite useful, they having nothing else [42], and which he called J in the
hopes that the nearness in pronunciation to German ears would confuse
interrogators of Air Force prisoners [43]. From whatever cause, secrecy
was retained.
On 22 February 1942 Air Marshal Arthur Harris became the
Commander-in-Chief of Bomber Command. He brought to this position
a vigor and ability that his long service had amply demonstrated. He also
brought to it the conviction that strategic bombing was to be decisive and
the determination to make it so. He knew the destructive ability of his com-
mand but preferred to ignore its manifest weaknesses by pushing ahead,
surmounting difficulties as they arose rather than worrying overly about
them in advance.
On the other side of the Channel leadership was virtually absent.
Kammhuber had devised what was to develop into a good initial mode
of defense, but he was allowed the power of a technician rather than of
a commander and was swept aside when new tactics were demanded.
Göring bore complete responsibility, but his long series of wrong decisions
crippled the arm that had been so terribly feared in 1939 and led to its ulti-
mate destruction. Unwilling to turn authority over to Erhard Milch, alone
among Germany’s ruling clique with the competence and drive necessary
to retrieve her fortunes in the air, and retaining the dictator’s support,
Göring blundered to the end. The air defense of Germany settled onto
subordinate commanders, radar engineers and brave, resolute air crews
who fought each new attack with whatever resources were given them but
without a sure, guiding hand directing a strategic response.
The air war against Germany could now begin in earnest. Both sides
had prepared for the event and completed their preliminary actions. The
Luftwaffe had perfected its air warning, was improving its night-fighter
direction and Flak rapidly and Bomber Command finally had a method
of navigating a few hundred kilometers into enemy territory and enough
bombers to begin. The first attack with Gee on Essen on 8/9 March 1942
opened Harris’s campaign of strategic bombing.
289
And so it was that March 1942 bore a strong resemblance to July 1940.
Britain, later joined by America, was to attempt what the Luftwaffe had
failed to do. In this they were successful because the invasion succeeded
two years later, whereas the German invasion had not been attempted.
But they failed to bring victory without invasion—the central doctrine on
which the proponents of air power had based demand for resources—and
the methods employed in the attempt generated more contention than any
other part of the Allied war effort.
6.2. COUNTERMEASURES
Watson Watt’s gigantic CH towers offered mute testimony to early concern
about radar countermeasures. The steel transmitter towers had many an-
tenna configurations during the course of the war, with horizontal dipoles
mounted along the sides in the beginning. The radiation pattern from the
side-mounted dipoles was sufficiently poor that curtain arrays became the
standard transmitter antenna for the later, West Coast, stations. Many of
the initial, East Coast stations were re-fitted with one or at most two cur-
tain arrays hung between the fortuitously placed towers that had resulted
from the original CH design of four operational frequencies. Owing to the
narrow bandwidth of these antennas, extra arrays had to be available for
the rapid changes of frequency needed to evade jammings. (Jamming is
an enemy transmission on the radar frequency of a signal modulated to
fill the operator’s indicator with obscuring traces.) This fitted well with
the need to have spare capability in the event of units not functioning, ei-
ther from malfunction or enemy action. Multiple wooden receiver towers
mounted dipoles of wavelengths corresponding to those of the transmit-
ters. A design anticipating countermeasures was but one characteristic of
British radar that set it off from early American and German equipment:
it had been intended from the first to be part of an air-defense system.
Colonel Martini attempted to jam CH during the Battle of Britain from
ground stations across the Channel but with little success [1].
Fighting the radar war required a knowledge of what devices had
to be countered, which became the province for a new specialty, electronic
intelligence. The information so gained then went to the laboratories where
equipment was designed to thwart what the enemy had brought forth.
Destruction of the enemy radar was the natural and traditional way of
dealing with any kind of new weapon, but countering it electronically was
often easier—and somehow seemed a more appropriate, less brutish way
for the engineers to fight—and if one was clever the enemy might be fed
choice bits of misinformation and thereby led astray.
Use of radio countermeasures slacked during 1941 after the lessening
of night attacks on England. Bomber Command had begun its attacks on
Germany but was still unaware of any serious radar threat and had no
interest in electronic warfare. As R V Jones’ scientific intelligence began
290
The Great Radar War
piecing together the dangers of German radar through the actions of 80

Signals Wing and the RAF Y-Service, the need for countermeasures be-
came more apparent to the people at TRE. It was the knowledge that the
Channel Dash of the Scharnhorst and Gneisenau on 12 February 1942 had
been aided by Martini’s jamming of the CD and CHL equipment expected
to monitor navigation that grasped the attention of the technical people
on both sides and showed them what was coming. The radar-intelligence
reason for the Bruneval raid that followed on 27/28 February served to
emphasize matters. An official organization, the Radio Countermeasures
Board, met for the first time on 24 March with all manner of service orga-
nizations represented, including one from the United States. It was only
an organizing group; TRE had to do the work.
It was Fighter Command, trained and skilled in the use of radar,
that made first use of a radar-spoofing device. Moonshine was TRE’s first
offensive interference. It responded to a Freya interrogation by returning
a much stronger echo, thereby giving the radar operators the appearance
of a large formation. It was first used mounted in an otherwise useless
two-man fighter, the Boulton-Paul Defiant; it was first used operationally
in a feint to cover the US 8th Air Force’s debut in a daylight raid on Rouen
on 17 August 1942 and remained in service for another two months, after
which the ruse no longer deceived with assurance and was set aside to be
brought out again for the invasion [2].
Bomber Command’s first active countermeasures were undertaken
unexpectedly and wrongly by the flight crews. This also initiated Bomber
Command’s greatest electronic weakness, one in which it persisted for two
more years—filling space with unnecessary radiation. If their commander
was not alert to the need for countermeasures, the flight crews were. They
knew by then that Flak was radar controlled, at least in part; they also
knew that it could be deadly accurate. In their informal discussions the
flight crews came to the conclusion that the IFF mark II, which should
have been switched off when away from possible interrogation by Allied
radar, worked as an anti-Würzburg jammer, if left switched on. In this
the desperate crews followed the paths of others having terminal diseases
and who seek delusive cures, basing their faith on the statistics of small
numbers highly skewed by the selection effects in acquiring them.
Astoundingly, the Command found no harm in this, which was true
inasmuch as none of the German sets of the time were interrogating on
wavelengths to which IFF mark II responded and hence the transponder
emitted no radiation or did anything else, but they compounded this fool-
ishness by permitting the installation of a switch, called the J-switch, that
caused the set to transmit even when not interrogated [3]! The object of this
was to boost morale by showing understanding for the crews’ anxieties;
fortunately for them it came too early to become the delight of the newly
organizing Y-Dienst, but it destroyed any remaining air crew respect for
the value of radio silence, although it also removed the last resistance of
291
Bomber Command to the use of active countermeasures.

Active airborne jamming of the German 2.4 m air warning equip-
ment, at the time restricted to Freya but soon to include Wassermann and
Mammut on the same wavelength, took the form of the TRE-designed
Mandrel, first used during the night of 5/6 December 1942. By June 1943
the activities of Y-Dienst were known, and Mandrel’s utility as a jammer
had to be weighed against its disservice as an alarming beacon. Various
modifications in its use never resulted in a suitable balance between the
two. The hope that Mandrel would prevent the Himmelbett Freya from
directing the Würzburg GCI onto the bombers proved illusory when there
were so many bombers that the Würzburg could locate a target unassisted
[4]. Another device, called Tinsel, went along with Mandrel. It was sim-
ply a slight modification of the communication transmitter tuned to the
German night-fighter-control radio frequencies and modulated by a mi-
crophone that picked up engine noise, the first of many attempts to disturb
the German pilots’ communication with their controllers.
The radar designers wanted to do more than just interfere with en-
emy electronics, and they did this by furnishing two unfortunate pieces
of equipment: Boozer and Monica. The former was a passive receiver for
German radar signals that set off a warning alarm for the crew, the latter a
small radar in the tail to warn them of the approach of a night fighter. Nei-
ther remained welcome for long aboard the aircraft. There were soon so
many radar emissions over and on the way to Germany that Boozer contin-
ually gave alarms to crews perplexed as to what to do with the information.
When Bomber Command began to enter Germany in streams rather than
the earlier extended formations, Monica began to alert the crews about
the presence of nearby bombers. The Germans extracted Monica from a
grounded bomber, and night fighters began to home on it with the simple
receiver Flensburg [5]. Not all inventions help.
American countermeasures began with Luis Alvarez taking the ini-
tiative of having one of his staff at Rad Lab, the Canadian Dr Don Sinclair,
work with the General Radio Company in modifying their field-strength
measuring set P-540 into radar intercept receiver SCR-587 (Army) and
ARC-1 (Navy). Its extraordinarily wide band, from 10 to 300 cm, was
to make it useful throughout the war, but it suffered at the time from not
having single-knob tuning, so it was not yet ready for the field. It was the
second American receiver to prove of value in the new field of electronic
intelligence; the Hallicrafter S-27, available in England for sale to amateurs,
had picked up the emissions of Seetakt and Freya in early 1941. In an early
lesson Rad Lab had been made receptive to the idea of countermeasure
work rather early when the transmission by a nearby amateur had over-
ridden a poorly shielded intermediate-frequency amplifier of one of the
experimental radars.
Four days after the attack on Pearl Harbor American countermea-
sures began at a meeting among representatives of the National Defense
292
The Great Radar War
Research Committee, the Rad Lab and the Navy. They decided without dis-
sent to establish a special laboratory for this purpose and proposed Alvarez
as its head because of his intercept-receiver initiative and general interest
in the subject. Commitments to work in progress at Rad Lab caused him
to decline the offer but with the strong suggestion to offer the job to Pro-
fessor Frederick Terman, head of the Department of Electrical Engineering
at Stanford University. This proved a fortuitous choice owing to Terman’s
knowledge and stature in the field of radio and his ability to approach his
many former students in building a staff. In August 1942 he secured an
especially valuable addition in the team that was working for Columbia
Broadcasting developing color television. Better yet, they brought with
them their own laboratory equipment, exactly the type needed and al-
ready familiar to them [6]. Terman’s general arrangement for obtaining
industry people was to have them remain employees of their parent cor-
poration, thereby retaining seniority and pension rights, and to pay the
corporation for their services. In addition to the satisfaction of contribut-
ing to the war effort, the company thereby gained electronic skills through
their engineers that they might easily have missed, had they remained on
the sidelines. By January 1944 Terman’s lab had 744 employees, 214 of
whom were research personnel [7].
Terman went to Britain in April 1942 for six weeks and established
excellent rapport with Robert Cockburn, who was chief of RAF counter-
measures, while he learned the characteristics of German radar and British
response. On return he organized work into three sections: building a jam-
mer for Freya under John Byrne, a jammer for Würzburg under Bob Sorrell
and continued work on the intercept receiver by Sinclair. By then his orga-
nization had been given the obligatory deceptive name, the Radio Research
Laboratory, and had taken up quarters in a wing of the Biological Labora-
tory at Harvard [8]. Whether for security or to confuse future historians,
the jammers for Freya and Würzburg were given the same names used by
TRE, Mandrel and Carpet, but they were different designs [9].
Early jammers on both sides emitted the radar frequency modulated
with a sine wave, but experience showed that it was better to modulate with
noise, as this gave the appearance of a deterioration in the performance
of the targeted radar receiver rather than some disturbing agency. This
was the technique Martini had employed so successfully in the Channel
Dash. Gas regulator tubes were common circuit components that formed
a moderately constant reference voltage when current within some range
of values passed through them. Without filter capacitors they showed
significant noise levels and took on a new function as noise generators.
Another noise generator was the photomultiplier tube. This device was
capable of producing an extremely short output pulse on absorbing an
optical photon and was the heart of all television cameras. By keeping
the multiplier in the dark and increasing the voltage beyond that required
for normal operation, thermal electrons from the photocathode and first
293
electrodes produced a shower of pulses at the output and a wonderful,

high-amplitude source of high-frequency noise was obtained very simply.
Both the Mandrel and Carpet jammers of the Radio Research Laboratory
used the RCA 931 photomultiplier for this. Specially designed gas tubes
later proved equally good and more convenient, as they needed much
lower voltages than photomultipliers [10].
The Radio Research Laboratory produced one of the most curious
jamming devices of the war, Tuba. The Lab’s American contingent at TRE
noticed a prototype ground jammer called Ground Grocer that worked on
50 cm, designed to get as much power as possible out of the Micropup tri-
odes and intended to interfere with the new Lichtenstein radar of the night
fighters. Beams were directed from the coast of England at regions where
they would encounter the night fighters pursuing returning bombers [11].
The Americans recalled the very-high-frequency beam tetrode called the
resnatron, invented by Sloan and Marshall, that had almost entered the
open scientific literature1 It could generate decimeter waves of very high
power with a modest range of adjustable wavelength. Radio Research
Lab proposed a jammer based on it, and the British ordered one imme-
diately [12]. The tubes that formed the heart of Tuba were demountable
and vacuum pumped; on-site repairs and maintenance were possible from
a truck-mounted machine shop. The power was radiated through wave-
guide-fed horns of the large size that 50 cm required, and its 80 kW of con-
tinuous power could light a fluorescent tube a mile away. A canvas cover
once burst into flame when placed over the exit horn. Initial deployment
proved troublesome to German fighters near the Channel that headed into
Tuba’s radiation but not to those further away, regardless of their direction.
All effort was soon pointless when the airborne radar wavelength changed
to 3.5 m just as Tuba became operational, a wavelength completely outside
its capabilities [13].
At times nature provided radar operators with disturbances far more
intense than did the enemy. This was particularly true of long-wave equip-
ment. Engineers and Air Force officers often saw alarming signals on
the oscilloscopes of CH that the seasoned women operators summarily
dismissed as ‘ionospherics’ or some such, yet saw them fasten onto an
imperceptible blip for which a squadron scrambled to intercept. On the
other side, the versatile Telefunken engineer Wilhelm Stepp commented
that ‘Initiating new equipment, and the long-wave set "Heidelberg" comes
to mind, was like an expedition into a wonder world of new and at the
time unknown phenomena and difficulties’ [14]. Out of these anomalies
emerged some new science. J S Hey located the ions in the upper atmo-
sphere that marked the path of an incoming meteor [15] and discovered
the greatly enhanced radio energy of solar flares using GL equipment; he
at first thought the solar radiation was very clever jamming by the Ger-
294
The Great Radar War
mans [16]. Radio propagation peculiarities near the ocean surface often
produced unaccountably and unexpectedly long ranges to targets that con-
fused the operators and sometimes generated phantom targets—fired on
by batteries from both sides of the Channel. Clouds and humidity gradi-
ents added to the confusion, and sidereal noise limited the signal-to-noise
ratio obtainable with CH [17].
When Watt and Wilkins made their calculations in 1935 of the inten-
sity of the re-radiation by a dipole, intended to approximate the resonance
behavior of the wing structure of an airplane, they inadvertently investi-
gated a very effective means for interfering with radar. A modern radar
operator is by no means exempt from the confusion that this early coun-
termeasure produces when a shy target fills the space surrounding itself
with dipoles cut to lengths that resonate at the wavelength of the incident
radiation. There are ways for him to deal with this nuisance but a nuisance
it remains, one that can saturate his equipment. When one considers how
obvious the idea is and how many persons thought of it independently,
how puzzling seems the extreme secrecy placed on it before its first use in
1943.
Lindemann, ever ready to find fault with early radar, had seized on
the effect of releasing dipoles in the CH beam that his student, R V Jones,
who was studying infrared detection, had pointed out to him [18]. One
of the virtues of CH, however, was its resistance to this form of interfer-
ence. The long wavelength and horizontal polarization required that the
dipoles had to be some meters long and suspended horizontally, which
never proved feasible. Knowledge of the efficacy of dipoles spread when
the Bawdsey researchers used dipoles suspended by balloons in testing
the 5 m GL sets [19]. American operators of SCR-268 learned early to
have someone climb a tower with an attached dipole in order to test their
equipment.
Thinking about these things was not restricted to the learned and was
especially encouraged among those most closely trapped by events. Thus
when 148 Squadron was making electronic intelligence flights in North
Africa in September 1941 and found themselves subjected to intense AA
fire, they suspected that their antennas were enhancing their radar echo. In
a subsequent raid on Benghazi they had the bomber crews throw out 46 cm
long aluminum strips, the dimension being that of the antenna suspected
of having drawn fire, but found no effect on the accuracy of the Flak [20].
Obvious speculation about what might have resulted had the crews been
informed about the Würzburg wavelengths and cut the strips to 27 cm
must be tempered by the knowledge that there were no Würzburgs in
North Africa until April 1942, indeed no German radar at all until January
of that year [21].
In March 1942 Joan Curran, the only woman scientist at TRE, re-
ported her own investigation of the idea, which she found practical and
compelling. By 4 April its use had Air Ministry approval and the code name
295
of Window. Fighter Command requested that its use be held back until its
effect against British radar had been established. These tests did, in fact,
show that the AI marks VII and VIII, the first 10 cm sets, were affected. Their
beams spiraled around the direction of the fighter, and Window confused
the radial indicator display; the tests also showed the GCI sets then being
introduced, which used 50 cm radiation, could also be seriously confused,
greatly complicating night interception [22]. Bomber Command was be-
ginning to find the losses attributed to Würzburg-directed Flak unaccept-
able, which led to a dispute between the two commands, and Window in-
troduction was postponed when it was learned that AI mark X or SCR-720,
which had a different kind of indicator, was much less susceptible. Nei-
ther command was aware at the time that the German airborne interception
radar was being designed for the Würzburg wavelength. Cockburn was
permitted to experiment but under the strictest security.
Window was the perfect countermeasure for the Würzburg wave-
length. To be effective Window must fall through the radar beam to insure
that the operator’s indicator is overwhelmed. If the radar beam is nar-
row, as is the case for microwaves, the chance of a sufficient number of
dipoles lying in it will be correspondingly small. The Würzburg 50 cm
beam was large enough to pick up thousands of the little dipoles as they
drifted through the sky, whereas a 10 cm beam would intercept only about
1/25 as many. The Würzburg used a rotating dipole at the feed, which pro-
vided a continuous sweep of all polarizations, so the random orientations
of the falling strips were well scanned [23].
Telefunken tested the idea at the Luftwaffe research station at Rechlin
in early 1940 with results so shocking to the top levels of the Luftwaffe that
the tightest secrecy was imposed on the knowledge; a ban was even im-
posed on further experimentation, effectively cutting off the development
of protective measures. In accordance with Telefunken’s use of geograph-
ical locations as code names, Leo Brandt called the metal foil technique
Düppel after a Flak battery stationed on an estate by that name near Berlin-
Zehlendorf [24], an exceptional way of being secretive—the German word
for dipole is similar to the English [25].
When Terman returned from his stay with Cockburn he brought
knowledge of Window with him. He instituted research in what the Amer-
icans chose to call Chaff and assigned Dr L J Chu, a noted antenna expert,
to study the matter theoretically. Theory was important because the ra-
tio of the width of the foil to its length determined the bandwidth of its
resonance. A very sharp resonance could miss the wavelength of the inci-
dent radar and generate only a small echo. To extract the practical aspects
Terman turned the theory over to Fred Whipple, an astronomer. Practical
manufacture set constraints on a design for the huge quantities that would
be needed, and Whipple soon worked out formulae that gave radar cross
section for a given bandwidth per kilogram. His results were confirmed in
experiments at TRE, and large quantities were stockpiled against the day
296
The Great Radar War
of use [26]. By the end of the war three-quarters of all American aluminum
foil production went for Window [27].
The introduction of Window caught the Würzburg-equipped gun-
ners and Lichtenstein-equipped night fighters completely by surprise.
Ever since the Bruneval raid, jamming had been expected for Würzburg,
and a modification, Wismar, that allowed relatively rapid changes of wave-
length was ready, but an air filled with dipoles had been officially banished
from thought and plan. What might have been accomplished in calm now
had to be done in extreme haste.
The night fighters were already in the process of changing their 50 cm
equipment for 3.5 m sets (for unrelated reasons) and thereby became im-
mune to the new clouds obscuring their prey, but a substantial change
of wavelength was not an immediate option open to the AA radar. The
basis for working through Window came from the very first reflection ex-
periments of the early 1930s—the Doppler effect. The wavelength of the
reflected signal was altered by motion of the target, and bombers moved
fast whereas the drifting dipoles moved with the wind. The Würzburg
used special high-frequency triodes in its transmitter, which were stable
enough in frequency to allow filters in the receiver to distinguish shifted
from unshifted echoes [28]. This was the first use of the technique that has
come to be called ‘pulsed Doppler’. Unfortunately, echoes from aircraft
were not just shifted, they were shifted to either side of the transmitter
wavelength depending on whether the targets were approaching or de-
parting and by an amount dependent on the velocity.
Asuitable device, called Würzlaus, invented at the Max-Wien-Institut
within two weeks after the alarm [29], suppressed the unshifted signal that
came from the cloud of dipoles, which made it easier for the operator to see
the unsuppressed echoes from the target. It allowed some degree of success
in distinguishing bombers from aluminum foil, and in skilled hands under
the right conditions this restored much of Flak’s accuracy. It worked on
the approach, where there was a Doppler shift to shorter wavelengths, and
on the departure, where the shift was to longer wavelengths, but failed,
of course, in the important mid-course region when the bombers flew at
right angles to the radar line-of-sight and where the Doppler shift was too
small for discrimination. But Würzlaus had the unfortunate characteristic
of requiring a fixed wavelength, which made it incompatible with Wis-
mar’s frequency agility. This meant that formations throwing out packages
of Window and operating their Carpet jammers, something that became
common in fall 1944, made radar-directed gunfire sometimes impossible.
When overcome in this manner the Würzburg made use of Stendal A and
became a passive device that at least determined the direction to the Carpet
transmitter [30].
Another expedient made use of an audible signal produced on the
reflected pulses by the propellers and general vibration of airframes, an
effect noted by Lorenz investigators when observing a windmill in exper-
297
298

The effect of countermeasures on the loss rate of RAF Bomber Command. The percentage loss averaged over each 3000 sorties is plotted
against the number of sorties for attacks on Germany. Note that the number of sorties was not uniform in time, the consequence of
the Command having been given missions other than over Germany and from restrictions forced by circumstances. The introduction
of the various radio countermeasures is noted by their code names (not all of which are described in the text) as are additional factors
that affected losses. The introduction of each new countermeasure generally resulted in an immediate reduction of losses followed by the
defenders countering the countermeasure. The loss rate dropped significantly after the liberation of France, after which the Luftwaffe
was greatly weakened and their air-warning capability significantly reduced. Adapted from RAF Signals vol 7, p 190. Crown copyright.
The Great Radar War
iments just before the war. The radar receiver output was passed through
an audio-frequency filter that suppressed the pulse-repetition frequency
before transmitting it to a pair of headphones; the operator then attempted
to ‘hear’ the airplane. The device, called Nürnberg, was manufactured
between September and December 1943 [31].
In early 1944 Dr H Pöhlmann of the Reichs-Luftfahrt-Ministerium
removed the frequency rigidity and other deficiencies of Würzlaus with
an improvement called Tastlaus, which worked with Wismar and thereby
greatly improved the ability of radar control to work through Window and
Carpet [32]. Despite all Allied precautions, Flak took ever greater numbers
of the attacking formations. During the last seven months of the war the
US Army Strategic Air Forces in Europe lost 1566 aircraft to AA fire [33].
What would the total have been without Carpet and Window?
The unsatisfactory nature of the methods to counter Window was
reflected in a competition offered by Göring for the best technical solu-
tion with tax-free prizes up to 300 000 Rm depending on the value of the
invention. Entries were to have been delivered to the Air Ministry Tech-
nical Office GL/C-F4 by 1 April 1944. Staatsrat Professor Abraham Esau,
who would shortly replace Plendl as Plenipotentiary of High-frequency
Research, was to head the judges [34]. There is no record of an award.
The Pacific War was almost a separate war, and this held particularly
for radar; thus the introduction of Window by the Japanese during a night
raid on Guadalcanal in May 1943 went unreported to Europe and did not
enter into the discussions that led to its introduction in the July attack
on Hamburg. The Japanese dropped 75 cm strips of ‘deceiving paper’
(Giman-shi) to disrupt the SCR-268 that was proving to be a problem.
They considered the use successful in reducing losses [35].
Observing enemy radar and radio navigation equipment in order
to discern its characteristics was the earliest function of countermeasures
people. The Royal Navy entered quickly into these transactions, setting
up an electronic countermeasures station at Dover during summer 1941
[36]. Such services quickly became airborne, and their missions, though
inherently passive, were by no means safe. The direct observation of a
Lichtenstein during the night of 3 December 1942 proved to be a very near
thing for an air crew that had deliberately positioned themselves as bait
[37]. Much more tangible information about the Lichtenstein came the
following May in the form of a Ju-88 equipped with 50 cm B/C model,
thanks to the carefully planned defection of a night-fighter pilot and ra-
dioman who secured the acquiescence of the mechanic with the aid of a
pistol. This gave Britain not only the details of the radar and radio altime-
ter but allowed the machine to be used in simulated fights with British
aircraft, which helped them locate weaknesses to exploit [38].
The vast expanse of the Pacific resulted in a number of intercept-
receiver-equipped long-range aircraft, called Ferrets, probing the Japanese-
held islands, but the first solid data came by way of a submarine, not an
299
airplane. USS Drum went on patrol in September 1942 with an ARC-1

receiver on board. The radio operator was told how to use it when the boat
was on the surface near Japan, but only if its employment did not interfere
with other operations. The strip chart records brought back were the first
interceptions of Japanese radar. A newly equipped Ferret in the South
Pacific at about the same time failed to be first, presumably because of a
poor receiver. The Ferrets soon improved their technique and successfully
mapped the radiation from the Japanese mark I model 1 set on Kiska,
Alaska in March 1943 [39]. These kinds of flight became a normal part of
fleet reconnaissance.
When radar sets pointed their beams according to the will of the
operator, intercept receivers could be tuned by hand, as the interrogating
beam could be expected to remain long enough to be recorded. When
the PPI technique became widespread, the receiver could expect to be in
the sweeping beam only a small fraction of the time, and this required
automatic scanning receivers. The ultimate of such receivers was the TRE
frequency-indicating receiver that presented received signals on a cathode-
ray tube with their amplitudes shown as straight lines radiating from the
origin and the frequencies given by the angles of trace orientation. Signals
that might be missed during most receiver scans stood out and were easily
identified by the operator [40].
Such were the principal means by which the adversaries came to grips
over Germany during the night and, after the arrival of the Americans, dur-
ing the day. There was variation that will occupy us yet, and deception was
not restricted to radio waves. German radar engineers found themselves
devoting ever more time and resources to devising the means of countering
Allied countermeasures, but they understood the peril of the time and gave
a full measure of ingenuity to protect their homes while the regime that
controlled their country seemed to the end incapable of comprehending
the need for an extensive development of defensive weapons.
6.3. AN AIR WAR OF ATTRITION

Lübeck is an ancient Hanseatic city closely tied through commerce to the
other old ports of northern Europe. Until the Nazis incorporated it into
adjoining Prussia in 1937 it was a separate political entity with a republican
government, expanded after 1848 to include democratic representation.
The trading families of Lübeck felt more akin to people in the sister cities
of Hamburg, Amsterdam, Antwerp and London than to those in Berlin.
But the cultural stamp that attracted Arthur Harris in March 1942 was
Lübeck’s old buildings, not their splendid medieval architecture but their
half-timber construction, for the city had been selected to be destroyed.
In addition to its combustibility the distinctions that made its selection
obvious as a target were a relatively light air defense and a location near
the sea easily recognized on a moonlit night, necessary because it was
beyond the range of Gee. It was small enough for Harris to be able to
300
The Great Radar War
demonstrate the destruction of a city with the force available to him, yet
large enough to count in the headlines.
So mankind increased the nightmare of total war by yet another de-
gree, but by then who would have urged the planners at Bomber Command
to desist? It is doubtful that such restraint would have come from many
inhabitants of Coventry or burghers of Rotterdam or Londoners who had
spent the winter nights in the underground stations, wondering whether
they would find their homes or work places standing when they made
their way out the following morning.
There were skeptics to Bomber Command’s capability, and Harris
saw the need to have results that would impress top levels of the Air Force
as well as political leaders. This was the reason for destroying Lübeck and,
shortly thereafter for the same reason and under similar circumstances,
Rostock. He was also clearly playing to the press in making the attack
with 1,000 bombers on Cologne at the end of May 1942, for which he had
had to strip the training squadrons in order to achieve the magic number.
The city was severely damaged, and it was a success for Gee.
Gee was a great navigational advance but had three weaknesses: (1)
its range was limited by approximate line-of-sight transmission, (2) its ac-
curacy was of the order of kilometers and (3) jamming, expected from the
start, began on 4 August 1942 by jammers called Heinrich [1]. Foreknowl-
edge of these faults had set TRE planning other navigational aids, two
of which become important: Oboe and H2S, both introduced in Decem-
ber 1942. Oboe would prove to be a remarkable blind-bombing device
that provided skilled users with the ability to drop bombs from very high
altitude to within a hundred meters of the target. H2S2 was the most in-
accurate of all—indeed accuracy is probably not the right word—but was
not restricted in range by the curvature of the Earth as were Oboe and Gee.
Gee remained in service, as it was not always jammed and was excellent
for navigation, a use for which the Luftwaffe soon found affection.
Using one another’s radio navigation systems became relatively com-
mon. When the Germans introduced Elektra Sonne and set up a station in
Spain to work with a second in Brest, Britain quietly accepted this neutral-
ity violation because it was useful to Coastal Command flying over the Bay
of Biscay, where Gee became inaccurate or did not reach. Elektra Sonne’s
intersecting beams required no special kind of receiver and were extremely
simple to use; it remained in civilian service after the war as Consol, its
British wartime code name [2].
Oboe made use of two widely spaced radar stations in England, called
Cat and Mouse, that could control one aircraft, generally one of the re-
markable Mosquitoes that flew very high and very fast. The aircraft was
equipped with a transponder that allowed the Cat to determine its range
with high accuracy. On approaching the target the plane was constrained
2 See Chapter 4.5 (pp 188-90).
301
to fly at a constant radius from Cat, whose transmitted signals were coded

to give the pilot a pure tone if on course, dots if he were on the near side
and dashes on the far side. The Germans noticed the curved approach
and called the new system Bumerang (boomerang). Mouse measured the
range between it and the aircraft, determined the ground speed and cal-
culated the release time for the bomb. It transmitted coded signals that
allowed the crew to prepare and indicated exactly when to release [3]. The
reader need not be told that skill, practice and cool nerves were required
for success, but all became available.
Oboe originated with A H Reeves, whose enthusiasm was picked
up by F E Jones, who helped him work out details in spring 1941. Given
knowledge of the usefulness of Oboe, especially in the destruction of the
Ruhr industries, it is difficult to comprehend the vehemence with which its
development was opposed at TRE and in the RAF. One objection was that
it would be suicidal to fly an aircraft on the required preparatory course.
This would have been a valid objection except for the remarkable char-
acteristics of the Mosquito that flew at altitudes and speeds that made it
almost impossible to be intercepted by fighters or shot down with AA fire,
and the Mosquitos that flew Oboe had astoundingly light losses. The ob-
jection that only one aircraft could be controlled at a time was overcome
by having the Mosquito drop marker bombs, a method which came to be
accepted in summer 1942 [4]. Another objection was the general impres-
sion that Oboe would be extremely easy to jam, which was true for mark I
that operated on a meter-wave band but not for marks II and III that used
10 cm radiation. The other problem Reeves and Jones faced at TRE was
the general acceptance that H2S would do the job and at ranges far beyond
400 km from the English coast.
Despite opposition, the director of TRE, A P Rowe, allowed the work
to go forward, and in September 1941 optical tests made at a distance of
130 km from Cat gave bombing accuracies of 50 m. These were followed
by blind bombing that showed Oboe superior to release by optical sights in
good visibility. A demonstration on 2 July 1942 before a number of senior
officers yielded a bombing accuracy of 65 m [5]. It was an accuracy that
was transferred to the Krupp works.
In December 1942 Mosquitos began to try Oboe on the enemy. There
had been reason to suspect problems of accuracy because of the poor con-
trol linking British and continental grid systems, but when Clifford Corn-
ford, who was responsible for the mathematics of Gee and Oboe, looked
into the matter, he found that the Germans had supplied him with the
data he needed. During World War I they had triangulated points across
the Channel in order to bring the two grid systems into agreement with
long-range artillery fire the objective, and the results had come into British
hands. A Mosquito raid on the easily recognizable Wehrmacht Headquar-
ters in Florennes, Belgium had demonstrated the validity of the correc-
tion [6].
302
The Great Radar War
In practice things were a little more complicated than when displayed

to the senior officers. In addition to the hostility of the recipients there was
the need to operate at extremes of altitude and range, 10 000 m and 400 km.
There was the much more difficult matter of having the main force find
and see the marker bombs, which cloud cover might completely obscure
although industrial haze would not. The first part of the practice used
isolated Mosquitos dropping sticks of bombs on targets clearly identified
on maps to confirm that Oboe was properly calibrated.
The defense was puzzled over these isolated flights and at first as-
sumed they were just harassment to stop production through the air alarm.
On 7 January 1943 Hauptmann Alexander Dahl, artillery officer from World
War I, free balloonist and radio amateur when he had dared, noted that
the explosion of a stick of three bombs had been quickly followed by an
immense flash bomb high in the sky. A quarter-century earlier Dahl had
observed artillery fire from a kite balloon and thought the RAF was ‘ad-
justing fire’ [7]. The photograph exposed at the time of the flash would
show the exploding bombs in the target area, and he assumed this was be-
ing used to determine and remove systematic errors. It was a reasonable
conclusion but wrong. The accuracy already sufficed; it was a training
flight.
After training his crews for two months Harris began his Battle of
the Ruhr on the night of 5 March. He had created a Pathfinder Force that
contained a small number of heavy bombers to lay many marker bombs
visually on those dropped by the Oboe-equipped Mosquitos leading the
way. This insured a sufficient number of markers for the main force to aim
at. The Battle of the Ruhr continued through July, although other parts of
Germany received attacks during that period [8]. The Battle of the Ruhr
ended as an unquestioned victory for Bomber Command, a victory for
which the necessary conditions were the accuracy of Oboe and the inability
of the air defense to destroy or even harass the high-flying Mosquitos.
The validity of both conditions had been completely rejected by Air Force
command two years earlier and viewed with considerable suspicion even
one year before.
H2S was introduced at the same time as Oboe and was used in some of
the Ruhr attacks as a means of gaining experience with it. Initially, at least,
H2S used Pathfinder techniques, not because it was restricted to a single
plane but because there were not enough sets for the many bombers. Its
function was strongly dependent on ground characteristics, especially the
presence of bodies of water, which reflected the 10 cm waves like a mirror
and presented the PPI operator with a welcome dark image of a river or
estuary. Thus the easiest targets had bodies of water as part of their make-
up; the next best were small cities surrounded by farmland; the worst were
large, distended industrial regions, and Berlin was the worst target of all.
H2S was not in the same class as Oboe for accuracy regardless of the target
conditions. If conditions were right, one could hit the town, occasionally
303
doing better than that.

The growing success of the defenders’ Himmelbett system and the
need of the main bomber force to release on marker bombs combined to in-
troduce the bomber stream. This sent so many aircraft through a Himmel-
bett cell that its two Würzburgs and one night fighter were overwhelmed.
In the 1000-bomber attack on Cologne in May 1942 only eight Himmelbett
cells were touched and only 25 night fighters guided to targets. Fighters
and Flak destroyed 44 aircraft, a large number for a night’s work but small
relative to the number of attackers. It also left a large number of fighters
tied to inactive Himmelbett cells.
The immediate response was Kammhuber’s call for more radar and
more Lichtenstein-equipped night fighters. The call was answered at well
below the requested amount. Neither Göring nor Hitler, having taught
themselves that war must be won by attack and ignoring air defense for
which their understanding was slender at best, had yet come to perceive the
seriousness of the threat. Kammhuber received much less than requested;
nevertheless by August there were 96 night-fighter stations and 30 000 sig-
nals personnel. Each night-fighter sortie required 140 persons operating
radar sets, working communications links and evaluating the vast amount
of data. With a second summer in Russia demanding extremes of man-
power, ground radar and control centers began receiving large numbers
of women; by January 1943 there were 14 000 Luftnachrichten-Helferinnen
in service [9]. The searchlights that had allowed Helle Nachtjagd to take
over after the planes had left Dunkle Nachtjagd were removed and placed
at the Flak batteries near target centers.
By early 1943 fighter pilots were increasingly critical of the Himmel-
bett system. Part of their complaints had roots in the inadequacy of the
Würzburg-Riese. It had gained the additional range over the small set sim-
ply by increasing the size of the dish, but transmitter and receiver had not
been improved, and its range was inadequate for a proper GCI radar, so
the resultant small size of the Himmelbett cells constrained the maneuvers
of the night fighters.
New tactics began to take a freer form over the objections of Kammhu-
ber, who favored tight control. He was initially bypassed and then re-
placed. Rather than guiding individual fighters to individual targets the
control officers located the bomber stream and guided the fighters to it.
There were enough targets that the flight crews could pick them up with
their Lichtenstein sets, something at which they became skillful. Over
burning cities the single-engine day fighters would fly above the bombers
to find them illuminated from below by the fires they had created. This was
particularly successful, if there was a thin layer of clouds below bombers,
whose outlines were easily seen on this illuminated screen. It was wild for
these fighters, who had to dodge their own Flak and called the tactics Wilde
Sau (wild boar). The tactic of night fighters joining the bomber stream took
by contrast the name of Zahme Sau (tame boar). The Himmelbett system
304
The Great Radar War
remained. It was effective on the retiring bombers that were scattered from
their original streams.
One of the curious outcomes of these new tactics was a call by the
fighter pilots for radar beams that had a wider angular pattern and greater
range. The 50 cm Lichtensteins had functioned well enough when the
Würzburgs had guided them to within a few kilometer of the target, but
now they had to seek their targets with less-accurate guidance. Tele-
funken’s engineers were quick to come up with a satisfactory new design,
Lichtenstein SN 2, which worked in the 3.5 m band; a wider angular spread
and increased range came naturally from increased wavelength. The first
of this type was introduced in July 1943 [10] and became a very effective
piece of equipment. The reader will recall that the main British induce-
ment for pressing the development of microwaves was the idea, seldom
questioned at the time, that they were essential for airborne interception
(AI). The earlier argument that the range of meter-wave AI was limited by
the ground-return of meter waves and thus by the altitude of the fighter
became irrelevant when combat was so high. The air war in 1943 was quite
different from the air war in 1941.
The air war against Germany was consuming a large portion of British
and American resources by January 1943 and took its form from the con-
straints that radar imposed. Radar now guided the bombers at night and
increasingly by day, as the Americans began to realize that not only can one
not see at night but frequently not during the day either. As an offensive
weapon radar was proving weak, but as a defensive weapon it was still
strong. That it was strong by day had been accepted by both combatants by
1941; that it was strong by night was coming as an uncomfortable surprise
to Bomber Command. The losses of the attacking bombers would grow
until some new countermeasure was introduced; the defense would then
master it and losses would grow again. It became a repetition in the sky
of the trench warfare that air power advocates had hoped to avoid. The
RAF decided losses up to 5% in an attack were acceptable; crewmen could
calculate their chances of completing the required 30 missions. For the
defending pilots and their radar operators the arithmetic was less straight-
forward but just as deadly. Just as German and Briton climbed out of their
trenches with steadfast loyalty in 1916, so their sons just as coldly flew
towards their deadly, confusing encounters in 1943.
When Prime Minister and President met at Casablanca in January
1943 the air war had been shaped by events. The conference stated the
purpose of the air offensive to be ‘the progressive destruction and dislo-
cation of the German military, industrial and economic system, and the
undermining of the morale of the German people to a point where their
capacity for armed resistance is fatally weakened’ [11]. It was policy fol-
lowing events. It was policy that formalized a war of attrition. The air
power talk of a ‘knock-out blow’ was forgotten. Another doctrine came
from the Casablanca meeting, almost, it would seem, as an afterthought:
305
unconditional surrender as Germany’s only way out of the war. This doc-
trine, combined with the unremitting destruction from the sky, the threat
from the east and the memories of 1919 stiffened the backbone of the Ger-
man people markedly.
As spring changed into summer of 1943 the Lichtenstein-equipped
night fighters became ever more proficient. Attempts to counter them with
Beaufighters equipped with a homing device for Lichtenstein, called Ser-
rate, was a failure because of the deficiencies of the Beaufighter relative to
the German night fighters, and losses from AA fire continued to mount.
This situation finally forced the introduction of Window. It had been re-
peatedly delayed as a consequence of the fear of how German bombers
might use it in attacking England, a fear that tells one that the superiority
in the air achieved by the Allies had not yet been fully grasped.
The German bomber force of 1943 could not be compared to the
British or American even taken separately, but memories were fresh as
time raced, and British night fighters were busy, nevertheless. German
night raids had never ceased, and the RAF fought each one. When GCI/AI
began to take a heavy toll of the He-111s that came over, the faster Ju-88
and fighter–bomber FW-190 were substituted with almost daily changes
in tactics, sometimes flying high, sometimes low. A particularly difficult
tactic to counter was the intermixing of Me-410s equipped with the radar
Neptun R-2, a 1.8 m AI set of limited production that allowed the detection
of a very closely pursuing fighter [12]. These fast, agile planes, the equal
of the Mosquitoes in speedy acrobatics, drew the night fighters on merry
chases often allowing the bombers to pass unmolested [13]. It all mim-
icked in small format the kind of radar war being fought over the Reich
with each measure generating a countermeasure.
During the Battle of the Ruhr estimates of causes of Bomber Com-
mand losses were 35% to radar-equipped night fighters, 35% to other night
fighters, 20% to radar-controlled Flak and 10% to other Flak [14]. (This
demonstrated the growing efficiency of the night fighters; 87% of the losses
during the 1000-bomber attack on Cologne were attributed to Flak [15].)
It was reasonable to assume that 55% of the losses would be countered by
Window. The debate in council was divided, and Churchill took personal
responsibility for the introduction of the new countermeasure, which was
used against Hamburg for the first time 24/25 July 1943. Damage to the
city was extreme, resulting in the first fire storm, and losses of the bombers
were remarkably light. It was a major defeat for the Luftwaffe. It not only
justified the use of Window but showed the value of H2S when ground
conditions were favorable. Despite the severe damage done, half of the
bombs missed the city. Attacks by night and day followed to complete the
destruction, which deeply affected the German high command.
When the American 8th Air Force began assembling in England under
Major General Ira C Eaker to join the RAF in the attacks on Germany the
two forces shared the common belief that strategic bombing could win the
306
The Great Radar War
war alone but disagreed completely on the next item of dogma, how to do
it. Bomber Command had been forced from its early belief in the ability
of self-defending bombers to penetrate German air defense, which had
left them nothing else but saturation bombing at night using navigation
techniques inadequate for hitting a target much smaller than a city. The
Americans were convinced that the B-17s, which flew higher and were
more heavily armed than the British bombers, could beat off the attacking
fighters during the day by the combined guns of tight formations [16]. By
early 1944 both delusions had been dispelled, and both air forces had been
defeated by the air defense made possible by radar. They were to defeat
the Luftwaffe yet, but the means for doing it, though available from the
beginning, had been rejected earlier and was not ready.
The Americans had no alternative to daylight bombing. Their crews
had no training in night flying and the much lower bomb load of the B-17s
could not compete with the Lancasters, a disadvantage they compensated
by an ability to hit the target instead of scattering explosives and incendi-
aries almost indiscriminately—compensated, that is, if they could see the
target. Disagreements naturally arose between Harris and the Americans
and intensified when Lieutenant General Carl Spaatz became commander
of the US Strategic Air Forces in Europe in December 1943. Spaatz would
concentrate, or such was the plan, on specific parts of the German economy
that were expected to have great effect: petroleum, transport, ball bearings.
Harris rejected these ‘panaceas’ out of hand. He would break the German
will, not their economy.
The Americans encountered an uncomfortable fact: the atmosphere
of northern Europe was nothing like that at the Bombardier School in west
Texas where ‘the skies are not cloudy all day’. The skies over Germany
were not only cloudy, they were richly filled with smoky haze, especially
around industrial targets, and could sometimes be made impenetrable by
smoke generators. They found themselves dependent on radar navigation
and bombing much as Bomber Command.
By October 1942 the overcast of European skies had convinced Eaker
that visual bombing might be the exception rather than the rule and the only
alternative was the blind-bombing techniques of the RAF. Oboe and H2S
were in advanced prototype stages, and the Americans asked their hosts
for some of each. This was not a welcome request. Production of both was
rather low and would be for a time; besides that, there was reluctance to
put Oboe into anything but the Mosquitos in order to delay its inevitable
capture, but in March Air Chief Marshal Portal agreed to Eaker’s request
for eight H2S units for American pathfinder units. The Americans quickly
accepted Gee as the standard for navigation and ordered 2000 sets for 1943
for which British production was adequate [17].
At about this time Assistant Secretary of War Robert A Lovett and
radar advisor David T Griggs were studying British blind-bombing tech-
niques. Their discussions with the upper command levels of the 8th Air
307
Force produced a near commitment based on Grigg’s knowledge of things

at Rad Lab for American H2S equipment, available tentatively in Septem-
ber 1943 [18].
Rad Lab elected to produce a 3 cm set rather than an improved du-
plicate of H2S, as experimental work at that wavelength was advanced
and an improved resolution would accrue. Twelve B-17s with this equip-
ment, called H2X [19] but officially carrying one of the new Army–Navy
nomenclatures, AN/APS-15, arrived for the 482nd Pathfinder Group at
Alconbury, located on a Roman Road 25 km northwest of Cambridge. The
new radar was not long out of the experimental stages and was accompa-
nied by a significant number of design and maintenance personnel.
The 8th Air Force had made its first use of H2S on 27 September in
an attack on Emden, a principal port since the severe damage inflicted
on Hamburg two months before. They made their first use of H2X on 3
November at Wilhelmshaven. By the standards of the method the attacks
were considered successful [20].
Both Americans and British strove to improve the performance of H2S
and H2X. Much depended on the rapid identification by the navigator of
landmarks from the ‘map’ that the PPI presented. Comparison with a
topographic map seldom proved particularly helpful and more often was
confusing. Operators used photographs of PPI displays of English towns
and cities in their training to develop the near-intuitive skills required to
trace their course deep into Germany, and photographs of PPI displays
from raids were pieced together to make radar maps. In mid-1944, when
the fabulous SCR-584 became available, they could practice on English
towns and cities. The 584 would track the test aircraft accurately in three
dimensions and record when the bombardier said he would have made
his release, allowing the error to be determined [21].
An analysis in September 1944 of relative accuracies for various
bombing methods showed Oboe and visual aiming to be about the same;
H2X was slightly better than H2S, hardly significant. Oboe–visual had a
typical error of 350 m, H2S–H2X 1800 m. Transformed into areas this made
Oboe–visual approximately 25 times more effective [22].
As Rad Lab’s equipment began to come into wide use by Ameri-
can forces there was a clear need for a British Branch of the Radiation
Laboratory, and on 9 September 1943 Lee DuBridge submitted a memo-
randum to Vannevar Bush requesting its establishment. In less than two
weeks L C Marshall had arrived in England to establish the outpost. The
choice of location was between the TRE labs at Malvern and a site near
the US Air Forces Headquarters. Technical concerns won over tactical, so
Malvern was selected. Douglas H Ewing became the director, administra-
tively heading one of Rad Lab’s Divisions. This group formed extremely
productive and cordial ties with the TRE personnel. They helped adapt
American equipment for the RAF and British equipment for the Army Air
Forces. They contributed greatly to the development of the second, mi-
308
The Great Radar War
crowave generation of Oboe [23].

While matters progressed smoothly in outfitting the two air pow-
ers with the latest in radar, things were going badly in the skies over
Germany—day and night. The first American attacks were a form of ad-
vanced training and were on targets in the occupied countries, which were
not nearly so heavily defended as those in the Reich and which were often
within range of escorting fighters, something that obviously made a lot of
difference. As the 8th Air Force bombed targets within Germany beyond
the range of escorts, losses assumed rates impossible to sustain. There
should have been no questioning the fallacy of the doctrine of penetration
by tight formations of self-defending bombers after the attacks on Schwe-
infurt and Regensburg of 17 August 1943 in which 60 out of 315 planes
attacking were lost, but Eaker was not yet ready to accept defeat. After
two months devoted to peripheral targets and re-thinking tactics a second
attack on Schweinfurt was attempted on 14 October in which 60 out of 229
were lost [24].
Bomber Command watched Eaker’s dilemma with the sympathetic
understanding of a teacher who had correctly predicted the consequence
of the disregard by his pupil of proffered guidance. But things were not
going well in the night either. The introduction of Window at Hamburg had
thrown the defense into serious confusion, and another success in the 17/18
August attack on Peenemünde, the flying bomb and rocket experimental
station, introduced an optimism that brought from Harris’s pen a statement
that was to be quoted in nearly every book written on the European air
war: ‘We can wreck Berlin from end to end if the USAAF will come in on
it. It will cost us between 400–500 aircraft. It will cost Germany the war’.
Whatever possessed Harris to think the Americans would join him after
their recent experiences is difficult to fathom, but Bomber Command had
reached a peak in its strength, and he set out alone on the Battle of Berlin,
a continuing action that began in November, went through March 1944
and was extended to many other cities. It did not cost Germany the war
and almost wrecked Bomber Command, but it gave a decided boost to the
morale of the British public.
The night fighters had been equipped with 3.5 m Lichtensteins that
were unaffected by Window and were well suited for individual hunting
on the long road between England and the deep targets. They acquired
upward-firing cannon that allowed them to fire on the bomber from an
unexpected quarter. They preserved communication between controllers
and pilots in a continual radio battle of voices, noise and music. Flak gun-
ners found that the bombers were not always enveloped in aluminum foil,
another consequence of the long passages. Ground radar improved con-
tinually and had centralized control by February 1944 [25]. There seemed
to be some kind of counter for every new countermeasure, even if not per-
fect. Losses grew beyond the ‘acceptable’ 5%, finally becoming 12% during
a March attack on Nuremberg. In April 1944 Harris terminated the battle
309
because of ‘casualty rates which could not in the long run be sustained’.
The results may have been far short of what had been expected but they
were disastrous enough for the Berliners [26] for whom the worst was yet
to come.
Crew morale in the bomber forces of both nations had been dealt
hard blows. Pathfinders reported far too few bombs hitting their markers;
planes landing in neutral countries had not always lost their way or been
seriously damaged. It was just as well that the bombers had to prepare
for the invasion, which required bombing targets in France that had much
lighter defense. The end of the Battle of Berlin came at the beginning of a
new phase of the air war—the introduction of the long-range fighter that
led to the absolute defeat of the Luftwaffe.
6.4. ARBEITSGEMEINSCHAFT ROTTERDAM

The extraction by technical personnel of the remains of an H2S from a
Stirling bomber downed near Rotterdam on 3 February 1943 sent a shock
wave through the German radar community, one that triggered a number
of people to say ‘I told you so’. Kühnhold together with Dr Anton Röhrl
had continued microwave research into the first two years of the war, but as
the heavy requirements of an unexpected air defense for the Reich became
apparent the sparse engineering resources went to designs having better
prognoses of success. After the 1939 failure to detect CH with airship
espionage, Martini had expanded the listening stations along the coast
both in number and in bandwidth, but none had microwave capability
and consequently had missed observing the 10 cm radiation emitted by
the coast defense stations across the Channel that had the NT 271 sets (also
called CD mark IV). Thus H2S realized Martini’s dire fears, painted all the
blacker because of his inability to secure the research he had favored.
Telefunken had continued projects for the development of 26 and
5 cm tubes, although without noticeable success. There remained those
who continued to dismiss the value of microwaves for radar because of
predictions that far too little of the incident radiation would be reflected
back to the radar set, being mirrored off to the side instead. These were
old, widely held arguments that rested on simplistic theory and idealized
experiment: reflections of such wavelengths from a flat plane did show
specular reflection—but airframes had far more complicated surfaces for
which the computational capabilities of the time were inadequate.
On 22 November 1942 Karl Rottgardt, Director of Telefunken,
presided over a demonstration of the latest Würzburg equipment and as-
serted that 50 cm radar was, given the shortage of research personnel, the
best answer for gun-laying radar. Shortly thereafter he terminated the com-
pany’s 26 cm research, and less than a month later the Technical Bureau
of the Air Ministry, over the objections of Martini, ended all microwave
investigations because of the shortage of technical specialists, this despite
310
The Great Radar War
Martini’s and Hans Plendl’s success in having a few thousand [1] electronic
specialists returned to industry from Air Signals [2].
At just the moment Germany had decided that microwave research
was no longer worth the effort, Bomber Command introduced H2S in
flights over the Reich. At just the moment the Wehrmacht realized a cru-
cial defeat at Stalingrad, the radar men realized a crucial defeat in the
laboratory. The extracted H2S equipment was severely damaged, but the
function of the magnetron was recognized, indeed all essential design el-
ements were quickly understood. The first question was: what was its
purpose?
Not that uses failed to spring to mind now that the British had sum-
marily disposed of specular reflection. Gen Martini, who had acquired
in December 1941 the post of Special Commissioner for Radar in addition
to his duties as Chief of Air Signals, understood the gravity of the situ-
ation even without knowing the exact purpose of the new weapon. He
responded quickly to the request of Leo Brandt, who had replaced Runge
by then as Chief of Development at Telefunken, to form a special commit-
tee to exploit the Rotterdam-Gerät, as it came to be called. Plendl, by this
time Staatsrat and Plenipotentiary of High-Frequency Research, concurred
in this. The Nazi state produced organizations that tended to overlap and
that prove difficult for an historian to untangle, and electronics was not
immune to this. Plendl and Martini shared responsibilities with General-
major Erich Fellgiebel, Supervisor of Technical Communications [3].
The committee, Arbeitsgemeinschaft-Rotterdam [4], first met on 23
February 1943 at Telefunken in Berlin with Brandt as chairman. Brandt had
already demonstrated qualities of organization and leadership that were
hoped would turn the committee into a useful mechanism. Twelve persons
attended the first meeting, two of whom, Plendl and Runge, had figured
importantly in radar but were to be removed from active participation
within a year. Two companies, Telefunken and Lorenz, were represented
as was the Physikalisch-technische Reichsanstalt, the German standards
laboratory, which like its British and American counterparts had excellent
general scientific competence. Wehrmacht and Air Ministry representa-
tives completed the group. Representation expanded in subsequent meet-
ings, 43 attending by the end of the year.
GEMA was not represented at the first meeting and took essentially
no part. At that time GEMA was heavily involved in providing early warn-
ing radar for which microwaves had little to offer. That Rotterdam was
obviously going to be dominated by Telefunken provided no incentives
for GEMA to participate. Less easy to understand is the complete absence
from the meetings of Kühnhold and Röhol, who had conducted microwave
research at NVK longer and more tenaciously than any of the others. Also
absent was Hans Hollmann, who had written the authoritative text on mi-
crowaves, used to advantage by the British [5]. He had withdrawn from
military work after observing the first years of Hitler’s Third Reich.
311
The operation of H2S was thought well enough understood at the

first meeting for them to order six similar devices to be built at Telefunken
for experiment. In the meantime Lorenz, Pintsch, Blaupunkt, Telefunken
and the Reichsanstalt pulled out all of their old experimental microwave
equipment, generally split-anode magnetrons, to start laboratory work.
Passive receivers to detect the approach of H2S received the highest
priority, and it was here that engineers met what was to be the main ob-
stacle for the whole project—crystal diodes! Two receiver designs were
proposed: a simple, low-gain device called Naxos and a heterodyne, high-
gain set called Korfu, both of which required diodes. Naxos, like nearly
all German radar equipment, was to appear in various forms and sub-
variations. Naxos Z was primarily a homing device for night fighters that
would use H2S as a beacon for locating the formations, although not in-
dividual aircraft; Naxos U was to warn U-boats of the approach of patrol
planes [6]. The Reichsanstalt succeeded in making crystal diodes, but high
production proved elusive and vibration caused many to fail. Naxos in
all its forms was for several months a very fragile and disappointing thing
and would soon be at the center of serious technical confusion in the Bay
of Biscay. It owed much of its fragility to a ceramic band-pass filter at the
front end, which often broke from shock, a defect not often recognized at
the time that gave Naxos an evil name. The cure was simple: remove the
filter. This also made Naxos capable of detecting 3 cm radiation. Korfu
used a magnetron as local oscillator, as it was initially easier to produce
than klystrons.
Next to the magnetron the most interesting item pulled out of that
wrecked Stirling was some co-axial cable. German high-frequency cables
had used Opanol (polyisobutylene) or for the highest frequencies cup-
shaped ceramics linked together in a daisy chain that was flexible but
fragile [7], so finding a plastic insulator used for these highest frequencies
was startling. Analysis showed it to be I G Farben’s Lupolen H, known to
the Allies as polyethylene, whose splendid electrical properties had been
overlooked3 . I G Farben promised an initial production of 100 kg per
month [8].
Interrogation of prisoners established within a few weeks that H2S
was a navigation and blind-bombing device [9], which added methods for
jamming and camouflage to the fund of problems for which solutions had
to be found. That the PPI gave a maplike representation of the ground
came as a surprise. The first thoughts had been that it would pick up a
few readily identifiable targets. Its use by attacking bombers was clear.
Much less clear was why Brandt continued to put so much effort into
duplicating a device of little value for Germany, as the Luftwaffe was in no
position to carry out night strategic bombing, while Flak had serious need
of microwave radar for gun laying because of the effects of Window.
312
The Great Radar War
The principles of H2S design may have been appreciated early, but the
construction of a functioning device did not come quickly, despite strong
support. The magnetron and the duplexing tube were the easiest com-
ponents to fabricate, and by May the company Sanitas produced the first
German production equivalents: the magnetron LMS10 and the TR switch
(Sperrohr) LG76. By the middle of June a complete Rotterdam made of
British and German components was completed after a delay imposed by
air-attack damage to the Telefunken shops where the work was being car-
ried out. It was mounted in an He-111 at the Rechlin Laboratories and
Development Establishment and gave a reasonable representation of the
ground when flown at altitudes below 6000 m, not an outstanding perfor-
mance but at least a start. Telefunken’s production of six Rotterdam sets
continually fell behind the expected schedule with the last delivered in
December 1943 [10]. Plans were made to extend the new techniques with
improved designs of their own in a series of equipment called Berlin. Mag-
netrons for other wavelengths were obviously to be tried, and LMS11 (5.8
cm, 15 kW) became available in December [11]. During these months tech-
nicians searched wrecked bombers not just for equipment disclosing new
techniques but for components to use. They especially valued the nearly
indestructible magnetron magnets because the Deutsche Edelstahlwerke,
the normal source of permanent magnets, had been seriously damaged
by air attack. Electromagnets with the inconvenience of another power
supply became the norm.
The wide use of countermeasures undermined a human element of
German radar. The high engineering quality of German equipment had al-
lowed station personnel to be sub-standard, as operation and maintenance
were simple. Now there was a need for a much higher level of ability to
work through foils and noise with the rapidly designed countermeasure
equipment. The question was where to obtain them. In addition to comb-
ing people from the military, Hans Plendl saw three sources: teenage boys
who were radio hobbyists and who were being inducted into AA defense
as Flakhelfer, prisoners with technical knowledge who were now residing
in concentration camps and technically trained men in the occupied coun-
tries. His actions hint that he not only saw these people as a resource for
wartime radar but as a resource for the future that needed preservation.
Britain and America had seen very early the great value of radio am-
ateurs, whose knowledge and enthusiasm permeated whatever quarter
of the electronic war they fought. Amateurs were not forbidden in Ger-
many, but unrestricted communication was outside the narrow limits of
a society in which everything not explicitly allowed was forbidden [12].
But there were still hobbyists who built and repaired radios, and Plendl
as Plenipotentiary of High-Frequency Research intended to draw on this
group. He had first-hand knowledge—his son was in this group. In the
summer of 1943 Flakhelfer were allowed to compete in examinations for
positions in a course of extensive radar training to be held at a camp on the
313
Stegskopf in the Westerwald. On 23 October 1943 the first of four classes of

‘Stegskopfer’ began their training as radar mechanics and operators [13].
That scientists and electrical engineers were being removed from so-
ciety by the Gestapo could hardly be kept a secret from men such as Plendl
and Martini, despite the terrible individual isolation imposed by a police
state. One of those known to Plendl was Dr Hans Mayer, Director of the
Central Laboratory of Siemens und Halske, who had been denounced by
the maid of a neighbor for unguarded talk [14]. Following his arrest in
August 1943 Plendl was able to have a high-frequency research institute
formed at the Dachau concentration camp with Mayer in charge of about
25 technically trained prisoners [15]. There is no evidence of it having pro-
duced significant high-frequency research, but it probably helped Mayer
to survive [16]. Had it been known that he had written the Oslo Report,
he would have never even reached Dachau.
By securing contract work for engineers in the occupied territories
Plendl obtained tangible research results, and those so engaged were saved
from forced labor in Germany [17].
Whether the top levels of the Nazi state were favorably impressed
by these measures is impossible to say, but Plendl was dismissed from
his post in March 1944 for disregard of an order of the Führer: ‘No one,
no office or officer, may learn of a secret matter unless this is absolutely
necessary in the line of duty’ [18]. The reason behind his dismissal has been
given by Plendl and is instructive about the attitudes toward research and
development in the Reich. Plendl had naturally become very concerned
about the effect of the air attacks, about which he was, of course, extremely
well informed. He conceived the idea of an improved shell for AA guns
and assembled a small group of ballistic experts to work up the idea. Initial
tests were favorable, and he took the design with the results of the tests
to the Chief of Flak, Generaloberst Weise, whose reaction was that Flak
ammunition was his affair and that Plendl should concern himself with
radio. The resulting discussion was not marked by restraint, and Hitler
dismissed Plendl from his post.
Plendl was not alone in trying to protect colleagues from a regime that
combined malevolence toward individuals and institutions with indiffer-
ence to the future well-being of the people. Hans Hollmann was able to
prevent deportation of people at the Kammerling-Onnes-Institute in Ley-
den by having them awarded a contract for investigating photographic
film development at low temperature. The SS saw no distinction between
temperatures suitable for photography and those near absolute zero, the
Institute’s specialty [19].
The terrible pounding Germany was taking from the air was creating
uncomfortable times for the radar men, aggravated by serious mistrust of
the whole scientist–engineer caste by the Nazis. There was also a growing
group of young, Nazi-indoctrinated engineers at the Air Ministry Technical
Bureau, party members, some members of the SS, who had been express-
314
The Great Radar War
ing strong criticism against Martini and Telefunken, which they took to be
under the influence of its earlier, Jewish leadership. The result was Himm-
ler’s intent in early 1944 to indict for treason Martini, some of Telefunken’s
management and Vizeadmiral Erhard Maertens, the former Chief of Naval
Communications, who had been sacked in May 1943 as a consequence of
the radar debacle in the Bay of Biscay4 . Martini did not lack defenders,
and Göring, not that he did not share in the enmity addressed to these
parties, stopped the process, partly because he realized the weak basis of
the charges and partly to keep Himmler out of his territory [20].
These accusations against intermediate management levels in
Telefunken—they were certainly not directed at Rottgardt—arose from the
atmosphere that the company’s founder Graf Georg Arco had instilled and
that were expressed in the acknowledged friendship of many on the staff
for Emil Mayer, the Jewish director who had been forced out in 1933 with-
out even a ceremonial farewell, as well as for others who had had to leave.
The accusations must have included Runge, who by 1944 no longer had
specific responsibilities and who left the company and war work toward
the end of the year. An incident had clarified his attitude on this key
Nazi point. After Otto Böhm left Telefunken, Runge kept a photograph of
him in his office. One day he found it on the floor with ‘Böhm—Jew!—
Bloodsucker!’ written on the back. Runge then hung the reversed picture
showing the inscription. Soon he was visited by a colleague he respected
who told him things were getting out of hand and that he should take the
picture down. Runge later wrote ‘and that is what happened’ [21].
The action intended against Telefunken was in fact out of charac-
ter, for the Nazi regime’s evaluation of political reliability was the inverse
of that of the United States, which screened industrial research workers
with care but took no interest in university faculties. Political purity was
highly valued for Germany’s universities, party membership becoming a
requirement for new academic appointments, but scientists and engineers
who had had during their student years problems owing to membership in
organizations hateful to the government generally found a desired seclu-
sion in industry [22]. Göring’s quashing the whole thing fits with his use
of jewish scientists at his Rechlin research institute and protecting their
families: ‘Wer Jude ist, das bestimme ich!’ (I determine who is a Jew!) [23].
While Brandt’s group attempted to understand and design 10 cm
radar equipment and passive detectors, those engaged in air defense had
immediate and pressing wants. Alexander Dahl, who had grasped the
significance of the isolated Mosquito flights over the Ruhr in January 1943,
had received a Naxos receiver for observing the approach of the attack-
ing formations by summer but found its directional accuracy and range
so poor that it gave no really useful information. He discussed the mat-
ter with Feldwebel (Master Sergeant) Robert Kaufmann, another former
315
radio amateur, and they quickly came up with an excellent but illegal so-
lution: combine the Naxos and a Würzburg. Such field modifications
were definitely not allowed, nor were violations of these regulations taken
lightly, especially when they involved radar, which was entangled in near
paranoic secrecy, but these men were defending their homes in the literal
meaning of the word. Dahl’s apartment in Wuppertal had been destroyed
in September, although his wife had escaped injury.
They saw that what was needed was a high-gain antenna, such as a
Würzburg dish, but also saw that the Naxos, although small, could not be
accommodated at the focus of the dish, so they decided to extract the crystal
diode from the receiver, incorporate it into a dipole for 10 cm radiation
and place this assembly at the focus of the dish, which they had been able
to ‘acquire’ because of a local Flak battery’s recently developed Window
blindness. The resultant signal, if it proved to be strong enough, would be
audible at the pulse repetition frequency of the H2S set, coming in bursts
corresponding to the rotation speed of the sweeping antenna. Nothing but
a high-gain audio amplifier following the diode was required. For testing,
Dahl made a 10 cm sparking dipole that he found he could pick up easily
at 1000 m. On 23 September they set up on a high point and observed
Mosquitos over the Zuider Zee with directional accuracy of 1◦ and a range
that soon proved to be limited only by the curvature of the Earth.
The next step was obvious. Similar equipment, soon to be called
Naxburgs, had to be established at widely spaced locations to allow tri-
angulation. This required cooperation from air defense and this required
confession of the destruction of a Würzburg and a Naxos. The disclosure
made the next layer of authority uncomfortable but common sense tri-
umphed over bureaucracy. Kaufmann and four other amateurs set up a
small factory in a barracks. By 16 October a second station allowed the first
triangulations, and the network soon joined the honest ranks of air defense
[24]. Air Signals installed a chain of Naxburg observation posts stretching
from the northern tip of Jutland to the Swiss border. It provided Y-Dienst
an extremely reliable early warning system against the Mosquitos, which
gave weak radar echoes and whose approach was the first indication of an
attack [25]. Naxburg was without question the simplest and most elegant
piece of equipment used in the radar war, and probably the cheapest. It
continued to the end of the war to give reliable positions of the pathfinders
and may have been the most effective use to which the Germans put their
knowledge of microwaves.
Having made magnetrons of shorter wavelength than Rotterdam’s
10 cm, the German engineers were hardly surprised when a British H2X
and an American AN/APS-15 came into their hands at the end of 1943.
They called the British set Rotterdam-X; there is no evidence that this name
came from any knowledge that 3 cm equipment was classified as X-band
by the Allies, but for whatever reasons it was aptly named. The Amer-
ican set acquired the name of the Dutch village of Meddo where it was
316
The Great Radar War
found. Knowledge that the enemy was using X-band radar was immedi-
ately important for the need of suitable passive receivers, which was not
an additional problem once the production of crystal diodes was in hand.
By the end of 1943 the Germans understood the basic elements of
microwave radar. The production of a few copies of H2S diverted engi-
neering skills from other development, possibly through uncertainties as
to how the magnetron might best be employed in satisfying Germany’s
needs. After a failed attempt Telefunken turned the design of a gun-laying
set to A-E Hoffmann-Heyden. They succeeded in picking up a He-111
at 8 km with an 80 cm paraboloid, then changed to the 300 cm dish of a
Mannheim set and achieved a range of 30 km. Their experiments against
Window were so successful that they recommended microwaves for ex-
tensive use. The gun-laying set took the name of Rotterheim and finally
Marbach, which eventually saw a successful use in the defense of Hanover
and Hamburg before the war ended [26].
An advanced AA radar system, Egerland, combined the Marbach
with a 10 cm panoramic search radar, Kulmbach. This set used a cylindri-
cal parabolic reflector fed from a slotted waveguide that, owing to phase
considerations, produced a beam oriented 30◦ relative to the cylinder axis.
Peak power was 10 to 15 kW with ranges of 20 to 30 km [27]. Introduc-
tion of this system in early 1944 in sufficient numbers might have changed
the course of the air war, but only two were deployed before the end. By
January 1945 there were prototypes of airborne sets of advanced designs,
but production, if it could have been fulfilled, was scheduled for spring
1945. A few Berlin N1a sets were successfully used by night fighters in
March [28].
In May 1944 the decision-making functions of Arbeitsgemeinschaft
Rotterdam were taken over by the Sonderkommision für Funkmesstech-
nik (Special Commissioner for Radar), a new sub-branch of Albert Speer’s
War Production Ministry and headed by Karl Rottgardt, Director of Tele-
funken [29]. The Rotterdam Group held meetings until 1 September, but
the function became one of providing instruction to high-level persons
through lectures on various radar topics, and the number of participants
grew markedly.
6.5. THE DESTRUCTION OF GERMAN CITIES COMPLETED

‘London can take it!’ Churchill used this exultant phrase as the name of a
chapter in Their Finest Hour wherein he described the Blitz, the unrestricted
bombing of Britain during 1940–1941. Those words must have expressed
extreme relief at the disproof of the air power doctrine that had crippled
western politicians for two decades, that the bombardment of cities would
produce panic and uncontrollable pressures to sue for peace. It had not
happened. Just the opposite. The people wanted the Air Force to ‘give it
back to ’em!’. But this elemental fact had made remarkably little impression
on British strategy and certainly none at all on Arthur Harris, leader of
317
Bomber Command, for London was not alone in being able to take it, so
could Berlin and Hamburg, which would provide no grounds for the Prime
Minister’s jubilation. On the other hand, it was irrelevant whether Britain’s
top levels believed the German civilians could take it or not, because the
imperatives of war required that the RAF bomb the enemy’s cities. If not
by day, then by night; if not precisely, then by saturation; if not effectively,
then as a gesture of defiance.
The air attacks on German cities grew in intensity while German pro-
duction grew in mocking proportion. This was primarily because German
production did not go on a war footing until after the defeat of the Wehrma-
cht before Moscow in December 1941 and Hitler had declared war on the
United States, but it also grew because the bombings had remarkably little
effect on industrial output. Those bombed out returned to their ruined
homes and built some kind of shelter in them. City services returned and
work went on. Heavy machinery in many smashed factories generally
functioned after minor repairs even though the buildings were in ruins.
Provisional shelter was provided, and work began again. None of this
showed on the aerial photographs [1].
The bombing hurt the civilian population well enough, but it stiff-
ened their resistance, just as the same treatment had affected their English
cousins. As the attacks became ever more terrible and the defeats at Stal-
ingrad and Kursk indicated the war was lost, a fatalistic determination
to fight to the end arose. Many gave thought to the meaning of the terror
from the skies, the unconditional surrender ultimatum from the Casablanca
conference, the menace from the Soviet Union and the gradually growing
knowledge of Nazi crimes for which the whole country would be called to
account. When these thoughts were compounded with memories of the
terrible post-war year of 1919 they yielded the grim evaluation: ‘Enjoy the
war; the peace is going to be tough’. Gestapo terror stiffened more than a
few backbones.
In early 1944 the Luftwaffe could mark a victory over the attackers.
They had shown the Americans the folly of their belief in the ability of for-
mations of Flying Fortresses to defend themselves against fighters and to
bomb accurately, and they had stripped the cover of darkness from Bomber
Command through the ingenuity and industriousness of their radar engi-
neers and fliers. It was a victory that did not, could not hold against the
thousands of aircraft coming against them, and the turn came with the in-
troduction of long-range fighters to protect the bombers in mid-1944. When
the Americans returned with fighter protection to deep daytime flights into
the Reich, the defenders rose to meet them and in so doing suffered the
attrition that wore them down. Replacement aircraft came along, but re-
placement pilots were more difficult to produce; there was ever less time,
less fuel, fewer instructors. The daytime attacks became so serious that the
specialized night fighters were thrown into the struggle with consequent
loss of these skilled men.
318
The Great Radar War
The introduction of the long-range day fighters had to be matched

by long-range night fighters as well. The Luftwaffe tactics of Zahme Sau
became particularly effective after the introduction of Lichtenstein SN-2,
the 3.5 m set with wide beams and a relatively long range. For Zahme Sau
the fighter directors would use ground radar and Y-Dienst triangulations to
guide the fighters to the bomber stream, where they followed the bombers
and located targets by airborne radar. Extreme efforts were made to block
or confuse the directions the controllers gave over the radio in an electronic
war of its own that changed from week to week, but the fighters were
equipped with Naxos-Z that allowed them to find the bomber streams
from the H2S emissions and without controller direction. They also had
the passive receiver Flensburg that homed on a bomber’s tail-warning
radar Monica. Adding these two devices to the Lichtenstein resulted in a
deadly combination.
Beaufighters proved too short ranged for countering the night fight-
ers, so Mosquitoes were used instead. They carried the 1.5 m AI mark
IV, which was replaced by the 10 cm AI mark X, the SCR-720, as use of it
over Germany was allowed [2]. They also carried a version of the passive
receiver Serrate tuned to the new 3.5 m band that homed on the Lichten-
stein. Knowledge of this had come about through a bit of ill luck for a
German pilot. He had flown in July 1944 his Ju-88 on a reciprocal course
and landed his machine with a Lichtenstein SN-2 and a Flensburg on an
RAF airfield, thereby ending the mystery of the new airborne radar in a
most direct manner and further alerting the RAF to the dangers of using
Monica [3]. Another device, Perfectos, actuated the German IFF thereby
obtaining range as well as direction. Knowledge or suspicion soon re-
sulted in the IFFs being turned off, putting an end to that advantage but
complicating the task of the German controllers.
When Bomber Command began intensive attacks on the Reich after
the invasion, they imposed radio silence on aircraft communication, radars
and jammers until close enough to the targets to make them useful. Each
attack was conducted with an intricate pattern of feints to steer the night
fighters away from the main force, sometimes by nuisance raids preceding
the main attack, sometimes by Mosquitoes pretending to mark the path to
a city that was not bombed, sometimes by using the large force of training
bombers that turned back at the last moment [4]. That Britain had sufficient
bombers to do this tells much about the course that the air war was taking.
Bomber Command also had an increasing number of B-17s flown by
100 Group whose only function was jamming with all of the power and
finesse that could be brought to bear. Their equipment allowed the frequen-
cies of German radar and communications channels to be determined and
disrupted immediately [5].
When deep daylight bombing became effective in late 1944 and 1945,
the destruction of Germany’s synthetic gasoline plants—the only sources
of fuel after the loss of Romania—became a reality, and the plight of the
319
Luftwaffe became nearly hopeless. Fighters rose to defend their homeland

to the very end, but most were flown by inexperienced pilots whose spirit
and determination were no match for the well trained and well equipped
Allies. The introduction of a modicum of precision bombing not only
brought down the synthetic oil industry, it also brought the railways to
collapse. The result of these two ‘panaceas’ was the end of significant
industrial production.
It came to pass during the last months of the war that radar began to be
less important for the defenders. Not that there were not plenty of ground
stations. In the territory that one might describe as the core of Germany’s
air defense in 1944–1945, including the Benelux countries, Denmark, Aus-
tria, Czechoslovakia, western Hungary and western Poland, were approx-
imately 65 first- and 126 second-class stations. First-class stations had one
or two Wassermänner with some coastal stations substituting a Mammut
and some inland stations a Jagdschloss in place of a Wassermann, two
Freyas and one or two Würzburg-Giants; a second-class station had one or
two Freyas and one or two Würzburg-Giants [6]. Germany’s air-warning
and fighter-control systems functioned with ever growing proficiency, but
with fewer and fewer fighters its function came to be more and more to alert
Flak and the cities. Flak became increasingly strong as guns were pulled
back from the occupied territories retaken by the Allies and accounted for
the major proportion of Allied losses. Gunners learned to shoot through
jamming and Window and obtained electronic aids for doing so.
Flak suffered from the combination of Window and Carpet, which
became extremely strong from the US forces toward the end of 1944, re-
ducing the effectiveness of radar-controlled gunfire to about a quarter of
what it was without these countermeasures. Frequently batteries were or-
dered to resort to predicted barrage fire when controlled fire was no longer
possible. This reason for filling the sky with undirected bursts was justified
much as had similar expedient on the other side during the Battle of Britain
and the Blitz: (1) make US air crews think their countermeasures were not
working, (2) maintain civilian and military morale through sound effects
and fireworks and (3) to achieve an occasional lucky hit. When caught
in good visibility Flak was up to its best accuracy, which was somewhat
better than their purely radar-controlled fire [7]. Jamming the air-warning
and fighter-control equipment was much less effective than jamming Flak
because of the redundancy of the information available to the defenders.
Keeping radar going became more difficult. Stations were manned
by what was left after numerous ‘combing’ operations to procure men for
the front. Old men and young women operated the sets, often developing
commendable skills, but maintenance personnel for faults that could not
be removed by replacing modules were in short supply. The Stegskopfer,
the teenaged radio enthusiasts who had received crash courses in radar,
helped bring ailing equipment back to service with the most difficult re-
pairs done by senior mechanics who raced from station to station by train,
320
The Great Radar War
bicycle and hitch-hiking with lodging secured as they might. Feldwebel

Kaufmann, the co-inventor of the Naxburg, found himself during the last
months setting up a jamming station and keeping it and others operating.
On occasion he obtained meals by repairing the radios of the farmers with
whom he stayed. Travel in those days entailed a high probability of being
strafed by long-range fighters, who shot at anything on the ground that
moved, if there were no Luftwaffe planes to fight. Kaufmann had his tool
box shot to pieces from this source [8]. Not the least problem such men
had to face was the danger of the Feldgendarmerie, the vicious military
police who pursued deserters, real and imagined, whom they left hanging
with a sardonic sign to deter future malefaction.
By late 1944 the Flak units defending the Reich had taken on an irreg-
ular character. Personnel were 45% foreigners, teenaged boys and young
women (Flakhelfer and Flakhelferinnen) with assistance from the Labor
Service (Reichs Arbeitsdienst). There were, however, enough experienced
men to keep the Allied fliers unaware of the loss of elite status [9].
German emphasis went toward the radio-navigational technique
Gee. The most effective station for this was located on the Feldberg in
the Taunus Mountains about 25 km north of Frankfurt am Main, which
utilized a pre-war television station. It also jammed meter-wave Oboe,
but with time Oboe worked on microwaves and the meter-wave transmis-
sions were only a cover. The jammers used Gee-like pulses to confuse the
receivers, but these transmissions disclosed the position of the station [10],
and photo reconnaissance planes confirmed it. In a deviation from the nor-
mal, soft procedures of radio countermeasures, the Feldberg station was
attacked by a force of P-47 fighters and put out of action on 2 March 1945
[11].
Jamming radio navigation equipment that operated on meter waves
was fairly straightforward, but microwave Oboe proved essentially un-
jammable. Indeed jamming microwaves in any application proved diffi-
cult and was never satisfactorily accomplished by the defenders. It was
the fundamental problem of placing the jammer in the narrow beam of a
microwave set. For H2S an alternative was to illuminate the target area
being observed by the bomber with radiation from a ground jammer sta-
tion, thereby scattering extraneous radiation into the H2S receiver. For a
while large spark transmitters with their great power looked hopeful, and
Professor E Marx at Braunschweig High Voltage Institute, whose tremen-
dous spark demonstrations have fascinated—and deafened—visitors to
Deutsches Museum in Munich for years, worked on this method. It failed
because the broad spectrum of Hertz’s oscillator injected too little energy
into the relatively narrow wavelength band of the H2S set.
Professor P. Gorcke worked with Telefunken to use the most powerful
magnetron available, the LMS100 that gave 100 kW peak power, and might
thereby circumvent the spark transmitter problems by concentrating the
radiation in the H2S bandwidth. An He-111 with a German-made H2S unit
321
allowed testing of their equipment, which was able to cover the PPI with
numerous little spots, but these dubious results came too late to have any
effect on the bombers [12].
It was not until the first copy of H2S, Rotterdam No 1, had been tested
in flight that the possibilities of radar camouflage could be examined in
any serious way. The first flight in June 1943 showed the contrasts between
water, towns and flat land on the PPI scope, which automatically presented
the question of altering the reflecting conditions sufficiently to confuse the
H2S operator. By September a number of experiments had been tried,
both to determine what might weaken the return signal and what might
enhance it. They learned that flat land appeared darker than plowed fields
and that meter-sized corner reflectors fashioned from sheet metal were the
best means for enhancing a reflected signal; it was further determined that
mounted on floats these might prevent lakes and streams from appearing
dark. Spacing them about 150 m apart did remove the darkness and hide
the distinguishing features of bodies of water, but they showed up as points
of light, if the radar receiver sensitivity were reduced [13]. It is difficult to
say what radar camouflage contributed or even in whose favor.
The extent to which German electronics was devoted to countermea-
sures toward the end of the war was estimated by Professor Abraham Esau,
who replaced Plendl as Plenipotentiary for High-Frequency Research, as
90%, a total of 4000. This prevented any serious development of 10 cm
capabilities [14].
If radar became less important to the Germans, it became more im-
portant to the Americans and in a way that completely turned their air
policy upside down. As they initiated deep daylight raids again the US
8th and 15th Air Forces found that visual bombing was the rarity not the
rule. Only four clear days in a month was not unusual during fall and win-
ter. This produced a remarkable change in American bombing doctrine:
the vaunted precision bombing allowed by the Norden bombsight, which
had held fast the minds of the Army Air Corps during its entire prepa-
ration for war, was used only as occasional opportunity allowed. Radar
bombing was to be the rule, as it had been for Bomber Command, and the
inherent inaccuracy of H2X meant that this would be carpet bombing. In
practice the 8th Air Force found the error of radar bombing was even worse
than expected from the controlled practice studies made in Britain, being
3 km on the average, a rather large pickle barrel [15]! So it was that the
two bomber forces approached one another in tactics as the war groaned
to a close. Bomber Command began attacking in daylight along with the
Americans, and both bombed by radar as a rule [16]. Increased daylight
operation made it easier for Flak.
Some improvement in the accuracy of radar bombing came from the
extended range achieved for new Oboe stations placed in France and the
lowlands. Oboe was enhanced by a new technique from TRE called Gee-H.
This made use of the existing networks of Gee chains. The transmitting
322
The Great Radar War
stations were modified so they could function as beacons, and an addi-

tional unit was added to the aircraft’s Gee receiver to allow it to operate as
interrogator of the ground stations. Ranges could be determined as accu-
rately as with Oboe and used to the same effect. It had the advantage of
permitting several aircraft to use the system simultaneously by interrogat-
ing with coded pulses yet leave Gee remaining functional for its normal
hyperbolic navigation [17].
The H2X systems, AN/APQ-13 and APS-15, were designed to work
with ground beacons and had, in addition to the PPI display, an accurate
range unit used in a manner called Micro-H [18]. It required, of course,
special 3 cm beacons that responded to coded interrogation. Both Gee-H
and Micro-H allowed blind bombing as accurate as Oboe, but were limited
by the distance to the radio horizon. However, as the war progressed and
the Germans fell back this allowed ever deeper penetrations.
Both air forces sent electronic intelligence flights frequently over Ger-
many. The British chose the Mosquito, the Americans a two-man version
of the P-38. One of the P-38s was equipped with the new AN/APR-7 mi-
crowave receiver developed by Terman’s Radio Research Laboratory, but
German interceptors using microwaves were very rare [19].
The news of the bombing of Dresden and its resulting destruction by
firestorm attracted little attention in the United States, where audiences
at the newsreels would soon be cheering the fire bombing of Japan, but
the reaction in Britain was surprisingly strong. In happier times Dresden
had been a favorite city of the English; some had maintained residences
there and many had visited to savor the music, architecture, art, history
and general charm that had gained the capital of Saxony the reputation as
the German Florence. The contending forces had worked out an informal
agreement to spare Italy’s treasures. Could not the same have been done
for Dresden? And for the Allies to have willfully destroyed a marvel of
European culture shocked people into examining the horror that was being
conducted in their name. By February 1945 area bombing had been altered
in the minds of those conducting the war from an act of barbarism to an
accepted practice; it had become the ultimate expression of outrage at
Germany’s refusal to give up—Ludendorff would have accepted the logic
of the situation after the invasion had succeeded. As an example of radar
bombing with H2S, H2X and countermeasures, however, Dresden was a
notable success.
So ended the Great Radar War. It was a struggle lofty in its technical
sophistication yet vile in its indiscriminate destruction of population and
civil organization. It was a struggle in which an air battle was won or
lost according to the appearance of an otherwise insignificant trace on an
oscilloscope located far from the scene of combat. It was a struggle that at
times seemed to remove control from the military commanders and give
it to an academic caste untrained in war who forced alterations of tactics
with bewildering rapidity. All of these remarkable changes in the nature
323
of war came about from radar.

In the Battle of Britain radar had been exclusively a defensive weapon,
one scarcely appreciated by Germans in command positions. As a conse-
quence of this ignorance they employed only trivial or half-hearted coun-
termeasures and did not press home attacks on the stations. When the
weight of bombs began to flow east and south rather than west and north
the new defenders put their own radar to immediate use—for not every-
one on the German side had been oblivious to its potential—and the new
attackers set about just as quickly to counter it. The four-year struggle had
begun.
At squadron level the Luftwaffe had begun to experiment with
ground-controlled fighter direction during the first weeks of the war, well
before their superiors showed interest or understanding. These individu-
als soon caught the attention of Generals Martini and Kammhuber, who
began to construct the Reich’s air defense around radar. The engineers at
GEMA built the necessary long-range equipment, Wassermann and Mam-
mut, and those at Telefunken improved the Würzburg as a gun-laying set
and adapted their 50 cm skills to the airborne equipment, Lichtenstein,
that Kammhuber requested. During 1940 Bomber Command had been
able to fly over Germany at night with light losses but little effect, ow-
ing to their severe navigational problems. As Bomber Command began to
have the concealment of darkness taken from it, TRE proceeded to invent
countermeasures to restore it.
In all this, radar was a defensive weapon being countered, but the
attackers needed more than the ability to protect themselves while over
Germany; they needed some way of hitting the enemy’s vitals. To this end
they strove to make radar an offensive weapon. The most effective outcome
of these efforts were the radio-navigational equipments Gee, Gee-H and
Oboe. The latter two proved to be blind bombing equipment of startling
accuracy—so long as the bombers were above the radio horizon of their
control stations, a criterion that left most of Germany’s industrial potential
safely out of range. Had these techniques been amenable to long-range use,
World War II would have had a different, possibly happier conclusion,
because the selective destruction of key industrial elements would have
been possible, as adequately demonstrated in the Ruhr.
The hopes that H2S would provide such a possibility did not long
survive the tests of reality. It was not ‘the turning-point of the war’ as
A P Rowe had pronounced [20]. Under ideal bombing conditions H2S
could assure that a particular quarter of a large city might be hit, but under
combat conditions it could give only moderate assurance that the city as
a whole could be hit—a decided improvement over ‘celestial’ navigation
but hardly what was needed.
If H2S was unable to deliver much more than saturation bombing,
was it worth it? If German morale did not collapse and industrial pro-
duction was not seriously affected until long-range fighters allowed some
324
The Great Radar War
degree of accuracy through visual aiming, was radar bombing anything

more than a way of punishing Germans? The answer to this is yes. Whether
militarily or industrially effective, the attacks hurt the population and had
to be fought. The air offensive against Germany amounted rather early
to a second front, against which the defense required Flak batteries for
the cities, elaborate air warning and fighter control, squadrons of highly
skilled night fighters, bunkers for the population, greatly enhanced fire-
fighting organization and equipment, services to assist those who had lost
their homes and continual readjustment of transport and industry. These
things required a huge military force and large numbers of administrative
and support personnel.
If the war could not be won in the air, then it had to be won on the
land, and for this the invasion was necessary. The success of this unparal-
leled undertaking turned on control of the air over the ships and beaches,
and this happened because the Luftwaffe had been severely weakened by
defending the Reich. That the air offensive was a descent into barbarity
cannot be denied. That the Allies had an alternative is a matter that only
subsequent generations have had the serenity to consider.
325
PHOTOGRAPHS: RADAR FOR AA ARTILLERY
An American SCR-268 1.5 m gun-laying radar on Kwajalein Island. This was the
Signal Corps’s first radar design. Its original function was to give AA batteries
air-warning capability and to point searchlights, but experience showed it capable
of directing blind fire. The center array of dipoles was the transmitter antenna, the
left array determined the azimuth (horizontal direction) to the target and the right
array the elevation (vertical direction). It remained the US Army’s AA radar
until replaced by the 10 cm SCR-584 in 1944. National Archives photograph
80-G-400984.
326
The Great Radar War
An example of the technical imperative, Flakleit g (FuMO 201). This radar was
built for the Kriegsmarine by GEMA. It used the 80 cm Seetakt wavelength and
was capable of directing fire on surface or air targets. It had an antenna configura-
tion very similar to SCR-268 but was developed independently. It was mounted
on a rotatable, underground, armored optical range finder for shore batteries. Pho-
tograph courtesy of Fritz Trenkle and Bernd Röde.
Receiver for British GL mark II. This

early AA set worked on the 5 m band.
The wooden structures at the sides of
the cabin held two dipoles that allowed
the determination of horizontal direc-
tion through rotation of the cabin and
lobe switching. The transmitter was a
separate unit located about 100 m dis-
tant. Elevation could be determined
through the elevator-mounted dipole.
This set was noted for its robust con-
struction and reliable electronics but
not for blind-fire capability for which
it had little. It proved a great fa-
vorite among the Russians, who built
copies known as SON-2, but British
imports overwhelmed native produc-
tion. National Archives photograph
111-SC 242266.
327
An American SCR-584 at Montalbano, Italy on 4 December 1944 and used by

Battery B, 403rd AAA Gun Battalion. This auto-tracking 10 cm set is easily
recognized by the paraboloid antenna sitting on top of the van that housed the
electronics and secured the dish during transport. The separate antenna in the
foreground is for IFF. The 584 was the most versatile radar developed during the
war. The best tribute to it is that when this was written (1998) hundreds were
still in operation and that it could be purchased from stock as could supplies for
maintenance. National Archives photograph 111-SC 229897.
British Army GL mark III. This 10 cm gun-laying radar saw deployment in 1944
when it became the standard for British AA batteries. It employed manual tracking
and, lacking panoramic capability, generally had to be ‘put on’ by other data. It
was linked to gun directors by selsyn transmission. Manual tracking greatly
diminished its accuracy when following the rapidly moving, low-flying V-1s. That
its qualities were inferior to the SCR-584 and that it came into service so late
after the British invention of the cavity magnetron were the direct result of the
low-priority given to AA artillery in Britain. Historical Radar Archives. Crown
Copyright.
328
The Great Radar War
Royal Navy type 284 50 cm radar mounted on a mark 4(GB) HA director for
both main armament and AA fire-control. The navy began fitting ships with sets
of this basic design in 1940. High power came from the Micropup triodes, and
beam formation came from combinations of Yagis, called fishbones. It was with
this kind of radar that the Suffolk tracked the Bismarck during the initial phase of
a memorable surface action. Churchill Archives Centre, Royal Navy photograph.
Crown Copyright.
German Würzburg C
(FuMG 39T-C) gun-laying
radar. This 50 cm set was
originally designed for fol-
lowing aircraft by having a
large number scattered about
the countryside. When its
directional accuracy, even
without the rotating off-axis
dipole shown uncovered here
at the feed, indicated its value
for directing AA guns and
searchlights, its function
changed; it went through
various modifications. It
proved the best radar until
the Allies introduced 10 cm
equipment for this in 1944.
Bundesarchiv photograph
594/266/31A.
329
German Würzburg at a typical 88 mm AA battery position. AA batteries generally

had four to six guns firing with identical or near identical settings from the same
director, whether optical or radar. Bundesarchiv photograph 356/1845/8.
Railway radar. Poor road conditions in the Soviet Union led the Germans to use
radar trains. A small Würzburg D, rather than a giant probably meant that it
was used for a rail AA battery. Such batteries were also useful in rushing AA
defense to important locations, especially harbors, until permanent defense could
be established. Bundesarchiv photograph 621/2943/24.
330
The Great Radar War
Japanese Army 1.5 m Tachi-2 searchlight and AA radar. This set had four dipole
receiver antennas placed before a metal screen and connected to the receiver through
a rotating capacitor that generated a conical scan. A fifth dipole was located at
the center for the transmitter. This set was successful enough in directing AA
fire at the B-29s bombing Japan that serious countermeasures were undertaken. It
had similarities with the British SLC but was designed before the capture of one of
these at Singapore. Tachi-4 grew out of this set and incorporated aspects of SLC,
replacing dipoles with Yagis. National Archives photograph 111-SC 290064.
Japanese Army Tachi-3. The transmitter of a 4 m searchlight and gun-laying radar

having design elements suggested by the British GL mark II. The dipole array was
mounted above an underground shelter that rotated in azimuth. Altering the phase
allowed the elevation of the radiation pattern to be adjusted. National Archives
331
Japanese Army Tachi-3. The receiver of a 4 m searchlight and gun-laying radar

having a design suggested by the British GL mark II. The dipole array was mounted
above an underground shelter that rotated in azimuth. The signals from four
dipoles were connected through a rotating capacitor to the receiver forming a
conical scan. National Archives photograph 111-SC 290062.
Japanese Army Tachi-4 searchlight and AA radar. This set had four Yagi anten-
nas, removed here to prevent capture, placed before the metal screens for vertical
and horizontal adjustment. The horn is a speaking tube for oral communication.
National Archives photograph 111-SC 231308.
332
The Great Radar War
Americans operate the Japanese Army Würzburg, Tachi-24. The construction of

this prototype was not completed until just before the end of the war and saw essen-
tially no useful service. Its requirements for components, especially the Telefunken
LS180 decimeter triodes, strained the capabilities of Japanese electronics indus-
try. Manufacture was by Nihon Musen (JRC) under the supervision of Heinrich
Foders. Photograph courtesy of Marvin Hobbs.
American and Japanese radar men discuss equipment and their recent belligerency.
The location of the meeting was the Aircraft Control and Warning Center inside the
grounds of the Imperial Palace, which had been off limits to US attacks. Photograph
courtesy of Marvin Hobbs.
333
CHAPTER 7
ALLIED VICTORY IN SIGHT
7.1. THE BATTLE OF THE ATLANTIC, 1939–1945

On entering the Second World War the four great naval powers were sur-
prisingly unmindful of the menace or value of submarines, despite the
instruction provided in the waters around Great Britain during 1917–1918.
For their part the Royal Navy accepted their unqualified victory over the
U-boats with escorted convoys as reason to discount them as a major threat
to the nation’s vital supply lines in the future, a view shared by Grossad-
miral Raeder and most of the upper levels of the Kreigsmarine. The Im-
perial Japanese Navy kept ‘the decisive battle’ uppermost in their plan-
ning and gave little or no thought to the use of submarines as commerce
raiders whether by Japan—hardly the proper calling for a warrior—or by
the United States. The Americans thought they had little to fear from sub-
marines attacking their shipping and rejected a form of warfare that had
been the cause of their joining the Allies in 1917.
In the high levels of world-wide naval command one man, Karl
Dönitz, later Grossadmiral, alone held a contrary opinion. Dönitz was
cut from the same kind of cloth as Air Chief Marshal Sir Arthur Harris.
Each was confident he was capable of winning the war alone with the
weapon he had so assiduously refined; each sought to destroy the enemy’s
capability to fight, the one by stopping the flow of supplies, the other by
destroying industry and the people’s spirit; each was a superb officer able
to inspire his command and organize its complex activities into a coherent
whole; each had in his nature a hardness founded on his concept of duty
that gave no pause in using the terrors at his disposal, if necessary pushing
the hideous logic of war to extremes that most soldiers reject in their hearts
even though accepting in their minds; each led bands of young men pos-
sessed of a courage fashioned from the idealism of youth, a courage that
allowed them to reckon coldly their poor chances of survival yet to con-
tinue unflinching; each vigorously opposed the diversion of his forces from
what he insisted was their true mission. Both finished the war in disrepute.
Dönitz was imprisoned for ten years as a war criminal, although for the
use of slave labor in shipyards, not for the activities of the Kriegsmarine.
334
Allied Victory in Sight
Harris found himself omitted from the honors generously distributed at

the end of the war, but probably as much from personal animosity within
the Labour Government as from reaction to his policy of saturation bomb-
ing. One of them saw in radar the cause of his failure; the other had placed
in it his hope for the success that eluded him.
Dönitz based his confidence in the submarine on tactics, not on tech-
nical improvements in the weapon, as the characteristics of his boats in
1939 were but little advanced over the ones in which he served in 1918.
His approach had two elements: (1) attack by a group of U-boats, the wolf
pack, ordered to the scene by radio from a central headquarters and (2) at-
tack on the surface at night rather than under water by day. He had tested
the latter during the final weeks of 1918, noting that at night a surfaced sub-
marine was almost as hard to find as one submerged during the day and
that its surface speed enabled rapid positioning for launching torpedoes. If
discovered, a quick dive could follow. He also noted that asdic and sonar,
the acoustical location techniques developed by the British and Americans
after World War I, were useless against a surfaced vessel. He had tried
out the wolf-pack tactic in pre-war exercises. He could be sure that Ire-
land would remain neutral, removing valuable bases for anti-submarine
patrols, although he probably expected this inconvenience to be dealt with
brusquely by Britain.
By mid-1940 Germany’s naval position had changed in a manner that
shook the British command: the coasts of Norway and France had become
bases for U-boat operation. By mid-1941 commerce raiding by surface
vessels was clearly a failure, and Dönitz was being given greatly increased
support and doing quite well.
The Battle of the Atlantic, as Churchill named the struggle of the
Allies against the U-boats, was without question the most complicated and
technical form of warfare the world had ever seen. From relatively quiet
beginnings it grew to a raging crescendo both in violence and cunning
until Dönitz’s defeat in 1943, yet continued for two more years. It was a
technical battle, but no less cruel for that.
Intelligence was the key for both sides. Dönitz directed his boats from
a shore station that collected information about Allied sailings from two
principal sources: decryption of enemy signals by the B-Dienst and reports
from the U-boats themselves. Agents listening to seamen’s conversations
in pubs, the subject of countless admonishing posters, figured not at all.
The British had succeded in working the German encrypting machine, the
Enigma, through a major effort at Bletchley Park. This was Britain’s most
closely guarded secret and is generally referred to as Ultra, a name taken
from the special security classification introduced to protect this source—
Top Secret Ultra [1]. The Germans balanced things by generally being able
to read the British Merchant Navy codes. There were periods when signals
were secure, but for much of the time encryption was pointless.
Radar entered the battle early but had little effect until comparatively
335
late. The 1.5 m type 286M was a combination air–surface search set based
on ASV mark I that began to be mounted on British vessels toward the
end of 1940, and by September 1941 177 destroyers and 40 sloops and
corvettes were so equipped [2], but it presented little danger to the U-boats,
owing to its lobe structure, fixed-direction antenna and simple indicator.
A submarine running with little more than the conning tower above water
was observable no farther than 1 km [3]. In March 1941 the destroyer Vanoc
sighted U-100 with a 286M and rammed it. Thus the first sinking marked
to radar’s account was effected by the method of Roman galleys [4]. The
type 286P had a directable antenna, which was a slight improvement, but
only twelve destroyers had it in September 1941.
The Navy had rushed to exploit the advantages of 10 cm radiation
as quickly as possible, succeeding in placing about 30 sets of this type 271
into use by September 1941 [5]. Microwaves extended the radar range, but
the lack of PPI indicators until mid-1943 [6] severely reduced its use in the
confusion of a convoy action. Possibly a more important contribution of
PPI for convoys was its enormous help in keeping such large collections
of ships on station at night [7]. The 271 gained endearment from its ability
to locate lifeboats at night or in fog.
An ASV mark I had succeeded in picking up a submarine from an
airplane on 2 December 1939 [8], but even the introduction of ASV mark II
radar, which was delayed because of the priority given the production of
radar for night fighters, had not had a dramatic effect. In general it was
more useful during days of poor visibility because it was difficult, nearly
impossible to attack a surfaced U-boat at night from meter-wave radar
data alone. This came about because the target reflection became confused
by the ocean-surface reflection as the aircraft approached the submarine;
this generally gave a minimum range of 1.5 km. It was the same effect that
limited the maximum range of meter-wave AI equipment. Anotable excep-
tion was the success of a radar-equipped Swordfish squadron at Gibraltar
that combined the slow speed and the better night vision offered by open
cockpits with the predictable courses of U-boats attempting to enter the
Mediterranean on the surface at night; in November and December 1941
they sank one and damaged five others sufficiently to force them to return
to their French bases for repair, deterring further passage of the straits [9].
But Swordfish crews always seemed to do impossible things.
Coastal Command Squadron Leader Sydney Lugg, remembered for
his invention of the radar beacon, came into discussion with Squadron
Leader Humphrey de Verde Leigh, a pilot with anti-submarine experience
in the previous war, and told him of the excitement and failure of ASV.
In doing this Lugg violated security regulations, as Leigh’s duties were
administrative without any need to know things about the highly secret
equipment, but out of this violation came a memorandum that Leigh sub-
mitted on 23 October 1940 that was to alter the duel between aircraft and
submarine.
336
Leigh proposed mounting a searchlight on the airplane and trying to

illuminate the target when the radar image became obscured by surface
reflection. The following month was filled with meetings and exchanges
of letters and memoranda; at the end of it he found himself charged with
developing the idea, and on 4 May 1941 he made several successful runs on
a submarine. He then experienced an infuriating delay of months when
Coastal Command considered another searchlight technique, the conse-
quence of an unfortunate and later regretted decision by Air Chief Marshal
Sir Philip Joubert when he took charge of Coastal Command, but saw his
design proved the better and in production by the end of summer [10]. By
June 1942 five Wellingtons had been equipped with Leigh Lights and crews
trained for their use. During the night of 4/5 June one attacked and seri-
ously damaged an Italian submarine at the southwest corner of the Bay of
Biscay, initiating a series of adventures for her and her stalwart crew, none
of which led to her sinking. It was the first of many such engagements, but
the first sinking did not take place until a month later [11].
To follow the Battle of the Atlantic one must have an understanding of
U-boat tactics. The World War II submarine was a surface raider that could
hide under water. On the surface it was faster than many merchantmen,
submerged slower than the poorest freighter. Thus a submerged attack
required prior positioning, which barring luck meant running on the sur-
face. This was not particularly dangerous, if there were no airplanes on
patrol, because the U-boat could not be seen by the convoy at distances that
allowed its lookout to see the top hampers and smoke of ships. Spotting a
convoy could be relatively easy when B-Dienst had extracted its route.
The tactics of defense swung between two opposites of thought: win
by sinking submarines or win by saving merchantmen. In practice, com-
binations of the two eventually did win, but it was a long time in the
doing. The first U-boat successes were countered by organizing ships into
convoys, the age-old method of protecting ships from raiders of any sort.
Procuring sufficient escorts and trained crews held back the start, but they
were in hand by late 1941. A properly protected convoy combined the two
approaches by forcing the submarines to fight—and thereby be sunk—
while protecting the freighters and tankers. It was not perfect for the de-
fense, but the convoy fights, often conducted in raging winter gales, ended
the first of the submariners ‘happy times’. Ship-mounted radar was one
of many weapons in this phase, all of which were desperately needed.
President Roosevelt’s bold and hostile extension on 18 April 1941 of
the boundary of the Pan-American Neutrality Zone, placing it closer to Eu-
rope than to America, was followed by his ordering the US Navy to escort
convoys within this region. Hitler did not want a repetition of 1917 and
ordered his forces not to attack ships so accompanied. It is difficult to ascer-
tain what the Navy learned during this on-the-job training, but whatever
it was, it was not apparent to the American merchant seamen who died
in record numbers during 1942. The unescorted, indeed completely un-
337
protected coastal shipping of the United States suffered astounding losses

after Germany’s declaration of war, making it a second, extremely happy
time for U-boat skippers, with occasional relaxation in tropic climes.
(The extreme ineffectiveness of American antisubmarine activity in
1942 led to the most contentious disagreement of the war between Britain
and America. All British and most American historians have laid the blame
on Admiral Ernest King, who held the post of Commander in Chief, US
Fleet. An evaluation of the controversy clearly lies beyond the bounds of
this book, but the reader must be informed that King has found an able
defender in Clay Blair [12].)
During 1941 Dönitz had greatly extended the range of his raiders
through the use of supply ships, and when the Royal Navy eliminated
these, as it had the surface raiders, he replaced them with large supply
submarines. During 1942 this mode of supply allowed the attacks to be
changed quickly from convoyed regions to those of individual sailings. For
a time, arrivals and departures from New Orleans became risky matters.
For Coastal Command the main battlefield in the war against the U-
boat was the Bay of Biscay, where the boats entered and left their patrols
from the French bases. By the end of summer 1942 enough aircraft had
been equipped with ASV mark II and Leigh Lights to have impact. At
the beginning there were never enough planes for the job, and very few
of the four-engine bombers that Joubert eagerly sought. A German radio-
listening station at Boulogne had detected 1.5 m ASV radiation from patrol
aircraft in the fall of 1941 [13], and Telefunken is reported to have exam-
ined an ASV set in May of that year [14], so it is not surprising that the
radar secret of the Leigh Light was quickly recognized. Submarines began
to mount the radar receiver R600A, made by the French companies Metox
and Grandin and generally called the ‘Metox’, that swept a wavelength
band from 0.9 to 3.8 m. The pulse repetition of the radar yielded a char-
acteristic tone at the receiver output that was monitored by headphones.
The first operational boat to carry a Metox left Brest on 9 August 1942, only
a month after the first sinking generated by the light [15]. Metox was a
great success, as it could pick up patrolling planes at ranges far beyond the
capabilities of the radar that was looking for them. Fliers soon began to
notice radar contacts disappearing from their screens as they approached,
and in October only one U-boat was sunk using the radar-light technique,
bringing the total for the method to only four, although many more had
been attacked [16], an event that did not contribute to feelings of well-being
of the U-boat crews, on whom the light was particularly unnerving.
The advantages of microwaves over meter waves for ASV were as
compelling as for AI. The potential was so strong that the Radiation Labo-
ratory had begun the design of the low-altitude bombing radar that was to
allow bombers to attack Japanese ships in the Pacific at night using radar
unassisted by light. Use of microwaves against submarines became in late
1942 a matter of cabinet level policy in the British government because
338
the system proposed for anti-submarine warfare differed but slightly from
H2S, the navigation radar that was the hope for bombing German cities
beyond the range of the radio navigation systems Gee and Oboe.
Those favoring initial use of microwaves by Coastal Command alone
argued on the basis of a rapidly advancing science that used physics and
statistics to analyze problems of engagement and that became a permanent
military discipline called operational analysis. P M S Blackett, physicist
and naval officer, demonstrated that the merchant ships saved by a few
long-range patrol planes would contribute far more to victory than their
use in bombing Germany [17]. Based on that analysis he, Tizard and the
Admiralty proposed diverting a large number of long-range bombers to
the Bay of Biscay action, but the plan was rejected as unnecessary by Air
Marshal John Slessor, who had just taken over Coastal Command, and by
Lord Cherwell and the Air Ministry [18]. In considering this dispute one
must be aware that Blackett and Tizard were temperamentally opposed
to the bomber offensive and that Slessor favored it. Tizard, Blackett and
their associates further argued that using H2S over Germany insured that
enough of a set would be recovered from a wrecked bomber to disclose
its essence to the engineers of Telefunken within days of its recovery, and
although copies might be slow in coming, receivers could be at sea in short
order. The proponents of H2S replied that it was by no means certain that
the Germans had not already learned of the microwave work and might
have receivers ready as quickly as they had countered the success of the
Leigh Light. (Neither party would have guessed that this receiver would
take so long to produce or be so unsatisfactory.) The resolution of this
conflict between Tizard on one side and Lord Cherwell on the other was
resolved by the War Cabinet Chiefs of Staff favoring immediate use of H2S
by Bomber Command.
Additional long-range aircraft for the Biscay fight came to the aid of
Coastal Command from another, unexpected source, one that was quite
agreeable to Churchill—the United States Army. Admiral King’s inaction
at the massacre of American coastal shipping had so affected his Army
counterpart, George Marshall, that Army bombers had been organized to
protect off-shore shipping [19]. By the end of 1942 American antisubmarine
efforts had become sufficiently menacing to force the U-boats to seek ships
in less hostile waters than the American coasts. By that time the American
Army was operating in North Africa and had the desire to give their ocean
supply lines some protection of their own, so one of the groups of Liberator
bombers was ordered there, the other to follow in time. The movement of
these aircraft, which were equipped with the new SCR-517 microwave sur-
face search radar, caught the attention of Churchill who proposed through
Harry Hopkins, Roosevelt’s chief confidant, that they be based in England
instead to help Coastal Command patrol the Bay. Roosevelt agreed, so the
units substituted a harsh winter in Cornwall for a mild one in Morocco but
with clothing intended for the latter [20].
339
Leo Brandt heard reports on 17 March 1943 at the second meeting of

the Rotterdam Committee of design efforts for a series of microwave re-
ceivers intended for 10 cm air warning [21]. They were needed at ground
stations to help work through countermeasures, to guide night fighters to
the bombers, and to alert submarines; all were named Naxos, Naxos-U
being for U-boats. All sets were delayed because of the difficulties in fab-
ricating the crystal diodes for the detectors, and the submarine design had
additional problems, ones that had not troubled Metox. The 1.5 m receiver
had a wooden frame to hold the antenna, the Biscay Cross, which could
be taken below in the few seconds allowed for a crash dive. A similar ar-
rangement for Naxos-U, issued to the U-boat men in October 1943, required
coaxial cables capable of operating at higher frequencies, and the cables
available had fragile insulation easily damaged by the rough treatment of
moving the antenna below in a crash dive [22]. The mistreated cables gen-
erated standing waves with lowered transmission the consequence. This
added measurably to the difficulty of poor sensitivity in the experimental
sets tried in the summer of 1943 and helped compound the confusion [23].
Coastal Command began using the 10 cm ASV mark III in the Bay of
Biscay in March 1943, the month after Air Marshal Sir John Slessor replaced
Joubert. Joubert had built Coastal Command into a splendid weapon and
led it through its worst times but could not evade the onus attached to its
performance in the escape of the capital ships from Brest, and he disappears
from the commanding pages chronicling the history of World War II.
The first aircraft equipped with microwave radar for the Bay of Biscay
were Liberators of the 480th Group of the Army Air Forces, which began
operating in cooperation with Coastal Command from late December 1942
to early March 1943. They had the 10-cm SCR-517 that had emerged from
a collaboration of Rad Lab and Bell Labs, but their planes had no Leigh
Lights, so radar only augmented eyes. Crews had difficulty in maintaining
the sets at first and found the range-bearing indicator [24] confusing and
of little value in navigation. Of their ten sightings only one was by radar
first, although two were simultaneous. An attack followed each sighting;
one submarine was thought to have been sunk [25].
Owing to the priority given H2S, Coastal Command did not have an
aircraft fitted with mark III until the end of January 1943 with no more until
March, when 20 were installed [26], and the first attack by the new radar-
light combination was on 18 March [27]. Mark III performed much as had
the mark II before being countered by Metox; it brought more sightings
than attacks and far more attacks than kills, but confidence in the Metox
receivers, which now gave no warning for microwaves, was shattered. A
fierce secrecy restricted knowledge of Allied microwave capability within
the Kriegsmarine [28]. Nothing was withheld from the top command lev-
els, but they lacked the electronic knowledge and experience that would
have allowed them to evaluate the rapidly changing stream of information
from which they had to make decisions. Those who did have the knowl-
340
edge and experience lacked an easy channel of communication with those

at the top and were unable to help them sort out fact from fiction. If those
at the top were confused, it is understandable that the boat commanders
were mistrustful. The records of the Befehlshaber der U-Boote show no
mention of microwaves until late 1943. The earliest evidence of Dönitz’s
comprehension of the microwave problem is in his report at meeting with
Hitler on 13 January 1944 [29].
This crisis led to the dismissal of Vizeadmiral Erhard Maertens,
Chief of Naval Signals and caused the Kreigsmarine to organize the high-
frequency research of naval and industrial groups into the Organisation der
deutschen Schwingungsforschung für den U-Boot-Krieg having nine co-
working sections [30]. Conspicuous successes of this effort have escaped
record.
The 480th Group was sent to Morocco before the Battle of the Bay
began, which was generally fought during the day, for which they were
well equipped. They came under the control of the US Navy in North Africa
where acrimonious dispute replaced cordial RAF cooperation [31]. By then
their radar skills had greatly improved, and sightings were generally by
radar first. An analysis of the realistic ranges for sightings by ASV marks
II and III and by vision on a clear day shows them to be equal at about
8 km [32].
Two possibilities came to the minds of the German radar engineers
who had to deal with the problem in ignorance of Rotterdam: the Allied
ASV set was switched on only very briefly so that the signal on the Metox
was so brief as to pass unnoticed or the pulse repetition rate had been
increased so that it generated a tone in the observer’s earphones above au-
dible limits, conceivable because of the short usable range of ASV, which
allowed short periods between pulses. The answer to both of these pos-
sibilities was determined by replacing the earphones with a ‘magic eye’,
a small tube with cathode-ray lighting that responded to high frequencies
and that was used for tuning radios visually. It quickly eliminated these
reasons as the cause of the submariners’ distress [33].
A third possibility was that the aircraft were using some means of
homing on the surfaced boats, and emanations from the Metox receivers
immediately came to mind. All heterodyne receivers radiate to some de-
gree at the local oscillator frequency, detectable if one seeks it. At about
this time an unknown Coastal Command prisoner told his interrogators
that the patrol planes were indeed homing on the receivers. Coming at
the time that it did, this removed all doubt about the origin of the mys-
terious attacks, and on 13 August Dönitz ordered all Metox sets removed
from service. A replacement receiver that radiated substantially less and
that was also specially designed to detect short bursts of radar signal, the
Wanz G-1, was quickly put into service and just as quickly removed, yet
most patrol aircraft were still equipped with the 1.5 m radar. By the end of
1943 U-boat skippers mistrusted all radio equipment [34]. The identity of
341
the prisoner who ‘confirmed’ the Metox radiation theory has never come
to light. Evidence points to individual imagination and action, not to an
intelligence plant [35]. Patrol aircraft crashed in the ocean, so examination
of the equipment they employed could not be made.
There is another prisoner interrogation that contrasts with that of the
crewman from Coastal Command. An American, almost certainly from
the Army Air Forces antisubmarine service, with a radar specialty and
considerable experience, did not just answer questions, he gave a short
course on microwave radar and its use against U-boats that fills 11 pages
of report [36]. He explained the operation of both 10 and 3 cm equipment;
he described the under-water microphones (sonobuoys) that transmitted
the sound picked up to aircraft in the vicinity through small radios; he
gave the sensitivities for various targets at different ranges; he explained
the tactics used in attacking a submarine; he even chided them for having
stopped an earlier 10 cm jamming to the west of Spain in October 1943 when
it had been very effective, a statement difficult to reconcile with German
capabilities1 .
The most puzzling aspect of the entire local-oscillator affair was that it
had been proposed by none other than Wilhelm Runge, sent in July from his
technical exile at Telefunken to examine the failure of the Metox receivers.
He suggested the local oscillator as a homing signal for the patrol aircraft
and detected it in an airplane at a distance of 80 km. Runge had attended
the first four meetings of the Rotterdam Committee and was well informed
about H2S and of the work being done on the Naxos microwave receiver
and yet appears to have approached the problem with the remarkable idea
that the Allies would not employ this new weapon at sea [37].
A grueling battle in the Bay began in late winter 1943 with tactics
changing from week to week. The terror of the lights greatly exceeded
their danger but led Dönitz to order surface passage during the day with
enhanced antiaircraft (AA) armament to fight off the attackers. When this
failed for the passage of single boats he ordered them to move in formations
so as to offer maximum AA fire. These changes were serious mistakes
because attacks during daylight were much more effective than those at
night with Leigh Lights, and Coastal Command was not deterred by AA
fire; the grim arithmetic of war favored the sacrifice of a few patrol planes
to sink one submarine.
Luftwaffe Ju-88s from France joined to attack the bombers, which
were soon protected by Beaufighters. It became rough over, on and under
the Bay’s surface. During June and July 14 boats were sunk in the Bay
by aircraft (out of the total of 54 sunk everywhere) [38] but mostly from
visual, not radar sightings. Since March 1942 there had been five Freya
stations on the west coast of France [39], which assured that at least the
last hundred or so kilometers would be free of patrol planes. A map [40]
342
showing the locations of submarines sunk for June 1943 through May 1945
indicates how the Bay was a dangerous part of the sea but less so than the
mid-ocean convoy routes.
High losses soon put an end to surface passage defended by AA
fire, and Dönitz adopted the method of maximum avoidance, having his
boats creep along submerged except for the four hours in the 24 needed
to charge batteries, and finally had them hug the coast of Spain until in
the open ocean. Maximum avoidance and the path by the coast of Spain
proved moderately successful. In addition to allowing occasional incur-
sions into neutral territorial waters, the Spanish route had inherent radar
protection: the coast itself caused confusing reflections, especially for the
still plentiful mark II sets, and the region had far more fishing boats than
the open Bay. These looked like U-boats on any kind of radar screen and
had to be examined with the Leigh Light, which could be seen at great
distances, warning watch officers that a patrol plane was in the vicinity
[41].
Such was the famous Battle of the Bay. Its main characteristic is a
panic induced in Dönitz and his brave captains by the Leigh Light and the
Metox local oscillator business. It is impossible for anyone now to deter-
mine why Runge did not immediately grasp that the shift to microwave
radar was the cause of the Metox failure. Perhaps he was so intrigued
with a clever technical solution and its experimental verification that he
accepted the prisoner’s ‘confirmation’ without confronting a simpler real-
ity. Perhaps the startlingly long range of 80 km at which he had observed
the radiation—was it a fluke?—caused him to think that homing on the
local oscillator was a better method of attacking the U-boats than radar, a
method so clever that the Allies were using H2S to conceal it. It is impossi-
ble for anyone now to know the extent to which secrecy prevented a wider
discussion. It is one of those times in history when one just does not know
what to think.
May 1943 marked the turning point in the Battle of the Atlantic. This
can be seen through graph or table in every book published on the matter.
Sinkings of submarines rose. Almost as many were sunk during the last
eight months of that year as during the entire war before May [42]. Sinkings
of merchant shipping decreased accordingly. These undisputed facts are
used by radar enthusiasts in asserting that the introduction of microwave
radar was decisive. Given the coincidence in the turn of the battle and the
introduction of ASV mark III, it is a conclusion worth entertaining but not a
true one, the testimony of Dönitz and Hitler notwithstanding. For evidence
siding with ASV the reader is referred to a paper by Russell Burns, who
relies primarily on the correlation of U-boats sunk and merchant shipping
not sunk with the introduction of microwave radar [43].
By the spring of 1943 the U-boats had acquired just too many adver-
saries, who were equipped with a remarkable array of new weapons. Of
these adversaries the airplane and the convoy were foremost [44]. When
343
nearly all shipping could move in convoys with air cover, the submarine’s
day was over. The reason lies in the surface nature of U-boat warfare. The
issue of the Bay of Biscay aside, the raiders required undisturbed daytime
movement on the surface in order to locate their prey and maneuver into
position. This could not be done under water because of low speed and
limited vision, and by spring the number of long-range patrol aircraft and
escort aircraft carriers kept the water around convoys under nearly con-
tinual visual surveillance. A sighting led to an attack, more often than not
frustrated by a crash dive, but often followed by surface vessels and more
aircraft.
Sighting was a game for two, however, and a submarine’s watch
usually saw an aircraft in time to submerge discretely or prepare to fight
it out; the mere presence of a patrol plane in the sky—with or without
radar—had accomplished the objective of forcing the raider down. Some
aircraft flying outside the Bay of Biscay began carrying Leigh Lights in 1944
[45], but daylight sightings predominated. The U-boats found themselves
having increasing difficulty getting into firing position and found their role
changing from hunter to hunted. These were also the same months when
ship building overtook ship sinking, an ultimate reason for U-boat defeat.
Worse yet, the U-boats encountered an enemy who was finally orga-
nized, equipped and trained for the task. The Americans had taken their
whipping, eventually listened to their elders and become rather profes-
sional as a consequence. The Royal Canadian Navy had shed illusions
about a cruiser fleet and put a formidable escort force together. There was
finally the necessary minimum of escort vessels and trained crews. Much
of Allied success came from using Dönitz’s own methods. Dönitz applied
strict control from map-filled rooms in Paris, and Commander Roger Winn,
a reserve officer crippled by polio and as such disqualified from more ac-
tive service, directed Britain’s war against the U-boat from similar rooms
in London.
Unified command, the outstanding mark of the British–American
alliance and the single most important element of the struggle, was slow
in coming, however. There were in 1942 six ‘nations’ fighting the U-boats:
the Royal Navy, the RAF Coastal Command, the US Navy, the Army Air
Forces, the Royal Canadian Navy and RAF Bomber Command, the last
trying to destroy submarine production. Having Admiral Ernest King as
US Chief of Naval Operations was not conducive to the correction of this
awkward business, but cooperation did improve as one descended the
chain of command and with time at higher levels, although true unified
command never came about.
Allied technical advances were there too [46]. Aircraft had
torpedoes—so secret they masqueraded as the mark 24 mine—that sought
the sounds of U-boat propellers, duplicated by a similar German device
for seeking surface vessels. Aircraft could drop buoys having underwater
microphones and pick up the sounds by radio, so that a U-boat sometimes
344
found itself in an ocean filled with ears. Aircraft flew with the ability to
locate a submarine by its magnetic field, although it was a marginal tech-
nique. Escorts could fire depth bombs from mortars to the front, removing
the problem of sonar losing the target just when the attacker was getting
close. Rockets fired from aircraft became as effective, perhaps more so,
than aircraft depth bombs. Not surprisingly they were fired first from a
Swordfish, which had proved itself a useful machine flying from British
escort carriers.
High-frequency direction finders, HF/DF, for which only a minimum
of inspiration was required to dub Huff-Duff, began triangulating sub-
marines rather early from shore stations and from aboard ship. The method
adapted Watson Watt’s old cathode-ray tube method of obtaining the di-
rection to a lightning strike. Having been designed for recording transient
events with a wide band of frequencies, it was altered into a device capable
of responding to messages on a single frequency, if ever so short in dura-
tion. Direction finding is accurate for low frequencies or for distances only
slightly beyond the horizon for all frequencies, but suffers for high frequen-
cies from erratic changes in the wave reflected from the ionosphere, correc-
tions for which are imperfect. Its use on ship by skilled operators for close
contacts was as valuable as ship-board radar, as it gave an accurate bearing
on which to dispatch an escort vessel [47], something that proved depend-
able because of the incessant radio chatter that U-boat tactics required.
Shore-based HF/DF was useful, if not particularly accurate, but
had an important secondary function in hiding the source of information
learned with Ultra [48]. HF/DF was secret, of course, but had to be known
by the many people operating the equipment and using the data. It made,
therefore, the ideal cover for Ultra and gave shore-based Huff-Duff an
enhanced reputation. Naval officers with knowledge of Huff-Duff’s di-
rectional accuracy of about 5◦ began to become suspicious of the locations
attributed to it [49]; nevertheless it continued to cover for Ultra in the
decades after the war. Mutual concealment was inverted when Dönitz’s
slowness in realizing the effectiveness of radar changed into an exaggera-
tion of its capability and so caused radar in turn to help hide the value of
shipboard HF/DF from him until June 1944 [50].
The technical and tactical complexity of the Battle of the Atlantic
is made all the more bewildering by its extent over space and time.
Statements concerning radar’s effectiveness necessarily rest on imperfect
records wanting statistical surety. Y’Blood’s study of the American escort
carrier hunter–killer groups comes as close to satisfying the desired statis-
tical criteria as any. Such groups were comprised of a carrier and a few
escort vessels and were used offensively on the open ocean. The aircraft
were a mixture of fighters and bombers equipped with the 3 cm AN/APS-3
designed by Rad Lab and manufactured by Philco. A hunting tactic had
been favored early in the war by those unfamiliar with operational analysis
and skeptical of the defensive doctrine of convoying. It was tried without
345
carriers and was a complete failure because the ocean is big and a subma-
rine small. Later success had its origin in the escort carrier that greatly
extended the area of observation and Ultra that localized the region to be
searched [51].
Y’Blood describes in detail the 55 sinkings by American carrier
groups, and from them one learns that the initial contact was made vi-
sually for 29, with sonar for 11, with HF/DF for seven and the remaining
eight with radar. In the subsequent, often extended battles every kind of
technique was used, but eyes were invariably the most common means of
observation [52]. In one of these fights the carrier USS Card and four 1919-
vintage destroyers engaged U-boats sent to protect the blockade runner
Osorno carrying an invaluable cargo of rubber; in the fight the destroyer
Leary was torpedoed and sunk by U-275 during the night of 23/24 Decem-
ber 1943 [53]. The reader will remember that she was the vessel on which
Page had tested his rudimentary radar for the first time afloat in 1937.
It was not in the nature of the Germans, and certainly not of Dönitz,
to hold back in the unequal struggle that marked the last two years of the
war. Microwaves did not block passage through the Bay, although they
certainly enhanced the danger there, and U-boats fought on the open sea
until the end. Losses were high but accepted. Desperate improvisations
went forward, and radar, Dönitz’s special technical devil, received what
ingenuity with limited resources could provide. The simplest countermea-
sures were decoys, buoys set at sea with reflectors to lure patrol planes onto
false tracks. One in which dipoles for 1.5 m radiation formed the tails of a
kite balloon carried the attractive name Aphrodite. Although such objects
were pursued from time to time, they were at most trivial perturbations
on great events.
The advantages that might accrue from equipping submarines with
radar, used with such excellent effect by the Americans in the Pacific War,
had not escaped German thought, and in March 1942 five boats had been
equipped with a modification of the 80 cm Seetakt. It had a six-dipole array
each for transmitter and receiver mounted in arcs on the conning tower.
The fixed antenna required swinging ship for direction, although phase-
shift circuits to the dipoles allowed 10◦ swings to the left and right [54]. In
operation its range proved to be significantly less than the 9–13 km specified
by the manufacturer. It also suffered from the unreliability that generally
attended the introduction of new naval radar, the combined result of design
that did not properly account for operation aboard a submarine and of not
adequately training members of the crew in its operation. The bulk it added
to an already crowded interior and its failure to achieve notable success led
to its rejection by boat commanders, as was a modification with a steerable
antenna tried out a few months later [55]. A 50 cm airborne sea-search
radar, the Hohentwiel, was adapted to submarines in December 1943 and
had good use in a few cases, but in general the skippers refused to turn it
on, so great was the fear of detection through emanations [56].
346
In efforts to determine what detection methods the Allies were using,

U-boat command routinely ordered submarines equipped for monitoring
radio and infrared emissions to sea under the code name Feldwache. As
the 1943 crisis developed, one was sent in the summer followed by two
more in February and March. The February cruise ended badly with the
vessel sunk and the specialist taken prisoner [57].
Research in camouflaging the conning tower of a submarine by cov-
ering it with a radar absorbent material, a ‘net shirt’, to reduce the reflected
signal gave way to something more distinct, the schnorkel submarine. This
kind of submarine had a tall breathing pipe that extended to periscope
height, allowing air for the diesel engines to be drawn while the boat was
under water. It was invented by the Dutch before the war and known
to both the German and British navies, but only one of them became so
needful of extended under-water navigation as to build it. The under-
water speed was much slower than on the surface; the diesel fumes could
be observed from aircraft; vision was much poorer through the periscope
than from the bridge; the all-important abilities to listen for ship screws
and move silently were eliminated when the diesel engines were running;
radio communication was greatly restricted and they were hated by their
crews because of sudden and unexpected reductions in the inside air pres-
sure caused by automatic valves on the air intake that shut to prevent
water from a too-robust wave drowning the engines. Nevertheless they
were used, as anything else had become suicidal by the last months of
the war [58]. The Allies countered with 3 cm airborne radar, which was
moderately effective against the schnorkel but equally so on small water-
spouts, blowing whales and the many items of detritus that covered the
war-stricken sea.
Radar camouflage techniques succeeded toward the end of the war
in hiding schnorkels—the only success of the Kriegsmarine in the elec-
tronic war—but it was a success of no consequence; the war was lost.
Professor J Jaumann developed with I G Farben Co. a combination of
semiconducting layers arranged for destructive interference into a rubber-
like substance that not only proved capable of greatly reducing an already
small radar echo but was also practical in application. The first of 150
U-boats so equipped went on patrol 5 October 1944 [59].
Radar’s greatest value may well have been its psychological impact.
When ASV mark III began pointing the Leigh Light, the submariners fell to
the defensive, and the confusion caused by false information, ignorance,
and overwrought imaginations created an atmosphere that made regaining
the initiative difficult. Most of their sinkings took place on the open ocean,
but the psychology of defeat was reinforced with every tedious crossing
of the Bay of Biscay. One of many idle speculations one can make about
the war questions the difference that a rapid and successful design of the
Naxos-U might have made—ten months elapsed between knowledge of
Allied microwave capability and the outfitting of submarines with even
347
substandard receivers. All that was needed were good semiconductor

diodes! A less idle speculation questions what effect a clear explanation to
the U-boat men about what was known of microwaves might have had.
One cannot but wonder about the purpose of this rigid secrecy.
Would Germany have won the Battle of the Atlantic, if the Allies had
had no radar? The answer is no. The turn would have been delayed a few
months, more ships would have been lost, but as soon as shipping traveled
in convoys with continuous air cover, the submarine had lost as surely as
had the surface raider. It was the airplane that was decisive. During the
years when air cover was sparse, Allied defeat had been held off by Ultra
and HF/DF in determining the location of the wolf packs and routing the
convoys away from them. When this failed, as it often did, the fate of the
convoys was determined by the efficiency and pugnaciousness of the es-
cort, characteristics that grew with experience—but complete air cover was
best. A reasonable estimate of the relative value of the observational meth-
ods can be best estimated from Y’Blood’s report of escort carrier groups
[60]. Radar, HF/DF and sonar proved about equally effective, but alert
eyes were equal to all of them together.
7.2. RADAR IN ARCTIC WATERS

When Stalin’s gigantic military force showed every sign of defeat, possibly
of an early collapse, Churchill pledged help to his ally. Stalin wished this in
the form of supplies and equipment but specifically not in the form of non-
Soviet troops within Russia. Indeed the aversion to foreigners made life so
difficult for the liaison and technical personnel required for providing aid
that they often felt more like prisoners of war than comrades in arms. His
appetite for supplies, however, was great and calls for them loud. That he
had offered nothing to Britain in her hour of need was now conveniently
forgotten—by both parties. There were only three ways to send goods
to Russia: (1) an extremely long sea route to the Persian Gulf with trans-
shipping onto a long and hopelessly inadequate railroad through Iran to
Armenia, (2) a shorter sea route through the Norwegian and Barents Seas
to Murmansk and (3) the extraordinarily long route by sea to Vladivostok
and the trans-Siberian railway, placing undue strains on British shipping
and Russian rails.
Convoys passing the northern waters in summer had to contend with
almost continual daylight with little but fog and otherwise unappreciated
foul weather in which to hide. The winter had brief, weak daylight and
storms that concealed them from attack but which were nearly as bad as
the enemy. Arctic mariners needed radar, and it was cherished even in
its primitive forms by those who had it. It proved crucial in one arctic
engagement.
A strategic element having a strong effect on these northern convoys
was the disposition of German capital ships. A battleship or cruiser let
348
loose among a convoy without comparable protection was a serious matter

for the defenders, and the mere presence of such vessels in nearby ports
required the deployment of substantial naval force. Hitler had convinced
himself that the Allies were planning to invade Norway and had ordered
most of his big ships to be stationed there. This had been the reason behind
the Channel Dash in February 1942, which had relieved the requirements
for defending Atlantic shipping but augmented those of the north.
Early in 1942 Luftflotte V began organizing anti-convoy operations,
training the first German squadrons skilled in torpedoing and dive bomb-
ing ships [1]. None of their aircraft had radar at the time [2], an awkward
disadvantage above seas where poor visibility was frequent, but during
the long summer days these fliers proved extraordinarily dangerous. In
July they sank well over half of convoy PQ-17. The losses were made much
worse through an order for the ships to scatter as the best response to an at-
tack expected from a battleship surface force—one that never sailed. Scat-
tering gave the best opportunities for air attack, as individual freighters
had little with which to fight off the raiders [3]. The next convoy was
well provided with AA defense, including the carrier Avenger and two AA
ships, all equipped with air warning radar, but nevertheless the bombers
and torpedo planes sank ten of 39 merchant ships at a cost of only five of
the 40 attacking aircraft [4]. Radar provided the defenders early warning
but little else, as none of their AA guns were radar controlled. The onset of
winter with its few daylight hours and dreadful flying weather curtailed
the efforts of Luftflotte V, which was sent to the Mediterranean to make
life difficult for the inhabitants of Malta and helpful for Rommel.
Convoy JW-51B, composed of 16 merchant ships escorted by seven
destroyers and five lesser vessels, left Iceland on Christmas Eve 1942 bound
for Murmansk [5]. The destroyers were equipped with type 286 (1.5 m
fixed-antenna search) and type 271 (10 cm surface search) radars; of the
smaller vessels a minesweeper had a type 271. None of the merchant ships
was radar equipped. Two British cruisers, Sheffield and Jamaica, that came
to the convoy’s rescue had type 286s, 50 cm gun-laying type 285s and the
new 10 cm search type 273s [6]. Attacking the convoy were the pocket
battleship Lützow and the heavy cruiser Admiral Hipper, both with 80 cm
Seetakt sets on forward and aft gun directors.
On the 28th and 29th the convoy endured gales that made station
keeping without PPI sets extraordinarily difficult, and five vessels became
separated. The ships became so coated with ice that several might have
capsized had the storm not abated. The attack by the two large ships
accompanied by six destroyers had been planned by the Kriegsmarine in
direct consultation with Hitler, who shared at least one characteristic of his
predecessor, Wilhelm II: the loss of a capital ship was considered to be a
serious loss in prestige, to be avoided if at all possible. The orders for the
attacking force were thus coached in cautionary terms that took initiative
away from the commander. Their opponents followed an old rule of the
349
Royal Navy: ‘When in doubt, steer for the sound of guns’.

The attackers had overwhelming strength against the escort but had
suffered from long periods of inactivity during which the members of the
escort had been at sea almost continually. The escort was led by Captain
R St V Sherbrooke, who had thought out a plan for just the eventuality that
took place and had prepared his command for it.
Naval battles are complicated by continual changes of position and
confusion about who is an enemy; they are consequently complicated to
describe. The details of the attack and defense of JW-51B need not con-
cern us here. The escort prevented the heavy ships from getting among
the merchant ships, employing destroyer tactics much admired, even by
Sherbrooke’s opponent Vizeadmiral Oskar Kummetz. British radar range
extended only to 10 km, and the opening of the fight followed the visual
sighting of three destroyers from the Hipper group. As gunfire erupted
from both sides during the next few hours radar was used effectively for
ranging, but none of the equipment present allowed blind fire. This was
particularly hurtful to the Germans, as the convoy was quickly hidden by
smoke. It is unlikely that any of the officers in the German squadron knew
that the Kriegsmarine had rejected lobe-switching in 1936 as too compli-
cated for combat use. The confusion of identity was not helped by IFF, as
the few sets installed generally did not respond. Radar ranging made fire
by both sides remarkably accurate, and initial fire was invariably close.
Deficient as it was, ship commanders kept radar reports in mind continu-
ally. Both sides were fortunate in having equipment that experience had
made reliable. Only one set, a type 273 of the Jamaica, failed as a result
of the ship’s own gunfire [7]. Both sides had well trained operators and
mechanics.
Radar did not save JW-51B: Sherbrooke’s bold and well executed plan
did. Radar was relied on for accurate ranges and confirmations of sightings
in poor visibility, but human eyes made the critical first sightings. Radar’s
main contribution was to make the battle bloodier. The Germans lost a
destroyer with all hands, and Hipper suffered damage with casualties. The
British lost a minesweeper with all hands, a destroyer with severe loss of
life and a destroyer damaged with moderate loss of life. All this came from
brief but accurate cruiser gunfire. There is no evidence that the outcome
would have been different had none of these stricken vessels been hit [8].
The damaged British destroyer, Sherbrooke’s flagship Onslow, might have
been sunk but for the alertness of her radar men. They observed the salvoes
leaving the Hipper and determined whether they were coming towards,
going right or going left, allowing the ship to steer away and possibly
avoiding more serious injuries—at least they thought so at the time [9].
This small action had annoying consequences for the Kriegsmarine
because the failure to have sunk a single merchant ship had put Hitler into
a fury. He summoned Raeder and announced that the big ships would be
paid off and their guns used in coast defense. Raeder resigned and was
350
replaced by Dönitz, who with time was able to have the imbecilic order
modified.
The Allies were not the only ones having convoy difficulties in the
far north. German Radar had been posted on the north cape of Norway—
desolate, isolated stations—to detect enemy aircraft or direct those of the
few Luftwaffe bases [10]. These German bases received most of their sup-
plies by costal convoys, which encountered trouble primarily from Soviet
submarines. After April 1943 these attacks increased significantly, bring-
ing the suspicion that there was a secret submarine base nearby. Remedy
came about by means of a Würzburg located at Vardø on the northeast tip
of Norway. It observed morning and evening that a Russian plane dis-
appeared into the ground clutter at the same place. Aerial photography
disclosed the base in an isolated Norwegian fjord, and a surprise army–
navy operation not only eliminated the base but uncovered a Norwegian
underground group that had been providing reports of German activity.
They also captured a Russian code book with which misleading informa-
tion was propagated for a few weeks [11].
When Lorenz lost to Telefunken in the competition for the best gun-
laying and searchlight radar they did not choose to leave the field entirely.
In particular, they had improved the decimeter triode, DS310, into the
much more powerful RD12Tf for the otherwise unsuccessful Kurmarkt to
give 50 kW pulses at 55 cm. Around this tube they built, at the urging
of their principal radar engineer, Gotthard Müller, the excellent airborne
sea-search radar Hohentwiel. It used a variety of dipole antenna config-
urations, generally arranged to give either forward or lateral coverage.
Display used the standard dual-beam oscilloscope with the right and left
traces showing alternately right–left coverage or lobe switching for the for-
ward direction. The time base was logarithmic. Hohentwiel could observe
a 5000 ton ship at 80 km and a submarine periscope at 6 km [12].
As the end of 1943 approached and the days became short, convoys
again began to go to Murmansk, and both sides prepared for attempts
by surface ships to break them up. When reconnaissance planes, now
equipped with Hohentwiel [13], spotted JW-55B on 22 December Dönitz
decided to have Scharnhorst and six destroyers attack. The Royal Navy
expected the appearance of surface raiders under such conditions and came
well armed and forewarned by Ultra; their available strength consisted
of the battleship Duke of York with cruiser consort and an independently
maneuvering squadron of three cruisers. The Scharnhorst lost touch with
her destroyers early; they had no radar and were unable to join the fight
or find the convoy, so the big ship fought alone [14].
The British battleship and cruisers had types 281 (3.5 m air warning),
284 (50 cm main armament), 285 (50 cm heavy AA) and 273Q (10 cm surface
search), and the Duke of York’s 273Q had a PPI indicator [15]. The Scharnhorst
had the long-standard arrangement of an 80 cm Seetakt on the forward
and aft gun directors [16]. Radar had had little influence on the outcome
351
of the action the year before, but this time it dominated and spelled doom
for the German ship more surely than the disparity in artillery. The British
radar was much improved over the previous year. Scharnhorst was sighted
by Belfast’s radar 20 km distant at 0840 hours from which an opening,
star-shell-directed volley carried away the Scharnhorst’s forward Seetakt,
leaving her half blind in a battle that took place almost entirely in the
dark. Scharnhorst eluded the pursuers until 1221 hours when she again
encountered the squadron of three cruisers and headed south to lose them
with her 5 knot speed advantage. She could watch the pursuers with her
aft Seetakt but was blind forward. The Duke of York lay directly in her path
and had seen her on the oscilloscope screens well in advance. The position
of the cruisers to the north and the battleship to the south left a free path to
the east, but her electronic blindness allowed no warning of the danger into
which she fell, after which her speed advantage was unable to save her. At
1650 hours Scharnhorst was completely surprised by illumination followed
by accurately placed 14 inch shells. It was one of the dramatic moments in
naval history, etched in the memories of all who saw it and remembered in
a painting by Charles Pears at the National Maritime Museum, Greenwich.
No one knows how many hits the Scharnhorst took, but they were
many during the two-hour struggle. Like the Bismarck and her World
War I namesake, she went down fighting courageously to the end. Only
36 survivors were taken from the dark, icy waters.
The invasion of North Africa put an end to the possibility that the
Allies might invade Norway, which together with the failure of German
surface ships to break up the convoys, removed much of the function of
the big ships in the north. Lützow was sent to deal with more pressing
needs in the Baltic and the Gulf of Finland, and Dönitz dismantled the
badly damaged Hipper as he maneuvered to retain big ships without direct
disobedience to Hitler’s orders. After the sinking of the Scharnhorst only
the Tirpitz, sister ship of the Bismarck, was left to threaten the northern
convoys.
Bringing about the end of the Tirpitz strained British ingenuity. It
began with carrier planes of the Victorious in March 1942 to little effect,
followed in October with some curious devices called sea chariots that had
been used with great success by the Italians, but British imitation failed.
Two of six midget submarines succeeded in September 1943 in placing
large bombs beneath the vessel and causing serious damage, although the
degree of hurt was unknown to the British. The Fleet Air Arm tried again
in April 1944 and succeeded in inflicting casualties but not in removing the
ship as a threat. The approaching air strike had naturally been observed
with a Freya, but the warning only allowed the crew to man guns and start
smoke generators, as Göring had no fighters to spare for Tirpitz [17].
Sinking the Tirpitz then became a special project for Bomber Com-
mand that was to occupy them for three months. The task was assigned to
Squadrons 617 and 9, the former famous as the Dam-Busters. Lancasters
352
were modified to take special 6 ton bombs that were certain to penetrate the
armored deck. Three attacks, each in excess of 30 bombers, were needed to
complete the job with two fatal hits finally secured on 11 November 1944.
In order to obtain the maximum of surprise for the third and last raid, elec-
tronic intelligence flights mapped the radar of the Norwegian coast and
found a location through which aircraft at 450 m could pass across Norway
into neutral Sweden, whence they attacked from the east [18].
This effort wrote a final chapter to the controversy between naval
and air power men that had raged during the two decades before the war.
An immobile battleship, defended only by her own AA guns and smoke,
required three attacks with 100 superbombs to be sunk. Although marked
up as a success, it in fact underscored the fallacy of believing surface ships
had been made obsolete by airplanes. Of the demonstrated failures of
air power to deliver on pre-war promises, the failure of land-based heavy
bombers to destroy naval power is the most grotesque.
7.3. THE MEDITERRANEAN, 1943–1945

‘For all its awesome history as a battleground between civilizations, the
Middle East did not strike American strategists as an area in which the
European war could be expeditiously won. On the other hand, they recog-
nized it as an area in which the global war could be very speedily lost’ [1].
Given the British stake in the region these sentiments were certainly not the
kind to cause Anglo-American friction and were to lead to useful lessons
in cooperation. The shores of the Mediterranean also proved excellent
places for the much needed apprenticeship of the green American forces.
There was enough resistance to the North African landings, greatly helped
by a defense lacking conviction or spirit, to provide useful intermediate-
level training. When the French forces surrendered the Germans occupied
Tunisia and quickly gave the Allies—the British ground troops there had
had no battle experience either—a taste of Luftwaffe ground support with
demonstrations of Rommel’s desert skills not far behind.
An understanding of radar did not accompany most US commanders
ashore. The 560th and 561st Signal Corps Air Warning Battalions encoun-
tered officers who had no conception of how these units were to protect air
bases, and they did not obtain their equipment from the holds of transports
until eight days after the first landing, when they were able to set up an
SCR-602, the new lightweight air-warning set [2].
RAF radar landed proficient and organized. A light-weight air-warn-
ing (LW) set came ashore with the first wave followed by four COL/GCI
sets (COL designated the overseas version of CHL) three days later. Addi-
tional equipment of the same kind soon covered the coast of Algeria and
two MRUs arrived after four weeks. All stations reported to a filter center
established in Oran [3].
An understanding of radar did accompany the commanders of the
US Army’s AA units. Among the first units ashore was the 62nd Coast
353
Artillery, which had long furnished the air defense of New York and was
the regiment that had provided the first troops for training in radar use in
late 1937. They had received a production SCR-268 for each gun battery
and had had time to hone their skills. With them came other units that
provided gun and automatic-weapon defense of North African ports and
air bases [4]. Battery commanders soon found shooting under radar control
much more effective than with searchlights [5], and their accuracy grew
with practice. The SCR-268 provided batteries with local air-warning as
well as gun-laying capability, justifying the Army’s decision not to adopt
the Bell Laboratory 60 cm CXAS.
The pattern of air attack on sites defended by American 90 mm guns
became fixed throughout the Mediterranean: a few trial raids at night on
a new target tested the defenses and found losses too high to be sustained,
after which raids were rare. An attack on Palermo on 4 August 1943, for
which the enemy numbers and losses were confirmed by prisoner inter-
rogation, showed what the SCR-268 could do. Shooting on radar control
at 29 night bombers, defending guns brought down five aircraft over the
port and damaged two sufficiently that they crashed on the return flight.
The prisoner asserted that the AA fire was the most accurate these fliers
had ever encountered. Despite the importance of Palermo as a port, there
was only one more raid; in it two out of 20 planes were seen to come down
and bombs were jettisoned before the target [6]. Microwave engineers,
steeped in the heady mysteries of the magnificent SCR-584, often speak
with disdain of the 268, but from November 1942 until March 1944, years
when the Luftwaffe was a potent force in the Mediterranean, it was the 268
that gained respect for American gunfire [7].
A large air-warning and fighter-direction system was soon in opera-
tion with a record amount of equipment. Night attackers had to deal with
GCI stations that guided AI-equipped fighters. Radar was secret and air-
borne radar was really secret, so the Air Ministry required this equipment
be sent by sea, and the attempts to bring down night bombers with GCI
alone failed. The first raid on Algiers on 20 November 1942 was met by
gunfire and did little damage, but General Eisenhower emphasized the
need and an AI-equipped Beaufighter squadron arrived and shot down
five unsuspecting bombers a week later [8]. A British GCI unit near Bône
was credited with aiding Beaufighters in the destruction of 23 aircraft, but
Freyas helped make December 1942 a wretched month for the RAF. The
1.5 m AI sets displayed the ground-return limitation that had originally
indicated the need for microwaves. Night raiders, Allied or Axis, learned
to fly very low to lose themselves in ground clutter. It was a different war
than the one over Germany where heavy bombers were forced to the high-
est altitudes to escape gunfire and in so doing made themselves perfect
targets for the 50 cm and later 3.5 m Lichtenstein sets.
The star of North African radar was unquestionably SCR-582, a 10 cm
harbor-surveillance radar that arrived at Oran on 27 January 1943 [9]. This
354
set, the second Rad Lab set to see combat (SG had entered a couple of
months earlier off Guadalcanal), startled all the American and British radar
men who saw it. Although intended for harbor defense, its 120 cm diame-
ter dish allowed it to pick up low-flying aircraft at 40 km. Its PPI indicator
allowed the operators to guide ships entering harbor through the protect-
ing minefields, and it proved perfect for detecting German motor torpedo
boats. All of these functions were nearly impossible for meter-wave sets.
SCR-582 was soon modified in the field for air defense functions [10].
US Secretary of War Henry Stimson was an early and enthusiastic
apostle for radar and had had Dr Edward L Bowles appointed as his spe-
cial radar advisor. Bowles requested that a senior engineer from Rad Lab
examine the use of radar on the North African front to keep design in
close touch with the realities of combat, and DuBridge sent Dr Louis N
Ridenour. He confirmed the correct emphasis of Rad Lab on microwaves;
he noted the deficient understanding of proper radar use by Army Air
Forces personnel; he learned the extent and quality of German radar and
countermeasures [11].
Rather than extract the Axis forces from Africa, Hitler chose to build
up his forces in Tunisia. The painful experience of supplying the much
smaller Africa Corps was thought to be offset by a much shorter supply
route, and the troops that came by the hundreds of thousands by air and
sea brought radar with them. Of particular note were three stations set
up for fighter control around the Gulf of Tunis, each using the Himmelbett
system of one Freya and two Würzburgs and directed by an ‘ace’ controller.
Like their Allied counterparts, they found using meter-wave radar in the
mountains to the west more exasperating than rewarding. The error of the
decision to defend Tunisia became all too apparent in May 1943 when a
larger force surrendered than at Stalingrad. A few pieces of radar equip-
ment were evacuated, as were the aces, but most sets had to be destroyed
and their personnel became prisoners [12].
Infantrymen on both sides experienced little comfort and less knowl-
edge of the wonders of radar. They were always convinced that their own
air power was gone. ‘Where is this bloody Air Force of ours? Why do we
see nothing but Heinies?’ Except for the defense of ports and air bases,
ground troops had little reason that they could see to thank radar—in the
unlikely event that they knew about it [13]. Radar was the weapon of an
air fight; its introduction to ground warfare had to wait two years.
The French surrender of North Africa caused not only the German
occupation of Tunisia but also of the remaining part of Metropolitan France
controlled by the Vichy Government. This brought French radar develop-
ment to the end. With defeat in spring 1940 radar research had moved from
Paris to Toulon, where it continued in cooperation with the Constructions
Navales, and to Lyon at the laboratory of the Société LMT. Two designs
reached prototype stage: a 16 cm sea-search set capable of detecting cap-
ital ships at 25 km and torpedo boats at 10, and a 50 cm set for sea and
355
air search that attained ranges on aircraft of 17 km. The ships at Toulon
were scuttled before German forces arrived and the radar equipment was
destroyed. Some clandestine research was continued in Lyon [14].
With Tunisia secured by the Allies the next step was Sicily. Here
was a more difficult operation, one that required a higher degree of ability,
almost as if it were the next step of a training schedule. North Africa
had been defended entirely by French units, many of whom—the Foreign
Legion no doubt excepted—had no wish to fight. Sicily was defended by
a mixture of Italians and Germans. Except for the units that had benefitted
from Rommel’s command, Italian troops had made an extraordinarily poor
showing, and those defending Sicily were green, ill trained, ill equipped
and ill disposed toward the Axis. The very opposites of these descriptions
described the Germans there.
Commanders planning the invasion took radar as a serious threat to
their air and naval operations, so electronic intelligence set about to map
out the Axis radar screen. The Americans dispatched B-17 Ferret aircraft
loaded with equipment provided by Frederick Terman’s Radio Research
Laboratory on 22 April 1943. Their equipment proved of value before it had
even reached the coast of Africa, as the flight’s navigation was in serious
error, and they located the mid-destination, Ascension Island, by homing
on its radar. They set up operations at Blida, south of Algiers, where their
strangely painted aircraft and super-secret nature filled the beer conver-
sation of neighboring fliers. Secrecy played them false later when planes
filled with the jamming sets intended for the invasion arrived only to have
maintenance crews, who were not instructed about their special nature,
begin making them ‘normal’ by removing non-standard electronics.
Flights in June established the locations of Freyas and a Wassermann
or two on Sicily; another flight circumnavigated Sardinia. After the inva-
sion the locations of the stations mapped by the Ferrets were confirmed,
either by a wrecked set or evidence of an earlier presence. The effects
of American and British jammers, Mandrel (for Freya) and Carpet (for
Würzburg), were never determined. Any effect escaped mention in avail-
able German records [15].
The Italian fortress island of Pantelleria had German radar that had
to be eliminated before the invasion could proceed. The Italian garrison
failed to fulfill Mussolini’s boasts of the impregnable strength of the base—
indeed they did not fight at all—and it was quickly converted into a British
radar post. A floating filter center allowed fighter control during the land-
ings [16]. IFF was its usual sorry self with fighters wasting much time
following up unidentified plots [17]. Ground radar units went ashore with
the landing forces.
Sicily saw the introduction of a new technique contributed by TRE
that made possible the landing of airborne troops behind enemy lines at
night. It consisted of a very light interrogation radar, called Rebecca (call-
ing her children) and a responding beacon, Eureka (‘I have found it!’),
356
thereby providing the troop-carrier pilot with range and bearing. Eureka
differed from beacons generally in responding only to Rebecca’s call [18].
The interrogator could be mounted in any army support plane and the re-
sponder was light enough to be carried by a parachutist. Advance parties
set up Eureka beacons in Sicily to mark the dropping zones for parachute
and glider troops. The mere idea of attempting such a navigational feat
without Rebecca/Eureka would have brought the most accomplished flier
to despair. Success in Sicily ensured it an important place in the airborne
component of the Normandy invasion and before that in the infiltration of
secret agents and their supplies into occupied Europe [19].
Completing the occupation of Sicily was a ground operation with
extensive air activity. Allied air was by then dominant and remained so
thereafter, but the Luftwaffe fought back tenaciously and skillfully; the
Mediterranean remained a major front for them and a major drain on their
resources. Radar was now a routine component of the air operations for
both sides, although secrecy often kept fliers from knowing the source of
their guidance.
The occupation of Sicily precipitated the fall of Mussolini and his
Fascist government followed by Italy’s surrender on 3 September 1943
followed in turn by the invasion of the Italian peninsula nine days later.
Whether this was the best application of Allied resources has remained a
matter of dispute, but the question of whether the Germans should have
resisted this new invasion was not considered by Marshal Kesselring. He
rightly saw Italy as a dangerous base for air attack on the Reich and decided
such an operating area must be kept as far away as possible. That the Alps
formed an impenetrable radar barrier for concealing attacks from the south
featured in his reasoning. This decision turned Italy into a major theater of
ground fighting for the next 21 months. German ground radar covered the
entire sea coast of continental Europe from Spain to Turkey and including
Sardinia, Crete and the Greek islands with 86 Freya, 18 Wassermann and 54
Seetakt stations [20], all carefully located by Allied electronic intelligence.
The sources of this information do not disclose the number of Freyas and
Würzburgs, great and small, that belonged to Flak and fighter direction,
but they must have been correspondingly great.
British forces landed on the sole of the Italian boot, and the Americans
went ashore at Salerno, 45 km south-east of Naples. Both landing places
were within fighter range of Sicily; German air power was not adequately
situated for a strong response and was unable to hinder the landings seri-
ously [21].
Much of British and American radar was concentrated in the 1.5 m
band: airborne AI mark IV and ASV mark II; ground-based radar
COL/GCI, LW and SCR-602; naval type 286, CXAM and SC; AA SCR-268.
Inasmuch as Window had been used in the July Hamburg attacks it should
have hardly come as a surprise that the German bombers made a distribu-
tion of Düppel cut to 75 cm at the Salerno landings, but it proved nearly
357
as effective in blinding radar as Window had been to the Würzburg and

Lichtenstein equipment months earlier. The call went out for microwaves,
and Allied pilots and gunners began to learn how to distinguish clouds
of foil from aircraft, as had their enemy counterparts a few weeks earlier
[22]. The 7.5 m MRUs and naval type 279s were immune to 75 cm Düppel,
which redeemed the otherwise increasingly obsolescent sets. Düppel cut
to 75 cm became part of the Luftwaffe technique throughout the theater
[23].
After four weeks the advance came to a halt along the Velturno River
and, air superiority or no, with or without armor, progress became very
slow, finally coming to a complete stop at the German fortified line called
Gustav that stretched from the Tyrrhenian to the Adriatic Seas. This was
just the kind of war that the tank and the airplane were supposed to have
relegated in 1918 to that location of favorite cliche, ‘the ash can of history’,
but the line held.
To break this stalemate another invasion was planned to get behind
the Gustav Line. It was a technique now well exercised, and the newly
acquired territory allowed fighters to range up the coast to cover it. The
place selected was near a small port 50 km south of Rome called Anzio.
The landing proceeded very much as had the one at Salerno but failed to
accomplish its objective of turning the enemy flank, in part through hesitant
American command and by a timely and effective German response. It was
marked by two radar matters of note: a radar feint and the introduction of
the SCR-584.
The Luftwaffe was in much better position to react to this landing
than the one at Salerno, and Kesselring had suspected such a move. The
landing was preceded by strong air and naval bombardment and estab-
lished a bridgehead in the early hours of 22 January 1944. The Luftwaffe
immediately began shifting units from other theaters—one had bombed
London on the 21st—to attack the fragile concentrations on the beach. The
US Twelfth Air Force removed much of the power of this blow before it
had a chance to strike. The most remarkable of these preventive raids took
place on 23 January when five American bomber groups, well escorted,
flew on courses obviously intended for the landing field of some of these
squadrons. They were intended to be picked up by radar and they were,
but a group of P-47s went to the scene of combat by another route flying
close to the ground to avoid radar and so timed as to arrive as the defend-
ing fighters were beginning their take off to meet the radar-heralded force.
The defenders were thrown into complete confusion, and the bombers car-
ried out their attack free of harassment. A total of 140 German aircraft were
destroyed in the air and on the ground at the cost of six bombers and three
fighters [24].
Allied forces failed to break out of their constrictive location on the
beach and expected a strong counterattack. Hitler ordered it for 1 February,
but Allied bombers had been working on the Italian railways, making
358
tedious work of the assembly. The counterattack, planned with personal

but not helpful intervention by the Führer himself, began on the 16th. Air
strikes were naturally a part of this action and were matched by radar
countermeasures. Martini had set up powerful land jammers for the 1.5 m
wavelength of the SCR-268, the airborne sets of the night fighters and any
naval radar that got in the way. Ground-based jammers that are near the
action can be particularly effective, owing to the high power and the ability
to remain in place, characteristics not possible in airborne jammers. The
Anzio AA guns and night fighters were blinded. To complete things the
bombers carried airborne jammers and let out large quantities of 75 cm
Düppel.
This situation brought response from Signal Corps personnel in Al-
giers where new SCR-584s and 545s (the Bell Labs alternate 10 cm gun-
laying set) were at hand, equipment immune to the 1.5-m jamming and
to Düppel cut to that wavelength. Under intense pressure crews were
trained in the use of these two, and by the 24th a night attack by 12 Ju-88s
lost five of their number to the newly arrived microwave radar [25]. The
counterattack had failed by then because American and British infantry
held their ground, but relief from bombers was appreciated for all that. A
British GCI station got use of a 584 and found it excellent for their function.
Countermeasures became increasingly effective against SCR-268 both as a
gun-laying and a ground-control-intercept radar, but months were to pass
before there were enough SCR-584s for even vital positions.
The war continued in the Mediterranean Theater without further nov-
elty in radar. Allied forces pounded their way up the Italian peninsula until
the bitter end, securing more advanced bases for bombers and tying down
a significant German force while American and British commands argued
about a ‘soft underbelly’ approach. The invasion of southern France suc-
cessfully followed Normandy and used all of the paraphernalia of radar
and its countermeasures [26].
7.4. JAPANESE SHIPPING DESTROYED

Japan discarded an ancient feudalistic past in preference for the modern
world in a truly breathtaking change called the Meiji Restoration and set-
tled its government in 1889 on a constitutional monarchy based on that
of 1871 Imperial Germany. As the century progressed power fell into the
hands of the Army, negating any traces of representative government; min-
isters who entertained delusions about having real control, and who did
not respond to instruction were assassinated. By the 1930s the Prime Min-
ister was a dictator of the Empire chosen by the Army, an office that Hideki
Tojo combined with that of Chief of Staff from 16 October 1941 until the
loss of the Mariannas led to his resignation. Aside from the manifest evils
of this regime, found in the brutal subjugation of Asian people unfortunate
enough to come under its heel, it maintained the credulous belief that the
359
warrior spirit would carry the Empire through war and that little thought
need be given such servile matters as the economic and commercial basis
of industry with its extreme dependence on shipping. This blindness is all
the stranger given an insular need of imports, and is stranger yet when the
seriousness of the growing blockade went well past alarming indications
of ruin. But Japan’s adversary was almost as slow in recognizing the fatal
weakness that lay in shipping. When the American Navy finally under-
stood what was to be done, it proceeded with efficiency to destroy Japan’s
merchant shipping.
7.4.1. Destruction with submarines

At the end of the Pacific War the US Navy’s submarine service could be fa-
vorably compared to the U-boat fleet that Admiral Dönitz had commanded
three years earlier. Boat commanders were young and daring and had
accomplished against Japan what Dönitz’s men had failed to do against
Britain. But things had not begun that way. The US began with a fair sized
fleet of submarines, many of construction growing out of World War I but
serviceable nevertheless, and to which a large number of modern vessels,
called ‘fleet submarines’, were joined. Compared with the older American
S-boats and the German U-boats these were almost luxurious. Two defects
of the service were serious and slow of correction. (1) Submarine command
had followed traditional peacetime promotion patterns, and by 1941 most
boat commanders were into their middle or late thirties. The baleful ef-
fects of age on audacity were compounded by the Navy belief that sonar
and aircraft had made the submarine a highly vulnerable weapon and that
caution was an important virtue. The near-complete failure of the Navy
to meet the challenge of the U-boats in 1942 had much of its origin in this
doctrine. (2) The new mark XIV torpedo was poor. It failed in three ways,
all of which had to be independently discovered and corrected in the field
when the Bureau of Ordnance refused either to admit there were problems
or even test torpedoes for alleged faults, asserting until counter-proof was
forced on them that the deficiencies lay with crew performance. First, the
mark XIV ran significantly deeper than set; next, the vaunted magnetic ex-
ploder frequently either failed to explode at all or detonated prematurely;
last, the contact exploder failed for a broadside collision—the condition
for a perfect shot! Some of the new mark XVIII electric torpedoes proved
capable of exasperating the users with exciting new ways to fail, such as
running straight down or following terrifyingly erratic paths. Its defects
were corrected in less time and with a higher degree of Bureau cooperation
than for the mark XIV. Boat commanders were relieved after an unsatisfac-
tory patrol or two but were all too often replaced according to old rules of
seniority. It required two years before a first-class group of boat comman-
ders was assembled and about the same time for the torpedoes to be made
reliable.
360
The American submarine war was initially hindered by wrong goals

perversely made worse by the availability of decoded Japanese mes-
sages. The Navy had immediately pronounced that the international
agreement forbidding submarine commerce raiders from attacking with-
out warning—the basis for America’s declaration of war in 1917—had been
invalidated by the Japanese in their attack on Pearl Harbor, yet the de-
struction of enemy shipping was not given urgency despite Japan’s obvi-
ous vulnerability to a submarine blockade. Instead boats were dispatched
on the basis of decoded messages to sink important naval units plus any
freighters or tankers that might get in the way. This policy produced many
wild chases but few sinkings of the targeted ships. About two years were
needed to have a submarine blockade take top priority.
The Navy was greatly helped by the ineptness with which the en-
emy fought back. Individual destroyer commanders were aggressive and
competent in the use of sonar, but they seldom pressed their attacks until
there was clear evidence of a sinking and seemed to think that laying a full
barrage of depth bombs around the sonar contact destroyed the boat. Thus
it frequently sufficed for the submarine to lie still and wait. The Japanese
never organized anti-submarine warfare in a way that evaluated the many
bits of information that led to a boat being tracked and sunk. Tactics drawn
from operations research were never employed.
The Naval Research Laboratory began thinking of providing sub-
marines with radar before the CXAM was in production. Their first con-
cern was something to warn the sub of the approach of aircraft, allowing
sufficient time to dive. The first set, the SD, was with minor modifications
the standard air-warning set through most of the war; it began production
in late 1941 and was the ultimate in simplicity. Having warning as its func-
tion the antenna gave no bearing information and range accurate to only
1000 m. In order to provide the 100 kW power needed for its broadcast
mode it had a pulse repetition frequency of only 60 Hz. The wavelength
was 2.65 m, the longest used by the Navy [1].
The SD was useful for guarding the boat from surprise while on the
surface but of little value in the quest for maritime victims at night or in
fog or as an aid to navigation. Given the constraints on antenna size for
a submarine this function could hardly be met with meter waves, but the
unexpected appearance of microwaves in 1940 changed that immediately.
Bell Labs, which designed most of the Navy’s microwave radar, began
working on a submarine radar almost immediately after learning of the
magnetron. It was designated SJ.
The achieved design goals of the SJ were (1) sweeping the horizon
with a narrow beam from an antenna at periscope position with the boat
submerged or on the surface, (2) observing a destroyer at tactically use-
ful distances, (3) determining the range with an accuracy useful for the
torpedo data computer and (4) presenting the observations on a PPI dis-
play, although this last followed the first deployment by about a year. The
361
wavelength selected was naturally the 10 cm band being so actively ex-

ploited by Bell and Rad Labs. The antenna took a radiative form similar
to the Rad Lab Raytheon SG but had the more troublesome requirement
of having to keep water out of the waveguide that connected it with the
electronics in the hull and doing that for the pressures experienced. The
Naval Research Laboratory contributed the necessary stock of knowledge
to make SJ seaworthy. The prototype, which was completed in December
1941, had a beam 9◦ wide and 29◦ in elevation, could detect a destroyer at
10 km with an accuracy of 25 m and a low-flying bomber at worthwhile
ranges. The cut-parabolic reflector could be subjected to strong forces from
depth bombs and from collision with the detritus that covered the wartime
ocean, so it was made to pivot harmlessly out of the way, if a replaceable
shear pin snapped [2].
Boat commanders had mixed reactions about SD, which left on its
first war patrol on 13 December 1941. They were restricted to radio si-
lence except under special circumstances as a precaution against direction
finders [3], so orders that prescribed having SD emit a steady stream of
very-high-frequency pulses yet forbade even the briefest of radio messages
impressed many submariners as another example of the weakness of intel-
lect they had long suspected inhabited certain elements of naval command.
It was thought by many that Japanese signals intelligence did learn of the
SD emission and provided homing equipment for patrol planes, but the
evidence is apocryphal.
Boat commanders had quite a different reaction to SJ, which went
on its first patrol in August 1942 aboard USS Haddock. The commander
used it to follow a ship at night, concluded with a successful attack, and
sank another in the same way two weeks later. The new radar had not
performed flawlessly, being out of calibration at times and out of operation
for extended periods, but these shake-down ills were recognized as such
and known to be correctable with adequately trained personnel. There was
apprehension of the 10 cm radiation being used as a homing signal, but its
narrow beam diminished this concern, and the set received an enthusiastic
endorsement from its first users [4]. The SJ radar was probably as important
as the periscope, as it turned night into day for the raiders in addition to
providing accurate range and a more secure air warning.
The Japanese fleet and its merchant shipping were destroyed in 1944.
Two epic battles saw the end of the principal fleet units: Philippine Sea
in June, Leyte Gulf in October. By the beginning of 1944 the American
submarine service had removed defective torpedoes, defective skippers
and defective tactical doctrine from its baggage. These changes brought
about a remarkably fast strangulation of Japan’s oil supply with immediate
strategic consequences and sank or disabled most of the remaining fleet
units.
In the Battle of the Atlantic the technical struggle changed from month
to month, measure followed by countermeasure. The Pacific war was not
362
so marked. The American boats began to be equipped in large numbers

during 1942 with SD and SJ sets and retained them with slight modifica-
tions until the end; the SD was often ignored, the SJ treasured [5]. American
boats were warned in December 1943 that Japanese escorts had ‘excellent
radar’ [6], the 10 cm mark 2 model 2, but its introduction simply added
another form of alertness that commanders had to have. An evaluation
of the Japanese radar’s effectiveness requires analysis of data that do not
exist. It certainly had no dramatic effect. In this way it was very much like
the introduction of the 10 cm type 271 by the Royal Navy in the Atlantic
without PPI.
A submarine–escort fight, although certainly not a typical one, illus-
trates the way in which the SJ affected operations. Near noon of 18 October
1944 USS Raton under Commander Michael Shea encountered a nine-ship
convoy with three escorts southwest of Manila. Raton was forced down
almost immediately by an escort and on resurfacing found herself in a
completely obscuring torrential downpour, the onset of a typhoon. Shea
regained contact through the SJ and had maneuvered on the surface to the
middle of the cluster of ships by about 2030 hours, concealed by darkness,
driving rain and a first rate electrical storm. On obtaining a favorable po-
sition he began to turn the boat, sending torpedoes at targets at different
points of the compass until his six bow tubes were empty. At ranges of
less than 1000 m it required his radar skill not only to aim torpedoes but to
avoid ramming or being rammed in the confusion caused by his attack and
the increasingly heavy sea. He then emptied his stern tubes and cleared
the dangerous traffic to take stock and reload. He pressed home two more
attacks in seas with waves approaching 10 m. When the intensity of the
typhoon forced him to break off, he had watched five ships disappear from
the PPI [7]. One might suspect that one or more of the escorts had 10 cm
radar, but the Japanese set did not have a PPI display and, as similar ex-
perience had shown in the Atlantic, anything less is not much help in the
confusion of several moving vessels.
Japanese escort vessels were equipped first with radar, but the Bureau
of Naval Construction opposed putting it on the submarines even though
it was the same 10 cm set as used for surface vessels and required little
modification. Commanders, now familiar with its use by the Americans,
prevailed and the boats sent to the Battle of the Philippine Sea were so
equipped. They also received a modification of the mark 1 model 1 for
air warning in which the array was replaced by a single vertical dipole
for the transmitter and a steerable Yagi for the receiver. Japanese boat
commanders found their radar more useful in surviving than in locating
the enemy, as an example illustrates. Commander Hashimoto took I-58
with human torpedoes (kaiten) for suicide attacks on Allied fleet units
during the Battle of Okinawa, but his progress was slowed to such an
extent that he could not keep his batteries charged as a result of the large
number of crash dives necessary to avoid the aircraft that he picked up on
363
his radar. He finally gave up and headed out to sea to find other targets
for his spirited kaiten pilots [8].
The attitude of Japanese boat commanders contrasts with that of the
German, who wanted nothing to do with the Seetakt sets that were initially
adapted to U-boats in small numbers. It is true they were not as suitable as
either American or Japanese sets, but with a little encouragement GEMA
could have probably made them useful. Absence of surface-search radar
robbed them of the use of the night for maneuvering into firing positions,
which was fatal when they lost the day. American and Japanese boats
found night and day equally useful.
7.4.2. Destruction with bombers

It was obsolescent at the beginning of the war but still in production,
although a new design was in the works; it was slow, an advantage in
launching torpedoes but not in making bombing runs; it had weak defen-
sive armament and preferred to work at night; it could carry a heavy load;
it had the long range required for reconnaissance; it could absorb a lot of
punishment and still fly; although not ugly, as some claimed, it certainly
did not have the graceful lines that distinguished such wartime stars as the
Spitfire, the Me-109 or the B-17; equipped with ASV mark II it proved to be
one of the most valuable aircraft to fly for the Allies. An alert reader will
associate these words with the Royal Navy’s Swordfish torpedo bomber,
but they apply just as well to the US Navy’s PBY Catalina flying boat.
When E G Bowen wanted to demonstrate ASV mark II as part of the
Tizard Mission, it was only natural to select the PBY for the demonstration.
The Navy was so impressed that they ordered 7000 from Philco, designated
ASE [9], so PBYs were the first Navy planes to carry radar, and they made
it felt in the Battle of Midway and the Guadalcanal campaign. In the South
Pacific PBY crews had to learn how best to fight their great birds and
survive. To be caught by almost any Japanese plane in the daytime was a
serious affair, so they restricted such flying mostly to rescue work for which
radar could not substitute for vision and in so doing filled war diaries with
examples of heroism. But to attack an armed enemy during the day was
foolish, barring unusual circumstances, so they acquired night skills. In
doing this they had something nearly as important as their radar, the radio
altimeter that allowed them to approach the surface of the sea close enough
for a night torpedo attack.
Their skills in attacking ships at night evolved during the early part
of the Guadalcanal campaign, when cutting the Japanese line of supply
to that island was vital and any method was worth trying. Radar could
find the target but because of the reflection from the surface of the sea
lost it as the aircraft approached. This was the same problem that Coastal
Command had had in the Atlantic and that had led to the introduction of
the Leigh light. Such a light was not within the improvisation capabilities
364
of the crews in the South Pacific, but unlike a submarine a surface ship
cannot escape combat by diving, so the attacking bomber has time to make
repeat attacks. The difficulties of seeing an air target at night are much
greater than seeing a ship on the ocean surface. Obviously, this kind of
fight would result in AA fire from the targeted ship, which without radar-
controlled guns would be shooting at shadows unless the plane could be
fixed in the beam of a searchlight, yet bloody duels were not rare.
One of the first adaptations to this new role was the painting of the
PBYs black, with the appellation Black Cats following almost immediately.
This made them much more difficult to see and gave them a sinister ap-
pearance that reinforced the normal aversion to surprise nocturnal visits
by an enemy. The torpedo was the most accurately aimed missile for this
work, but the mark XIII aerial torpedo had as many faults as the mark
XIV submarine torpedo, so good aim even at the favored close ranges fre-
quently failed to cause an explosion; bombs could be expected to explode
but were hard to aim for masthead attacks. The withering fire of four
0.50 inch machine guns mounted in the bow proved more than enough
for some small craft, giving the Cat some properties of a fighter but none
of the agility. These homemade techniques were extended throughout the
South Pacific and accounted for the loss of hundreds of thousands of naval
and merchant tonnage [10].
Most of the credit for the destruction of Japanese shipping deservedly
goes to the Navy, but a large fraction must be credited to Army Air Forces
bombers. If anything had been proved concerning air power during 1942,
it was that high and medium level bombing of any nautical target was
useless. Army fliers and their Australian comrades had demonstrated
in the Battle of the Bismarck Sea that low-level daylight attacks by large
numbers of aircraft using skip bombing and strafing with heavy machine
guns and cannon were very effective against merchant ships and escorts,
but this method was ill advised for single bombers seeking to sink the many
ships in the southwest Pacific that did not move in convoys. In order to
have assurance of a hit, the bomber had to approach at altitudes that made
the AA fire of even a freighter deadly. The Black Cats had shown the way,
but the Army fliers thought they could do better and had cause to do so,
for reasons other than as part of the overall Allied effort to sink enemy
shipping. Single vessels, generally traveling at night and hiding during
the day were supplying the island garrisons that the ground troops had to
fight, and the Army fliers wanted to isolate the battlefield [11].
As the 10 cm search radar SCR-517 came into use in that theater, crews
of patrol planes noted the ease with which these vessels could be observed
at night yet could not use it to place a bomb on one. This was not the
fault of the radar, which did not suffer the surface reflection limits of ASV
mark II; there was just no way to aim a bomb with the 517. To remove this
deficiency Bell Telephone Labs began in July 1942 the design of auxiliary
equipment to be used with various microwave sets to provide automatic
365
bomb release at altitudes above 20 m, designated AN/APQ-5 but generally

called LAB for low-altitude bombing. This was somewhat similar to the
problem that Alvarez was facing at that time in designing Eagle for high
levels, but general aerial bombing required a mechanical analog computer
of considerable complexity and weight, and Alvarez settled on the Nor-
den. At low altitudes the slant range as measured by the radar is a good
representation of the ground range, which allows an approximation to be
made that is much simpler to calculate than the general case and suffi-
ciently accurate. After this approach was accepted in July 1942 the project
moved rapidly through the use of an electric computer designed along
lines that Bell had employed in the M-9 AA gun director. The bombardier
observed the target on a scope display showing range against bearing for
a beam swept from side to side. Range on the scope could be 30 km for
search and reduced to 1.5 km for the final bombing run; a marker on the
scope indicated the moment of release. Steering commands to the pilot
ensued the same way as for visual bombing [12]. It proved to be remark-
ably accurate. The approach of the attacker at night made defense without
radar-controlled guns, which the Japanese lacked, impossible. Generally
the only warning obtained by the ship was the bursting of bombs.
The reason LAB was such an accurate blind-bombing technique and
H2S and H2X so poor resulted from a ship being a single object on a rel-
atively smooth surface, much like an airplane in the sky. Radar bombing
of land targets suffered from an over-abundance of reflections from the
objects that make up a civilized landscape and that confuse the person
looking at the PPI scope.
A wing of B-24s was organized under Colonel Stuart P Wright, who
had vigorously encouraged the project, and equipped with 22 pre-prod-
uction models of these devices adapted to the new SCR-717 radar. After
intensive training at Langley Field they went, together with a Western
Electric engineer, to Guadalcanal and dispatched during the night of 27/28
August 1943 their first operational ‘snooper’, which located and destroyed
a small freighter a few hours after takeoff [13]. During the period from
December 1943 through April 1944 the ‘Wright Project’ destroyed 47% of
the enemy shipping credited to aircraft in that region with only 20% of the
machines and personnel having that assignment [14].
The LAB aircraft on Guadalcanal became part of the 13th Air Force,
and a similar unit was established in the 5th Air Force operating out of
Buna over the Bismarck Sea, but the champions of maritime destruction
were those flying with the 14th Air Force from Liuchow, China over the
Formosa Straits and the South China Sea. In the five months after their first
flight on 24/25 May 1944 this unit sank 250 000 tons of merchant shipping,
three cruisers and three destroyers in addition to damaging or possibly
sinking another 100 000 tons of shipping and seven naval vessels. Shipping
was so dense that targets frequently had to be selected from several, and
they had to ignore 60% of the ships sighted during their first two months.
366
Destruction of these ships cut off supplies sorely needed for the defense of
the Philippines. And this from a base where operations had to be ‘measured
in pints of gasoline and ounces of bombs’, owing to the extreme difficulty
of flying supplies from Burma over the Himalayas [15]. The destruction
caused by this handful of bombers became so painful that the Japanese
Army forced the Chinese back and captured Liuchow on 11 November
1944 thereby ending the 14th’s commerce raiding [16]. That such a measure
was necessary to stop the sinkings demonstrates the complete absence of
Japanese night-fighter capability, which could have worked easily from
Formosa and would have found no simpler adversary than a snooper. For
want of airborne radar an army had to move.
In early 1945 the United States had brought Japan to the state that
Dönitz had planned for Britain. It was a condition for which there was no
correction, as Japan had no allies to intervene and no conceivable means of
breaking the blockade. The hopeless condition of the nation had become
obvious to the Imperial Navy and the civilian ministers of the government
after the Battle of the Philippine Sea, but rule was not by reason but by
the bushido code of honor that sent hundreds of fine young men to their
deaths in knowingly suicidal gestures against an all-powerful enemy.
7.5. THE WIDE PACIFIC

By the end of 1942 the Allies could begin to think about victory rather
than about staving off defeat, but their resources in the Pacific hardly al-
lowed extravagant plans for the coming year. It was during 1943 that
the US Navy grew to great strength and presented the Japanese with the
ugly circumstance of which Admiral Isoroku Yamamoto had forewarned
them. The United States had not been defeated by the wave of victories
that had followed Japan’s forces during the first half of 1942, and now
the dreaded industrial capacity was making itself felt. The ‘decisive’ vic-
tory at sea might still come, or the war in Europe might favor the Axis
and draw resources from the Pacific, but worry about the future was now
predominantly an Axis, not an Allied occupation.
For Admiral Nimitz the arrival in Hawaii of the new fast carrier Es-
sex at the end of May 1943 was a turning point. Every movement of his
command during the preceding 18 months had been adjusted by fears for
his carriers. Not that they had been held back; there had been plenty of
action, as evidenced by five having been sunk. Replacements were follow-
ing, but Essex had been laid down as part of the pre-Pearl Harbor building
program that had given battleships and cruisers priority, and not many big
carriers were on the way. On 10 January 1942 a change in plans sought to
relieve what was obviously going to be a serious weakness by converting
a number of cruisers under construction into light, fast carriers, a design
that came into being quickly because cruiser hulls were available.
Much more ominous for Japan was the blunder in having not reck-
367
oned the high attrition of their elite corps of naval fliers. Intended, as the
strategy demanded, to carry the Empire to early and dramatic victory, they
had been superbly trained but now few remained. Their skill had taken
advantage of the long range and agility of light-weight aircraft having
neither bullet-proof fuel tanks nor armor, but their unskilled replacements
found these advantages elusive when sent against well trained and rapidly
seasoned adversaries flying heavy, tough machines. The grand mistake of
not having provided adequate flying schools, while America turned out
thousands of pilots, spelled doom for the island Empire.
7.5.1. American radar, 1943–1944

As Nimitz’s fleet expanded in ships, planes and men, it also grew in the
capabilities for electronic warfare. The 1.5 m XAF that had so impressed
line officers during the 1939 Caribbean exercises had gone into production
as CXAM and made a simple evolution into the SK with peak power in-
creased to 200 kW and a PPI display for its new panoramic searching. It
could pick up a single bomber 3000 m high at 160 km; installation began in
January 1943 [1]. The ‘flying bedsprings’ were recognized by every seaman
as sure guardians.
Valuable as the CXAM had been in the carrier battles of 1942, it was
incapable of helping the fighter direction officers when things moved close
to the ship. Missing were resolution of even moderately spaced aircraft
or any height information. It was a task for which only microwaves suf-
ficed. When the problem reached the Radiation Laboratory the solution
was obvious. The 10 cm SCR-584 was moving toward production, and its
ability to track a plane in three dimensions was exactly what was needed.
Transferring the 584 to sea duty confronted the designers with two require-
ments not encountered for the Army application: (1) a carrier is not stable,
and stability is essential for obtaining meaningful tracking data, and (2)
an antenna aboard ship must be strong enough to function after having
endured a storm, the soldier’s option of retracting the dish into the trailer
not being practical. The first problem was solved with a platform that
maintained stable directions for three axes so as to eliminate roll, pitch
and yaw, achieved through servo-mechanisms guided by gyroscopes; the
second was solved by making a strong paraboloid. The resulting proto-
type, the CXBL, was mounted on the new Lexington in March 1943, and the
first two General Electric production sets, designated SM, went on Bunker
Hill and Enterprise in October [2]—several months before SCR-584 reached
a similar stage of production. Unfortunately, the solution of the stability
and strength problems made SM rather heavy, specifically 9 tons. It was
just one radar too many for the light carriers, making them top heavy, a
property not admired by seamen, and the light carriers rolled enough as
it was, so only the heavy carriers received this fighter-direction capability.
Rad Lab reduced the weight of SM by half in the SP, which was installed
368
on lighter vessels in late 1944 [3].

Radar use progressed so rapidly during these early years that many
things were left to the users. Admiral Nimitz had established schools for
operators, mechanics, fighter-direction officers and commanders at Pearl
Harbor, but doctrine was not settled. Radar was so new that even the most
recent graduates of the Naval Academy had learned nothing about it, and
this made it increasingly the province of reserve officers. This led from
time to time to the delicate task for the radar men of educating officers
who were both regulars and their superiors in rank [4].
On carriers the Combat Information Center had grown out of the
Radar Plot room that had naturally formed. The radar set was standard
issue, but everything else had to be worked out by the new users. All
required a method of plotting; some chose horizontal tables that mimicked
in much reduced space the Battle of Britain filter rooms, some vertical
sheets of clear plastic on which a seaman wrote mirror-image text. Radar
personnel arranged their accessories for most efficient operation according
to their own concept, much as one arranges the furniture of an apartment
for ease of living, and practice varied from ship to ship. Alert teams soon
had their maintenance shops outfitted with lots of non-issue tools and test
equipment, and those who knew anything about doing tedious work in
the tropics managed to get air conditioning [5].
By the end of the war Combat Information Centers had become big
business. They were provided large compartments with space for as many
as 50 inhabitants. It became the location where knowledge of an action
was concentrated and replaced the bridge as nerve center [6].
Determination of target height with air-warning sets was an art stud-
ied assiduously, as it was the most difficult coordinate to determine. Height
had to be extracted from the manner in which signal amplitude changed
with range as the target passed the different vertical lobes of the SK. There
was a general theory, but it had to be checked against calibration flights
in which an aircraft flew at a given altitude, then again at another altitude
and so on. This was a time-consuming affair that was not popular with the
flight officers, and success of a ship’s radar group turned on the eloquence
of their expression of the need and importance of this flying [7]. This kind
of height determination needed a not-furnished stable vertical axis for the
antenna or the motion of the lobes would confuse the lobe patterns, but the
surface conditions of the Pacific were on the average the best encountered
for meter-wave equipment at sea.
Secure line-of-sight communication was a minor side benefit of radar
employed by some vessels in the Pacific. The CXAM had a switch for
changing the pulse repetition frequency as an aid to determining whether
a pip had been returned from beyond the maximum display range. Re-
placing this switch with a telegraph key and listening to the receiver with
ear phones allowed two CXAMs to pass signals between one another in
code. The very high frequency and the narrow beam made interception
369
by the enemy unlikely [8]. When submarines began to fight as wolf packs
they used the SJ in this way to communicate with one another when sub-
merged [9].
The SG had become the loved one of deck officers for new and un-
expected reasons. The charts of the Pacific islands were distinguished for
their gross and dangerous inaccuracy. Close approaches to shore were the
rule not the exception, which made the maplike PPI display of SG a beau-
tiful sight for a skipper proceeding, perhaps a bit faster than he would
have preferred, toward an unknown, poorly mapped coast. It could be a
tricky even with a perfect radar because low-lying land in the foreground
could be missed and higher ground farther back mistaken as the shore [10].
The adoration of SG did not withhold criticism, if anything it amplified it;
waveguides sometimes became lossy when misaligned by the shock of
gunfire, and more power was needed to increase range. Raytheon began
production of the SG-1, having tightened the joints and raised peak power
to 50 kW, in May 1943 [11].
Main battery fire control received the 10 cm mark 8 (a Bureau of Ord-
nance designation that replaced the seldom used FH), which had gone to
sea in prototype from Bell Labs as CXBA. It was the first phased-array
microwave radar, allowing electrical rather than mechanical scanning. Its
15 to 20 kW peak power allowed any vessel within gun range to be taken
under blind fire. Electrical scanning gave the operator a continuous presen-
tation of the 30◦ sector toward which it was pointed and simplified lobe
switching. It allowed fall-of-shot corrections to be made while keeping
other targets under observation, preserving thereby in more convenient
form an advantage of the FC and FD. The beam was only 2◦ wide, and the
pulse width only 0.4 µs, giving excellent resolution and removing the con-
fusion that often accompanied multiple targets with the 40 cm sets. Data
were presented on two scopes: one gave target amplitude as a function
of range, the other range plotted against bearing. The mark 8 drew en-
thusiastic responses from all gunnery officers who had the opportunity to
compare it with the FC and FD. Production began at Western Electric in
October 1942 [12].
The antenna had a curious appearance that resulted from an inge-
nious design. Bell had used a phased-array antenna for steerable short-
wave transatlantic telephony before the war, and the engineers saw pos-
sibilities for using the basic idea with microwaves. They also introduced
a new kind of radiating unit, the polyrod, a properly dimensioned rod
of polystyrene inserted into the open end of a waveguide, which radi-
ated continuously along its length. George Southworth had observed the
propagation properties of dielectrics in his earliest microwave work. A
few experiments showed that rods could be shaped to give the directivity
needed for the mark 8 [13]. The fixed array of the mark 8 consisted of
14 identical horizontally arranged elements, each a vertical array of three
polyrods; the phase to each element was controlled by mechanically rotat-
370
ing phase changers. The design accomplished nicely the important goal of
suppressing unwanted side lobes [14]. It also seems to have been the only
use by the Allies of the polyrod radiator. It is worth noting that the limited
German microwave work employed it [15].
The Navy’s version of ASV mark II, the ASE, had become standard
equipment for the PBY flying boats but was ill suited for the torpedo
bomber TBF Avengers. NRL had had their own 60-cm air–surface radar in
development when the ASE was adopted and completed it for these new
carrier planes. Like the ASE this set, the ASB, used two Yagis directed 7.5◦
right and left of the forward direction [16].
The introduction of radar-equipped night fighters onto carriers was a
problematical thing. The first sets available were the 3 cm ASH (AN/APS-
4), search equipment that needed a separate crew member as radar operator
because of the complexity of operation. Attempts were undertaken with
the Avenger, which satisfied the requirements as the radar platform but was
too slow to serve as the gun platform. Fliers from the Enterprise attempted
to correct this by having two F6F Hellcats accompany the radar plane in
breaking up a night torpedo attack on the task group in November 1943.
The result was successful in shooting down two bombers and thwarting
the attempts of the third, but the confusion presented by the two fighters
without radar may have led to one of the American fighters, flown by a
distinguished Ace, E H O’Hara, being shot down in error [17].
Later a new set, AIA-1 (AN/APS-6), was mounted in single-seat fight-
ers and designed as an airborne interceptor. The set was simplified so that
the pilot could operate it alone and a pair of red-filtered goggles preserved
his night vision when viewing the 50 mm diameter scope on the instru-
ment panel [18]. Carrier night fighters were used sparingly because keep-
ing them aloft put additional requirements on already overworked deck
crews. Other ways were found to be effective against the dangers of night
torpedo-bomber attack.
A weakness of American electronic arms that was closely associated
with radar was the continued use of high-frequency radio for combat
aircraft—more than four years after the Royal Air Force had replaced it
in Fighter Command with crucial very-high-frequency equipment! These
old radios had been the curse of the carrier operations off Guadalcanal be-
cause of their inherent noise, erratic transmissions and the limited number
of channels they could carry. Change was in progress, but fighter directors
still had to deal with mixed systems in June 1944. Fortunately, communi-
cation discipline had been greatly tightened by then.
7.5.2. Japanese radar, 1943–1944

Japanese surface units were being equipped during these years with 1.5 m
mark 2 model 1 air warning, a dipole array following the design of the
land-based set, mark 1 model 1, and as such was similar to the general
371
principles of Freya and SCR-270, although of shorter range. These instal-

lations were not notably successful, owing to their not being sturdy enough
for use aboard warships. The set mounted on the 18 inch gun battleship
Musashi suffered broken antenna insulators and the failure of transmit-
ter tubes when the vessel underwent firing practice off Katsuru Island in
September 1942, which naturally enough sent criticism to the Naval Tech-
nical Research Department (NTRD) [19].
The 10 cm mark 2 model 2 had characteristics that distinguished it
from Allied microwave equipment. It used a low-power magnetron in-
stead of a klystron as local oscillator, the anode of which had nearly the
same arrangement of holes as the main oscillator but was made of thin
metal that could be deformed; this allowed the frequency to be altered just
enough to hold the intermediate frequency of the heterodyne receiver con-
stant. The waveguides used in all Japanese sets were circular; transmission
and reception were in separate horns; parabolic reflectors were not used.
This unit had made a good name for radar in the Imperial Navy.
As a result of the insistence of Commander Yoshio Sakura, the earnest
Communications Officer of the Chief of Staff, First Fleet, an experimental
model of mark 2 model 2 had been installed in the cruiser Atago while she
was undergoing repairs for injuries received in action off Guadalcanal in
November 1942. This radar had been of great use during an attempt by
seven cruisers to disrupt the November 1943 Allied landings at Empress
Augusta Bay on Bougainville. The cruisers had been driven away by air-
craft, but the Atago’s well trained radar crew had allowed them to thread
their way unscathed out of the extremely hostile zone [20].
The strains of production and development caused reorganization
at the NTRD. The Electric Wave Research Department became its largest
element and the one with the largest share of troubles. Having only begun
the first serious radar work in 1941 this group now had to compete with an
enemy that had begun in 1935. The Imperial Navy’s concern is reflected in
the placement of Vice Admiral Takeshi Nawa as NTRD chief in mid–1943.
His appointment resulted in part from dissatisfaction in the way Ito con-
ducted the more practical aspects of radar, and Nawa made an extensive
reorganization. He faced the problem of keeping design up to the latest re-
quirements and production adequate for a huge demand as well as training
operation and maintenance personnel [21]. It was a substantial order.
The new organization did not produce the results desired, and the
Navy Minister created in March 1944 the Electronics Development De-
partment to assume responsibility for all research and testing of radar,
communications and sonar equipment. This move was strongly opposed
by the units affected, their argument being that slow progress had come
from the lack of qualified personnel and material, not from organizational
defect. The new department still controlled the same operating labora-
tories as before; the absence of noticeable progress would bring another,
equally futile change the following year [22].
372
Matters were not helped by the attitudes of the authorities. Shigeru

Nakajima was responsible for microwave development at Japan Radio
where the cavity magnetrons had been invented in cooperation with
NTRD. At the beginning of the war his staff numbered 800 engineers but
was progressively reduced by inductions into the services to half that by
the end [23].
Knowledge of the Metox receiver arrived in Japan aboard the German
U-511 in August 1942, and more than 2000 of them were soon distributed
within the fleet. Inasmuch as they did not respond to microwaves, the
same deficiency that soon proved so troublesome in the Bay of Biscay, they
were of no more use against the ever increasing numbers of SGs than the
U-boat sets were against ASV mark III and quickly lost favor [24].
In February 1944 the Americans learned that the Japanese twin-
engine Betties had radar, the 2 m mark 6 model 2 (known in prototype
as the H-6), with a Yagi antenna on the nose and horizontal dipoles on
each flank of the fuselage [25]. This general search set was developed at
the Naval Air Technical Depot at Oppama to a great degree independently
of NTRD. It had a peak power of 2 kW and weighed 60 kg. The Depot also
produced the 60 cm FD-2 for night fighters having a peak power of 2.5 kW
and weighing 70 kg. There is little evidence of the employment of FD-2
[26].
It is much more difficult for countermeasures people to identify air-
borne radar. A ground radar station remains fixed while its wavelength,
pulse width and repetition rate are studied, but an airborne set is quickly
gone. The usual method is for a Ferret to fly as bait and, if attacked, use
the advance warning to evade destruction. Hunting for Japanese airborne
radar occupied numerous Ferret flights in late 1943, precipitated by re-
ports of night attacks of mysterious origin, but no Japanese planes were
so detected. The first information about the mark 6 model 2 mounted on
Betties, which was ill suited for use as an interceptor, came from the same
source that disclosed the existence of the 10 cm mark 2 model 2, a large
number of documents captured in February 1944 on Kwajalein Atoll in the
Marshalls [27].
Japanese radar suffered seriously from a shortage of most important
electronic components and of the outstandingly poor quality of much of
what could be obtained. To meet the increased demands vacuum-tube
manufacture had to be subcontracted to small companies with little ex-
perience and with unsatisfactory results. Nickel was so short that coins
of that composition were bought pretty much to extinction and recast. In
August 1943 the two services were so alarmed at the shortage of materi-
als that they set aside their grudges and formed the Army–Navy Radio
Technology Committee (Riku-Kaigun Denpa Gijutsu Iin Kai) made up of
Professor Hidetsugu Yagi of Osaka University, Professor Yasushi Watan-
abe of Tokyo Imperial University and Professor Hantaro Nagaoka of the
Institute of Physical and Chemical Research [28].
373
7.5.3. Carrier raids

The carrier raids that had preceded the Battles of the Coral Sea and Midway
had left a strong and very positive impression on nearly all members of
the Pacific command. They had kept the enemy guessing, had kept the
precious carriers away from prying eyes and had provided an unparalleled
form of training for well schooled but green fliers—all possible because of
the CXAM at the foremast. So carrier raids became the norm when an
invasion or fleet action was not impending. They also served to reduce the
aircraft population of the island bases. The process picked up a name in
South Sea pidgin of ‘makee learn’.
The element of complete surprise seldom came with these strikes
because the islands soon had air-warning radar, but the defenders were
still rattled when weeks of quiet duty were shaken by a 20 minute radar
warning. One of these—and it was on what Americans incorrectly thought
to be the Pacific Gibraltar of Truk—caught the defenders with most planes
on the ground. After the war it was learned that a lonely freighter loaded
with radar for Truk had gone down from a torpedo of USS Trout a couple
of months earlier [29]. The radar defenses of these bases did not include
fighter control, not even for the fortress that Rabaul had become [30]. The
reason for this probably lay with Japanese doctrine that was similar to the
German of 1940. Furthermore, no radars capable of returning the three-
dimensional coordinates needed by controllers were available until 1944.
No operational Japanese set had PPI.
One of these raids afforded an unexpected and valuable technical
evaluation. Astrike on Saipan and Tinian was ordered for Mitscher’s group
on 22 February 1944. On the afternoon of the preceding day a snooper had
spotted them and got away, so it was obvious that surprise had been lost,
and experience had taught them to expect a night torpedo-bomber attack
for which the Japanese had demonstrated competence. One of the early
1942 carrier raids had been called off when it was realized that they had
been seen, but two years had altered matters. There were night fighters
aboard, but Mitscher did not wish to miss the advantages of a dawn attack
by reversing course to launch and recover by sailing into the prevailing
easterly winds. He decided that night would force the attackers to fly
relatively straightforward paths, perfect for his FD radars, directors and
guns firing proximity fuzes. He ordered the carriers surrounded by an
inner circle of cruisers and an outer circle of destroyers—all having the
most modern AA weapons. As the enemy aircraft appeared as expected,
the fleet maneuvered by order from the knowledge gained by radar to
present the most difficult torpedo targets and the guns opened fire. None
of the ships suffered damage, and only five of the 20 attackers returned
to the home base. The raid proceeded on schedule and accomplished
the desired results [31]. This way of dealing with night torpedo bombers
became standard [32]. The improvement of AA artillery over that of only
three years earlier was startling to an extreme degree.
374
7.5.4. The invasions of islands

A consequence of the Doolittle Raid on Tokyo was a residue of doubt in
the Japanese Command as to the origin of the attack. Aircraft carriers were
certainly considered as the most likely base, but because the planes were
the normally land-based B-25s the thought remained that they might have
come from the Aleutians. The desire to remove such a threat affected the
planning of the June 1942 offensive against Midway by supporting the idea
of sending a diverting force against these islands. Two carriers were sent
to attack the known garrison and airfield at Dutch Harbor while troops
were to be landed on the western-most islands of Attu and Kiska. The two
carriers accomplished little and were sorely missed at Midway.
The attack on Dutch Harbor did not catch the defenders by surprise,
as they had been alerted by intelligence from Nimitz and a timely warning
by the SC radar of a seaplane tender. The attacking planes from the Ryujo
met lively AA fire, which distracted them enough from their work to keep
the bomb damage minor, if smokey. Planes from the Junyo were unable
to find Dutch Harbor, hardly surprising given the weather conditions. It
was, in fact, remarkable and showed excellent skill that all but one of the
aircraft launched were able to return to their ships, the one having been
shot down by the waist gunner of a PBY [33].
Kiska became the first enemy radar to be accurately mapped by a
fledgling electronic countermeasures unit. A B-24 equipped with an SCR-
587 intercept receiver flew three Ferret missions around the two western-
most islands of the Aleutians in early March 1943 and returned with the
knowledge that Kiska had two radar sets but Attu none. They produced an
excellent contour map of signal strength that showed the most favorable
directions of approach. Their efforts were followed by attempts to destroy
the stations, but the Japanese were able to repair the damage relatively
quickly [34]. Determining the radar potentials of island bases became a
standard component of military intelligence for the remainder of the war,
generally using cast-off aircraft but with growing importance and support.
So began a three-way war between the United States, Japan and the
weather, with the weather the most vicious and dangerous. It was a cam-
paign greatly assisted for the Americans through the rapid equipping of
PBY flying boats with ASV mark II, and it was from them that Admiral
Charles McMorris learned on 26 March 1943 that a large, well escorted
convoy led by Admiral Boshiro Hosogaya was making a major attempt to
break the blockade imposed on Attu and Kiska. Action, called the Battle
of Komadorskis, opened at 20 km with accurate early salvos of the Amer-
icans as a result of radar fire control. In the long-range gun battle that
ran for nearly four hours the out-numbered American squadron ended
badly damaged with the heaviest unit dead in the water but concealed by
smoke. Hosogaya, who had hurts of his own, was unaware of his superior
position, broke off the action and ordered the convoy back [35].
The Pacific War was to have many landings by Allied troops on the
375
beaches of defended islands. The first of these was the invasion of Attu
[36], begun on 11 May 1943 by the 7th Infantry Division, fresh from tropical
training that was intended to throw spies off the trail. Whether freedom
from spying served these men better than preparation for the conditions
they were to encounter does not seem to have been evaluated. The Japanese
and the weather provided the invaders ample resistance during an ugly,
18 day fight that showed what things would be like on such places as Iwo
Jima.
After the Battle of Komandorskis and the loss of Attu the Japanese de-
cided to evacuate the 1200-man garrison on Kiska. This was accomplished
on 28 July by 19 vessels under Admiral Masatomi Kimura that evaded the
American blockade, piloted themselves into the harbor, loaded all person-
nel and departed just as quietly. It was a brilliant performance followed by
an elaborately planned and executed American invasion on 11 August that
found nothing but a couple of dogs. Despite PBYs with ASV and patrol
ships, the evacuation had evaded interception. The key technical element
for the Japanese, used both for navigating through fog and evading the
blockaders, was the 10 cm mark 2 model 2 radar [37].
Associated with the evacuation of the Japanese garrison was a strange
radar naval engagement—the Battle of the Pips. On 22 July 1943 a PBY
reported radar contact with seven vessels, and an American group includ-
ing two battleships and five cruisers headed for the reported position on
a wide front. At 0043 hours on 26 July the Mississippi picked up three or
four large contacts on the SG at 30 km. This was followed shortly by the
FC fire-direction radar reporting five or six. The contacts looked similar
to those returned by cruisers or destroyers but were intermittent and un-
steady, which was reported to the bridge. At about the same time other
vessels reported similar radar sightings. Fire was opened shortly after
0100 at a range of about 23 km by those vessels capable of firing under
radar control. It became increasingly clear to the radar crew of the Mis-
sissippi that something was seriously amiss. As they had narrowed range
the SG contacts retained the same amplitude instead of strengthening, but,
much more serious, the echoes of the projectiles and the splashes were
stronger than those of the presumed targets. Fire ceased after 25 minutes
when it became clear there was no enemy there [38]. When visual observa-
tion cleared, moonlight and star shells disclosed no targets. Examination
of the ocean surface after daylight found no wreckage, no oil slicks, no
survivors—nothing but clean ocean surface [39].
This embarrassing action was caused by anomalous propagation to
ranges such that the blips on the scope had returned a few sweeps after
the original. They had encountered this phenomenon frequently with the
40 cm of the FC but never with the 10 cm of the SG. The Radar Officer of
the Mississippi later reconciled the contact bearings as echoes from Kiska
Volcano and the islands of Segula, Little Sitkin and Semisopochnoi [40].
Had the SG and FC sets had a switch to alter the pulse repetition rate as
376
did the CXAM and the SK, the Battle of the Pips would not have taken
place. Pulses from targets within the first sweep are unaffected by such a
change, but those returning from targets beyond jump on the display as
soon as the switch is pressed.
Observing critically the Attu and Kiska landings was Marine Lieu-
tenant General Holland Smith, who was to become the master of amphibi-
ous warfare. With army, marine, naval and air forces he would develop
this technique to its ultimate. The Gilbert Islands were to be his first test.
Two strategic directions had evolved by late 1943. MacArthur was
to clear the Solomons and parts of the Bismarck Archipelago in order to
move across the northern part of New Guinea through the Halmaheras to
the Philippines. Nimitz was to drive through the rings of island bases to
join with MacArthur for the final drive at a location to be decided. In the
original expansion Japan had garrisoned many locations throughout the
Pacific and was confident that each would demand a dear price for each
insignificant atoll. But once these bases lost their aircraft and could receive
no resupply, they became prison camps administered by Japan. It then
became clear that only those that afforded harbor and air fields for the next
step across the Pacific’s continental-sized distances need be taken.
The invasion of the Gilberts took place in late November 1943 as the
first step in Nimitz’s plan. The two islands of Tarawa and Makin received
a heavy pummeling by bombers and ships’ artillery, but the effect seemed
to have hardened rather than softened the resistance of the garrisons. The
attackers made no small number of blunders, and Tarawa became inscribed
as a Marine Corps memory of courage and blood. The Marshall Islands
were taken in early 1944 in a more workmanlike manner. The Marianas
were to be next, and it was assumed and hoped that attack on them would
precipitate a major fleet action.
While this was going on MacArthur secured the northern coast of
New Guinea with troops of the US, Australia and New Zealand. His path
was eased by much of the enemy’s naval and air forces that might have
faced him having been diverted to help check Nimitz’s drive from the east.
The FC and FD fire-control radars were of little use in identifying
land targets for the increasingly important artillery function of bombard-
ment preparatory to landing. Saturation bombing and rocket attacks were
spectacular and could be demoralizing, but the need was for accurate fire
on targets known to present danger. Such targets could often be identified
from aerial photographs but not recognized from the ship either visually
or electronically. A solution presented itself as forces began moving up
the Solomons in 1943. Prominent land features could be identified on the
aerial photographs and seen by radar. This allowed offsets to be calculated
so that fire could be directed on dangerous emplacements. The method
worked optically too, but accurate ranging made radar preferable even on
clear days [41]. It worked even better when the 10 cm mark 8 fire-control
radar became available.
377
Wherever a landing was made the Australian LW/AW was on the

beach soon, augmented in early 1944 by the light-weight SCR-602. They
were particularly valued in the New Guinea theater where control of the
air was more hotly disputed than in the Central Pacific. Once the islands
were secured, local air defense relied on a combination of the SCR-270
and SCR-268; the SCR-602 was considered expendable, but the LW/AW
often served when the heavy 270s and 268s could not be got into position.
The defenders liked the long range of the 3 m 270 for warning and found
the resolution and height data of the 1.5 m 268 good for fighter direction.
Australian Mobile Fighter Control Units and American Signals Warning
Battalions provided fighter direction whenever it was needed. They natu-
rally assumed a function that endeared them to all Allied airmen—guiding
lost fliers home and locating downed aircraft. The number saved is not on
record, but it is very large.
The SCR-268 may have had its most successful day in directing the
90 mm guns of the 9th Marine Defense Battalion in defending the Rendova
(New Georgia) landings in July 1943. Of 16 attacking bombers gun-fire
downed 12 with an expenditure of only 88 rounds, leaving the four that
escaped to be destroyed by fighters [42]. The island locations were ideal
for meter-wave equipment, and crews became so proficient in their use
that, unlike their comrades in Europe, who seldom had an excellent sea
surface to form consistent vertical lobe patterns, they greeted the arrival of
microwaves with indifference and relied on meter waves till the end [43].
7.5.5. The Battle of the Philippine Sea

Vice Admiral Jisaburo Ozawa began assembling the Mobile Fleet in May
1944 at anchorages at Tawi Tawi, an island off the easternmost tip of Borneo.
It was by any previous measure of naval might a powerful body and clearly
intended to fight the ‘decisive’ battle that would dispose of the American
fleet with one blow. It was obvious the United States was going to make an
important move, most likely the invasion of strategically important islands.
The Mobile Fleet would proceed to attack the landing, forcing the American
into the desired major action. The American Admirals planned things the
same way. Both sides sought a big fight, each confident of victory.
The Japanese were inferior in numbers of carriers and aircraft, dis-
advantages that could have been made good by adroit tactics, but they
suffered other disadvantages that were to make the outcome for them an
overwhelming defeat. The splendid carrier fliers with which Japan began
the war were largely gone, and their replacements were ill trained, and
training faced the same serious fuel problem as the Navy as a whole. Al-
though Japan possessed huge petroleum resources in Borneo, American
submarines were keeping it out of the homeland and the refineries with
the result that both ships and planes were desperately short by mid 1944.
Naval aviation suffered doubly because the skills needed by carrier pi-
378
lots had to be acquired by taking off and landing on carriers, something

that could only be done with the ships under way. Training carrier pilots
consumed enormous amounts of fuel.
The Japanese were also behind in the techniques of submarine war-
fare. Their anti-submarine forces never approached the skill of American
and British, which had proved themselves slow but assiduous pupils in
the hard school maintained by Karl Dönitz. Furthermore, the actions in the
Solomons had cost Japan a disproportionate number of destroyers. Thus
when an American submarine’s contact report of the assembly of the Mo-
bile Fleet off Borneo brought enough submarines to cause the fleet losses
and keep Nimitz well informed, the Imperial Navy could not keep them
off. A line of Japanese submarines posted to the east for the same pur-
pose was discovered and decimated by US hunter–killer groups taught by
Atlantic veterans.
Ozawa knew he was outnumbered in carriers and carrier planes but
was confident he held an important advantage in the large number of
land-based aircraft located at island bases capable of cooperating with the
fleet. These unsinkable aircraft carriers had long figured in Japanese and
American war plans, initially being highly regarded by both sides, but
the success of the carrier raids had reduced American fears. Unsinkable
they were to remain, but immobile as well and thus waiting for attack.
As the Battle of the Philippine Sea was to emphasize, a carrier without
aircraft is worthless, and unless it can be resupplied, as was unsinkable
Malta, it remains worthless. Ozawa’s unsinkable carriers were placed in
this category in a series of attacks over much of the southwest by bombers
of the Army Air Forces and the carriers, combining to destroy much of
Osawa’s air power as well as confuse him about the goal of the obviously
forthcoming invasion of someplace, revealed later to be Saipan.
There were five major carrier actions in the Pacific War, two of them
American victories of overwhelming degree, both under the command of
Raymond Spruance, a battleship admiral possessed of an austere person-
ality remarkably similar to that of George Marshall. The situations for him
at Midway and the Philippine Sea were inverted. At Midway he knew the
position of the Japanese and was able to make the crucial first strike; in the
Philippine Sea he was until the very end in the dark about the location of
Osawa’s ships, the consequence of the significantly shorter range of Amer-
ican carrier planes and the prevailing easterlies that allowed the Japanese
to launch and recover aircraft without reversing course—the modern sea-
man’s equivalent of the weather-gauge advantage of the age of sail. By
1944 Spruance no longer looked on this with the concern of 1942 because
he understood and was confident in the power of defense: radar would
give him adequate warning to intercept attacking formations well away
from the ships; planes that passed this line of defense would encounter
radar-directed 5 inch AA fire using the proven and respected proximity
fuzes; survivors would be received next by 40 mm guns having the excel-
379
lent Draper sights and finished with the 20 mms. His confidence proved
well founded.
On the morning of 19 June Osawa dispatched four major attacks,
all of which were spotted by the SKs in time for the fighter directors to
place ambushes of Hellcats, which removed a significant number of the
attackers. The play followed the script in each successive phase with little
damage or loss of life to the defenders resulting. Of particular importance,
all carriers remained operative. Much credit goes to Lieutenant Joseph
Eggert who functioned as task-force fighter director aboard the Lexington.
He was aided by Lieutenant Charles Sims, who was fluent in Japanese and
who listened to the commands of the airborne Japanese air coordinator
and informed Eggert of his every word.
Despite the serious losses Osawa was known to have suffered, the
Americans wanted desperately to sink enemy carriers. We can now say
this was a waste of effort for the battle was over; Japanese carrier power
was irretrievably ruined whether empty carriers existed or not, but such
talk is hindsight and would have been violently rejected on 20 June, when
the enemy fleet was finally located near the end of the day. A dramatic
attack was launched that succeeded in destroying a carrier. The retrieval
after dark of the returning planes marks a romantic peak in naval arms,
but the strike’s effects were not worth the losses. Three Japanese carriers
were sunk in this great battle, but two went down from submarines.
The 19th and 20th of June 1944 encompassed the most intense air
battle ever fought. It is impossible to give the number of aircraft involved
both from the uncertainty in deciding who was and who was not engaged
and from the uncertainty of the number of Japanese planes retained by
the island bases at the start, but one can give the numbers carried by the
two fleets with some assurance. Osawa had a total of 473 aircraft at the
beginning, Spruance had 956; of these Osawa lost 426 to which might
be added about 50 more from the island bases, Spruance lost 130, three
quarters in the attack of the second day with its high loss of returning
planes [44]. It is more difficult to establish what portion of Japanese losses
fell to AA fire, although significantly less than those shot down by fighters,
and much more difficult to separate the results of radar-directed proximity-
fuzed 5 inch guns from the 40 and 20 mm automatic guns. The radar-
directed fire was effective at ranges beyond those for the automatic guns,
which were also the ranges at which the courses were slow enough for
the manual-tracking FDs and especially for the mechanical-analog mark 1
director to follow. These directors had become by then the recognized
weak link in the radar-director–gun-fuze chain. Dramatic improvement
was on the way in the form of the mark 57 being designed at the Johns
Hopkins Applied Physics Laboratory by the team that had provided the
proximity fuze, but it would not be seen in the war zone until January 1945
[45].
Radar had transformed carrier warfare yet again. Its great strength
380
throughout the war was defense, and the SK, the improvement over the
XAF bedsprings demonstrated in the Caribbean five years before, had de-
feated the might of Imperial Japan off the Marianas as surely as the long
bow had defeated the French at Agincourt [46].
This great carrier battle had settled the fate of the Marianas. Without
control of the sea the garrisons on the various islands had no hope of relief,
but this in no way affected the determination with which they fought off the
invaders. In the early morning of 7 July the few thousand Japanese troops
that remained made a convulsive and suicidal attack on the American lines
that caught some Army units off guard but which was cut to pieces within a
few hours at the usual ten-to-one losses. After that their commander killed
himself. Separately the naval commander of Saipan did the same. His
command was sufficiently small, a few patrol boats and shore personnel,
that we would normally not record his passing in this mere outline of the
military struggle, but he was Vice Admiral Chuichi Nagumo, 26 months
earlier hailed as one of the greatest admirals the world had ever known.
What must have been his emotions in seeing the battleships he had sunk
at Pearl Harbor firing at Saipan from off shore?
Saipan had a large civilian population, a large fraction of which—
men, women and children, young and old—killed themselves in mass
suicides, although in nothing like the proportion of the military personnel.
Such was the Pacific War to be.
The word ‘radar’ entered newspaper articles in June 1943, having
been recently made public by Britain and America in a joint release. At that
time the Navy allowed an interview of NRL engineers that disclosed many
theretofore secret details [47]. In November 1944 a US Navy recruiting
poster stated: ‘Young men wanted by the Navy for training in RADAR,
one of the newest and most exciting developments of the war—with a great
postwar future [48]’.
381
PHOTOGRAPHS: AIRBORNE RADAR
An AN/APS-4 mounted under the wing of a Curtiss Seahawk. This 3 cm radar,

which functioned for surface search or aircraft interception, was enclosed in an
air-tight container that could be fastened to a bomb rack, indicated by an arrow.
It was designed by the Bell Telephone Laboratories and manufactured by Western
Electric. The photograph shows the aircraft hoisted aboard USS Chicago on 14
August 1945. National Archives photograph 80-G 700315.
1.5 m ASV mark II radar mounted on a Swordfish. This is a combination of two of

the most valuable weapons the Allies had in the war at sea. It was a combination
that cut off Axis supplies to North Africa in 1941 and 1942 and stopped the
Bismarck for destruction. When flown from escort carriers it became an important
weapon against the submarine. The ‘towel rail’ transmitter dipole is seen at the
center of the leading edge of the top wing. One of the two receiver antennas can be
seen attached to the outer starboard wing strut; the receiver antennas were Yagis
directed off axis to allow lobe switching. The photograph was taken aboard USS
Wasp in April 1942. National Archives photograph 80-G 7093.
382
1.5 m ASV mark II radar mounted on a Supermarine Walrus reconnaissance

plane. The ‘towel rail’ transmitter dipole is seen at the center of the leading edge
of the top wing. The two receiver Yagis are seen attached to the outer wing struts
and are directed outward to allow lobe switching for determining direction. The
photograph is dated 3 November 1941. National Archives photograph 80-G 25014.
1.5 m ASV mark II radar mounted on a PBY. This is a combination of two very
valuable weapons the Allies had in the war at sea. This British-designed radar gave
the US Navy its night vision when mounted in these long-range patrol bombers.
The transmitter antennas were mounted on both sides of the hull, below and behind
the pilot’s window. The receiver antennas were mounted beneath the wing on both
sides; they were Yagis that lay in a plane parallel to the one defined by the wing.
The date of the picture is 8 March 1942. National Archives photograph 80-G
403256.
383
The first 1.5 m ASV mark II radar mounted on a PBY. This is a combination of
two very valuable weapons the Allies had in the war at sea. This British-designed
radar gave the US Navy its night vision when mounted in these long-range patrol
bombers. The transmitter antennas were mounted on both sides of the hull, below
and behind the pilot’s window. The receiver antennas were mounted in various
configurations, often Yagis that lay in a plane parallel to the one defined by the
wing and beneath it on both sides, although it appears that in this case it was
mounted on top of the aft section of the hull. National Archives photograph 80-G
700269.
The antenna of a Japanese 2.0 m
mark 6 airborne ship- search
radar mounted on a Mitsubishi-01
two-engine medium bomber cap-
tured by American forces in
February 1945. This very suc-
cessful ship-search radar used a
common antenna for transmitter
and receiver. Shown here is the Yagi
used for forward search. A pair
of dipoles mounted on the sides of
the fuselage permitted scanning to
the sides. Although the Yagi was
a Japanese invention, they did not
incorporate it into their designs
until discovering it on British
Army searchlight radars captured
at Singapore. National Archives
384
A German Me-110 equipped with

Lichtenstein airborne radar. The
nose of the two-engine machine
shows four dipoles with reflectors
for each of two radar sets, the 50 cm
Lichtenstein C-1 and the Lichten-
stein SN-2, normally described as
3.3 m but shown here nearer 2 m.
The 50 cm version proved to have
too short a range for picking up tar-
gets independently of ground radar
but was better for the final stages
of the tracking. Imperial War Mu-
seum photograph CL 3299. Crown
Copyright.
A Bristol Beaufighter. This became the standard RAF night fighter and was
equipped with radars for airborne interception using 1.5 m and 10 cm. It is
not possible to ascertain the radar type from the photograph, except to say that it
was microwave, evidenced by the plastic nose that shielded the paraboloid reflector.
From the date, 27 March 1949, one would assume the set to be an AI mark X,
identical to the SCR-720. National Archives photograph 80-G 403264.
385
CHAPTER 8
THE END IN EUROPE
8.1. INVASION
One could, without great scholarly exertion, assemble quotations from
many leaders of World War II emphasizing the difficulty of landing and
maintaining an invasion force on defended shores. For the Americans
in 1942 such a stratagem for the European continent seemed a straight-
forward way of preventing Soviet collapse. For the British it recalled a
memory of the Dardanelles and Gallipoli in 1915, a memory burned par-
ticularly into the mind of Winston Churchill and reinforced with the later
horrors of the Somme and Passchendaele. Thus General Marshall’s call
for a cross-Channel invasion in 1942 by green American and British troops
was received with incredulity in London. The disastrous outcome of the
division-sized raid on Dieppe made a quick end to such plans but did not
remove the goal from American and, though retained more reluctantly,
British minds.
The cross-Channel invasion was Marshall’s unalterable intention
from the moment the United States entered the war, and it assumed the
same importance for his deputy in the War Plans Division, Brigadier Gen-
eral Dwight D Eisenhower. That they maintained for a period what in ret-
rospect would have been a highly flawed course can be understood from
the dire situation in which the Allies found themselves in 1942. There
was every reason to fear that the Soviet Union would be defeated by the
seemingly unstoppable drive of the Wehrmacht into southern Russia, and
a landing in France was a way to help prevent this collapse. The risks were
well understood by the two American generals; they estimated the chances
of a successful 1942 landing to be ‘only one in two and of maintaining the
beachhead at one in five’. These were desperate times [1].
When Allied troops finally went ashore in Normandy in June 1944,
they had behind them the experience of the North African and Mediter-
ranean landings as well as the early ones of the Pacific. The knowledge so
gained allowed the invasion to be a success, but of equal, perhaps greater
importance was the absolute command of the skies by spring 1944. For
whatever its strategic value, the bombing of the Reich by night and day
386
The End in Europe
had forced the Luftwaffe to fight, and although they had inflicted serious
damage on the attackers—one can say they had defeated the bombers by
early 1944—it had been accomplished at a terrible loss in trained air crews
and machines. Allied losses were also great but could be made up; German
losses could not.
By spring 1944 Allied superiority in radar was equally well estab-
lished, but its affect on the battle is less easy to evaluate. Its most important
contributions were in navigation and countermeasures.
The German radar chain occupied a special part of the invasion plans,
a part that produced some curiously contradictory actions. The task of
cataloguing the characteristics and locations of German radar and jamming
stations, much helped by the easily recognized fortifications with which
they had been provided after the Bruneval raid, was shared by R V Jones’
Scientific Intelligence and Claude Wavell’s Central Interpretations Units
[2]. By locating the invasion fleet and its air cover these sets could provide
crucial warning of where and by inference when the landing would take
place. Ideally, the stations should have been destroyed, but starting with
Göring’s half-hearted attempt to destroy Chain Home at the beginning of
the Battle of Britain, it had been learned that radar stations were not so
easy to eliminate as it might seem. The electronics were well entrenched
and required a direct hit by a heavy projectile, and the spindly antennas
of Seetakt and Freya were hard to bring down and often capable of repair.
Rocket- and cannon-firing fighter bombers attacked the sites and proved
much better suited to the task than had Göring’s dive bombers. It was a
nasty business for the fliers because the radar operator clearly read their
intention, and the stations were well defended by 37 and 20 mm automatic
guns [3]. Destruction was attempted during the preparation along the
whole coast where landings might be possible, so as not to tip off the
selected beaches, and of 92 naval and air sets operating before the assault
only 18 were either left in operating condition or restored during the critical
time between 0100 and 0400 hours of 6 June. On top of this the general
failure of the electric power grid forced the use of engine-driven generators
for which fuel was scarce to non- existent [4].
Knowing that complete elimination was not possible, the attackers
followed with jamming and deception. Jamming the many sets from
ground and airborne transmitters could only hope to reduce the range
at which detection would occur, and that with uncertain reliability. The
amount of warning radar gave the defenders is not certain, reports vary
from three hours [5] to almost nothing [6].
Three squadrons of the airborne Mandrel screen were assigned the
protection of the advancing fleet by flying racecourse patterns at ten loca-
tions along a line from the Bill of Portland to Littlehampton [7]. Areinforced
squadron set up a similar screen along the Somme estuary to protect the
very vulnerable 1000 aircraft transporting airborne troops for the landings
east of Caen and on the Cherbourg peninsula. These jammers dropped
387
for the first time Window cut for Freya, often called Rope because of its
1.7 m length, in addition to using Mandrel. They also jammed the radio-
telephone channels of the German night fighters dispatched toward what
they took to be a bomber stream. The total effect was so strong that the night
fighters returned to their control points. One complication for the counter-
measure aircraft was that their Mandrel jammers seriously interfered with
their own Gee receivers, so navigation had to be by dead reckoning [8].
The vessels of the fleet did not have to depend on airborne jammers;
they were outfitted so as to produce cacophony on the Seetakt, Freya and
Würzburg wavelengths. The Channel radar war had seen the use of the
Royal Navy’s type 91, which by 1944 was installed on major warships
as well as on the English shore. Lighter vessels were equipped with light-
weight jammers designed by the Harvard Radio Research Laboratory, orig-
inally for airborne use. This was the first time that the Royal Navy per-
mitted jammers modulated with noise to be used, rather than sinusoidal
and repetitive-pulse modulation, and the type 91 sets were modified to
conform to this American practice. Noise jamming is harder to filter out
and requires a transmitter with a wider pass band, which makes it corre-
spondingly difficult for the jammed radar operator to evade the noise by
making a small change in frequency. Noise modulation had been avoided
for fear the Germans would take it up too (a curious fear because Martini
had used it during the Channel Dash), but by June 1944 their jamming was
no longer so greatly feared. Some 800 jammers were distributed among
the assault vessels [9].
The limitations on destruction and jamming caused much effort to go
into deception. The closest point between England and France was across
from Dover. It was a natural place to cross, and was close to Germany.
Normandy was selected for various reasons, primarily for the ease with
which it was hoped to take ports with sufficient capacity to sustain a conti-
nental army, but from the beginning an elaborate ruse had been exercised
to make Calais seem the selection. This had involved the creation of a ficti-
tious army group in Kent, manifested by discreet radio transmissions, the
deceptive meanings of which were left for astute German intelligence of-
ficers to penetrate in a manner conforming to Allied wishes. At some time
the fleet approaching Normandy would be observed, more than likely by a
few functioning Seetakt sets, and the enemy had to believe it a feint. Here
Robert Cockburn’s skills came into full play.
To reinforce the impression that the real invasion would be at Calais
required that a fleet be detected by the radar stations that remained after
the fighter bombers had done with them. The requisite number of vessels
could hardly be released for such a purpose, so a radar spoof, the greatest
of all time and Cockburn’s masterpiece, took its place. To ensure that no
unsuspected radar wavelengths would be encountered, Martin Ryle, who
did the detailed planning, made an electronic espionage flight along the
coast [10]. Finding no surprises, he loaded the aircraft with Window cut to
388
The End in Europe
the half-wavelengths of the Seetakts and Würzburgs. Following a series of

parallel racetrack courses, eight Lancasters directed an ‘invasion’ toward
Le Havre; they dropped the foil at locations that continually advanced
at the speed of an approaching fleet with each turn of their course. Six
Lancasters provided the same spoof directed toward Calais [11]. Enough
Mandrel jamming was added to give the operation credibility without
completely obscuring the ‘ghost fleet’. The distribution of the drops in
time and space were matched to Seetakt’s beam width, pulse width and
lack of height data. It was another instance where Seetakt’s lack of lobe
switching hurt as the cloud of foil could not be distinguished from a number
of small individual targets. It was all interpreted by the radar operators as
designed, although one can well imagine that their analytical skills may
have been somewhat diminished by the recent visit of fighter bombers.
Insuring that the advance of the foil fleet proceeded at the stately
speed of water craft when dropped from aircraft flying forty times faster
required extraordinarily accurate dumping, which in turn required ex-
traordinarily accurate navigation. This had three components. One nav-
igator computed dead-reckoning positions; a second located the aircraft
with Gee; the third fixed the exact position for the drop with Gee-H, using
the approximate positions from the other two in order to make his fix at just
the right moment, as Gee-H required interrogation by the operator [12].
A few launches equipped with Moonshine repeaters replied to the
radar interrogation by an enhanced return signal, thereby creating the il-
lusion of the reflection from a large vessel and covering the possibility that
some of the equipment had enough resolution to distinguish individual
targets from the mass. Moonshine had been introduced earlier as a feint
for the first daylight flights by the US 8th Air Force. The method, a ruse
fairly easily penetrated, had then been withdrawn after a few successes for
use on D-Day1 . The launches also towed balloon-borne reflectors. As the
‘fleet’ came closer to shore powerful loudspeakers provided the sound of
ships’ winches and anchors, enveloped in copious amounts of smoke. A
fake airborne drop, complete with fireworks for sound effects, completed
the Calais spoof [13]. The naval bombardment force off Le Havre had a
diversion of its own made up of radar reflectors on balloons that were both
anchored and towed by minesweepers [14].
It was learned later that the feint toward Le Havre had not been
detected because the attacks on the radar stations had been too effective,
but the stations at Calais reported the menacing fleet [15]. A few hours
of confusion resulted from these games with the result that forces in the
vicinity of Calais were given an invasion alarm order at 2300 hours on 5
June whereas those in Normandy received no such order until the landing
was in progress [16].
By June 1944 navigation for surface vessels around Britain had come
389
to rely on Gee routinely. It functioned well in all kinds of weather and

gave positions accurate to less than a kilometer; jamming was less a prob-
lem at locations well removed from the air over Germany. One of the
few successes of the Dieppe Raid was the accuracy with which the as-
signed beaches were reached, the result of Gee. There would be confusion
enough on the beaches without units landing well away from their goals,
and Gee kept things sorted out. Indeed, the channels of approach followed
the hyperbolic Gee grid lines. The Channel had huge belts of mines that
had to be swept accurately just before the arrival of transports and landing
craft, and the minesweepers used Gee. These were navigational require-
ments that would have been quite impossible three years earlier and were
a fundamental requirement for 6 June 1944. Gee made it seem easy.
The importance to the landings of this navigational system, which
had been known to the Germans long enough for Allied bombers to en-
counter jamming routinely, does not seem to have been given major con-
sideration in the plans for defending the beaches. It figured highly when
planning the attacks on radar and jammers, and the five known jammers
were eliminated in the same fighter–bomber raids that hit the radar chain
and no jamming for the Channel was encountered [17]. One suspects that
the oversight of not preparing reserve jammers, kept quiet until the in-
vasion was looming, was the result of defensive plans having been made
primarily by men versed in war on land who had little appreciation of the
navigational difficulties of the landing craft. The attackers had expected
jamming despite all, so an alternative system was mounted on principal
ships, minesweepers and lead landing craft. This was another system us-
ing hyperbolic coordinates that was manufactured by the Decca Company
and called Decca Navigator. Although not without fault, it was more ac-
curate and easier to use than Gee. When no jamming was encountered, it
was removed from service because of its secret nature [18].
In the initial movement of the attacking fleet, radio emissions of any
kind were severely restricted, but once it was clear the enemy knew what
was under way, even if imperfectly informed of certain key elements, these
restrictions were removed. Transports moved to about 6 km from the shore
to transfer men and equipment to the landing craft. These were guided
by control landing craft that depended primarily on a gyro compass and
a 10 cm surface-search radar for guidance in the final approach but also
had fathometer, accurate clock and underwater sound equipment. They
had special radar maps of the shore that showed the effects of prominent
coast features and allowed the operators to compare their PPI screens with
a chart showing a reasonable expectation of how the shore line would
appear. Landings were expected to be accurate to within 60 m in position
and 1 minute in time [19]. Minesweepers dropped radar-reflecting buoys
to mark the boundaries of swept zones and shore parties erected corner
reflectors to serve as fixed navigation points [20].
The flanks of the landing were covered by parachute and glider
390
The End in Europe
troops, who dropped under cover of darkness. The aircraft used Gee for
initial navigation but followed radar beacons set up by undercover agents
to locate the drop sites. This was the great triumph of Rebecca (US manu-
facture AN/APN-2), a modification of ASV mark II that sought the portable
beacon, Eureka (AN/PPN-1). Eureka had not only the happy characteristic
of speaking only when spoken to, it also allowed the operator to transmit
signals at a slow rate to Rebecca, thereby providing crucial last-minute
information about local wind or enemy actions [21]. This airborne action
was fraught with the possibilities of disaster and was opposed by many
in high command. The success in getting the airborne forces moderately
close to their assigned target areas at night was possible only through navi-
gational accuracies considered impossible by many not privy to the secrets.
The success of these forces was limited and the casualties high. Landing
division-sized units from the air, which had seemed the most modern form
of warfare, was already obsolete, but unfortunately the lesson was not yet
learned and the great airborne disaster at Arnhem was not avoided [22].
Defense of the fleet and the beachhead against air attack had been
prominent in the thoughts of the planners. To this end the British LW and
its American copy, SCR-602 went ashore in early landings [23], although air
superiority left them little work. Fighter control radar units went ashore
on D-Day plus 1 [24], but during the interim their duties were assumed
by three fighter-direction tenders accompanying the fleet with RAF teams
manning them. Each tender had two 1.5 m GCI sets (AMES type 15) and
one 50 cm early warning set (AMES type 11). These ships proved of little
value. The confusion caused by so many aircraft with the usual IFF mix-
ups, by the large number of ships and by the reflections from land made
it difficult to identify German planes, and those that did inflict damage on
the fleet—the low-flying mine layers—were seldom detected with 1.5 m
and 50 cm equipment [25].
It was during the invasion that MEW (microwave early warning,
AN/CPS-1), the ultimate surveillance radar, made its first contribution2 .
With its great range, 300 km under good conditions, it provided a target
resolution in horizontal coordinates superior to any air-warning equip-
ment. Its 66 tons, four 12 kW engine-driven generators and crew of 30 to
50 provided unparalleled information at five indicator positions, each with
a 30 cm PPI scope and two smaller scopes for range plotted against signal
and range plotted against direction. The PPI scopes were not only large
but were so devised that the indication zone could be moved off center,
thereby allowing the active sector to be seen in more detail. MEW came
with a fighter-direction center equipped with a large vertical ground-glass
plotting panel on which coast, azimuth and range lines were marked along
with the air-controllers’ grids; data were recorded with colored chalk by
plotters standing behind and writing in mirror image [26].
391
Shortly after MEW had been set up and before it had operational
communications installed, it gave a dramatic demonstration of its capabil-
ity. On 20 March the operators in training noted a large formation at 270
km headed over the Atlantic from France. A telephone call to a nearby
GCI station, which could not see the aircraft, disclosed that 14 B-17s with
140 men had just reported themselves hopelessly lost and were planning
to ditch. They were advised of their true position and vectored back to
England [27].
Luftwaffe radar troops gave a demonstration of what the subjugation
of Germany was going to entail. A radar station near Douvres, about 12
km north of Caen and 4.5 km from the beach, was encircled as the invaders
moved inland. The garrison consisted of only five officers and about 200
men, but they were well dug in and had acquired a nice assortment of
antitank and machine guns. The station finally surrendered to a major
armored attack after 11 days’ siege [28].
In addition to bombardment by major warships, the British and
American bomber forces contributed their devastating power to soften-
ing the beach defenses. RAF Bomber Command dropped 5300 tons of
bombs on beaches and coast defenses during the night of 5/6 June using
H2S to aim. The US 8th Air Force had the task of repeating the exercise
five minutes before the first troops were to land. Although intended to be
a daylight attack, the deteriorating weather forced the use of H2X. Bom-
bardiers received radar maps and extensive instruction; the sharp water–
land boundary was helpful to give hope of accuracy. Given that their
bombs were intended to fall only 1000 m ahead of the landing craft and
that a one second error in bomb release translated into about a 100 m error
on the ground, it is not surprising that caution was in the air [29]. The
bombardment killed no Allied personnel, but then few bombs fell on the
beach defenses; they did not, however, contribute to German good humor
in positions immediately behind the beach.
In looking back on the course of World War II one tends to see the
Allies on a steady path to victory after the beginning of 1943, which is
indeed how it came to pass, but it was not foreordained. Failure in any one
of the campaigns preceding the cross-Channel invasion would have meant
costly delays, but failure of the invasion would have had consequences
that would have given us a completely different world. The success of the
landings makes it appear as if there could have been no other outcome, but
this was certainly not the view on either side in early 1944. Many in high
places of Allied command were openly skeptical of an invasion holding,
and just as many on the other side were confident of repelling it [30].
8.2. FLYING BOMBS

During the early years of the 20th century ordnance experts gave consider-
ation to what was generally referred to as the ‘100-mile gun’. The ballistic
392
The End in Europe
knowledge of the time indicated that it should be possible to construct a

cannon with such a range but also indicated that the size of the projectile
would be so small and the accuracy so poor that it would have no tactical
value commensurate with the great expense and the short useful life of the
tube [1]. So matters stood until the stalemate of the trenches of World War I
gave desperate ideas special appeal, and Germany built and employed the
famous Paris gun. Its purpose was to cause panic in the city, thus replacing
tactical justification with strategic. If the Paris gun had any effect on the
outcome of the war, it was to add to the French bill of particulars against
Germany.
It is therefore strange that the German Army kept the idea alive after
its proved worthlessness, but the belief remained that such projectiles,
greatly increased in size and in the number delivered, would produce
panic that would force the desired capitulation. The Army had lifted a
page from Douhet’s air power doctrine, although it is doubtful whether
they were affected by his writings or even knew of them. The ballistic
arguments against a gun had been accepted by then, and plans for a rocket
took its place. In 1936, when Rm100 000 ($24 000) looked big to the radar
people, about 100 times more was spent on the development of what was
to be the A-4 rocket [2], subsequently given the propaganda designation
V-2 by Josef Goebbels, the V standing for Vergeltungswaffe (retaliation
or vengeance weapon). The Luftwaffe became interested in the Army
project because of their interest in rocket propulsion for aircraft, and the
two services collaborated to build an experimental station at Peenemünde
on the island of Usedom off the coast of Pomerania.
The A-4 project quickly became gigantic. The Army built a spe-
cial town for the thousands of workers employed. When experiments
on rocket-propelled aircraft soured, the Luftwaffe turned their attention to
the militarily much more effective pilotless flying bomb, the Fieseler Fi-103,
propaganda number V-1. Its development was at Fieseler with production
scattered, but Peenemünde served well as a test station. The Oslo Report
had described, not too clearly, activities at Peenemünde, but the report’s
general reliability had attracted R V Jones’ attention, so he kept it ever
in mind as a guide for the future. Thus when reports of flying bombs
and rockets began to arrive from the wide range of sources on which he
had established his intelligence network, his studies took a serious look at
Peenemünde.
By December 1943 tests of the V-1, also known as FZG-76 began. (The
V-1 acquired a number of names. In addition to FZG-76, which stood for
Flakzielgeraet and rose out of earlier intentions, it also had the German
code name Cherry Stone; the British code name was Diver, and the recip-
ients of the device called it the buzz bomb and the doodle bug.) Flights
were launched from a site on the eastern side of the island and were di-
rected along the coast where Würzburg-Riesen could follow their courses
and determine their three-dimensional coordinates. As months went by,
393
modifications in design improved accuracy to the point where the device

could be used for targets the size of a city. These data were of extreme
value to the designers and just as valuable to Jones—for he received the
data immediately. The tracking stations transmitted by radio the results of
each flight in a very low-grade code that did not even require the services
of Bletchley Park for decryption. The data told all one needed to know
about the flight characteristics of FZG-76 [3].
Given what was learned it required little effort to recognize the many
launching sites being constructed across the Channel for what they were.
Members of the underground provided details that aerial photographs did
not show, risking and often suffering the fate such activities brought with
them. The radar plots of May 1944 showed the accuracy had reached a
stage satisfactory for attacking London [4].
Preparing a defense against the two V-weapons was a challenge of
the first order. Bombing attacks on the production and launch sites were
obvious and came first. Production of the V-1 was dispersed, so the at-
tacks concentrated in the launch sites, which proved difficult to destroy.
Production of the V-2 was concentrated at Peenemünde and was hit by a
major and costly attack by Bomber Command during the night of 17/18
August 1943, which seriously affected production. Except for the chief de-
signer of rocket engines, Dr Walter Thiel, few of the highly skilled designers
and workers were killed, but many of the foreign and concentration-camp
workers were [5].
These rockets, capable of placing only a few tons of explosives
daily on London, had become in the minds of Hitler and his accomplices
weapons of immense power. The measures taken to insure production
show the grip it had taken on the high levels of command. Most aston-
ishing was the rapid creation of an underground factory near Nordhausen
from an unprofitable gypsum mine. This plant retains a standing in indus-
trial history that may be unique—it had its own crematorium to dispose
of the corpses of thousands of forced laborers.
The flight characteristics of the V-1 allowed it, unlike the V-2, to be
brought down by both fighters and anti-aircraft (AA) guns. Its speed was
known to be sufficiently high to require the fastest machines. It was a nasty
target because the explosive used in the warhead was easily exploded by a
fighter’s projectiles with the resulting blast often destroying the destroyer.
The flight characteristics had come about from the aerodynamic re-
quirements that the designers faced, but for whatever reason, these char-
acteristics were unquestionably the best for avoiding the AA gun defenses
then in place. The altitude was high enough to be out of the range of
the automatic cannon but low enough to provide large ground returns
for meter-wave radar. For the heavy guns the situation was very bad.
The combination of high speed and low altitude meant that tracking by
whatever mechanism would have to be fast, too fast for most crews doing
manual tracking. The tracking problem was compounded: radar or opti-
394
The End in Europe
cal aiming equipment was manual, the directors were not only manual but
slow mechanical analogue computers, and the mobile guns also manual.
These units had greatly improved their efficiency since the Blitz but were
hard pressed to engage the V-1s [6].
Guns placed close to London were worse than useless because in
downing a bomb—unless they caused it to explode in the air—they caused
it to fall on metropolitan London, whereas it might have continued beyond,
had it not been molested. Since October 1943 some of the heavy guns had
begun receiving the GL mark IIIC, the Canadian 10 cm set together with
an electronic predictor. It was a tremendous improvement over the 5 m
GL mark II but was a manual tracking set [7].
In preparation for the coming onslaught, the Air Ministry ordered Air
Marshal Roderic Hill to prepare the defenses of RAF Fighter Command and
Army AA Command. Night- and day-fighter squadrons were detailed for
this service with machines stripped of armor and other non-essentials to
give them greater speed. The bulk of the heavy guns moved to the edge of
Greater London along an arc at the southeast perimeter. Moving them to
the coast was rejected because of the expectation of jamming from across
the Channel. This gun belt soon became well organized and entrenched [8].
The battle entered the first phase on 13 June and continued for a
month. During those days the fighters were more successful than the guns,
both together bringing down about 40%, but it was apparent that the gun-
ners were improving and that the position of the gun belt was unfortunate.
The CHL stations would guide the fighters onto the bombs while over the
Channel, leading to pursuit. If a bomb had not been destroyed by the
time the two had reached the gun belt, the fighter had either to relinquish
the chase or risk the fire of excited gunners. On the recommendation of
General Pile, the long-suffering AA Commander, Hill ordered the gun belt
moved to the coast and instituted new rules of engagement: fighters were
to operate over the Channel and between the gun belt and London and not
to enter the gun belt no matter how hot the pursuit. Thus fighters would
get first crack over the Channel; gunners would form a tough gauntlet after
which any escapees could be hunted down before reaching London; finally
there was a line of barrage balloons. The transfer of the gun belt on what
amounted to a moment’s notice was achieved during three days in mid
July; it required the movement of 4.5 million vehicle km, re-deployment of
23 000 men and women, and re-laying telephone cable sufficient to stretch
from London to New York. Dawn of 17 July saw all guns in action in their
new positions, and the second phase began [9].
During the weeks before the move to the coast General Pile had
pressed for American SCR-584 automatic-tracking radar with the Bell Labs
M-9 electronic director. In this he was helped by Churchill, and when
things began to get tough a special emissary of Pile’s to Washington se-
cured 165 of the new gun-laying radars. General Eisenhower took interest
in the situation and lent 20 American four-gun 90 mm batteries [10]. Un-
395
fortunately, most of this equipment came into the hands of troops, whether
British or American, who had to begin shooting at the Divers while study-
ing the operating manuals. Into this stepped a few Rad Lab men who
went from battery to battery adjusting equipment and instructing person-
nel in its use. Among them were Ivan Getting’s co-designers, Hurach (later
Henry) Abajian and Lee Davenport [11].
The scores of the guns went up markedly as robot began shooting
robot. The SCR-584 tracked automatically and its data were converted au-
tomatically to gun orders by the M-9 director, and the American 90 mm
and the fixed-mount British 3.7 inch guns followed these order automati-
cally. At the end of the chain came the final element in this completely new
air defense weapon—the proximity fuze. The fuzes had been adjusted to
the small size of the V-1 from firings in New Mexico at a model built from
information gained through Jones’ intelligence network [12]. The results
of all these technical advances reached a climax on 28 August when 97 V-
1s were launched: fighters shot down 23, guns 65, balloons removed two
more, and only four of the remaining seven reached London [13].
Advancing British forces soon occupied the launch sites. This did not
free London from attacks because V-1s were carried beneath He-111s and
launched over the North Sea, which required moving batteries again, but
the number reaching London was very small. Antwerp, however, was less
easy to defend and received a heavy bombardment, far more than England.
Except for relatively minor fighter action the defense of Antwerp relied on
the guns but without proximity fuzes, which had not been cleared for use
over the continent. The 584 with time fuzes destroyed 40those engaged,
but the gunners wanted proximity fuzes. Opposition to their use came
from Admiral King, who looked on them as private property of the Navy,
and it took a confrontation between him and Vannevar Bush to free them
for use on the continent—just in time for the Battle of the Bulge [14].
The MEW saw action against the V-1s; the first one sent to Europe
had served to monitor the sky over the invasion and was then moved to
track the flying bombs. The single complaint about MEW was its lack of
height information, which the SCR-270 provided, but height information
was unimportant for the V-1s because of the limited range of altitude in
which they flew. This made the MEW a perfect GCI set for the fighters as
well as giving accurate warning to the batteries [15].
That the capabilities of Britain’s AA artillery lagged so far behind both
Germany and the United States has its origins, as discussed in Chapter 1.5,
in attitudes found at various levels of British command, but Lord Cherwell,
Churchill’s scientific advisor carries no small amount of blame. The talent
found at the War Ministry’s Air Defence and Research Establishment at
Christchurch was adequate to have equalled the American achievement,
but it was blocked at every turn by Cherwell’s bias against that kind of
weapon. Less hurting to the war effort but adding to a record of poor
counsel was his attitude about the rocket bombs. As intelligence contin-
396
The End in Europe
ued to pour in from Jones’s network about the V-2, Cherwell continued to
dismiss it as a ruse to divert resources, a position he retained for a remark-
ably long time—a repetition of his earlier refusal to believe the Germans
had radar.
Once the danger was too clear to be ignored the matter of defense
became acute. The V-2 was immune from fighters and guns, although
there was a plan for use of the latter that was never implemented. Attack-
ing the locations for production and launching was the only immediate
recourse, but after the Peenemünde raid of August 1943 production be-
came dispersed and underground, and the launch sites were simpler than
those of the V-1s and harder to destroy. Some results came from attacking
train loads of the rockets, but the overall effect was not great. London and
Antwerp simply had to take it.
Radar played an inconsequential role in this part of the play. Some
CH stations were trained to watch for ascent of the rockets. The long
wavelengths gave satisfactory reflections, but the accuracy was too poor
for locating the insignificant launch sites. Coupled with tracking by spe-
cially altered gun-laying sets that gave last minute information about the
descending phase, this brief warning had an important function. A hit on
the Charing Cross river tunnel of the London Underground would have
flooded the entire tube portion, almost certainly with great loss of life and
serious disruption of economic and administrative functions. The tunnel
was equipped with floodgates that were shut when the radar warning was
received [16]. Oddly enough a German copy of CH, Elefant-Rüssel, a few
sets of which appeared late in the war, was used to help the engineers de-
termine where the bombs fell. It was able to observe descent at a range of
800 km [17] but with an accuracy of no use to either designers or users [18].
Elefant-Rüssel was an updated imitation of CH, built by the Reich-
spostzentralamt with Telefunken help. The reason for its introduction is
to be found in the increasing chaos that marked the high levels of German
radar direction late in the war and that led to many useless radar develop-
ments while ignoring the vital. There is no evidence that the Peenemünde
group asked for it. The transmitter, Elefant, broadcast over 120◦ on the
10 to 15 m band from an array mounted on a high tower. The receiver,
Rüssel (elephant’s trunk), was located about a kilometer away and dif-
fered from CH in using a high-gain steerable array for direction finding
rather than pairs of crossed dipoles. Its range was less than Wassermann
despite much greater power and direction was significantly worse, much
as its CH counterpart. Very few were deployed, all of them along the coast.
The means of guidance for the V-2 during its ascent—once the motor
stopped it was a ballistic projectile—was initially not known in England,
and there was reason to suspect a radio method. To this end a tremendous
jamming effort was undertaken [19] but to no effect of course, owing to the
German use of inertial guidance.
The development of the V-2 must rank as one of the most curious
397
military events of modern time. The total weight of explosives it delivered

to London, Antwerp and Liege amounted to little more than a single attack
by Bomber Command and was even less accurate. To achieve it Germany
expended talent and resources equivalent to the atomic bomb project of the
Allies [20]. Peenemünde was in a way a mirror image of Los Alamos. Both
projects achieved scientific and technical results that surpassed dreams of
1935. Both projects were driven in part by the fear that the other side
was pursuing a similar goal. Both projects were managed by generals
of uncommon ability and had technical leaders of brilliant intellect. Both
projects produced weapons that were, in so far as the outcome of World War
II was concerned, of minor importance. Both projects combined during the
postwar decades to provide two states with ultimate weapons and their
peoples with ultimate nightmares. The German project proved a serious
drain on manpower and resources despite the use of slave labor. In the
end it was merely the Paris gun writ large.
8.3. THE BATTLEFIELD TRANSFORMED

For all its overwhelming effect in the air and at sea, radar had altered
the war of infantry, field artillery and armor only secondarily through
its effects on the air war. The dive bomber and later the rocket- and
cannon-firing fighter–bombers had shown themselves to be excellent an-
titank weapons, but radar seldom had a part to play in their encounters.
Ground radar had been employed in the deserts of North Africa where
the RAF had relearned its army support role, but its function had been
constrained by an unyielding secrecy that prevented soldiers and even
other airmen from learning its capabilities and discussing among them-
selves how best to use it, so its impact on the ground fighting had been
peripheral.
All this began to change as armies moved from the Normandy beach-
heads toward Germany, although not as a part of the enormous amount
of planning that had preceded that campaign. It grew out of experience
and resulted primarily from the characteristics of two pieces of equipment:
MEW and SCR-584. These two 10 cm radars had already made a strong
impression on those who had encountered them. Both made strong con-
tributions to the destruction of the V-1s, MEW by accurately locating the
incoming low-flying bombs and the 584 by directing AA gunfire. Both
belonged to the new, wartime radar generation.
Ground troops received much closer air cooperation during their
movement through France than they had in previous campaigns, and MEW
was one way the American forces found to accomplish it. The British
Branch of Rad Lab, the Massachusetts extension at TRE, converted one of
the precious, laboratory-made MEWs into a portable rig during 11 days
in April, and it went ashore in Normandy six days after the first landing
parties. By August it was placed at Cherbourg at an elevation of 225 m
398
The End in Europe
from which its view of the terrain reached to the Eiffel Tower. Installation
at the highest elevation available was a complete break in the technique
used by meter-wave ground radar, which sought emplacement in a shal-
low bowl-like terrain in order to restrict ground returns to nearby objects,
thereby clearing the oscilloscopes of fixed reflections at more distant ranges
where targets were expected. To eliminate the oft-remarked deficiency of
no height information, a British 10 cm height finder (AMES type 13) was
joined to the station. The extremely thin horizontal fan-shaped beam of
this set allowed the height of selected targets to be determined with a res-
olution equal to that of MEW’s horizontal components. Absence of the
Luftwaffe frequently left MEW serving only as an air traffic control, but
the positions of even very distant enemy planes caused the few that did
venture over the battlefield to be attacked promptly [1].
8.3.1. Tactical Air Control

Infantry and armored forces received fire support from field artillery by
calling on forward observers who went with them and shared their fate.
It was a well proved technique, so Tactical Air Control copied the method
by forming teams of ground observers, who could discuss requests for air
strikes with the MEW crews that had the big picture of the air situation and
talk to the pilots of the fighter bombers once they had reached the spot. At
this point the other star performer entered the play, SCR-584 with a new,
unforeseen role [2].
Early tactical air support had generated a fragment of dough-boy
wisdom: ‘When you want them, they can’t come. When they can come,
they can’t find you. When they can find you, they can’t identify the target.
When they can identify the target, they miss it and drop the bombs on
you’. The employment of MEW and SCR-584 could not remove the first
part of this incantation, but they could remove the remainder—if the right
equipment was near the spot. The main problem for the pilots was seeing
the often well camouflaged target on a confusing and unfamiliar ground
during the brief period available. To be effective, indeed to survive the
attack the pilot could not invest much time studying the terrain. If the
ground observers had access to a 584 from which range and direction to
the target was known, then the radar set could track the incoming fighters
and radio corrections to their course that put them onto the target with
near infallibility. This not only insured that the enemy received the fire but
also gave the attack near complete surprise.
The organization of this new form of warfare was worked out by
the US 555th Signal Aircraft Warning Battalion. At the heart of a fighter
control center sat an MEW with AMES type 13 Height Finder that was
located 20 to 40 km behind the front lines. The center was connected to
three 584s located closer to the front and connected to the ground observers
through the center or through one of two intermediate forward director
399
posts. The 584s had been equipped with special plotting boards having
reverse illumination so the plots could be seen on a military scale map [3].
Calls for support could be examined in terms of the complete air picture,
and attacks dispatched with control originating with the MEW and ending
with the ground controllers and the 584s. The battalion deployed other
radars positioned as necessary for coverage in MEW gaps resulting from
hilly terrain. These were AMES types 11 (50 cm), 15 (1.5 m) and 22 (50
cm) with some light-weight air-warning sets near the front, a few of which
were the new AN/TPS-3 [4].
These auxiliary sets were, in fact, the result of the history of the or-
ganization, as they were the original equipment with which the battalion
had been organized and trained before the invasion and used in the early
weeks. A remarkable improvement in the performance of the battalion
came about when the key radars changed to the MEW and the SCR-584,
transforming the technique from desultory to vigorous [5].
Similar British units, made up of more experienced personnel, had
long used what by then had become auxiliary sets for the Americans. This
left habits acquired in North Africa that caused them to reject the capabil-
ities of their own AMES type 26, which had capabilities similar to MEW.
Specifically, the fighter control center continued to rely on the reports of
the intermediate forward director posts rather than looking at the screens
of the type 26. These procedures even denied the information to the fliers
that they were, or more correctly could be under radar control. There were
successful but surreptitious demonstrations by restive young officers of
the new techniques, but they were unable to add education of their supe-
riors to their wartime accomplishments [6]. How curious that the open
relationship between fliers and controllers in Fighter Command was not
picked up by Army Cooperation Command.
Officers skilled in GCI attained reputations among their comrades
much as fighter pilots became Aces. Such a one was Squadron Leader John
Lawrence Brown, who guided the first GCI interception on 26 February
1941. Brownie, as he was known to fighter pilots, had ‘that genial, well fed
look that one usually associates with gentlemen farmers’ [7]. His exploits
earned him the appellation among ground radar officers of ‘The Great
Brown’. He was killed in the fighting at Arnhem, that disastrous result
of Montgomery’s bad judgement, where he was directing Light Warning
units [8].
8.3.2. Radar observation of the battlefield

As alert operators of the forward-positioned SCR-584s began to accus-
tom themselves to service in the front lines they noticed that ground re-
turns sometimes showed movement at night or in fog of vehicles, even
individual soldiers. Thus when there was little activity in the control of
fighter–bombers, they aimed their beams on various bits of terrain to act as
400
The End in Europe
electronic sentinels, and occasionally put an explosive end to movements

thought to have been concealed by providing accurate coordinates to the
artillery [9]. Even MEW had a few successes of this kind, which helped
cement a sound relationship with the ground troops [10].
In February 1945 during action to secure a bridgehead over the Saar
a 584 was placed within 1500 m of the front. Its easily identifiable antenna
was exposed only at night, at which time it assembled information about
enemy and friendly troops: echoes from shells disclosed gun emplace-
ments, infantry and tank movements showed up, and in one case vehicles
were located at 26 km [11].
8.3.3. Counter-mortar capability

The prototype XT-1 had disclosed a capability that would alter the tech-
niques of field artillery. The early radars had observed the flight of shells
toward the target and the fall of shot, but the 584 had shown very early
that it could track the path of projectiles. This had led to the discovery that
the firing tables for the American 90 mm AA gun were in error, which nec-
essarily meant that the directors were calculating incorrect firing orders.
The error was verified by other means, and the tables were replaced [12].
It was noted on the Italian front that the 584 could pick up and
track mortar shells in flight, effectively plotting their trajectories in three-
dimensional coordinates. This allowed the projectile’s path to be extrapo-
lated back to its point of origin with good accuracy, so a judiciously placed
584 could locate enemy weapons otherwise well concealed [13].
The infantry mortar is a light-weight device that is easily concealed
in a hole or trench. Its accuracy and high rate of fire had made it one of
the great killers of ground warfare since its introduction in World War I. In
tracking the shell’s parabolic trajectory the 584 could determine the ground
position of the weapon onto which artillery fire could then be placed, but
the 584 was ill suited by its large size for this kind of work. Experimen-
tation showed that the new generation of light-weight air-warning sets,
which were much more mobile and inconspicuous, could accomplish the
task. The 50 cm AN/TPQ-3 had a wire-mesh paraboloid antenna, and al-
though it could not track the shell as did the 584, it could sweep the horizon
through a sector and obtain range-azimuth plots of it ascending and de-
scending. A special slide rule allowed the point of origin to be determined.
This powerful new method saw only experimental use during the war but
clearly pointed the way for the future.
8.3.4. Radar for AA guns

While these new uses for SCR-584 began to disclose themselves, uses that
soon had Rad Lab sending special modification kits for better adaptation
to the tasks, the set’s original purpose proved itself in countless actions.
The combination of 584, M-9 director, automatic tracking 90 mm guns and
401
proximity fuzes (during and after the Battle of the Bulge) had made AA
fire deadly accurate. This was the weapon viewed in some quarters just
a few years earlier as a hopeless waste of resources, valuable only for its
psychological effect on the enemy and the defended.
As an example, in August a bridgehead 40 km southeast of Paris was
attacked at night by 35 German bombers preceded by three pathfinders and
defended by the 109th and 413th AAAGun Battalions. The attacking planes
reached the bridgehead without drawing fire because they responded to
IFF interrogation as Allied aircraft, disclosing their true identity only after
dropping flares. The two battalions brought down all three pathfinders
and 13 of the main forces with eight probables, and this without proximity
fuzes.
On New Year’s Day 1945 the Luftwaffe launched a strong attack with
single-engine fighters against Allied airfields in response to requests from
ground commanders whose troops were being dealt with severely by tac-
tical air. AA batteries, by then armed with proximity fuzes, were credited
with bringing down 394 with 112 probables during action later referred to
as the ‘AA Battle of 1 January’ [14]. It proved to be the Luftwaffe’s last
major attempt.
When the US 9th Armored Division captured intact a bridge across
the Rhine at Remagen, it was protected by an extraordinary air defense
with SCR-584 a key element, but this equipment served another function
as well. One was placed upstream to watch for swimmers, boats, mines or
whatever else might be floated downstream to attack the bridge [15].
8.3.5. Saturation bombing

The German attack on Allied lines on 21 March 1918 was preceded by a
brief but paralyzing bombardment by 6000 guns that resulted in the near
extermination of the British lines. The initial penetration succeeded, but
the Allies’ rear lines of communication were intact and the attackers found
the transport of reinforcements and supplies extremely difficult over the
moon landscape that the bombardment had created. The Allied line gave
ground but ultimately held. The same technique was revived by the Allies
26 years later, but with bombing aircraft furnishing the explosives.
Artillery still provided the soldier his most dependable support fire,
both in its accuracy and availability, but shells from light and medium field
artillery are smaller than all but the lightest bombs. A 200 kg artillery shell
is considered large, whereas a 1000 kg bomb is not. Starting in Italy the
Allies began concentrating the bombardment of German entrenchments
by using hundreds of bombers.
Such massive aerial bombardments preceded the American breakout
from Normandy at St Lo, and the inability of the Germans to recover,
owing to the complete Allied dominance of the air, and the use of tracked
vehicles to move through the shattered land spelled the difference from
402
The End in Europe
March 1918. Similar bombardments were used in front of the British lines,
although with less success. Accuracy was of extreme importance and it
was with Oboe and Gee-H to which success could be ascribed [16].
8.3.6. The proximity fuze over land

From the time of its introduction into combat in the Solomon Islands in
January 1943, the proximity fuze remained an exclusive weapon of the US
Navy. This insured that any duds would fall in the ocean, virtually elim-
inating the possibility that one might be captured. Their use against the
V-1s during summer 1944 was the first deviation from this strict rule. Fuzes
had been designed for use in field artillery pieces against ground targets
with release to follow as the strategic and tactical situation demanded. The
German December offensive in the Ardennes, the Battle of the Bulge, was
the crisis that called for employment of the new fuze in ground warfare.
Artillery shells bursting in the air over the heads of infantry in the
open was long recognized as particularly dangerous, but such fire had to
be observed and adjusted to have any effect, because the effectiveness of
a shell burst is very sensitive to the height at which it explodes. Blind fire
was possible, but only in a static front where the artillery had had time
to determine the coordinates of battery positions and targets, conditions
that definitely did not apply to the retreating American forces and that
had allowed the Germans to advance openly under cover of darkness. On
release of the new fuze for use over land these attacks were broken up,
leading General George Patton to remark: ‘The funny fuze won the Battle
of the Bulge for us’ [17]. Henceforth the proximity became standard with
Allied field and AA artillery. It allowed air bursts to be placed on targets
beyond the effective range of time fuzes and under all kinds of condition. It
enhanced the already high reputation of US artillery held by the Germans.
8.4. POST MORTEM

The radar war between Germany and Great Britain began before the offi-
cial outbreak of hostilities and continued after the surrender. The British
sides of the before-and-after events were controlled, curiously enough, by
the same man: W P G Pretty. In August 1939 Pretty, then a Flight Lieu-
tenant, was on duty at Fighter Command Operations Room when LZ-130
approached on the airship electronic espionage flight. Just two months
short of six years later, Pretty, by then an Air Commodore, took command
of what must rank as one of the strangest military operations in history—
Exercise Post Mortem.
As German forces had retreated they had destroyed nearly all of the
radar equipment that had to be left behind, but the surrender on 7 May
1945 left the stations in Schleswig-Holstein and Denmark intact, and their
retention in operating condition with the crews held together was stipu-
lated in the terms for surrender. It had long been planned, of course, to
403
examine the equipment and interrogate personnel involved with air warn-
ing, raid reporting, night fighting and its direction, but the Air Ministry
decided to attempt something significantly beyond such routine measures:
have the Germans put this system back into operation with RAF officers
observing how well their recent enemy gauged the nature and weight of
attacks, which would be made by formations of heavy bombers using all
the cunning of electronic countermeasures.
German cooperation was essential and needed to be obtained volun-
tarily because the tests would be meaningful only if the Germans went at it
with a will. The response of the prisoners was remarkably positive, allow-
ing plans for the exercise to proceed apace. Two reasons can be advanced
for this degree of cooperation: first, there was the pride of professionals
in demonstrating what they could do; second, there was the conviction
among the Germans that the Soviets were bound to clash with the western
powers, and that they would make use of captured German radar. A choice
between Britain and Russia was no choice at all for the prisoners [1], who
now wanted Britain to know all their tricks.
The performance of radar against the night attacks was the matter of
primary interest, but it was decided to make the flights during daylight
because the bomber streams crowded aircraft so closely that collisions were
a significant danger at night, and it was all the same to radar. This made
tests with night fighters and their controllers meaningless, but the idea of
having fighters rise to meet the bombers seemed to be pushing things a bit
far anyway, so only ground radar was tested. Flak was also excluded from
the exercise.
The radar deployed amounted to 16 Freya, ten Würzburg-Riese, two
Dreh-Freya, six Wassermann, three Mammut, two Jagdschloss and one
Elefant-Rüssel. Also to be evaluated were two stations of the direction-
finding Y-Dienst and 21 visual and aural observation posts. The total sys-
tem could not be expected to work with top efficiency because the Denmark
chain had not had to deal with particularly heavy attacks during the last
few months and so were not as experienced as their comrades to the south;
furthermore, the confusion at the end of hostilities had made it impossi-
ble to reassemble all crews as before with a corresponding reduction in
efficiency. Nevertheless, the personnel were qualified and experienced.
They were commanded by Generalmajor Alfred Boner, Chief Signals Offi-
cer, who strongly favored the project [2]. Boner and Pretty had feared that
the personnel, after having been subjected to years of vicious propaganda
with hatred of the Allies as its object, would not apply the skills necessary
to make the operation a success. It was not a situation where forced obe-
dience would suffice. In this they were pleasantly surprised because for
whatever reasons cooperation was very good [3].
Fourteen exercises were planned of which 11 were flown during nine
days beginning 25 June and ending 5 July. Three were canceled owing to
bad weather [4]. Seven of the exercises flew more than 200 heavy bombers,
404
The End in Europe
and during four of these about 24 aircraft from 100 Group accompanied
them to test the effectiveness of concurrent airborne jamming and interfer-
ence. Five exercises were made up only of jammers from 100 Group, and
two were flown entirely by 31 fast, high-flying Mosquitoes [5].
Impressions came thick and fast to the observers of the system that
had been the principal weapon against them through so many hard air
battles. Two contradictory observations came first. The quality of German
electronic engineering was found to be remarkably good, but it had been
in part so designed to compensate for the generally poor quality of the
operating personnel. This was the inverse of the British approach, which
was to get new models of equipment into the field rapidly and rely on
personnel of the highest intelligence and training to overcome problems
with them. The German policy, initiated at the start and held to the end, was
to engineer the sets for ultimate simplicity of operation and maintenance.
The observers were surprised to learn that some of the long-range air
warning sets such as Wassermann and Mammut had the weak transmitters
of Freya and depended on high-gain antennas for range. It also came
as a surprise to learn how little had been done with microwaves. The
Würzburg equipment that was relied on for the Himmelbett close-in fighter
control data was still found to be overwhelmed by Window. It had been
expected that 10 cm sets would have been deployed by then to replace the
old 50 cm equipment, yet none were found, and this more than two years
after the Rotterdam H2S had been recovered and nearly two years since
the introduction of Window. The Himmelbett system was overwhelmed
by the bomber streams but worked well against returning planes that had
become separated from the stream and frequently had no Window cover.
Window and jamming of the night-fighter-communication radio were seen
to be the most effective of interference methods. Here the poor quality of
operating personnel made itself felt, as the anti-Window modules added
to the sets proved complicated to operate and were frequently ignored by
the crews.
The observers noted the relatively late application of PPI and only
for Jagdschloss and Dreh-Freya, the latter a Freya modified to rotate
and present panoramic display. That PPI had not been applied to the
Würzburgs for improved GCI, the first British application, was a definite
surprise. It was with a tinge of satisfaction that the observers noted the
miserable cabins in which the German operators had to work [6].
Compensating for these faults, however, was the sobering realization
of how easy it had been to pick up the attacking force at extreme range.
When Bomber Command terminated the continuous attacks loosely called
the Battle of Berlin in early 1944, which had debilitated them as well as
hammered the enemy, they had come to realize that many of their prob-
lems had resulted from the bombers emitting far too much radiation: radio
communications, jammers, tail-warning radars, IFF and H2S. On resum-
ing heavy attacks on the Reich after the invasion was secure, they reduced
405
these emissions to what was thought to be the absolute minimum, but there
were always signals for Y-Dienst to use for triangulation, and they showed
themselves masters of the technique. Mandrel jammers often made it dif-
ficult for the sets working on the Freya band to track the formations, but
Mandrel proved to be an excellent signal for Y-Dienst as did H2S! The
first two simulated attacks by main-force heavy bombers were conducted
without the use of any interference methods and the IFFs were ordered
switched off. It came as a shock to see one of the Freya sets, which was
equipped with the Flamme attachment, pick up the one single bomber that
had its IFF inadvertently operating as if to proclaim ‘I’m English, I’m not
a bloody foreigner!’. The IFF had functioned as a secondary radar for the
defender at ranges much greater than Wassermann, in this case at 310 km.
The main value of Mandrel for the attackers was found to be in confusing
the defenders as to the composition of an attack but certainly not in hiding
it. Mandrel also helped in deceiving the defenders as to which was the real
attack and which was a spoof, but the only certain way of avoiding early
detection was to approach very low [7].
There was an Elefant-Rüssel, the imitation of CH, at the Robbe Station.
It was found to have about the same range capability as Wassermann and
Mammut for bombers but was much worse for Mosquitoes; its receiver
array had no lobe switching and was, of course much worse in directional
accuracy than the two older sets.
A non-concentrated force of Mosquitoes flying without countermea-
sures at 7000 m was detected by Mammut and Wassermann at ranges in
excess of 200 km only minutes after Y-Dienst reported their H2S emissions.
Initial pick-up was for an individual aircraft; the estimate of the number in
the formation became confused as more followed. The long-wave Elefant-
Rüssel observed them first at only 50 km, a consequence of their wooden
construction and Elefant’s long wavelength [8].
The operating personnel and their officers cooperated fully and were
eager to demonstrate their system. The women operators at one station
were an exception [9]. This was generally attributed to their being more
thoroughly indoctrinated in Nazi ideology than their male comrades, but a
Danish source reports another reason: ‘During the exercise the female per-
sonnel were not very cooperative. This has later been interpreted as their
being Nazi-minded, but the truth is that an English soldier had assaulted
one of them immediately before the start of the exercise, and he was in the
bunker during the exercise’ [10].
With this exception the prisoners were fortunate, primarily for having
come into British hands. Of the four allies, Britain took the most relaxed
attitude toward prisoners, allowing many to proceed home quickly. Those
taken by the Soviets suffered the worst, although erratically, some released
soon, others after many years of hard servitude, if they survived. Those
taken by the Americans had either a fair time of it or suffered the hell of
the infamous—and oft denied—100 000-man pens in which huge numbers
406
The End in Europe
of men were enclosed for months without shelter, medical treatment or

adequate food.
The teenage radar technicians from the Stegskopf, who were scattered
all over the Reich, experienced the dreadful variety of the chaotic end of
German resistance. The detachment assigned to the Kriegsmarine was
pulled from their radar sets and sent in their blue uniforms with Mauser
rifles to defend the last ditch in Berlin, nearly all dying in it. One died
remaining behind to destroy a Freya against the calls of his comrades to
be sensible and get away while there was time. One simply got off the
British truck that was transporting him when it passed through his home
town. Some died in the pens. One secured early release from Russians by
repairing their radios. Hans Plendl’s son went from an American pen to a
French prisoner of war camp and was released in May 1946 [11].
When the exercise was complete the equipment was dismantled and
shipped to England or the United States for further examination with Freya
serial No 1 going to the Royal Aircraft Establishment at Farnborough [12].
Station Robbe on the island of Rømø was operated by the RAF for several
months in order to make detailed studies of Freya, Wassermann, Mammut,
Elefant-Rüssel and Würzburg-Riese [13]. For unknown reasons the RAF
gave none of this equipment to the Danes or the Norwegians [14], and
equipment not shipped out was destroyed, although a substantial amount
of German radar and electronics equipment was removed surreptitiously
for the Danish Army by members of the underground, who had been given
the responsibility for guarding it by the British [15].
As the war moved toward a conclusion that was obvious to all Ger-
mans with insight and whose minds were not enslaved by Party fanaticism,
the radar engineers feared the consequences of defeat for Germany’s tech-
nical history and possibly even its future. This gave rise to the curious
‘Aktion Kindersarg’ (Operation Child’s Coffin), which was organized to-
wards the end of 1944 by General Martini with the participation of Leo
Brandt and a very few top radar men. It was clandestine and dangerous,
initially from the Nazi government, which had draconian punishments for
defeatism in any form, and potentially from the Allies, whose benevolence
was a most uncertain quantity and who could easily be believed capable
of punishing any who withheld information they wanted.
The conspirators made a selection of key documents describing radar
and packed them into a tight metal casket. The original idea of it being the
container for a child gave way to a larger one for a soldier worthy of burial
with honors. A bit of scurrying during times of administrative confusion
that often held toward the end of the war allowed the material to be given
a military funeral with proper burial papers [16]. The name ‘Kindersarg’
was also the name of a minor piece of radar equipment, which allowed
discussion by telephone without arousing suspicion. Removing the coffin
from the Russian Zone in the early 1950s proved to have difficulties and
risks of its own, described, perhaps a little dramatically, by Cajus Bekker
407
[17]. Some of the material so preserved was later published by the Deutsche
Gesellschaft für Ortung und Navigation in a very limited edition [18].
The postwar fates of the leading German radar engineers was as var-
ied as those of the Stegskopfer. Some attained high positions in the new
West German military services, once they and their country got on their
feet again. Wolfgang Martini and Josef Kammhuber attained high rank in
the new air force and in NATO; Rudolf Kühnhold returned to the navy
[19]. Gotthard Müller returned to the successor-company of Lorenz, SEL,
after ten years as a Soviet prisoner [20]. Wilhelm Runge quit Telefunken
in late 1944 but returned to it after the departure into a Russian prison of
Karl Rottgardt, the company’s director whom Runge had despised so long.
The determination of Telefunken’s employees to save their company dur-
ing such difficult times created a working atmosphere that he described as
‘the happiest of my life at Telefunken’ [21]. GEMA’s factories fell within
the Soviet zone of occupation and ceased to be a part of western economy.
Freiherr von Willisen attempted to form a company with a number of for-
mer GEMA workers, but it failed in 1948 [22]. Hans Plendl emigrated to the
United States where he was employed at the Air Force Cambridge Research
Laboratories [23]. Leo Brandt directed to reconstruction and electrification
of railways in the Ruhr and lower Rhein districts [24].
408
The End in Europe
PHOTOGRAPHS: TUBES
Japanese cavity magnetron M-312.

The split-anode magnetron, which
was capable of generating mi-
crowaves, was the invention of
Kinjiro Okabe in 1927, and re-
search with magnetrons was con-
tinued in Japan. In May 1939
Shigeru Nakajima operated a 10 cm
water-cooled magnetron that deliv-
ered 500 W of continuous power
and that quickly evolved into this
tube. A smaller, tunable version
with the same electrode configura-
tion, the M 60, served as a local os-
cillator. The pair formed the heart
of the mark 2 model 2 sea-search
radar that the Imperial Navy de-
ployed in 1942. Photograph cour-
tesy of Dr Shigeru Nakajima.
Anode of the Japanese M-312

cavity magnetron. This an-
ode was water cooled, hence the
groove around the outside for the
tubing, and was capable of deliv-
ering 500 W continuous with 10
cm wavelength. Diameter was
52 mm. The alternation of cir-
cular with flask-shaped cavities,
called the Mandarin configura-
tion, provided frequency stability
much as did strapping on British
designs. Inaccurate production
machining of the large cavity
pulled the frequency sufficiently
to make the matching with a lo-
cal oscillator tube, the M-60, dif-
ficult. Anode courtesy of S Naka-
jima and N Koizumi; photograph
courtesy of John Bryant.
409
The Telefunken decimeter-wave triode, LS180. Transmitter triodes for the 50 cm

band were developed independently by five different tube manufacturers around
1939. Others originated at General Electric Company (Britain), Western Electric,
Lorenz and GEMA. In tubes of this kind the leads were positioned and dimensioned
so that they formed part of a resonating structure. The LS180 was not the most
powerful even among the German tubes but, as the heart of the Würzburg se-
ries, probably shares with GEC’s Micropup the honor of being most widely used.
Photograph courtesy of Bernd Röde.
410
CHAPTER 9
THE END IN ASIA
9.1. THE PHILIPPINES AND OKINAWA

In leading his squadron into Manila Bay in 1898 Commodore George
Dewey led his country into the affairs of the western Pacific, a region con-
cerning which few of his countrymen had ever given thought, and in so
doing altered America’s position in the world. Dewey’s defeat of the Span-
ish warships was followed by little more than a pro forma defense by the
Spanish army, and the Philippines came under American control. These
events had interrupted a rebellion of the Philippine people against Spanish
rule, and they merely transferred the fight on learning that there had only
been a change of masters.
The resulting Philippine Insurrection was fought according to the
martial traditions of the archipelago, and although the Americans scorned
them as the ways of an inferior people, they adapted with alacrity to local
custom. The result was an ugly war, but not an ugly ‘little’ war, for it was
the largest the United States had fought in more than a hundred years,
the Civil War being the obvious exception. It compelled quadrupling the
army and dragged on for seven years. It had abundant examples of courage
and endured hardship but has left no trace in the mythology of the Great
Republic, seldom even catalogued in the lists of its wars.
Surprisingly, the outcome was positive. The Insurrecto leaders were
taken into the new colonial administration, and the people were promised
independence in 1946. The Japanese invasion came at a time when things
American were highly valued in the Philippines, and their newly created
army fought bravely beside the Americans against the intruders; their
highly revered commander, a Field Marshal of the Philippine Army, was
General Douglas MacArthur, whose father had been instrumental in secur-
ing amicable relations between the two peoples. These historical matters
were now to determine the strategy of the last year of the war against Japan.
MacArthur’s paternalistic feelings towards the Philippinos had been
reinforced by the steadfast loyalty that they had maintained. Americans
who had avoided or escaped Japanese prison were hidden and incorpo-
rated into the growing number of guerilla bands that fought the forces of
411
occupation with a violence and unity of purpose far greater than anything
observed in the underground movements of western Europe. MacArthur
insisted that the liberation of the Philippines had to be the next step on the
road to Tokyo, initially opposed by the Joint Chiefs of Staff and by Admiral
King in favor of an attack on Formosa. A decision for the Philippines came
about through Nimitz’s analysis, which convinced King, of the need for
more service troops for Formosa than would be available.
9.1.1. Leyte and Luzon

The campaign began with an attack on airfields on Formosa, Luzon and
Okinawa by Admiral Halsey’s task force of nine fleet and eight light carri-
ers. Their purpose was to destroy as much land-based Japanese airpower
as possible and to cover the target of the invasion, the large island of Leyte
located between Samar and Mindanao. Large numbers of defending fight-
ers met the raiders but suffered severe losses and could not prevent the
destruction of many aircraft on the ground. Japan’s aviation manufacture
was still delivering aircraft in large numbers, but the quality of the replace-
ment pilots was by then very poor. These altered circumstances brought
about a change in Japanese tactics.
The small fraction of these green pilots to return from a raid and
the meager results they achieved combined to cause Vice Admiral Tak-
ijiro Ohnishi, the Navy’s senior air officer, to employ suicide fliers, the
kamikazes [1]. He reasoned that inasmuch as most of the fliers did not
return anyway, suicide missions would improve the probability of their
destroying the enemy by having them crash their bomb loads onto the
ships, a maneuver requiring much less flying skill than dropping a bomb
or launching a torpedo and causing a gasoline fire as well. He also rea-
soned that this would prevent them from dying in vain, logic that perplexed
western minds. This relaxation of training requirements for kamikazes also
inverted the logistics of Japanese airpower by providing more pilots than
aircraft. This decision, triggered by the Formosa raids, would soon give
the Pacific War a new and terrifying phase, one greeted by the Emperor
with ‘Was it necessary to go to this extreme?’ [2].
Japan planned to meet whatever landing took place with the full
remaining power of their fleet, despite it being by then woefully inferior
in numbers and efficiency and completely lacking in squadrons for the
carriers. The Americans expected this response, and the result was a battle
that saw almost every conceivable combination of naval units in conflict
with one another. The Battle for Leyte Gulf was the largest naval battle of
history. It was also decisive, although the decisiveness was not accepted
in Tokyo.
The US 7th Fleet under Vice Admiral Thomas C Kinkaid was to place
MacArthur’s soldiers ashore and provide bombardment from six old bat-
tleships and air support from ten escort carriers. Halsey’s 3rd Fleet with
412
The End in Asia
its ten fast carriers and six fast battleships would stand ready to counter
the Imperial Navy when it attempted to interfere. There was, however,
no local unity of command. The two fleets were subordinate to Admiral
Nimitz in Hawaii, and Nimitz and MacArthur were subordinate to the
President through the Joint Chiefs of Staff in Washington.
Japanese interference began as soon as it was clear that Leyte was
the point of invasion. Two forces of battleships and cruisers converged
on Leyte Gulf, where the transports and landing craft were to be found.
One, under Vice Admiral Shoji Nishimura, approached from the south,
to enter the Gulf through Surigao Strait; the other, under Vice Admiral
Takeo Kurita, which included the 18 inch gun super battleships Yamato
and Musashi, was to pass through the San Bernadino Strait from the west
and attack the landing sites from the north. The latter expected to meet
Halsey’s ships, hence the presence of the strongest units. A third force
came from Japan under Vice Admiral Jisaburo Ozawa with a fleet made
up of useless carriers intended as bait to draw Halsey from his duties of
protecting the landing.
The Battle of Surigao Strait took place during the night of 24/25 Oc-
tober 1944, more than two years after the US Navy’s defeat at Savo Is-
land. The two engagements had the similarity of action in a broad channel
formed by islands, but the outcomes were vastly different. The use of
radar illustrates how the two navies had changed by 1944. In 1942 two
American radar-equipped destroyers had been positioned as pickets, but
for uncertain reasons the 1.5 m SC on the crucial guard ship failed to ob-
serve the large fleet. No such incident marked the 1944 action. Nishimura
was first reported and harassed by motor torpedo boats equipped with
the Raytheon 10 cm SO [3], who then turned them over to attack by de-
stroyers [4]. The American radar had PPI, which removed the confusion of
the reflections that proved so troublesome for the A-scopes. The Japanese
ships had 10 cm radar but no PPI, a serious disadvantage in the narrow
waters filled with rapidly moving ships, islands and a jagged coastline.
Radar-equipped PBYs, the ‘Black Cats’, had searched for the fleet but had
failed to find it; one was shot down by friendly fire [5].
Forewarned, the Americans had lines of Kinkaid’s battleships and
cruisers at the top of the Strait, barring entrance to the Gulf of Leyte. The
force was stronger and the superiority of radar made it overwhelming, and
it rapidly dispatched Nishimura’s ships. Three of the battleships had the
mark 8 fire-control radar, the Bell Labs 10 cm phased-array set that swept a
30◦ sector and allowed the battle to be kept under observation when firing
at a single target. It worked perfectly. The other three had the 40 cm FC
(new designation mark 3) and had difficulty in sorting out the confusion on
their screens, which retained the A-scopes. USS Pennsylvania was unable
to identify a target and fired none of her 14 inch projectiles. To be without
effective radar was to fight without effect [6].
When Kurita exited the San Bernadino Strait he expected to meet the
413
full force of Halsey’s ships and aircraft. The previous hours had certainly
been difficult for him. In the approach to the strait he had encountered
aggressive submarines and had to swim from his sinking flagship. Then
he received the full force of the Third Fleet’s vast air power that cost him the
super battleship Musashi, sunk from the impact of bombs and air-launched
torpedoes. But when he entered the open ocean the overwhelming force
that he bravely expected to meet was not there. Indeed, there was nothing
there! Not so much as a motor torpedo boat or a PBY patrol plane circling
overhead. Halsey, who thought Kurita had turned back, had swallowed
the bait of useless carriers to a degree the Japanese could scarcely believe
and had raced north with everything he had. With Kinkaid’s heavy units
to the south of the Gulf of Leyte nothing protected the landing but the slow
escort carriers, being used for air support of the troops ashore, and their
protective screen of three destroyers and three destroyer escorts, all under
the command of Rear Admiral Thomas L Sprague, who had not been told
of Halsey’s departure from station.
The collision of Kurita’s overwhelming force with Sprague’s thin line
was a confusing fight in which the planes of the escort carriers went ag-
gressively after the attacking ships, smoke was applied to the scene in
liberal quantities and the destroyers and destroyer escorts pretended they
were capital ships. The vehemence of the defense by aircraft that attacked
with machine guns once their ground support bombs were gone and de-
stroyers that used 5 inch [7] guns once their torpedoes were gone caused
Kurita to think he had finally encountered the Third Fleet and to repeat
the Japanese mistake of Savo Island. He did not steam resolutely into the
Gulf and make quick work of the fragile ships he would have encountered;
he turned back and left the scene through the strait whence he had come.
Radar functioned as an accustomed weapon during these exciting actions,
although the heavy punishment inflicted on the American destroyers elim-
inated much of their electronic equipment before they were sunk.
The conquest of Leyte followed the long-established methods of Pa-
cific ground fighting but with a somewhat larger radar contribution. Leyte
put an end to the peculiar reluctance of Army anti-aircraft (AA) units to
make use of SCR-584. On previous island locations they had been satisfied
with the SCR-268, but Leyte was mountainous, and the ground returns,
which had been no problem when fighting off an attack from over the
sea, were now serious when planes came erratically and suicidally over
the hilltops. The combined 90 and 40 mm units of Leyte brought down
over 300 planes during the first few weeks, mostly from 90 mm fire, still
without proximity fuzes, which had not been released for use over land. It
was the most amazing example of AA fire seen in the Pacific and brought
admiration from General MacArthur [8].
The AN/TPS-2 was one of the most advanced radars to be placed in
the hands of troops in 1944: a portable 75 cm set designed and produced
by General Electric with a total weight of 300 kg, capable of observing a
414
The End in Asia
bomber at 100 km and being placed into operation in 20 min [9]. This
modern set saw deployment under conditions in austere contrast with
its sophisticated origins, for it accompanied a detachment from the 597th
Signal Air Warning battalion which was set ashore in Mindanao from a
submarine to monitor Japanese shipping and air traffic before the invasion
of Leyte, guided and protected by Filipino guerrillas. A similar action went
with the invasion of Leyte. The 583rd Battalion placed similar detachments
starting by submarine and continuing by jungle trails and native sail boats
to advantageous positions on the islands of Panay, Negros and Cebu, ef-
fectively putting an end to the radar blindness imposed on the east coast
of Leyte by the mountains to the west [10].
With practice the AAunits expanded their new skills and found a local
use for the 584 in the ground fighting. An artillery observation plane would
fly directly over a target well behind the front while being tracked by radar,
thereby furnishing coordinates of the target for a future concentration of
unexpected artillery fire. This technique became more accurate as it was
refined by having the spotting plane make multiple passes from different
directions.
American field artillery had become a surprise weapon during the
war by the technique of placing all batteries for a given front on a common
coordinate system, thereby allowing fire from any gun within range of a
target to fire on it. This created the terrifying ‘time-on-target’, wherein
every gun within range placed a shell on the target within a second or
two of the same time. The basis for this was a rapid and accurate survey
of the gun positions, not always an easy task, and the forests and hills of
Leyte increased the difficulty until the AA artillery units did the job with
the SCR-584 [11].
While the Army was beginning to demonstrate excellent AA fire, the
Navy’s performance with FD radar with the mechanical-analog predictor,
although greatly improved with the introduction of the proximity fuze,
was failing all too often. The high-flying, level formations that charac-
terized air attacks in the first years of the war were being replaced in the
Philippines, where there were many dispersed landing fields, by many
small formations [12] coming from who knew where and being hidden by
mountains from electronic eyes of the ships during part of their approach.
Attacks on ships often came in very low, evading air-warning radar until
late and making it impossible for the 40 cm FD to determine height accu-
rately. When air attacks began to come from kamikazes the deficiencies of
air defense became serious. The proximity fuze was of no avail, if the shell
was not placed within 20 m of the target.
Bell Labs had recognized the weakness of the FD in tracking low-
flying aircraft and designed a 3 cm adjunct height finder for mounting
beside it, a vertical ‘orange peel’ reflector to form the radiation pattern
into a thin horizontal fan. This fan- shaped beam had an advantage over
a narrow conical beam, such as found in SCR-584, because it was less
415
susceptible to locking onto the image of the target mirrored by the ocean,
the problem Luis Alvarez and Alfred Loomis had had to solve for Ground-
Controlled Approach. Bell also undertook to design a replacement for the
FD, the mark 12, using the same basic structure but on 33 cm and with
automatic tracking in range; automatic tracking in direction was to follow
in a later modification [13]. This still left the mark 12 with manual tracking,
not suited to the agile attackers and still followed by a slow, mechanical
predictor, and most ships retained the mark 4 (FD). No mark 57 directors1
appeared until early 1945. Once the kamikazes were within their range,
the non-radar 40 and 20 mm guns had to save the ship and often did.
This situation seemed nothing short of scandalous to Ivan Getting,
who had been working on a Navy 3 cm fire-direction radar, the mark 35,
and a compatible and highly advanced director, the mark 56. When there
was no evident move by April 1945 to place this system into service, he
wrote a sharp letter to the Coordinator of Research and Development for
the Navy Department, pointing out the superiority of the Army’s radar,
thinking that tweaking inter-service rivalry would suffice to bring about
the desired conversion from 40 cm manual sets to 3-cm automatic. Given
the size of the US fleet by the end of 1944 and the time required for conver-
sion compared to the expected length of the war, it is hardly surprising that
this suggestion was rejected, which with afterthought Getting conceded to
have been the right decision [14].
The Navy’s radar equipment was not their only air-defense prob-
lem. The five great carrier battles had been conducted at an almost stately
pace. The SK radar and its predecessors could assure a carrier’s Combat
Information Center (CIC) of 30 to 40 minutes warning before the attacking
formations arrived—at which time ‘stately’ hardly described things. This
warning was the carrier’s armor plate, because it allowed the assorted col-
lection of explosives and gasoline on and below the thin flight deck to be
removed. Bombs might hit, but the American carriers did not turn into
floating infernos as did the unwarned Japanese. Off the Philippines this
comforting situation vanished.
This use of radar did fairly well for the fights between carriers and
Luzon-based planes in the first phase of the Battle for Leyte Gulf, but
when air attacks came at almost any hour of the day with only minutes
warning, it was not possible for the ships off Luzon to function and keep
their decks free of gasoline and explosives [15], and on 24 October the
light carrier Princeton suffered the dreaded fate and became an exploding,
blazing hulk. The calm assurance of American air dominance that had
been built up earlier in the year and crowned at the battle of the Philippine
Sea was becoming distinctly less assured.
Things became worse with the invasion of Luzon, the principal island
of the Philippines. Preparatory air attacks on 3–6 January 1945 by carriers
416
The End in Asia
for the landings at Lingayen Gulf were met by a ferocious defense made
up of some experienced fighters and many kamikazes, which left one ship
sunk and 11 damaged. Further, the efficiency of the carriers for working
their assigned land targets was seriously disrupted by the confusion in the
Combat Information Centers that these many attacks caused.
The need to decentralize the radar defense was evident to all. De-
stroyers and other light craft equipped with SC and SG (useful against the
low fliers) were spread out to pick up the attackers as soon as possible
and vector fighters onto them independently of the saturated CICs of the
carriers [16]. Before, the CIC would marshal its forces for an ambush; now
the destroyer controllers would call upon available fighters to go after in-
truders much as a radio dispatcher sends taxis to passengers. In the past
destroyers had screened capital ships against submarines; now screening
against aircraft was added to their duties. Unfortunately, the new assign-
ment did not bring with it the increased space for the plotting boards es-
sential to well run CICs. Destroyers are crowded vessels. The new system
was just going into use when the Japanese air strikes stopped—they had
run out of planes on Luzon and could not get new ones. The fleet began
to breathe more easily—but not for long.
While preparing in December 1944 to support the Luzon landings, the
3rd Fleet was struck by the full fury of a well remembered typhoon, which
made the SG radar all the more beloved by seamen. When the storm hit,
the fleet was attempting to refuel, and the relative proximity of the ships
to one another became a serious hazard, one compounded when at times
visibility shrank to less than the distance from the bridge to the bow and
when some small vessels lost the ability to steer. During these hours station
keeping and avoidance of collision resulted through the efforts of the SG
operators [17]. It was the first such storm to leave behind a record of PPI
photographs that clearly showed its eye [18]. Three destroyers foundered,
146 planes were lost and 790 men died.
9.1.2. Okinawa
After the Battle of the Philippine Sea in June 1944 perceptive authorities
on both sides knew that Japan had lost the war. But Japan was ruled by
a dictatorship controlled by the Army, and the Army could appease their
war gods only if defeat ended in death. Were there doubts among the
Allies about this determination, they were dispelled by battles on islands
whose names, theretofore unknown, oppress the memory: Iwo Jima and
Okinawa.
Iwo Jima is a bit of unstable volcanic rock and proto-soil midway
between Saipan, the base of the B-29s attacking Japan, and Honshu. It
sustained airfields that were used to attack the bombers and the new Saipan
base, and bombardment from sea and air had not been able to neutralize
them. Seizing the island, an operation not expected to be more costly
than for other Pacific islands, would eliminate this and allow fighters to
417
accompany the bombers in their attacks and provide a refuge for damaged
B-29s. Seizing Iwo Jima had no parallel. There had been vicious fights
over Pacific islands, but nothing like the month-long struggle in February
and March for Iwo Jima. It was not a modern fight, and radar was of
little consequence, but the technical advancement of the times and the
importance of the island for the bombing of Japan insured that it would
have a strong radar complement. An AN/TPS-10, a light-weight, 3.3 cm
height finder named Li’l Abner, was quickly placed on the top of Mount
Suribachi [19].
A base in the Ryukyus was necessary for the invasion of Japan, and
Okinawa was selected. Experience with the kamikazes at Luzon had pre-
pared the minds of the invaders for a difficult time off shore; Iwo Jima
taught them to expect the worst once they were on shore.
The Battle of Okinawa engaged a larger part of the British Common-
wealth than the Australian units that had fought with the Americans since
1942. Naval requirements of the European war had slacked, and in De-
cember 1944 Admiral Sir Bruce Fraser conferred with Admiral Nimitz on
deploying the British Pacific Fleet, which included New Zealand vessels
and consisted of two battleships, four fleet carriers, five cruisers and 15 de-
stroyers. (One of the carriers was a veteran the reader will remember from
the tough Mediterranean of 1940–1941, HMS Illustrious.) They came with
different but effective radar and were greeted warmly. The overwhelming
naval superiority of the Allies would soon be put to the test.
Preceding the 1 April landing on Okinawa was a carrier strike against
all possible Japanese landing fields, helped by B-29 raids on bases in
Kyushu. During these preparatory raids radar failed the Essex-class carrier
Franklin, which was heavily damaged by two 250 kg bombs dropped on a
busy flight deck by a single Aichi dive bomber that approached low and
was undetected until too late [20]. Japanese aircraft losses were staggering
and provided the beachhead with four days relatively free of kamikazes,
the time the Japanese needed to restock their fields. This was expected
and predicted. The absence of resistance to the landing parties was neither
expected nor predicted. Also not predicted was a garrison almost double
the reported size. That it was deeply entrenched in the southern hills was
the final surprise. The defending garrison had its own surprise. They had
planned to watch the ships struck by an overwhelming kamikaze attack,
but none came [21]. What followed was morosely predictable, following
plans laid out by the two contending generals.
The key to defense against kamikazes was a ring of radar picket boats,
destroyers or other light vessels that remained at station 60 to 80 km from
the location of the main fleet. Each had a fighter-control officer aboard in
communication with fighters patrolling in the vicinity, whom he vectored
onto the approaching fliers. The method worked moderately well in pro-
tecting the beach and main elements of the fleet, but the pickets had to look
out for themselves. The Japanese occasionally made use of Window [22],
418
The End in Asia
probably limited by the availability of cut foils.

Göring had given little thought to Chain Home (CH) on initiating
the Battle of Britain, dropping the idea of destroying the stations after
a few attempts seemed to have failed, and instinctively mistrusting the
whole idea anyway. Five years later no commander held such beliefs, and
the radar pickets—sentries terribly alone—bore the brunt of the Okinawa
kamikaze and bomber attacks. This led to a tactic of hitting a picket with
multiple attacks, which succeeded all too often. Many sank from the hard
blows received but none can match the ordeal of USS Laffey, a 2200 ton
Sumner-class destroyer launched in 1944. She, her AA consort LCS-51 (a
landing support craft) and the combat air patrol fought 22 attackers, in a
wild mixture of aircraft and gunfire. Laffey was hit by six kamikazes and
four bombs in addition to receiving intermittent showers of machine-gun
fire. Radar may have been the object of the struggle but soon became
irrelevant to these desperate warriors. Only one of the attackers escaped.
The Laffey was salvaged [23].
To relieve the pickets the island of Ie Sima and Hedo Misaki, the north
tip of Okinawa, were taken and by mid-April functioned as air warning
stations, which allowed the number of pickets to be reduced to five [24].
Two other islands, Iheya and Aguni, were taken in June to relieve the
pickets further [25]. Great ships also suffered. The Essex-class carrier
Bunker Hill was the recipient of skillful use of rain clouds and Window that
planted a single kamikaze on a deck full of refueling and rearming aircraft
with the inevitable result [26].
The great battleship Yamato with a few remaining elements of the
fleet sailed from the Inland Sea that April and perished in an attempt to
hit the Allied ships off Okinawa. Neither fuel for a return voyage nor air
cover were provided. The pride of the Imperial Navy served primarily as
a decoy to lure carrier planes away from the kamikaze attacks.
On 19 June the last organized resistance of the defenders was over-
come. This last great battle of World War II, fought in the air, on and under
the sea, and on land, was the clash of two hardened, professional forces.
When the kamikazes failed to destroy the invasion fleet during the first
few days, Lieutenant General Mitsuru Ushijima, commander of the island
garrison, and his staff knew they had no hope of defeating the invaders
or of relief, circumstances that in no way altered their dogged resistance,
which was countered by an equally determined opponent, commanded
by General Simon Bolivar Buckner. Ushijima and Buckner died within
less than a week of one another, one by suicide, the other killed in ac-
tion. Few Japanese survived. An American who survived the 90 days of
combat ‘could consider himself a fugitive from the law of averages’. Both
sides supplied history with countless examples of courage and devotion to
comrades and country, seldom recorded, most now forgotten. The victors
pondered what the future held for them in the next battle—known by all
to be the invasion of the Japanese homeland.
419
9.2. THE DESTRUCTION OF JAPANESE CITIES

America’s relations with China stand at the heart of the Pacific War. Drawn
by the mysteries of its distant civilization, recounted by returning mission-
aries and traders, Americans found sympathy for China’s pain during the
disintegration of the ancient empire at the beginning of the 20th century
and enmity for Japan’s brutal attempts at domination in the 1937 invasion
that filled American magazines and newsreels with graphic accounts of
savagery, specifically the unrelenting bombing of defenseless cities and
the delivering of Nanking to weeks of indiscriminate slaughter, rape and
plunder. The sinking of an American gunboat at the same time by Japanese
bombers hardened American anger. These events, tangled by diplomatic
matters touching every imaginable issue, eventually led in December 1940
to the American embargo on metal and petroleum products that forced the
Japanese decision for a war directed to the south against the United States
and Britain rather than to the north against Russia.
If China was at the center of America’s entrance into the Pacific War,
it was at the periphery of the conduct of that war, the consequence of
communications strangled by the mountain passes to Burma that required
maintenance by air and an inadequate road. China’s Generalissimo Chi-
ang Kai-shek, a political leader of limited military skill who found internal
enemies more important than the Japanese, had the assistance of two Amer-
ican generals whose opinions conflicted. Major General Claire Chennault
had organized an effective group of American-financed volunteer fighter
pilots for Chiang before Pearl Harbor, which had been expanded into the
14th Air Force in March 1944. One of its objectives was to provide refueling
fields for the 20th Bomber Command, based in Kharagpur, India, to bomb
the Japanese homeland with the new long-range B-29 bombers. Lieutenant
General Joseph W Stilwell, fluent in the language and with years of experi-
ence in China, was the obvious but unwilling choice for commander of US
forces in the China–Burma–India theater. He organized Chinese divisions
in Burma and demonstrated that, given proper leaders and training, they
were a match for the Japanese. He considered the leadership of Chiang’s
armies facing the Japanese on the mainland to be inept and corrupt.
Chennault was confident that American air power from Chinese
bases would perform the kinds of miracle the prophet Mitchell had
promised, and Chiang found this an agreeable way to fight. Stilwell
pointed out that the Japanese had not overrun the locations for the bases
simply because there was no reason to do so, and if American bombers
began using them, the Imperial Army would have no trouble taking pos-
session. This is what happened, although the logistics of operations over
the eastern Himalayas, the fabled ‘Hump’, would have ended the B-29
raids anyway. Eliminating the 14th Air Force’s B-24 radar-directed attacks
on shipping in the Formosa Strait probably figured larger in the Japanese
420
The End in Asia
decision to take the bases2 . No one who correctly predicts a defeat is loved,
and Stilwell’s inability to be tactful with the incompetent soon made him
the enemy of most Allied commanders in the theater, General William Slim,
his British counterpart and Britain’s best field commander, being a singular
exception. Stilwell was recalled [1].
The decision to build the B-29 had been a great gamble. Initial discus-
sions with Boeing during the summer of 1939 begot a prototype in Septem-
ber 1942, at which time contracts for full production had been made but
for which factories were not yet completed. The Superfortress became
an unquestioned success, but the speed with which its production was
secured resulted in countless minor design faults—minor from the engi-
neering point of view but major to the crews of the many planes lost in the
first months because of them. For a significant time these ‘bugs’ were a
greater danger than enemy fire.
The B-29 had space reserved for radar-countermeasure equipment,
and these functions were uppermost in the minds of the fliers that prepared
for the 15/16 June 1944 raid on the Japanese homeland, the first since the
Doolittle raid of 1942. British electronic intelligence flights from India
and Ceylon had found meter-wave radar early in the year in southeast
Asia, and the B-29s that had attacked Bangkok ten days earlier had logged
radiation from nine sets. These were from Japanese Army Tachi-6 early
warning radars that worked on the 4 m band. It was a static device with
wide-angle transmission and up to four receivers, each with a steerable,
directional antenna so that multiple attacking formations could be tracked.
Peak powers of 10 to 50 kW and pulse widths of 25 to 35 µs gave it a range
of 300 km. Special interest was attached to the absence of any evidence for
the 50 cm radiation that might have indicated the presence of Würzburgs.
American forces had captured a 1.5 m mark 4 model 3 radar on Saipan
in June 1944 and three months later a mark 4 model 1 on Peleliu, both of
which appeared capable of functioning as searchlight or gun-laying sets,
and Allied planes had been shot down under unseen conditions [2]. So it
came as no surprise that the attack on the Imperial Iron and Steel Works
at Yawata, Kyushu launched from Hsinching, China, the longest bombing
run in history, disclosed a rich radar spectrum. The Tachi-6 came first as
they passed over mainland China, followed by the familiar naval mark 1
model 1 when they reached the coast. As they prepared for their bombing
run the receivers picked up a score of radars working on meterwaves, the
Tachi-1, 2, 3 and 4 searchlight and gun-laying radars. (A 1.5 m set made
by Toshiba, the Tachi-31 that incorporated the Würzburg indicator, would
make its appearance by the end of the year, although not distinguished
from earlier models on the intercept receiver.) The inaccuracy of the fire
and the few times when bombers were illuminated by searchlights did not
demand the conclusion that the equipment was capable of locating the
421
bombers in three coordinates. Again no sign of the Würzburg [3]. The

effect of their bombs on the Imperial Iron and Steel Works was trivial [4].
The weakness of the defense led to a reconnaissance by a modified
B-29 during the daylight of 21 August flying very high over Japan and
Korea. It returned with excellent photographs, which had been sorely
lacking to intelligence, and was the start of routine coverage. The plane
had the usual AN/APR-4 receiver with analysis equipment and directional
antennas, and it found plenty of signals to analyze, but no fighters rose to
intercept, presumably because the speed and altitude were too great [5].
Other attacks on Japan from China followed the Yawata raid, none
yielding results worth mentioning except to raise Chinese and lower Japan-
ese morale. In August 1944 Major General Curtis LeMay took command
of the 20th.
The Yawata raid had preceded the Battle of the Philippine Sea by four
days, so the Saipan base would soon replace the need for the circuitous
route to Japan. The 21st Bomber Command began to occupy Saipan in
October 1944 with Brigadier General Haywood Hansell commanding. At-
tacks from the Mariannas could be carried out more frequently than those
from India by way of China, but the results were in no way commensurate
with the cost of operating such a long bombing run with a supply line a
third of the circumference of the globe. Although carried out during day-
light hours, the problems of accurate bombing that had been ignored from
the planning of the late 1930s made themselves felt in the Asian theater
as in the European. The cloudy skies not only forced radar bombing, but
the unexpected winds of the jet stream at times made accurate bombing
impossible from the high altitudes needed for safety. Hansell, a strict be-
liever in precision bombing, did not deliver the expected results and was
replaced by LeMay on 20 January 1945, who faced the same problems as
Hansell and saw them as a crisis.
On the night of 9/10 March 1945 the bombing of Japan changed from
an expensive nuisance to the most destructive force the world had ever
seen. LeMay decided that the night air defenses were so weak that he
could risk low-altitude bombardment. There was no evidence of radar-
equipped night fighters, so he stripped his planes of machine guns to
lighten and streamline them. Combined with the elimination of the fuel-
expensive climb to high altitudes, the tactic allowed much heavier loads of
incendiary bombs. Its first application was the most destructive air attack
of all time, which made a holocaust of Tokyo. Other cities followed, and
Tokyo received more fire from the sky. LeMay’s estimate of the defense
was correct. There were no night fighters and the guns were no more ef-
fective than the British guns had been during the Blitz, but these were not
the puny bomb loads of a hundred or so He-111s, these were the immense
loads of hundreds of B-29s.
Japan’s air defense was weak in material and trained personnel, but
the rivalry between the Army and Navy confounded the hope of using
422
The End in Asia
efficiently what little there was. In the homeland the Navy provided the
defense of their own, extensive bases while the Army provided for the
defense of the cities. Each maintained separate air control centers, separate
fighter squadrons and separate radars. They had separate IFF equipment
and interrogation signals, so that in addition to the normal problems of this
device they had the severe one of not knowing whether an aircraft that did
not respond was the enemy or the other service. Each service maintained
its own gun engagement zones [6].
The weakness of the defense caused no letup in American coun-
termeasures activities. There were false alarms about Japanese fighters
homing on bombers by activating American IFF that had to be inves-
tigated. Some of the Japanese early-warning sets did turn the IFF into
range-enhancing secondary radar, but early warning was not the defend-
ers’ problem [7]. Gun-laying and airborne radar was. These elements
might become effective, so the attackers deployed countermeasures, us-
ing techniques honed in Europe. At first reliance was placed on Rope,
the meter-wave version of Window, which was dispensed liberally and
which interfered demonstrably with searchlight operation. Nevertheless,
the radar-directed flak improved and learned to work through the drifting
foil, bringing down 4.4% of the bombers during two May night attacks.
This was answered by special B-29s carrying high-power jammers, called
Porcupines because of their numerous blade antennas or Guardian Angels
because of their circling the target area during the attack while saturat-
ing the ether with powerful radio noise. Losses attributed to the gunners
dropped [8].
The cities were destroyed one after the other with very high casual-
ties, much higher in proportion than for the German cities that had sturdier
construction and an advanced civil defense organization. The bombing of
Japan was carried out by radar to a high degree and in many ways resem-
bled the night bombing of Germany by the RAF. It was, however, much
more destructive than for Germany. First, there were no navigational prob-
lems in locating the target cities. Loran navigation was in place sufficient
to guide the formations to southern Japan [9]3 , and the preponderance
of targets located on the coast yielded PPI displays that approximated a
map of the target city to a degree rarely encountered in Europe. Second,
the H2X (AN/APS-15) was slightly better for rendering detail than H2S.
Third, the weakness of Japan’s air defense allowed bombing at altitudes
half as high as had been used by RAF Bomber Command, making the PPI
displays sharper and the bombing more accurate. Finally, the inflamma-
bility of the targets made area bombing inherently destructive. Daylight
precision bombing was retained, in part because of a shortage of incendiary
bombs, but as over Germany, formations that intended to attack visually
often encountered unexpected cloud cover and had to aim by radar.
423
As the weakness of Japan’s defense against night attack became ob-

vious, LeMay began warning cities by radio and leaflet of impending at-
tack. This had the advantage of demonstrating the inability of the Army
to defend the people and removing some of the stigma attached to area
bombing [10].
Two of Luis Alvarez’s radar designs had been deployed in Europe
with outstanding effect at the time and for the future: Microwave Early
Warning (MEW) and Ground-Controlled Approach (GCA). The third con-
tribution of his brief radar period was the 3 cm blind-bombing radar Eagle
(AN/APQ-7)4 . This set had electronic scanning ±30◦ in the forward di-
rection that provided a much improved resolution over H2X. Its antenna
was a vane mounted below the fuselage, which required specially modi-
fied B-29s. This antenna provided a beam only 0.4◦ wide horizontally and
was the reason for Eagle’s superior resolution; the beam’s large vertical
radiation pattern was an advantage.
Eagle was given to the 315th Wing of the 21st Command along with
careful instruction by Rad Lab personnel before being committed to action,
something that had been lacking for the crews using H2X. The 315th Wing
conducted 15 strikes beginning 26 June and ending 14 August. They were
precision attacks of the kind normally assigned for daylight raids, but
being night attacks were carried out at low altitudes with accuracy equal
to visual. The unit flew a total of 1200 sorties of which 1095 bombed the
primary target [11].
During these trying times occurred one of the most remarkable parts
of the radar story, one which in a bizarre way returned it to radar’s earliest
years. One aspect of Japanese research that progressed with a degree of
satisfaction was the development of high-power magnetrons. Indeed it
so satisfied Yoji Ito, the head of electronic research at the Naval Techni-
cal Research Department (NTRD), that he extrapolated powers to levels
that convinced him a microwave death ray was possible, calling to mind
Harry Wimperis’s 1935 query of Watson Watt about the possibilities of a
death ray. Ito’s ideas found strong support from military authorities. The
Army had been investigating a number of possible death rays since 1939
under the direction Major General Hideki Kusaba, and a rare example of a
combined Army–Navy project resulted at the Mitaka Office of the NTRD
that involved so many experts that it became known as a ‘physicist super
laboratory’. Tests succeeded in killing a rabbit at 30 meters in 10 minutes
and stalling an engine, presumably neither protected by the sheet-metal
that would surround the contents of a bomber, and plans went forward
to build a 15 m reflecting dish to concentrate the beam. Needless to say,
nothing affecting the war came from this project [12].
For maximum destruction atomic bombs were to be exploded about
500 m above ground. To accomplish this the device was equipped with
424
The End in Asia
four radar fuzes, a barometric fuze and impact fuzes on nose and tail, in
case the other five failed. The radar was the 72 cm AN/APS-13, a modi-
fied tail-warning set with Yagi directional antennas pointing forward. The
squadron’s electronic officer, Lieutenant Jacob Beser, who flew both mis-
sions, monitored the spectrum of Japanese radar to learn whether any
matched the wavelength of the bomb’s sets, including possible harmonics.
If such were found, he could disable one or more of the bomb’s radar fuzes,
as such signals might detonate the bomb prematurely, possibly destroying
the plane, but no interfering radiation was encountered and the fuzes were
not altered [13].
On 6 August a B-29 carrying the uranium bomb took off from the
Marianna island of Tinian accompanied by two observation planes, one of
which carried Alvarez, by then a bomb designer, and preceded by others to
advise them about the weather. Cloud cover over Hiroshima, the primary
target, was light enough to permit the required visual bombing. Single
B-29s were common in Japanese skies and seldom attacked, owing to their
altitude, speed and apparent immediate harmlessness, but this one made a
perfect bomb run and destroyed much of the city. Among the attacks of the
past few months it would not have stood out in the statistics of destruction,
except that the casualty rate was high for the moderate amount of build-
ing damage, the consequence of the population ignoring the somewhat
confused alarms.
Three days later the same procedure was followed for the plutonium
bomb dropped on Nagasaki; it was 67% more powerful but caused less
damage, in part because less accurately laid. After Hiroshima the govern-
ment had warned the population to take cover for single bombers but had
not explained the terrible nature of the new explosive. As a consequence
the people of Nagasaki had not taken cover and suffered high casualties,
although significantly lower than Hiroshima [14]. Nagasaki was the sec-
ondary target; the primary, Kokura, was obscured by clouds. The bom-
bardier reported that he aimed visually, as required, through an opening in
the cloud cover, but the inaccuracy was typical of an H2X bombing. Given
the reluctance of the crew to bring the bomb home, many have thought the
hole in the clouds to have been fictitious.
The bombing of Hiroshima and Nagasaki ended the Pacific War. Of
this there is no question. Whether the end would have come as soon or
soon enough to prevent the dreaded invasion without the use of these two
explosives is a dispute that began immediately after the end of hostilities
and continues, growing in asperity as time passes. It is a curious dispute
because it centers on the nature of the bombs, not the injury done by them,
for there is little evidence that the arguments would rage, had Hiroshima
and Nagasaki been dealt equally hard blows by a few hundred of LeMay’s
fire bombers. Japan surrendered through the direct and brave interven-
tion of the Emperor, who had long wanted peace and found that the atomic
bomb provided him the means to obtain it. There is little doubt that had
425
the war continued only a few weeks longer Japan would have suffered
more from the destruction of nearly every remaining seat of population
than from the two atom bombs. That it was war degenerated from the
chivalrous rules to which western nations had romantically thought them-
selves to have attained is hardly a question. That it came to this through the
implacable working of military and technical imperatives is obvious. The
absence of strategic bombing with nuclear explosions in the succeeding
half century may have accrued from the awakening of humanity caused
by the tactics of LeMay and Harris and the atomic bomb.
These questions aside, it is clear that strategic bombing was decisive
in defeating Japan. Sea power was necessary to implement the bombing
and had blockaded the island kingdom to the degree that it was no longer
able to defend its skies or feed its people, but air power had finally de-
livered the promised victory. Nevertheless, the Pacific War had followed
the general plan formulated by the Navy during the previous decades:
(1) force the Japanese fleet to fight with the American strength for its de-
feat maintained through the establishment of island bases, (2) blockade
the home islands and (3) if that did not force surrender, attack the cities
from the air. None of these pre-war plans had foreseen an invasion of the
homeland—or an atomic bomb [15].
426
CHAPTER 10
THE MEASURE OF RADAR
10.1. NAVIGATION TRANSFORMED

At the end of the war the British carrier HMS Implacable, bound for Van-
couver, received the pilot for the passage through the assortment of islands
south of the port. Despite the urgency of the vessel’s mission, to land a
large number of repatriated prisoners of war in time for the special trains
that awaited them, the pilot refused to take the ship up because of fog.
A consultation between the captain and his navigation officer over charts
concluded with agreement that the passage could be made with radar and
the skipper rang for 20 knots, whereupon the pilot announced he not only
disclaimed responsibility but would not even remain on the bridge [1].
Such a decision by the captain resulted from the sure knowledge of the
capabilities of his radar. The PPI provided a clear definition of the shores
and established position, even if navigation markers were not visible; other
ships were easily avoided.
This clash illustrates the dramatic difference in navigation that radar
and associated techniques had brought about. In 1939 the position of a ship
in fog or under cloud cover came from dead reckoning—an estimate based
on speed and course with corrections for the effects of wind and currents.
After a few days of dirty weather at sea the calculated position might be a
hundred miles from reality. In 1945 the position of a ship equipped with
Loran (see below) would, depending on momentary radio propagation
conditions, be known to accuracy as good as a routine celestial fix.
10.1.1. Radar
Before microwave surface-search radar became a fixture on their masts,
liners on the North Atlantic route ran consistently in violation of rules
requiring reduced speed in fog, ‘. . . a speed which does not preclude getting
all the headway off a ship in a distance less than half the range of existing
visibility’. But to arrive in port late meant schedules disrupted and possibly
tug boats and gangs of longshoremen idle at company expense. Captains
who valued their berths took care to avoid this.
427
The postwar handbooks for navigators and pilots instructed their

readers on the relative merits of 3 and 10 cm radar and about atmospheric
effects on propagation. The shorter wavelength had the advantages of
reliably sighting buoys and other small targets combined with the disad-
vantages of more confusing sea return and of being blinded by a rain squall.
Small ships generally elected 3 cm, large ships frequently both.
With time almost anything that floated on navigable waterways car-
ried radar, but in the decades after the war many ships did not [2]. This
caused busy harbors, especially those troubled with heavy fog, to install
shore-based radar. In this a 3 cm station was located so as to have the
best possible view of the harbor and approaching waterways. Operators
provided pilots and masters with data that allowed them to find their way
rather than remain hove to, waiting for visibility. It could also advise ships
equipped with radar of traffic conditions beyond their line of sight, served
as a continual check on the condition of buoys and other floating seamarks
and proved valuable in rescue operations or in positioning dredges or
dumping. Liverpool installed the first such system in August 1948 and
was quickly followed by Sunderland, the Isle of Man, Long Beach, Van-
couver and Spitsbergen [3].
The use of radar for aerial navigation proved to be the inverse of
maritime. At sea it was the ship that wanted radar in order to see land,
other ships or the the rapidly growing number of radar beacons. Shore-
based radar was useful, but not often encountered. By contrast commercial
aircraft did not generally mount radar. There was certainly little in H2S or
H2X to recommend it as a system of navigation, and when an airplane came
within range of a radar beacon, it was probably subject to traffic control on
the ground. Ground radar never allowed a competing airborne system an
opportunity, it was simply too obvious a method of controlling traffic.
Pilots favored methods that allowed them to determine the bear-
ing to a ground station of known location, often their next destination.
The favorite after the war was Omnirange, officially VOR, a continuous-
wave directional system similar to Sonne–Consol, except that its receiver
actuated a visual indicator rather than relying on audible signals. The
name is flier’s parlance, as the device provided only direction informa-
tion. Distance measuring equipment, officially DME, was soon paired
with VOR. It drew directly on the design of Y-Gerät, with the exception
that the aircraft rather than the ground station transmitted the interro-
gating audio-modulated radio-frequency signal, which the ground station
re-transmitted on another radio frequency [4].
Under such circumstances there was a clear need for secondary radar.
In addition to enhancing the range and representing the target aircraft with
a strong mark on the PPI, the responding signal could be coded to give the
air controller such obviously important data as the aircraft’s altitude, iden-
tification and remaining fuel. It was all to happen but years after the end
of the war. Secondary surveillance radar brought with it all the old prob-
428
The Measure of Radar
lems of IFF and many new ones. Whereas primary radar depended only on
the characteristics of the ground set with no problems of compatibility, sec-
ondary radar is surrounded by a great compatibility problem that includes
commercial, private and military interests together with an ample portion
of international disagreement [5]. The technical constraints continued to
change as a result of the rapidly evolving electronics of the postwar years,
continually complicating negotiations. Finally, the International Civil Avi-
ation Organization adopted a standard based on the American IFF mark X
in 1958 [6]. It is a story with its own attraction, but it is not our story
[7].
Air controllers would have to wait for secondary surveillance radar
but were able to receive the benefits of a technique that would remove
many distracting signals from their screens, the moving target indicator. Its
invention had first been forced on the Germans as a way to distinguish the
echoes of moving bombers from near-stationary clouds of Window. This
had been done, as described earlier, by filters that passed only return pulses
that had had their frequencies shifted to some degree by the Doppler effect
of the moving target. Pulses on or near the transmitter frequency were
attenuated. Its value to the operator of surveillance equipment is obvious;
it removes the large number of echoes from fixed objects that clutter the
screen, allowing the aircraft to stand out.
TRE and Rad Lab worked on the same idea but without the air at-
tacks that spurred the Germans. Their approach did not use filters to reject
pulses that had not been Doppler shifted. Instead the received pulses
were compared with the previous pulse, rejecting both if they were iden-
tical in amplitude and phase. This was accomplished by requiring the
returned pulses to pass through a transmission line that delayed them by
exactly the time between transmitted pulses. Such a delay line was, inter-
estingly enough, not electrical but acoustical, generally being the transmis-
sion of sound through mercury. This proved capable of detecting planes
moving at near right angles to the line of sight much better than filters
[8].
The most highly valued radar for bomber crews returning to their
base in fog was the ground control approach, Alvarez’s GCA, which al-
lowed a ground radar operator to guide the blind flier down. When Amer-
ica and Britain began the supply of blockaded Berlin by air in 1949, GCA
allowed the transports to land irrespective of visibility. Indeed, the airlift
would have been impossible without it. Despite these achievements, GCA
was not used except for military flying, and commercial airfields were to
be shut down by fog long after blind landing was an actuality. The reason
was simple. An airline pilot is responsible for the safety of the plane, and a
landing with GCA in effect gave the controls to an unknown radar operator
on the ground. The pilots would never agree to this.
429
10.1.2. Radio altimeters

Barometric altimeters had a sufficient number of faults, potentially dan-
gerous ones, that commercial aviation quickly adopted the radio altimeter
in the postwar flying world. The United States left the war with two light-
weight and relatively inexpensive devices: the pulsed SCR-718C and the
frequency-modulated AN/APN-1, both on the 70 cm band. The pulsed
set proved best for determining high altitudes, the frequency-modulated
set for low altitudes [9]. Germany left the war with equally well de-
signed equipment. The pulsed FuG 102 worked well for high altitudes,
the frequency-modulated FuG 101 for low [10]. Both were adopted by
eastern block nations.
10.1.3. Loran
The organization of the Radiation Laboratory had called for three high-
priority projects: airborne radar, gun-laying radar and a long-range-radio
navigation system to meet the needs of both aircraft and ships. One might
object that the resulting Loran (long range navigation) was not radar and
hence does not belong in this account, but its use of pulsed high-frequency
waves, its close association with navigation methods that used radar di-
rectly and its development at Rad Lab make the inclusion reasonable. By
the end of the war there were a remarkably large number of radio naviga-
tion systems in use: radar beacons, Oboe, Sonne–Consol, Rebecca–Eureka,
ground radar control, Gee, Decca Navigator, Gee-H, Rebecca-H, Micro-H,
Shoran and Loran. Of these Loran remained in service the longest and had
the widest coverage.
Alfred Loomis proposed Loran in October 1940, and a Project 3 Com-
mittee was formed almost immediately to carry out the charge. In addi-
tion to the Rad Lab people its meetings included representatives of RCA,
Sperry Gyroscope and the British Embassy. When committee manage-
ment yielded what might be expected, the work was re-organized under
Melville Eastman in early 1941. Loomis’s original plan called for a hyper-
bolic coordinate system using pairs of master–slave stations in the same
way as the British Gee. The Americans had decided by summer 1941 that
the line-of-sight restrictions imposed by very high frequencies limited the
use too severely for the kind of navigation desired. They began studying
the possibility of using the reflections off the ionosphere of 40, 60 and 100 m
waves that allowed a greatly extended range.
Initial tests indicated positions determined from the sky wave to be
stable enough for accuracies of a kilometer or less. An operational test
with 100 kW peak transmitter power was undertaken in June 1942 using
a 150 m band from which amateurs had been ejected and which remained
the Loran standard [11]. At that time R J Dippy, the inventor of Gee, came
to Rad Lab from TRE to work out the compatibility of the aircraft mounting
of Gee and Loran equipment. The Loran sets were so designed that they
430
fitted into the slot where a Gee set had been removed, thus allowing planes
of the two nations to use both kinds of equipment interchangeably.
The first priority was given to navigation through the foul weather
of the North Atlantic, the route of many airplanes for transport, transit and
convoy protection, so pairs of stations began operating in Newfoundland,
Labrador and Greenland to form the North Atlantic Chain. By January 1943
the US Coast Guard and the Royal Canadian Navy assumed operational
responsibility.
The designers originally thought the slave station had to be so situ-
ated that it received the ground wave, but evidence accumulated that the
sky-wave pulse, which lags the ground wave in time, could be identified
and was stable enough to synchronize the two satisfactorily. This allowed
the two base stations to be situated 2000 km apart, giving wider coverage
and a grid capable of greater accuracy. It became known as sky-wave syn-
chronized or SS-Loran and had a range capable of encompassing most of
continental Europe. It was used operationally by RAF Bomber Command
in October 1944 [12].
The extreme distances of the Pacific placed heavy demands on any
radio-navigation system and led to the decision to exploit the good long-
range propagation of very long waves. During the winter of 1944–1945 an
experimental system using 1700 m operated in the south-eastern United
States but did not see war service [13].
By the end of the war 70 Loran stations and 75 000 receivers were pro-
viding navigational information for 30% of the Earth’s surface [14]. The
150 m band Loran continued as the postwar standard. The advantages
of low-frequency propagation, the frequency region favored for maritime
communications because of the absence of skip zones, compelled investiga-
tors to surmount the problems of greater atmospheric noise, and a 3000 m
system became operational on the American East Coast in 1957. The older
system was referred to as Loran-A, the newer as Loran-C. In Loran-C the
multiple sky waves can lag as much as 1000 µs, but the first arrivals were
readily identified and sufficiently stable for an acceptable fix [15].
10.1.4. Decca Navigator

A hyperbolic-coordinate navigation system similar to Gee, officially des-
ignated QM by the Admiralty but quickly named Decca Navigator for its
manufacturer, Decca Radio Ltd, was employed during the Invasion and
then withdrawn from service to preserve secrecy. After the war Decca
marketed the system and established chains of shore stations along the
Channel. The network of chains spread over most of Northern Europe
and to many coastal region in the British Commonwealth and became very
popular with seamen.
It was similar to Gee in using a pair of stations to determine a hyper-
bolic coordinate, but differed in using low-frequency continuous waves
431
around the 3000 m band. Positional data were carried by the relative phase
shift between master and slave and attained accuracies of tens of meters.
It was restricted to similar ranges as Gee but by confusion brought by the
sky wave, not by line of sight propagation. The phase difference was indi-
cated by a meter. Constant deflection indicated movement on one of the
hyperbolic coordinate lines [16].
10.1.5. Radio navigation and the velocity of light

Shoran was a hyperbolic-coordinate navigation system that used the same
principles as Gee-H but with an automatic pilot’s direction indicator. It
had an on-board computer allowing a continuous presentation of the data
that permitted blind bombing. It proved to have the accuracy of Oboe or
visual bombing but with each bomber capable of using it independently.
It was introduced into combat in April 1945 and consequently had little
effect on the course of the war [17].
The US Coast and Geodetic Survey learned of the accuracies sus-
pected of the various radio-navigation systems and wished to evaluate this
new method of measuring the Earth. With the Army Air Forces they set
up as a test, Shoran ground stations at locations where there was surveyed
control. The results showed precision of two parts in a million but sys-
tematic discrepancies from the control data of 50 parts per million, which
could be effectively removed if the velocity of light was altered from the
accepted value [18].
The tabulated value in 1941 was 299 776±4 km s−1 . This had resulted
primarily from extensive measurements initiated by A A Michelson and
carried to completion after his death by his assistants. The experiments
had used long, evacuated paths and incorporated what was thought to
be the ultimate in experimental accuracy [19]. It was significantly lower
than Michelson’s previous value, and the experimental uncertainties as-
cribed each measurement made them incompatible one with the other.
Any warning provided by this was swept aside by a confirmation exper-
iment at Harvard that used the Kerr-cell light switch rather than Michel-
son’s rotating mirror and that allowed electronic timing. The result was
299 764 ± 15 km s−1 [20]. The Coast and Geodetic Survey found that a
value of 299 792 ± 2.4 km s−1 made their Shoran data compatible with their
ground-surveyed data, very nearly Michelson’s older value.
William Hansen had proposed to use a resonating cavity to measure
the velocity of light. By knowing the frequency at which the cavity res-
onated, its dimensions and the rate of its power dissipation, one could cal-
culate the velocity of the waves. This experiment had been carried out at the
National Physical Laboratories in England before the Shoran experiment,
and gave 299 793 ± 9 km s−1 [21], a value also significantly in disagreement
with the then accepted value. Other confirmations of the new value were
to follow with microwaves and light. The laser was eventually to take ac-
432
curacy to the currently accepted value of 299 792.458 ± 0.0012 km s−1 , but
for a time the blue ribbon belonged to microwaves.
These products of the wartime radar laboratories present us with an
example of technical progress in its purest form. As a result of radar and
Loran fewer ships run aground or collide. As a result of radio navigation
passenger travel by aircraft has become as reliable and dependable as that
by the railroads before the war, and ground control by radar has allowed
traffic densities almost equal to the demand. This has transformed the
world more than anything since the age of discovery. To fly across oceans
for a routine business meeting has become commonplace. People of mod-
erate incomes take a week’s vacation in parts of the world that their parents
would have never dreamed of visiting. Nothing has shrunk the Earth as
has modern air traffic, and radar is its primary guide.
And yet this progress has come at a price. Within the lifetime of many
of the readers, the navigation of air and sea required skills acquired only
after exacting apprenticeships, skills for which men were justly proud and
for which at times required them to draw on their reservoirs of courage.
The automatic ship and airplane are now technical realities. But this is the
inevitable result of technical progress. A worker’s skills—skills that had
given him his sense of worth, skills whose application gave him his place
in the world, skills that made him a man—are being made of no impor-
tance. Whether in transport, the shop, the mill or the farm, occupations
that once gave meaning to life for much of humanity are being destroyed or
transformed into tractable jobs that give little satisfaction at the end of the
day. It is the technical imperative. It is the consequence of mankind having
taken the course of evolution into its own hands. It cannot be avoided.
10.2. SCIENCE AND THE ELECTRONIC AGE

Radar was derived from science, primarily physics, but the link was
through a path of electrical engineering that had already converted the
basic ideas about electromagnetic fields and the motion of electrons in vac-
uum and conductors into design elements. The link between physicists and
electrical engineers was strong during the inter-war years. Nuclear physics
was growing in the United States and Britain and demanded competence
in electronic design. Cyclotrons required high-frequency oscillators of ex-
treme power, and the radiation detectors needed pulse electronics, both
generally made by physicists. These skills were directly transferable to
radar and account in part for the large number of physicists employed
to that end by both countries. Having forced some of her best nuclear
physicists into exile and covered quantum mechanics and relativity with
odium as ‘Jewish physics’, Germany lagged in that discipline, and only
the Kaiser-Wilhelm Institut at Heidelberg had a cyclotron. Relatively few
German physicists went into radar.
Physicists picked up engineering quickly but brought very little new
433
physics with them. Scientists and engineers pushed the development of

microwaves together, and an attempt to assign credit one way or the other
becomes a game with words. Bell Labs was populated with engineers, Rad
Lab with physicists. Both designed advanced microwave systems. One
contribution science did make to radar during the war was the study of
semiconductors, which allowed the Allies to produce reliable microwave
detectors. In going the other way, in contributions of radar to science, the
examples are many and important.
10.2.1. Astronomy
Observing celestial objects through the electromagnetic radiation they emit
in wavelength bands longer than infrared had had a tentative beginning be-
fore the war. In determining experimentally the noise sources of a 14.6 m
receiver at Bell Telephone Laboratories, Karl Jansky noted a source that
could not be ascribed either to his equipment, atmospheric discharges or
man-made interference. In order to check his suspicions that he was receiv-
ing radiation that was not of terrestrial origin, he constructed a directional
antenna that could be rotated about its vertical axis and demonstrated after
a long series of observations that it had a maximum in the constellation
Sagittarius, which lies in the direction of the galactic center [1]. Jansky’s
discovery did not go unnoticed, for Bell made certain there was adequate
public announcement, but astronomers did not know exactly what to do
with these data, as they were neither able to apply them to any of the
phenomena then under investigation nor use them to suggest new obser-
vations. Furthermore, the radio techniques employed were simply too far
from their discipline to allow the matter to be properly considered. The
radio amateur Grote Reber did the only continuation of Jansky’s work and
succeeded in having his results published in the astronomical literature [2].
Receiver noise limited the sensitivity, hence the range of a radar set,
so noise studies were common, both as research and as parts of the routine
maintenance of equipment. Jansky’s ‘cosmic noise’ must have been ob-
served by them countless times without an appreciation of its significance,
but reports of it number only two: J S Hey saw it with GL mark II but had
to have a colleague identify it [3]. Wilhelm Stepp, the Telefunken engineer,
also encountered it [4]. Two new astronomical phenomena caught the at-
tention of alert operators of meter-wave radar receivers: solar emission
and meteors incident on the earth’s atmosphere.
Radio emission from the sun, particularly strong during sunspot ac-
tivity, had been observed before the war by radio amateurs, who reported a
‘curious hiss’ at times in their receivers [5], but there had been no scientific
follow-up. This source of radiation was rediscovered in February 1942 by
Hey, who had been assigned the task of evaluating the possibility of the
Germans jamming the Army’s GL equipment. The escape of the three cap-
ital ships in the Channel Dash a few days earlier had enhanced concern
434
about such techniques. As it turned out he set about his task during the
period of a solar flare and was rewarded by what appeared to be a very
cunning form of jamming, which he quickly identified as radiation from
the sun [6].
The 5 m GL mark II was the radar set with the longest wavelength
that had an antenna steerable in two angular coordinates, making it ide-
ally suited for establishing the source as the sun [7]. Hey found these
characteristics just right for making another discovery. When V-2 rock-
ets began falling on England some means of tracking them was sought as
part of the desperate methods being planned for countering them, and Hey
suggested equipping GL mark II with a larger antenna. None of these anti-
rocket methods came to trial, but in the process of tracking them Hey made
another discovery. The operators observed high-altitude echoes four to ten
times an hour that had nothing to do with V-2s but sometimes triggered
false alarms. As the danger of German air and rocket attacks receded, Hey
requested permission from Antiaircraft Command to investigate the phe-
nomenon systematically while an operating network still existed. General
Pile’s intellect was attracted to the idea, and permission came quickly. The
troublesome echoes came from the trails of ions left momentarily in the
upper atmosphere by meteors [8].
Bernard Lovell, who had led the TRE team in the development of
H2S, intended to return to his pre-war research in cosmic rays at the Uni-
versity of Manchester, where, just before leaving for war service, he had
observed with cloud chambers a cosmic-ray shower that had deposited an
impressive amount of energy in the atmosphere. In one of his early duty
assignments, he noted activity on the receiver scope at a Chain Home (CH)
station, which he took to be enemy aircraft, but was told by the operator
‘Oh, those are not enemy aircraft, they’re ionosphere’. Finding no basis for
such transient events in his knowledge of the ionosphere, Lovell thought
he might be seeing cosmic-ray showers [9]. The idea so appealed to his
professor at Manchester, P M S Blackett, that the two wrote a paper about
the radio events, without disclosing the method by which the radio ob-
servations had been observed, speculating that they might have had their
origin in cosmic-ray showers [10].
On learning of Hey’s meteor studies, Lovell reasoned that the same
equipment should show up the desired cosmic-ray showers, so he and
Blackett, both having acquired a substantial amount of gratitude in the
War Office, arranged for the transfer of Hey’s modified GL equipment
to the University. The electrical noise that inhabited a major industrial
city caused them to shift the work to a rural site in Cheshire named Jo-
drell Bank. Lovell’s observations showed that cosmic-ray showers were
beyond the sensitivity of his equipment, and he shifted his interest to me-
teors, the events he had seen at the CH station in 1940. His systematic
study of meteor trails after the war allowed their trajectories to be de-
termined and showed them to have come from within the solar system,
435
answering a question open since the time of Newton [11]. He and his asso-
ciates pursued other aspects of astronomy using radio waves, and a major
radio-astronomy observatory came into being. As indications of the var-
ied possibilities of radio astronomy began to be perceived, Lovell became
convinced that a large steerable paraboloid was needed as an antenna and
found enthusiastic support for his wild idea of a 75 m diameter dish in
Blackett.
Desires to enter the new field caused two other observatories to be
opened. E G Bowen had left Rad Lab in late 1943 to join the Radiophysics
Laboratory in Sydney, which greatly impressed him with its high level of
competence [12]. Australians had made some solar observations with 1.5 m
equipment as soon as the war had ended that indicated radiation from the
sun-spot regions of the solar disc [13] and were not long in planning a 63 m
diameter steerable dish for radio astronomy, whose similar appearance to
the SCR-584 was not accidental. Martin Ryle at Cambridge University built
a special array of dipoles that allowed him to observe solar radiation at 1.7
m in a clever way that insured that galactic radiation did not contribute
[14]. These three observatories, all operated by former radar men, were
for a decade the principal sources of knowledge in this new discipline.
The Moon offered a more substantial target for radar astronomy than
trails of ions. It also offered the possibility for instructing the public about
hitherto secret work by the Signal Corps, and the opportunity was not lost.
An SCR-271 at the Corps’s Evans Laboratory at Belmar, New Jersey was
altered for the task so that single returns would be visible on a scope dis-
playing range against signal. Two parameters controlled the experiment:
the travel time to the Moon and back is about 2.5 s, and the half-diameter of
the Moon from which signals can be returned provides a spread in return
times of 0.0116 s. The former led to pulse repetition about every 3 s; the
latter allowed the use of pulses of 0.05 s. The long pulses allowed the use of
a receiver with a very narrow pass band and consequently very low noise.
(The narrow pass band required compensation for the Doppler effect that
results from the relative radial velocity of the Earth and the Moon and the
Earth’s rotation.) The 271 was given an extra dipole array but was unable
to track the target in elevation, so the traces showed, rather dramatically,
the Moon entering and passing through the radar beam at its rising or
setting [15].
Two years earlier Zoltán Bay had begun to plan the use of a
Hungarian-designed and built 2.4 m radar to observe reflections from the
Moon. Bay was head of the laboratory of United Incandescent Lamp and
Electrical Co. (Tungsram), a large manufacturer of radio tubes. He had
an antenna steerable in both azimuth and elevation that allowed tracking
the Moon, but he did not have the equipment that would allow the high
degree of frequency stability required for a narrow-band-width receiver,
so his transmitter power and antenna gain were insufficient to observe
single traces on an oscilloscope. He devised instead a unique method of
436
integrating the return signals—electrolysis of a KOH solution into hydro-

gen. A rotating switch formed the 0.06 s transmitter pulse, protected the
receiver and routed the signals according to delay into ten 1 mm diameter
coulometers that bracketed the times at which signal would be returning.
This procedure left a record of receiver noise plus signal from data ac-
quired during repeated cycles of 30 min on the Moon and 30 min off. In
cell 6, where the 2.5 s pulses were expected, the gas volume from the Moon
observations exceeded that of the blind cycle by 4.4%. It was a small ef-
fect, only five times the experimental uncertainty, estimated from the blind
data. Wartime air attacks and the removal of his original equipment by the
Soviets delayed the experiment, but he persevered and succeeded with
reassembled equipment in February and May of 1946 [16].
These two experiments introduced two important techniques that
were to make radar astronomy a valuable method for studying the solar
system. DeWitt and Stodola showed how sensitive the observations were
to Doppler shifts, and Bay showed the power of integration. The later
combination of sharp time resolution, Doppler shift and signal integration
would evolve into a technique that would provide remarkably accurate
topographic maps of the surfaces of Venus and asteroids [17], but long
before such sophisticated results could be admired, radar astronomy de-
termined with great accuracy the astronomical constant, the distance scale
of the solar system.
But there were radar observations of the Moon before either of these
two experiments. In 1943 Telefunken was attempting to build a long-
range early-warning radar using their 50 cm electronics, which they ex-
pected would be an improvement over GEMA’s 2.4 m Wassermann and
Mammut, reasoning along lines similar to those advanced at Rad Lab for
making MEW. Unlike the Rad Lab project, the Würzmann, named from
the combination of Würzburg and Wassermann, was a cheap trial that
drew on a large stock of components used in Telefunken’s widely used
50 cm directed-beam relay communication links Michael [18]. The an-
tenna was made of 32 Michael arrays (a total of 640 dipoles), arranged side
by side 16 on each of two parallel 36 m masts capable of rotation about
their common vertical axis. The Würzburg electronics were used with the
transmitter power boosted to 120 kW peak power and a newly designed
high-frequency amplifier at the receiver input that greatly improved sen-
sitivity. It was located overlooking the Baltic Sea at the easternmost tip of
the island of Rügen [19].
After setting up this equipment Stepp left it with an assistant, Willi
Thiel, for further testing. When Stepp visited the site a few weeks later,
Thiel reported a strange disturbance that had troubled him from time to
time, which lasted about two hours and vanished without remedial action
on his part. In examining it together they noted that it disappeared when
the antenna direction was turned from the east; when the antenna faced
the east the signals would remain for 2.5 s after the transmitter power was
437
switched off. Clearing visibility disclosed the Moon and settled the matter.
Thiel demonstrated the phenomenon to various visitors, but Stepp made
no written report at the time.
He did make a report of the matter later, but in a remarkably un-
derstated manner. Immediately after the war he entered the Technische-
Hochschule Darmstadt, as his mentor, Wilhelm Runge, had done a gen-
eration before, and took his theoretical and experimental radar work as
the subject for his thesis. At the bottom of a page of his dissertation is
recorded: ‘Als erstes von uns jemals erfasste extraterrestrische Ziel wurde
von uns Anfang 1944 mit dem Würzmann auf Rügen der Mond beim Auf-
gang erfasst’ [20]. He also mentioned it at a conference about space travel
in 1951 [21] and again the following year at one on maritime navigation
[22]. This curious behavior has three possible explanations: (1) during the
war such a report might have been interpreted by morbid-minded officials
as wasting time on non-essential work, which could have been considered
sabotage; (2) Stepp was the consummate engineer, who saw the work as
too rough for a formal report and easily improved, given the opportunity,
or (3) he may have taken the attitude that astronomers already knew the
distance to the Moon better than he could have determined it, making it
nothing more than a nice stunt [23]. Years later when reports of the event
gained wider circulation, Stepp wrote a short article that gave credit for the
discovery to Thiel as well as providing some of the technical parameters
of the equipment [24].
Radio astronomy was not restricted to meter waves, and the Sun at-
tracted microwave experts. Bell Labs’s inventor of waveguides, G C South-
worth, led the way, observing the quiet Sun with microwaves in 1942 and
1943 [25], but a more important function of these short waves lay else-
where. During the war Dutch astronomers at the Leiden Observatory had
received the issue of Astrophysical Journal that contained Reber’s article on
radio astronomy and discussed its implications. The advantage of having
a spectral line at radio frequencies caused H C van de Hulst to demonstrate
the theoretical possibility of the hyperfine transition of atomic hydrogen,
the major constituent of interstellar space, being observed at 21 cm [26].
The group set out to build detection equipment, as did E M Purcell of
Harvard and Rad Lab. Both reported detection at about the same time in
coordinated publications that showed mutual assistance [27]. The Leiden
group used the 7.5 m diameter paraboloid antenna of a Würzburg-Riese,
and a few of the thousands of these dishes continued their existence in
this new peaceful service. Their data provided remarkable evidence for
the rotation of the Milky Way through the Doppler shifts observed, but
a hoped for interpretation in terms of the galaxy’s structure proved only
qualitative, owing to the absence of reliable data for distance.
Both groups used a microwave invention from Rad Lab that would
become the heart of radio astronomy, the Dicke radiometer [28]. In this
instrument Dicke addressed the problem of measuring thermal radiation
438
with a superheterodyne microwave receiver and devised a simple and

elegant solution in which he alternately compared 30 times a second the
noise-like radiation incident on the antenna with the thermal noise of a
resistor of known temperature [29], a method that eliminated drifts in gain
or noise within the receiver. This device, equipped with a horn antenna,
determined the temperature to a precision of 0.5 K of various objects around
Rad Lab from a small microwave band of their thermal radiation, often in
ways that amused his coworkers. The concept of antenna temperature was
not new, having been introduced by R E Burgess a few years earlier [30],
but the accuracy and simplicity of the radiometer fixed it as the unit of
radio astronomy. Dicke used the radiometer in its simple form to measure
the temperature of the Sun and Moon, doing with a small horn something
that had required a large paraboloid for Southworth just two years earlier
using a non-comparing receiver and only for the Sun [31].
10.2.2. Laboratory physics

Given that so many of the physicists joining radar projects in America and
Britain had left research on the atom and its nucleus, it would have been
surprising had they not put the skills and techniques learned in radar to
use in that field.
E O Lawrence’s laboratory at Berkeley, California had led the world
in the design of high-energy particle accelerators before the war, of which
the cyclotron had proved most efficient. As higher energies were sought,
bigger—and more expensive—magnets were required, which had given
rise to a parallel project by D H Sloan to use acceleration by radio-frequency
fields but dispensing with the magnet by stringing out the electrodes in a
straight line. The idea pre-dated the cyclotron and, in fact, had inspired it.
This mode of acceleration had a series of cylindrical electrodes connected
to a high-frequency source. Particles were accelerated when residing be-
tween electrodes and drifted when they resided within the shielding of
the cylinder during the other half cycle, emerging again to receive another
acceleration. The electrodes had to be positioned to account for the in-
creasing speed of the particles. This idea had been the basis of the earlier
Berkeley accelerator, but owing to the low-frequency and low-power oscil-
lators then available, its use was restricted to heavy ions at experimentally
uninteresting low energies [32].
Luis Alvarez began working on the idea of a linear accelerator during
the last months of the war, inspired by the thought that there were thou-
sands of SCR-268s in existence that had been made obsolete by SCR-584.
The 1.5 m wavelength, pulsed high power and number of transmitters
available allowed dreams to extend to fantastic particle energies [33]. A
31 MeV machine worked successfully using 30 radar transmitters, operat-
ing with 300 µs pulses, later increased to 600 µs, and a repetition rate of
15 Hz. These transmitters were replaced by specially designed units [34],
439
probably the result of dissatisfaction with the relatively short life when
driven hard of the 480 Eimac 100TS tubes in the ring oscillators and with
the time required to re-tune when they were replaced [35].
William Hansen’s microwave research before the war at Stanford had
been directed toward constructing a linear accelerator for electrons using
resonant cavities. The electron’s light mass required wavelengths best
measured in centimeters rather than meters, if the accelerator was to have
practical dimensions. The availability of high-power microwave gener-
ators led to a number of such machines being built right after the war.
Although priority can hardly be ascribed great significance [36], the Ra-
diophysics Laboratory in Sydney won the race [37] with a design using
25 cm wavelength, quickly followed by TRE [38] using 10 cm. Hansen
fulfilled his plan and completed a machine at Stanford [39] only months
before his untimely death [40]. Much greater electron energies would be
needed before this technique would be of use in investigating the nucleus,
eventually done at Stanford and completing Hansen’s pre-war goal.
I I Rabi’s laboratory at Columbia University had led the world in the
atomic and molecular beam experiments to which he had added radio-
frequency resonance for nuclear magnetic moments in a static magnetic
field, but they had been unable to perform one experiment that was high
on their list as being of fundamental importance: accurately determining
the hyper-fine structure of atomic hydrogen, the tiny energy difference de-
termined by whether the atom’s proton and electron magnetic moments
are parallel or anti-parallel. (This is the transition that is responsible for
radio astronomy’s 21 cm radiation.) From a knowledge of the already
measured magnetic moments of the proton and electron, quantum me-
chanics showed that radio frequencies were needed for which laboratory
capabilities in 1939 were just approaching. This restriction fell while the
members of the group were away at Rad Lab, and on return to civilian
habits they concentrated on hydrogen in studies that were to carry them
beyond atomic structure. By the end of 1947 they had measured the hy-
perfine splitting and had found a discrepancy with the theoretical values
ten times their experimental error, but it was the Golden Age of physics,
and virtue triumphed as the authors were able to add in the proofs of their
article that others had shown the magnetic moment of the electron differed
from the expected theoretical value, bringing experiment and theory into
happy agreement [41].
The hyperfine splitting of hydrogen was not the only mystery this
simplest of atoms had to offer. The rules of quantum mechanics that say
yes or no to its changes of energy allow the atom’s electron to be trapped
in an excited state for periods of seconds from which it can be induced
to decay by various external means. Here the electromagnetic effects of a
new discipline, quantum electrodynamics, disclosed themselves through
experiment in what became known as the ‘Lamb shift’ after its discoverer,
the details of which we must forego. While determination of the hyperfine
440
splitting went on in one part of the Columbia University Physics Depart-

ment, this delicate little structure was mapped in another [42].
In both of these experiments the magnetic moments of the nucleus
and the electron interact with one another and a static magnetic field im-
posed from without. Changes in the orientation of the little atomic magnets
alter the energy of the system and are brought about through stimulation
by high-frequency radiation. These experiments were performed on free
atoms in space, but there are also couplings of electron and nuclear mag-
netic moments to external static fields when the atoms are not free in space
but locked into solids or liquids. In such systems high-frequency radiation
can also produce measurable effects, but here the results did not carry the
observers to the dizzying heights of theory but to an analytical technique
that would transform chemistry: nuclear magnetic resonance or NMR. In
it the proton’s resonant frequency measures the total static magnetic field
which it experiences, a field that has two components: the field imposed
from without by the experimenter and the local field that results from the
molecular composition. This latter proved a useful quantity for unraveling
molecular mysteries, and many years later in the computer age, it would
become a useful medical diagnostic instrument, although cleansed of the
dreaded word ‘nuclear’ or even its initial, becoming magnetic resonance
imaging or MRI.
Two investigators invented independently complementary tech-
niques of making these observations. E M Purcell left Rad Lab for a position
at Harvard, and Felix Bloch left the Radio Research Laboratory (counter-
measures) to return to Stanford. Both had conceived of a resonance formed
in flipping the proton in a solid from one orientation in a static magnetic
field to another. By the end of 1945 both had concluded a successful ex-
periment. Purcell’s observed the alteration of the load experienced by the
oscillator when the external magnetic field passed through resonance [43].
Bloch used the vibration produced in the proton to induce an oscillatory
signal in pick-up coils so arranged that they received little or nothing from
the oscillator directly [44].
None of these experiments, neither the hyperfine splitting, the Lamb
shift nor the nuclear magnetic resonance, used klystron or magnetron. Fa-
vorite tubes brought from radar were the RCA triodes 2C40 and 2C43, de-
signed as microwave amplifiers but excellent for low-noise work at longer
wavelengths. Klystrons did find wide laboratory use, magnetrons much
less, although their magnets were encountered performing a wide variety
of functions. It is unlikely that anyone expected the magnetron’s future to
lie in the kitchen.
Molecules are capable of rotating or vibrating as solid bodies, al-
though restricted by the laws of quantum mechanics to discrete energy
states. The quantum energies that drive them from one state to another
are in the infrared or longer wavelength bands, and some of these tran-
sitions fall into the region of microwaves. A transition in water vapor
441
presented itself uncomfortably at Rad Lab as work progressed on the

1.25 cm H2K microwave system. Tests made in spring 1944 gave sig-
nificantly shorter ranges than had resulted in winter, and the cause was
absorption by water vapor in the air. This was not a surprise. J H Van
Vleck had predicted in 1942 possible water absorption in this region as
well as by oxygen for the 0.5 cm band. The problem was, as pointed
out in his report, neither the exact wavelengths nor the linewidths, deter-
mined by collisional effects, could be given with accuracy sufficient for
engineering needs. The value expected from the much finer resolution of
the H2K radar caused development to be pressed despite uncertainties.
Data were sought from four different approaches: (1) controlled radar-
range measurements, (2) infrared experiments on water lines of shorter
wavelength that should have the same degree of collisional line broad-
ening, (3) absorption in a resonant cavity with varying amounts of wa-
ter vapor and (4) atmospheric measurements with Dicke’s radiometer.
The result of all this was that the water line was found to be right on
the 1.25 cm wavelength. The limited amount of H2K equipment pro-
duced was restricted to low-altitude bombing and production was cur-
tailed [45].
Out of Dicke’s atmospheric absorption measurements came one of
the first experiments of what would be a large part of postwar experimen-
tal work on molecules [46], and out of Van Vleck’s studies came better
understanding of collisional broadening [47]. The water resonance study
was the beginning of a very large postwar field. From Oxford came the
measurement of a number of resonances between 1.1 and 1.5 cm in am-
monia [48], which was to lead to masers in the hands of Charles Townes
[49], thence to lasers and an entirely new and rich approach to the study
of atomic structure.
The microwave studies of ammonia demonstrated in a dramatic form
the difference that a new experimental technique can make. Before the war
the literature gave only a single broad resonance of 1.1 cm in ammonia,
tediously made with home-made split-anode magnetrons, notorious for
their poor frequency stability. Klystrons made the difference [50].
10.2.3. Meteorology
Mariners and aviators had noted early that fog and clouds were invisible
to 10 cm radiation but that rain storms showed up clearly, the consequence
of the radar reflection of raindrops and hail stones, which increased as the
sixth power of their radius. Snow flakes were visible on radar too, but the
description was less quantitative. With a bit of experience the plan position
indicator proved to be a useful weatherman. The wartime results of US
Navy experience were published [51] with a number of PPI photographs of
thunderstorms, cold fronts and a striking series of the infamous typhoon
of 18 December 1944 in the Philippine Sea. The Radiation Laboratory had
442
not neglected this subject and reported similar studies with both 10 and
3 cm equipment [52].
Meteorologists found radar sets available after the war and added
them to the traditional instruments of their craft. One of their first in-
vestigations mapped the vertical structure of storms with a transportable
10 cm height-finding set, AN/CPS-4, and reported a layer, called the ‘bright
band’, straggling the 0◦ C isotherm of stratified clouds, thought to consist
of an ice–water mixture, and a column structure, thought to consist of con-
vective rain or hail or both [53]. Others showed the strength of the radar
signal to be proportional to the amount of precipitation [54]. The field even
attracted one of the radar eminences, E G Bowen, who sought to under-
stand rain formation from ground and airborne observations [55]. Radar
was to remain a highly visual instrument of this science.
Closer to the workaday routine of the meteorologist was the use of
a radio direction finder, SCR-658, for tracking weather balloons [56] and
whose steerable antenna often has caused it to be identified as a radar set.
During the war meteorologists frequently made use of accessible radar for
this.
10.2.4. Semiconductors
Of the science whose origins can be drawn in some way from radar, the be-
ginnings of the understanding of semiconductors is easily the one that has
led to the most substantial changes in human life, yet it was the wartime
research that was easily forgotten at the time among the more sensational
ones already reported. The crystal detector, the non-linear conducting bar-
rier formed by a metal wire pressed against the surface of some kind of
semiconducting crystal, was one of the early means of receiving radio sig-
nals. The vacuum diode, followed by scores of vacuum tube types, quickly
relegated the crystal detector with its vagaries to a cabinet of discarded
equipment or to a child’s toy. The predictability of vacuum triumphed
over surface conditions difficult to control, much less understand.
As experimenters approached microwaves they found that the elec-
tron transit times and inter-electrode capacitances of vacuum tubes made
their use impossible and noted that whatever the faults the crystal de-
tector had, they did not extend to transit times and large inter-electrode
capacitance. Hans Hollmann utilized it as an element of a regenerative
receiver, which he described in his 1936 textbook on high frequency [57].
At about the same time Southworth was studying waveguides at Bell Labs
and had incorporated crystal diodes as rectifying detectors [58]. Rectifiers
for power circuits were devised during those years using various kinds
of layered surface but with indifferent understanding of the basis of their
operation.
Microwave radar demanded a mixer for heterodyne reception, and
crystal-detector methods were taken from vague memories or from the
library. The purest grades of elemental silicon were found to be the best
443
surface on which to place the cat’s whisker, but purity was not easy to
obtain. At GEC B J O’Kane and G C Edwards had used a silicon–tungsten
catwhisker in a coaxial line for their 25 cm work in early 1940, and GEC fur-
nished silicon to other researchers [59]. After experiments by H W B Skin-
ner, British Thompson-Houston Ltd manufactured them as did a section
at Rad Lab [60]. Engineers had little patience with the erratic behavior of
these early diodes, and basic research was the obvious solution. Henry
Torrey at Rad Lab coordinated a program that utilized outside research
groups, principally the University of Pennsylvania under Frederick Seitz,
who concentrated on silicon, and Purdue University, under Karl Lark-
Horovitz, who concentrated on germanium. Silicon became the chosen
element because of its temperature stability. The first requirement was ex-
tremely high chemical purity, obtained through fractional crystallization;
the second was the discovery of the importance of controlled, microscopic
amounts of an impurity, boron being found to be the best [61].
Few dreamed of the effect on the world of the subsequent invention
of a three-element semiconducting device, the transistor in the 1950s, and
none imagined the effect a decade later when the intrinsically small tran-
sistor was incorporated into the chip. To attribute all this to radar stretches
an historical connection to the breaking point, but radar was a station on
its technical progress.
Such was the rich legacy radar left the civilian world. Except for radar
astronomy, all would have come into being without radar, for all had begun
to grow, however haltingly, before 1939. What radar contributed was the
intellectual fire in the postwar investigators. They had made radar. It had
been their unique fate, and they acknowledged no limits. They sent men
to walk on the Moon.
10.3. SECRECY AND THE TECHNICAL IMPERATIVE

The affairs of men have ever been entangled with secrecy. A poker player
does not want the contents of his hand known to the other players nor a
general his plan of maneuver known by his adversary. But whereas hon-
orable card players do not try to penetrate the secrets of the others except
through the psychological guile that is part of the game, nations consider
the acquisition of such knowledge about hostile, potentially hostile or even
friendly nations an honest profession. Military intelligence accumulates
secrets by various means, sometimes adventurous and romantic but gener-
ally dull and repugnant; to the information obtained surreptitiously belong
huge amounts of information obtained by tedious study of open sources.
In modern military organizations intelligence is the occupation of staff of-
ficers at almost every level of command from battalion up; at the top it has
become a substantial bureaucracy.
Radar was from the first enveloped in great secrecy by all parties.
This was, of course, quite natural and an alternate course seems in ret-
444
rospect hardly possible, and yet it is not far from the truth to say that
there were only a few aspects that were worthy of the severe restrictions
imposed. For radar itself was not a secret, amply demonstrated by its
being pursued independently in eight different countries in 1939, whose
differing degrees of advancement had their origins in the support pro-
vided rather than in differing stores of fundamental and clever ideas. All
of the important inventions of radar were made separately as the need for
them arose: extreme peak-power transmission for pulsed signals, broad-
band radio-frequency amplifiers, common-antenna usage, lobe switching,
conical-beam scanning, triodes for decimeter transmitters, plan position
indicators, IFF directly and indirectly interrogated. There were three inde-
pendently conceived developments of radio-navigation systems that used
hyperbolic coordinates: Gee, Decca Navigator and Loran. One might in-
clude the German Sonne as a fourth, although it placed the master–slave
pair only a couple of kilometers apart and used the nearly radial asymptotic
grid lines.
A similar situation applied to countermeasures. The potential of
metal foils cut to dipole length for the incident radiation and thrown out
by attacking aircraft so terrified British authorities that their use against
Germany had to be decided at cabinet level in 1943, yet it had been sug-
gested five years earlier and was well and early known to the three major
radar powers and used first by Japan in May 1943—information that did
not reach the European theater. It so terrified Göring that he forbade even
research in methods to counter it. Other jamming and interference tech-
niques were almost self-evident. On the other hand each radar group
thought that it was unique in coming onto radio location, and there does
not seem to have been much discussion questioning this assumption.
Keeping things from the enemy begins by restricting knowledge from
all friendly personnel ‘who do not need to know’. This seemingly logical
principle assumes that those who set the restrictions know themselves
who needs to know, and it is with this that matters become discordant. It
is instructive to insert here Hitler’s obsession with this.
1. No one, no office or officer, may learn of a secret matter unless

this is absolutely necessary in the line of duty.
2. No office or officer may learn more about a secret matter than
is absolutely necessary for carrying out the task in question.
3. No office or officer may learn about a secret matter or the nec-
essary part of a secret matter earlier than is absolutely necessary
for carrying out the task in question.
4. It is forbidden thoughtlessly to pass on orders, the secrecy
of which is of decisive importance, according to some general
distribution list [1].
445
Even a casual knowledge of Nazi Germany will have led the reader
to the belief that secrecy was driven to extreme there. A radar man recalls
the very strict security under which people worked.
Even the slightest violations were punished extremely hard, so

it was best not to know about things, and people got into the
habit of forgetting everything they had heard when they left the
office. We had a slogan: ‘Secret! Burn before reading!’. And it
was more than a joke. It was dangerous for unauthorized per-
sons even to know that a secret document existed, especially
those marked ‘Geheime Reichssache’ (secret of national impor-
tance) [2].
In Britain the radar research that was carried out for the Air Ministry
at TRE, famous for Rowe’s Sunday Soviets, dispensed with any ‘need to
know’ rule for those cleared for the level of secrecy being discussed. Not
surprisingly the situation in Germany was different. A prominent engineer
for Lorenz writes in the company radar history of the situation created by
secrecy:
In the last years of the war there were approximately a thou-

sand [!] different radar projects. This degenerate situation,
which was completely out of control, resulted from extreme se-
crecy and from the orders pertaining to secrecy, which were
made by persons having little or no knowledge of the technical
situation. This made coordination of the needs of Army, Air
Force and Navy with research and development by industry
practically impossible.
Among other things this led various services to request the de-
velopment and even to purchase for the same purpose com-
pletely different pieces of equipment, secrecy covering extremes
of independent administration. Immature designs, even fail-
ures were covered under this protection. Reciprocal exchange
of information was practically impossible except in special cases
when it had been ‘filtered’ through the proper offices, which
generally robbed it of technical detail [3].
Especially puzzling was the slowness of the Germans to make use of

their excellent universities and technical institutes, which were employed
only slightly for war research until relatively late, in marked contrast to the
Anglo-American case. The Kriegsmarine prevented GEMA from utilizing
them until 1944 on grounds of secrecy [4].
Things were not all that great among the Allies. Production of SCR-
584 did not begin until a year after the first order, and it was not until early
1944 that the first set reached combat, delays in part attributed to a mis-
understanding of material priorities in the War Department that had its
446
origin in secrecy [5]. But for secrecy London might have been able to fend
off the V-1s with effective anti-aircraft (AA) fire from the beginning rather
than after weeks of destruction followed by frantic equipment shipments
and the hectic training of battery radar sections in its use while the bombs
flew. A close-in fighter-control radar for the US Navy, the SM, designed
at Rad Lab and based on techniques developed for the 584 went into ser-
vice six months before the Army set [6] because the Navy Department’s
assignment of priorities was not so entangled.
The United States showed signs of imitating Germany in the matter of
circuit diagrams and instruction manuals when distributing the 584 in the
Pacific. The new equipment began to arrive in March 1944, but the manuals
were marked ‘secret’ and for that reason did not reach the troops, so the
sets remained in their crates, unused. When knowledge of this situation
reached Washington, the Signal Corps dispatched in July H B Abajian, who
had worked closely with Getting in designing the set, to provide instruction
in the use and the great, but at the time unappreciated, advantages of the
new radar [7].
It would have been to either side’s advantage to relax secrecy re-
quirements significantly, but such a decision would have to have been
made at the very top level of government by someone with Solomon’s
wisdom and authority combined with an incredibly broad knowledge of
electronics. The grip of the conventional and the weight of responsibility
were simply too great. Nevertheless, a faint attempt in this direction was
made in 1938. The Deputy Director of the Scientific Research and Experi-
ment Department of the British Admiralty saw the article in the New York
Herald Tribune of 21 March 1938 that had described an approach to radio
location; he suggested that thought should be given to the desirability of
announcing Britain’s RDF capability and, inasmuch as radio location was
known in America, of opening exchanges with the American government
[8]. An absence of anything further on this tells us how far the ‘thought’
went.
Modern technical industry is beginning to appreciate this not ob-
vious attitude, as demonstrated by a comparison of the two American
centers of technical innovation. Silicon Valley, where there is almost no
attempt to retain industrial secrets, has left Boston’s Route 128, where in-
dustrial secrecy is highly valued, behind [9]. Such an attitude is rejected out
of hand by Akio Morita, noted leader and co-founder of the enormously
successful Sony Corporation [10], although his reasons seem to rest more
on his repugnance with espionage as a way of conducting business and
with employee disloyalty rather than from a careful evaluation of the ef-
fects.
For radar to be effective it had to be used by troops, and here the
cold hand of such rules as The Official Secrets Act with penalties for slight
infractions being ‘hanged by the neck until life is extinguished’ made them-
selves felt. It is remarkable to read in the memoirs of the commander of
447
an RAF ground-radar unit deployed in France after the Invasion for Army
Cooperation Command:
Generally we did not know the squadrons by their number but

by their R/T call signs Potter, Station, Jamjar, Iceberg or Wonder
and so on. We answered with our call sign of Bazar but we
did not know the pilots nor they us, they did not know of the
capabilities of our radar or that we had one, we were just a
control voice on the radio. I just hoped always that the orders
and information that I gave over the radio had enough ring of
confidence for the pilots to believe in what I was telling them.
Their lives could depend on it [11].
The free exchange that marked TRE’s Sunday Soviets became stran-
gled by regulation and fear as one descended the chain of command, but
as noted in Chapter 5.1 this attitude seems to have been common to Army
Cooperation Command and may have resulted from policy decisions un-
related to secrecy. By contrast Fighter Command continued its free inter-
action between GCI directors and AI operators in night fighters [12].
The personnel at CH stations were strictly segregated according to
their duties. The radar equipment was within guarded barbed-wire en-
closures, and the administrative staff of the station, some of whom were
superior in rank to the operators and mechanics and were responsible for
the station, were not allowed inside. If there were two kinds of equipment
being operated at the same station, crews were restricted to their own set
and not permitted to learn about the other. At mess, conversation could
not turn to technical matters [13].
It is not surprising to read of even worse restrictions on the other side:
Even more astonishing, Major Schulze was forbidden in March

1942 by the Commanding General of the XII. Fliegerkorps to
speak to the commanders of the air warning units about night
fighting techniques. . . . This ‘Geheimniskrämerei’ (keeping
shop with secrets) took on such forms that officers, who should
have been giving thought to ways of improving the air-warning
service, were insufficiently oriented about night fighter use of
radar [14].
Things were not much better in the Royal Navy. Derek Howse writes
in Radar at Sea:
While the very high level of secrecy about radar in the early
days had the virtue of denying information to the enemy, its
continuance once radar became operational—and particularly
after the outbreak of war—was a great hindrance to the proper
use of radar in the fleet. During the first year of the war, the
448
majority of those at sea were ignorant even of radar’s very exis-

tence, let alone its capabilities and limitations. And this applied
not only to the junior ranks, but to senior officers as well, unless
it so happened that they had recently served in a staff appoint-
ment where there was a need to know about it. For example,
when Rear-Admiral J G P Vivian hoisted his flag in Carlisle as
Rear-Admiral AA Ships before the landings at Aandalsnes in
April 1940, he knew nothing about radar, although it was fitted
in three of the AA cruisers he commanded [15].
The extraordinary ineptitude that marked German and British use

of radar in the Dieppe Raid was dealt with at length in Chapter 5.3; it
had much—although certainly not all—of its origins in the secrecy that
enveloped the technique. Command and staff at the levels of Lord Mount-
batten and General Montgomery and at comparable levels among the Ger-
mans were either ignorant of radar or ignored its potential. Fighting men
never realized that they had an enemy in their own ranks named secrecy.
The restraints of secrecy among Allied personnel were much more
relaxed in the Pacific than anywhere in Europe. Other than the air war
over Germany, no theater made such normal, daily use of radar, whether
for navigation, defense against air attack, fire control or in the struggle
with or against submarines. Everyone seemed to know about radar and
appreciate it, as indicated in a piece of scurrilous doggerel written and
quoted in Chapter 5.5 about Admiral Scott. The common feeling was cer-
tainly that lots of people ‘needed to know’. This contributed greatly to
the ever increasing efficiency and ingenuity of its deployment in that the-
ater. This relaxation did not happen from a command decision; it simply
grew among the personnel, to a large degree as a consequence of their
complete isolation from possible eavesdroppers, but it was not repressed.
Indeed by July 1944 this attitude had infiltrated official publications. The
introduction of a Navy intelligence document concerning Japanese radar
admonishes security officers:
The widest possible use should be made of the TECHNICAL

DATA ON JAPANESE RADIO AND RADAR EQUIPMENT.
Although the material is CONFIDENTIAL, that classification
should not be a bar to proper accessibility to this bulletin. Mea-
sures should be taken to insure availability of the report to Ser-
vice personnel who may profit by reading it [16].
Secrecy provided a potent weapon for the bureaucratic wars between

various services. We have seen how the Kriegsmarine kept DeTe II from
the Luftwaffe and even tried to prevent the Luftwaffe from purchasing
equipment from GEMA. In Japan the Imperial Army gave specific orders
to those working for Ikuta Research Office of NEC not to communicate
any radar information to the Imperial Navy. Those responsible for such
449
stupidities no doubt had reasons that seemed valid at the time but they
certainly puzzle an outside observer today.
Much was made in the United States by scientists and administra-
tors in the National Defense Research Committee about the Army and
Navy being ignorant of each other’s radar activities before the war [17],
statements based solely on administrative knowledge and that these pages
have shown not to be true for the designing engineers, but it was true at
command level.
Colton heard that the Navy was detecting airplanes by radio,

but was advised by his colleagues that the project was secret and
the Navy would not talk to the Signal Corps about it. Colton ap-
parently did not learn of the work that Hershberger was doing
at Fort Monmouth until after the first part of 1935 [18].
At Hershberger’s request, Page agreed to forward the brief NRL
monthly reports to the Signal Corps Laboratories. The first in-
dication that NRL had achieved success with the pulse method
seems to have been provided to the Army by the sudden and
unexplained cessation of these reports. This occurred when the
project was upgraded from a confidential to a secret classifica-
tion by the Navy. Apparently the Army was not permitted to
receive secret data from the Navy [19].
As demonstrated here, the engineers at the two labs managed to get

around these restrictions when the occasion arose. A striking example
of this is the Signal Corps use of Page’s ring oscillator for the SCR-268
[20]. This clever design allowed several commercially available Eimac
100TH triodes to be formed into a very powerful 1.5 m transmitter. Other
designers, world-wide, clung to the push–pull circuit, hence demanding
more powerful pairs of output tubes. Page alludes to the transfer as having
taken place in 1937 [21]. In another case, when Bell Labs demonstrated the
first CXAS, the NRL people wanted it equipped with the lobe switching
they knew was used on the SCR-268 but which they had not incorporated
into NRL’s XAF [22].
With authorities keen on preventing knowledge of even the very ex-
istence of radar from escaping, an incident in 1939 is amusing. A German
publisher issued annually a pocketbook describing the ships of the world’s
navies. When the 1939 volume appeared it caused consternation in Ger-
man radar circles [23] because of a photograph dated 1938 of the Torpedo
School Ship G 10 displaying a prominent Seetakt antenna just forward of
the foremast [24]. The photograph was passed for publication by various
naval authorities, all kept in the dark about the new technique and, of
course, unable to recognize the apparent mattress as the mark of a secret
weapon. There is every reason to assume that the British naval attaché
in Berlin purchased the book and that naval intelligence studied it, but
450
there is no record of them having grasped the significance of the antenna

either—very likely for the same reason that the picture had escaped in the
first place. R V Jones was unaware of the matter until informed of it by the
author [25].
This was followed, as we have noted in Chapter 3.1, about a year
later when L H Bainbridge-Bell, a competent radar man, was sent by the
Admiralty to examine the wreck of the Admiral Graf Spee in Montevideo. He
sent a description of the Seetakt antenna only to have its significance lost on
naval intelligence in London [26], very likely because of some convoluted
demand of security.
Such failings did not mark the scientific intelligence section started
by Jones in September 1939. Although immersed in Britain’s radar very
early, he had not incumbered himself with the idea that radar was theirs
alone and began hunting for evidence of it across the Channel long before
anyone else thought it remotely possible, supported in his quest by the
disclosures of the Oslo report. He finally convinced British officialdom in
early 1941 when photographs of a Freya were at hand. When a photograph
of a Würzburg-Riese on a Flakturm at the Berlin Tiergarten came to him
through the help of the American Embassy in Berlin, he was not long
in putting the correct interpretation on it. In this he had the help of a
Chinese physicist who had seen it and reported it to Jones—after suffering
indignities at the hands of suspicious British officials [27].
The cavity magnetron was one technical secret for which every con-
ceivable reason combined to demand secrecy, yet, as we have seen in Chap-
ter 4.1, it was following the technical imperative just as did other compo-
nents, although less uniformly with time. It probably eluded American
microwave designers because of the hold that the klystron had taken on
their thoughts. It probably eluded German designers because of their dis-
appointment in the early years with the poor frequency stability of the
split-anode magnetrons—and their failure to read German language engi-
neering journals where the Brown–Boveri work was reported. Randall and
Boot had the advantage of knowing nothing about magnetrons—a signifi-
cant gain inadvertently resulting from the Bawdsey work having initially
ignored microwaves!—and going back to fundamental papers by Hertz in
their reading.
Concern for the magnetron unnecessarily cost the lives of British
fliers. An effective way of reducing the number of Luftwaffe night fighters
attacking Bomber Command was to have British night fighters attack their
German counterparts as they left or returned to their bases, which could
only be done with 10 cm AI because of its reduced ground returns at low al-
titude. This was foiled through the prohibition of fighters with microwave
AI from supporting Bomber Command until May 1944 [28] even though
H2S had been in use over enemy lines since December 1942. (Night-fighter
support using 1.5 m AI for Bomber Command had been forbidden until
August 1943 [29] even though the German use of 50 cm Lichtenstein was
451
known to the Air Force well before.) In fact, although unknown at the time
of course, concern for the loss of the magnetron was misplaced because the
Germans made remarkably little use of the device when they did obtain it,
despite their immediate recognition of its significance and value. On the
contrary, it contributed significantly to the splintering of their efforts.
The Germans were just as convinced as the British of the value of
their radar secrets. An instructive example of the way an obsession with
secrecy can hurt were the measures taken at coastal radar stations after
the Bruneval raid by constructing fortifications around them for protec-
tion against future commando action. This made every station stand out
in aerial photographs, which greatly simplified their destruction before
the Normandy invasion. Prior to the emplacement of barbed wire and en-
trenchments it had been difficult to locate these stations exactly, although
approximate positions came from direction-finding techniques. It is a mon-
strous example of locking the barn after the horse was gone, as there was
no need for the Allies to take another set. It is not known whether radar
engineers were consulted on taking this step. Not very likely—it was se-
cret!
Given the enormous amount of espionage material shipped from the
United States to the Soviet Union during World War II [30], it would be
surprising if radar had not had a sizeable component, but only one spe-
cific instance seems to be recorded. Julius Rosenberg is reported to have
told David Greenglass, his brother-in-law and accomplice spy, that he had
stolen a proximity fuze while working briefly at Emerson Radio in 1945
after having left the employ of the Signal Corps [31]. Inasmuch as he was
bragging, it seems doubtful that he would have left off something he had
obtained on other forms of radar. The absence of evidence of any other
covert espionage about radar leads one to think it was minor, to Germany
probably zero. It is unclear what the stolen fuze yielded. A most important
component of the secret of its manufacture was in quality control, which
could not be so easily transferred. Radar did not have high priority in the
Soviet Union after the original flowering in the early 1930s, so there were
probably no specific demands for radar information as there were for the
atomic bomb. And there was little need when the SCR-584 was finally
sent as Lend Lease, for it contained about every important radar secret
that existed. The loving copy of it as SON-4 and long use tells of Russian
evaluation [32].
It is idle yet unavoidable to speculate what secret agents of the radar
powers might have procured had they been able to penetrate the electronics
research laboratories at the beginning of the war, an occurrence that has left
no record in a huge amount of published material, if it did take place. For
those using and countering radar, the radiative characteristics of frequency,
pulse repetition rate, polarization, beam width and power were sought
urgently. These, however, could be obtained by listening, although not
always an easy or safe procedure. The ability to resist jamming was the
452
principal design parameter that might have been obtained by a spy.

Designers would have naturally been curious about what the peo-
ple on the other side of the hill were doing, but it is questionable whether
possession of detailed circuit diagrams would have made more than a per-
turbation on the work being done at any given time of disclosure. Consider
what the consequences might have been, if a British master spy had ob-
tained in late 1940 or early 1941 detailed plans for the Würzburg. Bear in
mind that its superiority was only beginning to be recognized by the Ger-
mans. Months of adaptation by troops and engineers were indicating that
it was going to be an excellent set for both Flak and Luftwaffe, but these
were properties not yet felt by Bomber Command. In Britain the drawings
would have drawn from the many who had little confidence in AA artillery
a perfunctory dismissal, overruling General Pile’s instinctive positive re-
action to it. The Admiralty would have pointed out that they already had
decimeter triodes and were well on the way to making their own 50 cm
fire-direction set, the type 284, with them. In America, informed through
the exchanges established through Tizard, the Naval Research Laboratory
would have said that the Bell Labs CXAS, rapidly becoming the FD, was
as good or better; the Signal Corps would have still preferred SCR-268
for the same reasons they had declined CXAS except for limited use in
coast defense. Thus a tremendous espionage coup might well have been
received by design engineers with ‘Why do you suppose they did it that
way?’. Those planning countermeasures would have noted with inter-
est that the Würzburg—like nearly all radars of the time—had too small
a frequency range to resist jamming, and such were the reactions to the
Würzburg components seized in the famous commando raid.
The attentive reader will counter that the Japanese valued highly the
notes of Corporal Newman describing the SLC radar captured at Singa-
pore. The extent to which they helped the engineers transform it into the
Army’s Tachi 1, 2 and 4 searchlight and gun-laying radars is not clear. No
such manual was obtained for the GL mark II, which was the basis for a
somewhat improved copy in Tachi 3.
The most difficult task in dealing with secrecy is the official status that
it assumes quite independent of its original purpose. Elaborate procedures
are required for the control of secret documents, for the alteration of the
levels of classification and for the grading of personnel to limit access; the
conduct of this business becomes the concern of a special class of official
who guards his tasks and secrets jealously. When the decision to label
something secret is left entirely to him the results can approach the absurd.
This is self-destructive, but a cure addressed to a national requirement does
not allow brash treatment.
Without question the most profitable use of putting secrecy aside
came from the Tizard Mission [33]. This extraordinary and unprecedented
exchange of information between the United Kingdom and the United
States in September 1940, strongly opposed by high-level parties on both
453
sides of the Atlantic, came about through the keen insight of A V Hill and
Henry Tizard reinforced by an understanding support from Prime Minister
and President, both of whom realized the importance of science to modern
warfare and were not timid. America gained the long-sought microwave
generator; Britain gained microwave techniques and access to American
electronic industry. Perhaps more important was the habit of exchanging
secret information even before America was a belligerent. Had there not
been a Tizard Mission, the Allies would have won the war but only after a
longer struggle, spawning who knows what postwar complications.
The exchange of technical material during the missions of the Impe-
rial Japanese Army and Navy to Germany in 1941 present a barren contrast
to the openness of Britain and America, for this Axis exchange was no Ti-
zard Mission. Although the Japanese were pleased with what they learned,
their ally told them essentially nothing about radar, although the knowing
eyes of Yoji Ito grasped much just from observation, and Japanese radar did
not start in earnest until his reports of its importance reached Tokyo. The
Japanese, on the other hand, told the Germans nothing about the resonant
magnetron nor about how to make good torpedoes, critical knowledge the
Germans needed.
A surprising and highly beneficial decision on secrecy came at the end
of the war. In a particularly progressive decision the Radiation Laboratory
ceased to exist by the end of 1945. This dispersed its great store of talent
widespread across the land, filling universities and industry with the latest
in electronic knowledge. I I Rabi had insisted that they produce a final
technical report and had set grumbling people to work on it. Louis Ridenoir
became editor and succeeded in getting the 28 volumes of a microwave-
radar encyclopedia declassified [34], making it a technical best seller that
went a long way in transforming the national scientific base. Selected
volumes were to be found near the workbench or desk of nearly every Rad
Lab veteran to remind him of how a problem similar to the one vexing him
at the moment had been solved in Cambridge.
All this will bring a wan smile to the face of an official charged with
deciding classifications, for it is easy to pontificate when in possession of
the knowledge of how everything has turned out. In order to make an
administratively wise decision about what to place on or remove from the
secret list requires knowledge of what the enemy knows. This is uncertain
at best, so when in doubt, clamp down. Two comments are needed as a
counter-argument. First, technical secrets differ from other military secrets
and should be treated differently than such matters as war plans, order
of battle, status of supply or decryption capability. Second, successful
military operations have always called for taking risks.
Technical secrets differ from other kinds in two ways. Foremost is the
inescapable technical imperative. If you can do it, a competent adversary
can too and may even be ahead. Second is the inevitable time delay in
putting into practice what is learned from the enemy. A commander may
454
act immediately on learning of the enemy’s plans, but a technical chief may
require months, perhaps years to duplicate some clever new device—if he
thinks it worth doing. During that time one’s own secrecy may have held
back effective use of the device being so protected.
The prime example of this kind of thing is the resonant magnetron.
The Germans made very little use of this super-secret device in the two
years they had the opportunity. On the contrary it splintered their al-
ready over-burdened resources. GEMA recognized this by ignoring it com-
pletely; they had vital air-warning sets to deliver and improve even though
their close collaborator, Rudolf Kühnhold, had continued microwave re-
search for a long time. Yet during those years Allied decisions were made to
protect AI-10 that cost airmen’s lives, even though there was every reason
to believe that a magnetron was in German hands. Allied exploitation of
the magnetron had required a development and industrial effort that took
two years before equipment began to have any tactical effects. Given this
experience they must have thought German engineers had super-brains
and unlimited resources.
At this point remarks should be made about the atomic bomb, as the
degree of Soviet espionage pertaining to it is now widely known. The
atomic bomb is, perhaps, one of the best examples of the technical imper-
ative. Physicists everywhere made rough calculations of a uranium bomb
almost as soon as they learned about fission. The experiments required
to decide whether such a device was possible were obvious. The time be-
tween the discovery of fission and the construction of a bomb depended
on the effort put into the project and the industrial and scientific resources
available. The United States and Britain gave the project all possible sup-
port and completed it in less than five years. What espionage unquestion-
ably did for the Soviet Union was reduce the development time, compen-
sating thereby for a later start and for a substantially weaker scientific and
industrial base.
The worst effects of technical secrecy are the restrictions on knowl-
edge imposed on one’s own forces through strict ‘need to know’ rules.
Unauthorized knowledge of radar was widespread among personnel of
the US Navy in the Pacific, and this contributed markedly to the efficient
operation of the fleet. This is the form of risk taking that should be borne
in mind by those responsible for classification. In this case it was a deci-
sion made from the bottom without official sanction. It is well to prevent
technical secrets from being gained by the enemy, but when these decisions
handicap one’s own forces then these restriction are themselves the enemy.
Better to let the enemy learn a technical secret from time to time than keep
knowledge from those who may be able to put it to use, even the ones some
official deems ‘do not need to know’. Let the enemy puzzle over what it
means, or wonder why it was done that way, or debate whether it was a
plant, or argue about what to do with it. Remember the magnetron!
455
10.4. AN EVALUATION
And what have we learned from this long, detailed story of grand sci-
ence, dramatic battles and wanton destruction? We have acquired some
clarity about the origins of radar, a matter where a substantial amount of
error and misconstruction, dare we even say mythology, can be found.
We have encountered repeatedly the simultaneous occurrence of an idea
independently at various laboratories when a common intellectual basis
had been laid; this is hardly a new observation and carries the name tech-
nical imperative, but because of the secrecy imposed, radar probably has
the richest supply of such manifestations that can be documented. We
have also noted that radar owes its beginnings to the large development
effort devoted to television by the broadcasting industry. We found dif-
ferences in approach to prewar radar development but found them not in
national or democratic–totalitarian characteristics, not even in widely dif-
ferent cultural backgrounds, but rather in whether the impetus came from
engineers at the bottom or from officials at the top. We have encountered
many examples that allow us to examine the sterile question about the
relative merits of physicists or engineers. We have seen the deformities
placed on rational actions and planning by the demands of secrecy. We
have examined radar’s status as a determining element in the greatest war
of history. Finally we have looked briefly at the revolutionary change that
radar and radar techniques made in navigation, the most important since
the invention of the ship’s chronometer. Paired with this remarkable civil-
ian application came a burst of scientific activity based on the electronic
skills that the engineers and scientists took back to their laboratories. Let
us examine these matters one at a time.
Who invented radar? This question appeared frequently in the years
following the war and did not lack for voices giving an answer. It was
the subject of many articles, scholarly or otherwise, and was discussed
at formal meetings and in countless bench-top discussions. Watson-Watt
awarded the title to himself with complete assurance and disputed in print
any challengers. Kühnhold certainly thought the title was his. Taylor,
Young and Page had equally good claims. But it should be obvious to any-
one who has persevered this far that the question really has no meaning;
it is a question much like ‘which vote decided the election?’. We gener-
ally imply by the word ‘inventor’ one or more persons who by study and
experiment discover or produce for the first time a new device. Radar
certainly qualified in the 1930s as a new device, but which one of the sets
that appeared qualified as first and, if we select one, when did it actually
become a radar?
If you begin with the question of who first thought of using echoes
of radio waves for locating an airplane or ship, you will find, depend-
ing on your outlook, either no satisfactory answer or dozens. Christian
Hülsmeyer may not have been the first to have thought of it but he was
definitely the first to attempt to do something beyond dreaming. The
456
patents listed by Tuska and Kern [1] alone demonstrate the fertility of
imagination during the prewar years in the matter of radio location. The
hundreds working in technical capacities on television of the 1930s had
seen reflections from aircraft from the beginning of their work. They were
impossible to miss with the amplitude-modulated equipment of the time
and were as common as airplanes. It would have been a dull engineer who
did not connect such observations with the idea that became radar. Had
the realization been within the means of a radio amateur, there would have
been hundreds of radars developed by 1939, but radar required moderate
financial support, so there were only a dozen or so.
Leave aside for the moment the resonant magnetron and consider
the meter-wave sets of the 1930s. Each was a composite of standard el-
ements of communication engineering. It only required that a problem
be formulated in order to receive the solution, if the solution lay within
the capabilities of the time. All discovered within months of one another
that transmitter tubes could be driven in pulsed operation to powers far
beyond their normal maxima. That an array of dipoles could be made to
produce a narrow beam with consequently high antenna gain for meter
waves was something already entering textbooks [2]. The basic idea for
lobe switching had already been used for methods of radio devices that
brought fliers to their desired destination by ‘flying the beam’; it required
only slight adjustment to provide accurate direction capabilities for locat-
ing an aircraft with the primitive radar beams. Methods of protecting a
sensitive receiver from the transmitter when using a common antenna de-
pended on knowledge of the position of nodes on a transmission line and
proved easy enough in application. The PPI indicator has a long trail of
inventors in thought; practice came to TRE first because of the acute need
for it in their first GCI equipment. Telefunken’s fighter-control method
did not require it, so German use came later. Rad Lab saw the immedi-
ate need for it in the SG. When circuit designers demanded ever shorter
wavelengths, the tube designers delivered transmitter triodes with leads
located to minimize the input reactance; Telefunken, GEMA, Lorenz, GEC
and Bell Labs all delivered the ultimate tubes within months of each other.
Some were better than others, but all provided the hearts of the 50 cm sets
that appeared in 1939 and 1940.
The powerful generator of 10 cm waves does not fit so well into
this pattern, yet its exception, now known to the attentive reader, actually
emphasizes the rule being propounded here. Certainly all who knew of
the invention by Randall and Boot of the cavity magnetron cannot think
of it in any other way than as their great discovery. Only pedants would
deny them the distinction of having invented the cavity magnetron. And
yet the Japanese deployed in mid-1942 operational 10 cm radars using it.
The Russians published a description of it in the open literature in 1940,
and the Swiss company of Brown–Boveri had been working on it since the
mid-1930s and first published in 1937. This device was indeed following
457
the technical imperative as did all the other key elements of radar, just
not quite so smoothly. Arthur Samuel at Bell Labs and A B Wood at His
Majesty’s Signal School both recorded designs that might well have led
them to the cavity resonator, had their work been pushed.
Phased-array radars appeared whenever designers wanted electric
scanning. The technique was first used at about the same time by GEMA
and Bell Labs, the former for the 2.4 m Wassermann and Mammut long-
range air-warning sets and the latter for the 10 cm mark 8 fire-direction set.
Alvarez used the method in a more clever way slightly later.
And so it goes. One could tell similar stories about less significant
parts of radar design, but the important part of all this is the independence
of nearly simultaneous invention. There is no evidence that any of the par-
allel work just described was not, of itself, original. Furthermore, if one
permits himself the thought experiment of removing any of the principals
from our drama—Watt, Kühnhold, Taylor, Runge—or even the lesser char-
acters, does he seriously believe that radar would not have moved forward
just as fast? The answer must surely be that the pace would have been the
same. The equipment would have been different, perhaps worse, perhaps
better, but there would have been plenty of radar prototypes in 1940. That
such equipment would have been tactically ready for the crucial battles of
1940–1942 cannot be said with the same assurance.
Radar is unusual in having sprung from civilian roots. Funding for
electronic research at military service laboratories during the inter-war
decades was grudgingly given until 1935, not even enough to provide the
best communication equipment. One encounters careless statements about
the negligence of various armed forces in not providing radar in the decade
right after World War I. These criticisms are based on imperfect technical
and historical knowledge. Until the early 1930s radar equipment was im-
possible because the components required for ranging did not exist, and
radar without ranging was of no significance in World War II. The lack of
range information was the reason that the collision-avoidance equipment
of Christian Hülsmeyer and Henri Gutton was summarily dismissed by
seamen.
Ranging in a useful form required the timing of the echo pulses, and
this needed a receiver capable of delivering an output pulse of only a mi-
crosecond duration. Such receivers could not be constructed using triodes
as amplifying elements; multiple-grid tubes were essential. Ranging also
required a method of determining, generally in the presence of extrane-
ous signals and noise, the time between transmission and return of the
echo. This required a cathode-ray tube capable of displaying the signal
obtained from the receiver as a function of time and fast enough to follow
it. Although cathode-ray tubes had existed since the turn of the century, it
required the electron-optically focused, high-vacuum tube to do this.
Both of these circuit elements, the pentode, which became the most
important multiple-grid tube, and the focused cathode-ray tube, required
458
development beyond the means of the military service laboratories of the

between-war decades. They were developed because the requirements
of radar were technically the same as those of high-definition television.
During the 1920s subscribers to popular magazines about science and engi-
neering read avidly about attempts to provide television. All were based
on electro-mechanical optical scanners or in rare cases on the extremely
slow gas-focused cathode-ray tubes. Capital flowed to the developers be-
cause radio broadcasting and the movies were clearly financial winners,
even during the depression, and television was rightly seen to be just what
the public wanted. By 1930 pentodes and suitable cathode-ray tubes were
available, and by 1935 broadcast television was a reality. The radar men
had the components they needed, and radar was also a reality.
(The determination of the height of the ionosphere with pulsed radio
waves in 1925 has often been put forward as the basis for a radar set.
Those experiments used pulse widths a hundred times longer than those
necessary for radar and consequently were not so severely affected by the
slow response of the electronics of the time.)
Radar may thus be unique in being a military usage having sprung
from a technology developed for civilian purpose. It may also be the only
military technology that has been more beneficial to mankind than the
civilian use from which it came.
A comparison of the American, German and British approaches to
radar is instructive. The similarities of the American and German ap-
proaches were emphasized early and have been noted by others [3]. In
both nations radio engineers observed the phenomena of echoes and insti-
tuted research that culminated in prototype equipment constructed of com-
mercially available components that they built out of meager funds. The
first American work was done at the Naval Research and the Signal Corps
Laboratories, and both had to use (illegally) funds appropriated for other
purposes in their first experiments; their directors had the foresight to rec-
ognize radar’s value and the courage to take a risk. Both labs worked early
with private corporations to widen design and provide for later produc-
tion. The first German work, although stimulated by Rudolph Kühnhold,
a naval research scientist, was funded by private capital in the correct ex-
pectation that the government would contribute to the development and
purchase the equipment they built. In both countries radar came from the
workbenches of the engineers and had to be sold to the high levels of the
war apparatus. The American effort found enthusiasm from a navy greatly
worried about air attack and an army concerned about its inability to shoot
down bombers at night. The Army Air Corps quickly saw a device to pro-
tect its bases from surprise attack. The German government’s response
was positive but with less enthusiasm; radar was and is a superb defen-
sive weapon, and defense was not foremost in the minds of those planning
the coming war. In both American and German examples the research costs
of the first prototype sets were comparable to the price of a bombing plane.
459
Britain’s approach was the inverse of this. By 1934 some political

leaders and many officials of the Air Ministry had become concerned about
Hitler’s rapid rearmament and especially its emphasis on the air arm, con-
cerns amplified by the knowledge of Britain’s woeful weakness and the
Air Force’s stated inability to stop bombers. With the object of clearing
away the numerous naive suggestions of a defense with ‘death rays’ they
went to the Radio Research section of the National Physical Laboratory
for advice. They were presented instead within days with a clever plan
for radio location. The plan was evaluated by an Air Force officer of re-
markable insight, who saw it providing the warning necessary to position
defending fighters to break up the formations of attacking bombers; he
provided the funds needed immediately and paved the way for all that
was required for a complete air-defense system. The result was a national
effort that produced the heroic engineering, CH, the only part of Britain’s—
indeed of democracy’s—defense that was ready that fateful September of
1939.
CH was a technical anomaly. It was designed by physicists specializ-
ing in long waves, not electrical engineers experienced in the meter waves
of television, engineers of whom Britain had many of the very best. But
CH worked, and it saved Britain and possibly western civilization in the
Battle of Britain, but it was a dead-end design. Its ability to track bombers
once they passed the coastline was quite limited, which became particu-
larly onerous when it forced the German attacks into the night. For this the
methods already in use in America and Germany had to be independently
learned and adapted, and CH had little to offer those designing them. It
is idle and mean spirited to criticize something so obviously successful as
the CH radar in accomplishing its original goals; nevertheless, it froze the
Bawdsey designs for far too long as long-wave solutions for problems that
demanded shorter wavelengths. The result was a scramble in 1939 and
1940. In his analysis of structural engineering David Billington noted that
providing an engineer with unlimited funds is a guarantee of obtaining
an inferior product; the best designs have invariably had tight budgets
[4]. A basic fact remains that must silence such criticism: Watson-Watt
and Wilkens conceived an effective defense quickly from their knowledge
of radio. The plan was adopted and carried through with a will. Would
someone else with a technically better plan have succeeded? Billington’s
criterion does not have a life-or-death parameter.
Much has been written about the dominant role played by physicists
in the radar projects. Watt did not want electrical engineers introduced into
the work early because he feared their judgement would be tied to the con-
ventional, and he was not alone in this, but there is little to be found in these
pages to justify this prejudice. One can say that physicists became first-
rate radio engineers speedily and with an enthusiastic verve, but Hanbury
Brown recalls in his memoirs how startled he was to encounter the much
better techniques and tools of the EMI engineers with whom he worked
460
in 1939 [5]. Until 1940 the American and German radars were designed
almost entirely by engineers, who showed themselves in no way devoid
of imaginative ideas.
The resonance magnetron of Randall and Boot is a prime example
of a physicist’s design—complete with sealing wax—and it was certainly
not conventional. Another great radar-designing physicist, one active in
the field for only two and a half years, was Luis Alvarez. To Alvarez must
be applied the term ‘radar scientist’ rather than radar engineer. Whereas
other scientists became superb engineers—one thinks immediately of Ivan
Getting—Alvarez remained a scientist. He conceived three startlingly orig-
inal projects, found for each a capable engineer to get things moving, and
left for Los Alamos. But he was a distinct exception. One need only exam-
ine the pages of the Rad Lab’s 28-volume encyclopedic final report. Those
loving descriptions of circuits were written by engineers; they may have
been physicists before or after, but at Rad Lab they were engineers. What
was true there was also true at TRE.
The doleful price often paid for secrecy does not need to be reiterated,
but closely associated with secrecy is the matter of administration. One of
the most remarkable properties of the British–American alliance was the
free flow of information and the collaboration in planning. If secrecy was
baneful in its preventing serving military personnel from learning about
the valuable weapon that was at hand, it did not inhibit work at the design
and planning level. Top radar engineers gained access to the knowledge of
colleagues at other laboratories with moderate ease. Similarly, there was
a reasonable attempt to provide a coherent overall development program.
True, there was duplication, which with the assurance of hindsight one can
say was unnecessary. It is extremely doubtful whether the Allies needed
four different groups working on 10 cm gun-laying radars: SCR-584, SCR-
545 and GL marks IIIB and IIIC. Yet who could have selected in early 1941
the group that would deliver a 584? Even before America’s entrance as
an active belligerent there were important Allied technical agreements, the
adoption of IFF mark III is an important example; disregard that it was
the poorer of the choices available, one that led to its use by the German
air-warning service as a range-enhancing secondary radar. At least it was
a good try to take control of this horrible and ever unsolved problem.
When the question came up during the war, it was generally assumed
that the totalitarian powers had a significant advantage over the democra-
cies in their ability to control their military, scientific and economic appa-
ratus with great efficiency from the top. Radar shows the very opposite.
In the beginning German radar outdistanced both Britain and America
in advancing excellent prototypes, well engineered for production. This
was the result of a small group of designers employed at three electronics
firms. These men found themselves relatively unhindered in what they
undertook, but once their projects became national ones, the stifling hand
of bureaucracy soon ‘controlled’ everything.
461
The German administrative organizations that had something to do

with the planning and production of radar continued to grow and change
throughout the war. It is difficult, very likely impossible, to understand
their structure now. The Nazi bureaucracies were jealous of one another
and fought—even as disaster was obvious—for control. The result was
serious duplication and fragmentation. Germany never had a common
IFF standard between the services. The muddle created by finding a cavity
magnetron in the ruins of a Stirling bomber in February 1943 worked almost
as a secret weapon for the Allies, for no coherent plan of exploitation can
be extracted from the Rotterdam Protocols or anywhere else, only greater
fragmentation and wasted effort.
Did radar win the war for the Allies? In a way this question is just
as meaningless as ‘who invented radar?’. Wars are won by determination,
resources, manpower, leadership, courage and luck. Among resources
are weapons, and superior weapons have had remarkable effects in many
wars. The democratic states, Britain and France, certainly lacked determi-
nation during the years Hitler built his war machine. The appeasement
at Munich left France relieved that war had not come but impressed the
British that they had suffered a serious and humiliating defeat that trans-
formed the whole nation to the same spirit that the radar men had held for
years. Thus when the island later faced the monstrous tyranny alone, they
had more than just spirit. They had a working air-defense system of the
kind never seen before and never even imagined by other air powers. Did
radar win the Battle of Britain? No, but it would have been lost without
CH. This statement is as categorical as any one can make in historical mat-
ters. The battle was extremely close; radar made more than the difference.
What would have been the fate of western civilization had Hitler won? At
that point thoughts become decidedly morbid.
The crucial value of radar for Britain in the Mediterranean during
1941–1942 has been discussed at length in Chapter 5.1 with the conclusion
that it provided the method by which supplies to Rommel’s forces were
seriously restricted. This prevented him from taking Suez and putting an
end to British naval control of the eastern Mediterranean. German control
of the Middle East meant control of Britain’s oil and the interdiction of
supplies to the Soviet Union. Japan’s leaders might even have considered
pushing on to link forces with their ally. The consequences of this are less
dramatic than the invasion and defeat of Britain but nevertheless dire. It
probably would have happened had the Royal Navy and its Fleet Air Arm
not had radar.
Similarly, the importance of radar to the American fleet in 1942 has
been treated in Chapters 5.4 and 5.5 with the conclusions that the Ameri-
can carriers would have been highly vulnerable to air attack without the
CXAM that gave them the warning needed to clear planes from the deck
and secure explosives and fuel. Holding Guadalcanal required the contin-
ued functioning of the Cactus Air Force, possible only through the timely
462
warnings of their SCR-270 and 268. Had the United States been defeated in
the Pacific in 1942—and it was a close thing—it is doubtful that America’s
Germany-first policy could have been retained. The reader can supply the
plot for his own drama.
These are three major engagements in which radar kept the Allies
from defeat. Such a statement can always be contested because there is
no way to prove it, but it is as near to a ‘truth’ as history allows. It was
certainly radar’s heroic period. It was also the meter-wave period.
After 1942 radar became increasingly routine for all the contending
forces, taking on an importance shared with other weapons. To have fought
without it would have been the same as removing quick-firing artillery or
machine guns, but it did figure critically at times.
The Battle of the Atlantic was certainly crucial for the Allies, and the
defeat of the U-boats has frequently been credited to radar, often in state-
ments that allow no contradiction, but the conclusion reached in Chapter
7.1 does indeed contradict them. The submarine lost when merchant ships
traveled in convoys protected by adequately equipped and trained escorts
with air cover over their entire routes. Radar was useful, but useful only to
the same extent as were other Allied technical advances and probably less
valuable than ship-borne high-frequency direction finding. The basis for
the decisive importance ascribed by some to microwave radar lies in graph-
ical displays of merchant-ship sinkings as functions of time. The decline
of these losses after the introduction of the equipment in patrol bombers
is impressive, as is the counter-diagram showing the increased sinkings of
U-boats, but these plots are deceptive: correlation is not causality.
The failure of the Allied invasion of Normandy would have certainly
stopped the Allied path to victory, which seemed so steady after 1942, and
failure was a definite possibility. Two key elements in providing success
were the remarkable experience the Allies had gained by spring 1944 in that
most difficult of military operations, an amphibious landing on a hostile
shore, and the near absolute command of the air over the beaches. Radar’s
part in this unprecedented operation was important, but declaring it vital
or not, as has just been done for the other engagements, is more difficult.
The radar-like devices, Gee and Decca, that allowed the entire fleet to
navigate under difficult conditions can be given a substantial credit for
success. For boats to arrive more or less on time and at the right place
would have been one of the most difficult aspects but for Gee. The extremes
of countermeasures plus the destruction of Seetakt and Freya sets in large
numbers certainly made things easier for the attackers and confused the
defenders during crucial hours. Radar beacons made air drops at night
decidedly more accurate. Would the invasion have failed without radar?
It is not so easy to say. There was a big force off shore with a strong
determination not to fail.
One of radar’s great moments came in the defense of London against
the flying bombs. The Allies would not have lost the war had the SCR-584,
463
the M-9 director and the proximity fuze not been ready, but London would
have suffered great destruction.
The ferocity of the Pacific War reached a peak at Okinawa with the
attacks by suicide pilots on the fleet supporting the invasion. This became
an extreme test for radar fighter direction and radar-directed AA guns. It
required a complete new concept for the former and proved difficult for
the latter. It is conceivable that the invasion might have been repulsed had
the fleet not had radar. It would not have cost the Allies the war, but it
would have made the last year much more bloody.
The air offensive against Germany has been called here the Great
Radar War. It has taken up more words than any other portion of this
book because its radar activities were so varied and extensive. In the wide
use of radar only the Pacific War can be compared with it, but the Pacific
War, although radar became a standard component of the two contending
forces, was fought without a countermeasure struggle comparable with
what was found in the air over Germany, primarily because Japan was so
completely outclassed. Radar shaped the struggle over the Reich, essen-
tially determining its course.
We have seen how initially the Luftwaffe was unable to down at-
tacking bombers at night and the RAF was incapable of even finding the
targeted towns. Ground radar continually improved the defense and the
radar-like device, Gee, allowed target location, although not accurate blind
bombing. Airborne radar began to make night fighters really dangerous,
and Gee was replaced by Oboe, a blind-bombing device as accurate as the
best visual. Gee and Oboe were both limited by the distance to the horizon
and so could not guide deep flights, a failure for which the microwave H2S
was to provide a solution. The solution allowed by H2S was saturation
bombing.
With the introduction of H2S at the beginning of 1943 Allied radar
became technically superior to German, but it was a superiority that the
bomber crews could hardly appreciate because the Luftwaffe made good
use of radar’s strong bias toward the defender. By March 1944 deep night
attacks could no longer be sustained. Radar had helped the Luftwaffe
dispel the delusion of penetration through the cover of darkness as months
earlier they had dispelled the American delusion of penetration with self-
protecting formations. Germany won the first round of the air offensive;
when the second round began in late 1944, dominated by the long-range
fighter, radar was much less important.
In some theaters radar bordered on being irrelevant. It figured not at
all in the Blitzkriegs east and west that initiated the European war. Its im-
portance in the Great Patriotic War fought by the Soviet Union was minor,
except perhaps in the air defense of Leningrad and Moscow. The Japanese
used radar to observe the swarms of attacking bombers that systemati-
cally destroyed their cities, but they no longer disposed of an effective
air-defense force. Radar bombing became the rule in destroying Japanese
464
cities and was much more effective than when used against Germany be-
cause the weak air defense allowed bombing from half the altitudes to
which the RAF had been forced by German flak. The location of so many
targets on the coast, just where H2X provided the best PPI views, and the
inflammability of the cities made a terrible difference.
By the end of the war in Europe radar was transforming the nature of
ground warfare. The SCR-584 was guiding dive bombers onto targets with
the efficiency earlier found only in fighter control. Its automatic tracking,
something a small group at Rad Lab had to fight for during initial planning
stages, was even found to determine the positions of artillery by tracking
the projectiles. Its ability to locate those deadly but elusive infantry mortars
gave rise to a special set for that purpose that was better employed near
the front, although deployed very late in the war. The 584 was even used
to observe the movement of vehicles covered by darkness and fog.
It is obvious that radar transformed the nature of war more than
has any other single invention. It turned the entire concept of strategic
bombing on its head, replacing the dominance of the bomber with the
dominance of the fighter and the AA gun. Reality and radar had quickly
disposed of the air power fantasy of a short, decisive ‘knockout blow’, and
a war of attrition was fought not in the trenches of Flanders but in the air
over Europe with the same result as in 1918—the end came when Germany
was exhausted. In forcing this transformation radar also helped move the
conflict to a more barbaric level: trench warfare for civilians.
The changes imposed on fleet action were equally drastic. By provid-
ing aircraft carriers with a 20 to 30 minute warning it removed the fragility
that had restrained plans for their employment and allowed them to par-
ticipate in, indeed dominate major engagements. Without radar the Pacific
War would have been a battleship war. Carriers and radar-directed guns
with proximity fuzes had reduced to nothing Mitchell’s prediction of the
demise of surface ships to land-based air power.
In August 1945 the atomic bomb upstaged an announcement planned
of Rad Lab’s contribution to the Allied war effort, displacing the cover-page
article in Time magazine to page 78 and prompting Lee DuBridge to make
the oft quoted remark: ‘The bomb may have ended the war, but radar won
the war’ [6]. Hoyt Taylor expressed it slightly differently, perhaps more
accurately and in keeping with his personality: ‘The bomb finished the
war, radar fought the war’ [7].
465
APPENDIX A
A FEW RADAR ESSENTIALS
Radar is the technique whereby the position of some object is determined

by illuminating it with radio waves and observing the reflections. It is
a highly technical subject requiring a large dose of electronics, physics
and mathematics for a thorough understanding, but the basic ideas and
conceptions are not all that hard to grasp and can be understood by a
layman having a science education of the kind that should be gained in a
good high school. The purpose of this appendix is to introduce a number
of ideas necessary to the key elements. It can be skipped by many without
loss, read selectively by others or studied carefully by those with little
electronic knowledge. There will be brief portions of the text outside the
explanation of this little essay. They are there because their omission would
irritate the technically trained (for whom the book is also intended) and
may be skipped without significant loss.
A.1. ELECTROMAGNETIC WAVES

Radio waves generated by various activities of civilization impinge upon
us wherever we go and affect our lives to a degree that their discoverer,
Heinrich Hertz, would have found incomprehensible. They are waves of
co-dependent electric and magnetic fields whose motion shares some of the
propagation characteristics of mechanical waves such as sound, seismic
waves or waves on the surface of a liquid. They differ in requiring no
material medium for their propagation and in this were the cause of a major
crisis in physics at the turn of the century, resolved by the special theory
of relativity. They are the same as mechanical waves in having a simple
relationship between wavelength (the distance between two adjacent crests
of the wave), frequency (the number of crests that pass a given point per
second) and the velocity of propagation. For radio waves this velocity is
the velocity of light; they differ from light in wavelength and in the inability
of human senses to perceive them. Electromagnetic wavelengths extend
from the extraordinarily short (gamma rays) to radio waves of hundreds
of kilometers—the spectrum reproduced in countless textbooks.
466
A Few Radar Essentials
The velocity of electromagnetic waves in a vacuum is constant and

slightly slower when traveling through matter, such as the atmosphere.
The frequency is not so affected, being the number of crests passing a given
point per second, but the wavelength is consequently shorter in matter,
and for this reason electrical engineers adopted frequency as the method
of specifying a wave. Units of frequency were originally given as cycles per
second with the standard prefixes of the metric system for thousand cycles
per second (kc) or million cycles per second (Mc). The German practice of
naming the cycle per second after Hertz has been adopted internationally,
so we now use kilohertz (kHz) and megahertz (MHz). Wave velocity is
the product of wavelength and frequency just as a person’s velocity is
the product of the length of stride times the number of strides per unit
time. The velocity of light is approximately 300 million meters per second,
so to obtain the approximate wavelength in meters from the frequency
divide 300 by the frequency in MHz; to obtain frequency divide 300 by the
wavelength in meters.
Wavelengths have from the beginning been measured in metric, not
English units. For radar the wavelength is more useful to us than fre-
quency because the interaction of waves with antennas and targets, which
is strongly dependent on it, allows more convenient comparison and was
the preference of most World War II radar men. For this reason wavelength
is used rather than frequency throughout the book with two units of length
sufficing for our purpose: meter (m) and centimeter (cm).
All waves transport energy, mechanical waves by the motion of par-
ticles, electromagnetic waves through the electric and magnetic fields that
constitute them. The power radiated by a radio transmitter is measured in
watts (W); power also uses metric prefixes to allow convenient sized units;
MW, kW and µW (microwatt) will suffice.
When any wave is incident on a boundary between two different
materials, e.g. air and water, part of the wave’s energy is reflected and
the rest transmitted. Sometimes nearly all is reflected, as with light on
a mirror; sometimes most of it is transmitted, as with light on a pane of
glass. For the waves used in radar, conducting surfaces are good reflectors;
other materials also reflect but not to such a high degree and according to
material characteristics that are not important to us. If the wavelength
is much shorter than the size of the reflecting surface, reflection will take
place as commonly observed with mirrors, called specular reflection; if
the wavelength is comparable to the size of the reflector, the waves will be
emitted in a wide distribution of directions. In the early days of radar some
theorists argued that specular reflection would prevent wavelengths of a
few centimeters from being useful in radio location, but this did not prove
to be the case because of the irregularities of airplane and ship surfaces.
When jet aircraft, which had neither propellers, large external motors nor
radiators, were introduced, they produced much smaller radar signals for
microwaves (10 cm or shorter) than for meter waves.
467
When separate trains of electromagnetic waves encounter one an-

other their electric and magnetic fields add vectorially. Two waves of the
same wavelength that are in phase, i.e. that have their wave crests mov-
ing together, increase their combined amplitude; if they are out of phase,
they decrease their combined amplitude. This property, called interfer-
ence, is particularly important in shaping the radiation patterns of radar
equipment.
A.2. REFLECTED SIGNALS

All radar sets function by sending out a signal to encounter a target. This
radiated signal loses intensity just as does light leaving an incandescent
lamp, in proportion to the square of the distance. The signal reflected from
a target is, of course, a very small fraction of the transmitted signal, but
worse yet, the reflected signal is subject to the same law of propagation as
the transmitter. Put together, this means that the reflected signal received
by the radar receiver decreases as the fourth power of the distance. The
only reason that radar is possible for any reasonable range given such
diminution, is that transmitters can be made that radiate hundreds of kW
and receivers that function with small fractions of a µW.
World War II radar sets operated in three modes: pulsed, continuous
wave and frequency modulated. In the former the radio frequency signal
produced by the transmitter has a duration that is short enough to form a
wave train tens or at most hundreds of meters long. The speed of light is
such that a train 300 m long is formed during one microsecond of trans-
mission (µs). When operating in pulsed mode the time between pulses is
usually set so that the round trip time is enough for the maximum range
desired. The higher the pulse repetition rate the more reflected signals will
be available for analysis.
Continuous-wave radar is much less often used than pulsed radar
because distance information is more difficult to extract. It is excellent for
measuring the speed of the target and as such has become the favorite of
traffic police. If the target is in motion the wavelength of the reflected wave
will be altered by the well known Doppler effect, noted acoustically by the
change in pitch of the whistle of a passing locomotive or optically by the
shifting to the red in the spectra taken of distant receding galaxies. When
the reflected radio wave is mixed with the wave emitted by the transmitter,
a new wave is formed, the familiar beat note of music. This is the easiest
kind of radar set to build. Speeds of automobiles were measured this way
in 1932.
If one modulates a transmitted signal so that its frequency changes
with time, the reflected signal will return with a different frequency than
the one being dispatched at that moment. Mixing the two frequencies will
produce a beat frequency equal to the difference between the reflected and
the momentarily transmitted frequencies, and this beat frequency is easily
468
related to the distance of the target. This kind of operation is continuous, so

separate antennas for transmission and reception are required. Obviously,
motion of the target will produce a Doppler shift, which has generally
restricted the use of frequency-modulated radar primarily to radio altime-
ters.
Some radar targets do not wish to escape detection and furthermore
want the radar operator to know their identity. In war it is vital to recognize
your own ships and aircraft. In peacetime a controller wishes to know the
identity of the blip as well as other important pieces of information, such
as its precise altitude (generally difficult for ground radar to determine),
speed, remaining fuel and the like. This is accomplished by equipping
the plane with a radio transmitter that is actuated by the interrogating
radar signal or auxiliary signal. It responds on yet a different wavelength
for which the radar set has a separate receiver and to which it sends the
desired information in coded form. In war when enemy targets show
no cooperation whatsoever, and when enemy intelligence becomes overly
inquisitive about these signals, this gets to be a very messy problem. The
signal transmitted by the responding set, the secondary radar, is much
larger than the unaided reflected signal, allowing greater range.
A.3. ANTENNAS
An antenna is a metal structure that either transmits or receives electro-
magnetic waves. An antenna intended for transmission is equally good
for reception. Hertz invented the first antenna, which is widely used in
the radar equipment described here. It is called the oscillating dipole. Two
metal rods, each a quarter-wavelength long, lying on the same line are
connected at their common center to a high-frequency alternating current
generator, called an oscillator. The oscillator forces currents of opposite
polarity to flow in the two halves of the dipole, and these currents cause it
to radiate. A wave that is incident on the dipole will cause currents to flow
to the receiver that is connected to it. There is a glut of antenna designs,
but the oscillating dipole suffices for most of the radar equipment we shall
be discussing.
The radiation pattern of a dipole is at a maximum in the plane perpen-
dicular to the axis and evenly distributed around it. The pattern reaches
zero at the poles. This is satisfactory for broadcasting but not for a radar
set, which needs a directed beam. A radio analog to a searchlight was the
goal of many, although not all designers.
The first step in producing a directed radiation pattern is to place a
metal surface on one side of the dipole and parallel to it. This produces a
radiation pattern only on one side of the dipole but one that is distributed
over 180 degrees. Placing dipoles in an array side by side allows them to
interfere in such a way as to build up the radiation in the forward direction
and to diminish it on the flanks. The greater the number of dipoles in the
469
Figure A.1. Dipole radiation pattern. The drawing on the upper left shows
schematically the basic elements of a dipole antenna. It has two halves, each
one quarter wavelength long that are connected at the center to a high-frequency
alternating current generator. In practice the dipole is connected to the generator
by means of a transmission line. The drawing on the upper right shows the
radiation pattern of a dipole radiating into free space. The solid curve shows a
constant value of the intensity and merely indicates the shape; the pattern extends
out indefinitely. The maximum intensity is in the plane perpendicular to the
dipole. The drawing on the lower left shows the radiation pattern of a dipole that
has a reflector on one side. The drawing on the lower right shows the radiation
pattern of an array of four dipoles placed side by side with reflectors. These kinds
of pattern are called lobes.
array, the tighter the radiation pattern.

Another method of producing a narrow beam is to place the dipole at
the focal point of a curved reflecting surface, such as a paraboloid symmet-
ric about the beam axis. This is the structure of a searchlight: the metallic
paraboloid being replaced by a mirror of the same shape, and the dipole
replaced by an electric arc. This is the form of antenna that is often used
for microwaves, waves of a few centimeter wavelength. It used to be com-
monly seen at airports but with only a horizontal strip of the paraboloid,
470
an arrangement that produces a vertical fan-shaped beam that will be able

to interrogate aircraft at a wide range of altitudes. The same was long seen
aboard ships, where the function was to locate other vessels or land forms
regardless of roll and pitch. The paraboloid is generally being replaced by
a linear array of microwave radiators, as is discussed in Chapter 4.5.
A.4. LOBES
The radiation pattern produced by the tricks just described is called a lobe.
It determines the direction to a target with an accuracy limited by its an-
gular dimension. One notices that the forward end of the lobe is rounded,
which means that slight changes in the direction one points the beam
make small changes in the energy incident on or received from the target.
This is even more serious than it seems because aircraft targets change the
amount of energy reflected with changes in the aircraft’s orientation—they
twinkle—making the determination of the direction of maximum signal
difficult.
If one places two arrays of dipoles side by side but with slightly
different directional orientations, two slightly overlapping lobes will result.
The fronts of the lobes are blunt but the sides are steep. If the target is
illuminated alternately with one lobe or the other, each lobe will yield a
reflected signal, generally of different amplitude because one originates
nearer the center of its lobe than the other. If the antenna is positioned so
that the reflected signals are equal, then the target will lie on a line bisecting
the angle that orients the two arrays. Remarkable angular accuracy can be
had this way. In a radar set this technique is called lobe switching. The
same idea was used during the 1930s for air navigation.
If a parabolic reflector is used, the same effect can be had by mounting
the dipole slightly off axis and causing its position to rotate about the beam
axis, producing what is called a conical scan.
A.5. VERTICAL LOBE STRUCTURE

The examples of lobes just reviewed assume that the antenna radiates into
free space, which may be the case for a radar pointing upward to direct
a searchlight or an antiaircraft gun or for airborne radar at high altitude,
but if the radiation pattern is projected horizontally over land or sea a
complication arises. The lower half of the radiation pattern will be reflected
off the surface, which is an excellent conductor, and will interfere with
the waves that come directly from the antenna to produce a vertical lobe
pattern.
The consequences of this vertical lobe pattern are both detrimental
and beneficial. Consider an airplane flying horizontally toward the radar
set. At some point it will encounter the lowest lobe, from which it returns
a signal. As it flies on it enters the region between lobes and can disappear
completely from the radar screen only to reappear when it enters the next
471
Figure A.2. Lobe switching. Accurate direction can be obtained with the blunt
lobes illustrated in figure A.1 by making use of the rapid change in intensity on
the flanks of the lobes. The drawing shows the effect of a transmitter that radiates
alternately pattern 1 and then 2. A target in the direction C will produce signals of
equal amplitude (D) for either lobe, whereas a target in direction E will generate a
larger amplitude (F) in pattern 2 than in pattern 1 (G). The radar operator adjusts
the direction of the antenna so as to equalize the two signals.
lobe, and so on according to the number of lobes. This is naturally con-

fusing to an inexperienced operator. In principle it can be used—and was
used successfully as operators became skillful—to estimate the height of
the plane, which is almost as important a coordinate for an air controller
as the plane’s horizontal location. But over land this is complicated by
the ground not being uniform. Even for the important case of observing
the arrival of aircraft approaching over the sea, height estimation requires
many calibration flights by test aircraft flying fixed courses. The interpre-
tation of such data makes as much use of intuition as of science. But for
many sets during the war it was the only way of determining the height
of the target. Some long-wave sets were equipped with alternate antennas
placed at various levels in order to switch the vertical lobe pattern to help
the operator observe the target between lobes. The problem of the vertical
lobe pattern is primarily one for long-wave equipment.
472
Figure A.3. Vertical lobe patterns. The radiation patterns of figure A.1 are for
antennas radiating into free space. If an antenna has its beam pointed parallel
to a conducting surface, such as the ocean or very flat ground, and is located at
a height H above it, the lower half of the pattern is reflected and interferes with
the direct radiation, producing a pattern of vertical lobes. In the diagram the
dashed curve represents the pattern in free space; the solid curves the first four
lobes of the resulting radiation. The reader can easily imagine the confusion for an
untrained observer when an aerial target approaching from the right disappears
momentarily as it leaves the first lobe and proceeds to the second. The angle of the
first lobe relative to the antenna axis (in radians), , is approximately equal to the
wavelength divided by four times the antenna height above the surface.
A.6. THE YAGI ANTENNA

If one places a conductor that is slightly longer next to a radiating dipole,
this parasitic electrode will function as a reflector. If the parasitic elec-
trode is slightly shorter than the dipole, the effect will be to enhance or, as
generally termed, direct the signal on that side. Hidetsugu Yagi utilized
this in 1928 in a combination of elements that bears his name in America
but is called the Yagi–Uda in Japan. In it an oscillating dipole, called the
driver, has a reflector on one side and one or more directors, each somewhat
shorter than its predecessor, on the other side. The result is a directional
radiation pattern. The voltage at the center of all these elements is zero, so
the reflector and directors can all be mounted on the same metal support
at their centers, a matter of great convenience. Unlike the dipole arrays
described, the theory of the Yagi is difficult to apply and inaccurate, so
empirical design is the rule. It has come to dominate long-range, very-
high-frequency communication and is almost the only exterior antenna
for television. Many varieties are encountered.
A.7. ELECTRON TUBES

For anyone with an electronics knowledge acquired since 1970 the electron
tube, also called the vacuum tube, is an archaic device that has been almost
completely replaced by the transistor and its multifold descendant, the
integrated circuit or chip. The radar story of World War II requires some
understanding of tubes, or valves as they are called by the British, for the
473
Figure A.4. Schematic of a typical Yagi antenna. The driver is connected to the
transmitter by a balanced transmission line. To its left is a metal rod somewhat
greater in length than half a wavelength; to the right are directors that are somewhat
shorter. The exact lengths and spacings, which generally lie between 0.1 and 0.25
wavelengths, and the number of directors are determined by experiment. The
radiation pattern is strongly directed to the right.
transistor was a postwar invention, one which had origins in radar work.
In experimenting with incandescent lamps Edison noted that current
flowed to the hot filament from an adjacent electrode, otherwise stated,
electrons flowed from filament to anode. John Fleming noted that the
application of an electric potential between the filament and this second
electrode resulted in current for one polarity of applied voltage, none for
the reverse. He used this as a sensitive detector for early wireless and
named it the diode. By passing current only in one direction it formed a
variable direct current that could be heard in the operator’s head phones.
It took the form of a hot central filament, the cathode, surrounded by a
sheet-metal second electrode, called the anode or plate. Lee De Forest
inserted a grid between the two and found he could control the stream
of electrons from the cathode to the anode, which allowed signals to be
amplified. His invention, the triode, transformed radio. Triodes presented
circuit designers with problems as they pushed to ever higher frequencies
and shorter wavelengths. These problems were solved to a great degree
by the introduction of one or two more grids between the control grid
and the anode, tubes called tetrodes and pentodes. Tubes, along with
components already in use, such as the resistor, capacitor, inductor and
transformer, allowed the design of a huge number of electronic devices by
1939, generally for radio and audio amplification.
474
A.8. TRANSMITTERS
Transmitters are devices for producing an alternating current, such as one
uses for household electric power, but at much higher frequencies than
the 60 Hz of America or the 50 Hz of Europe. The wavelength of a 60 Hz
oscillator is 5000 kilometers (km), hardly suitable for radar. A transmitter
for communication by Morse code, a common application at the beginning
of the radio age, generates power at the desired wavelength whenever
the operator depresses the telegraph key. In early radio broadcasting the
amplitude of the radio frequency signal varied according to the magnitude
of the audio signal, the amplitude modulation that remains today on the
AM band of radios. Television of that era also used amplitude modulation,
so that the signal intensity varied according to whether the picture was to
be bright or dark. The radar transmitter for the pulsed mode requires
abrupt changes from zero to maximum signal and to zero again, design
requirements in common with television transmitters of the time.
The final tubes in a transmitter, the output stage, must be the most
robust, as high voltages and large currents are demanded of them. The
early radar designers all quickly learned that these tubes could be driven
to much more than their nominal power for the few microseconds required
to form a pulse, because the limit of a tube was its ability to dispose of the
heat formed when the electrons struck the anode, energy not radiated by
the antenna. During the short period of the radar pulse the anode rose in
temperature rapidly but had a relatively long period to cool by thermal
radiation.
A.9. RECEIVERS
A radio receiver is a sophisticated instrument, but what we need to know
here is relatively simple. Its function is to select the signal of interest from
the jumble of wavelengths picked up by the antenna and to amplify it to
a useful level. The input of a receiver is some device that resonates at the
frequency of the signal to be amplified. For Morse code a very selectively
tuned detector is used because the sharpness of its tuning suppresses the
amplifier and antenna noise that is not at the resonant frequency, and the
sharp tuning presents no problems. There is a problem in using such a
receiver in radar, one that was discovered quickly by all the early experi-
menters who were not experienced enough to avoid it.
Consider this analogy. A guitar is a resonant device, so is a banjo.
Pluck a guitar string and the tone continues for a some time; pluck a banjo
string and the tone quickly disappears. The technical term for this dif-
ference is damping: the guitar has low damping, the banjo a great deal.
There is a lot of wonderful theory about all this that is based on the math-
ematics of Joseph Fourier, whose work is thoroughly studied by electrical
engineers. We shall, however, be satisfied with mentioning the concepts
and deal with the results qualitatively.
475
Figure A.5. The simplest form of radar indicator, the A-scope. The drawing shows
the appearance of an oscilloscope screen in which the horizontal trace begins at
the left when the transmitter is pulsed. The spot moves across the screen with
the output of the receiver applied in the vertical direction. Generally a large pulse
is recorded simultaneously with transmitter output and target pulses occur at
various times along the trace. The pattern repeats at the pulse repetition rate of
the radar set, usually hundreds or thousands of times a second. This repetition is
very useful in allowing the operator to examine carefully the very weak signals.
When a short wave train is incident on a sharply tuned receiver, it

responds much as does a plucked guitar, setting up vibrations that can
hide subsequent plucks that are weak. Because of their damping banjos
are better suited for fast tunes than guitars. For radar one needs a receiver
that has the right amount of damping. It must resonate enough to select the
radar frequency but damp the vibrations so they will not obscure the next
signal. Engineers prefer to speak of ‘pass band’ rather than ‘damping’. A
communication receiver has a narrow pass band (sharp tuning), whereas
a radar receiver requires a wide pass band, as does a television receiver,
in order to follow the rapid changes of amplitude. The shorter the radar
pulse, the wider must be the receiver pass band.
A wide pass band comes with a price: the wider the pass band, the
larger the receiver noise, with consequent difficulty in observing small sig-
nals. This leads to the curious result that the maximum range of a radar
set is not enhanced by decreasing the pulse width, thereby utilizing the
increased amplitude of the transmitter that accrues from shorter pulses,
because it requires a wider receiver pass band and with that comes en-
hanced receiver noise; the two effects work against each other.
A.10. INDICATORS
The information gained with a radar set must be made available to its oper-
ators. The crucial element in this is the cathode ray tube because it allows
one to measure microsecond time differences. Such tubes are the display
elements of television receivers. In one a highly mobile electron beam
476
Figure A.6. The plan position indicator (PPI). This is the most widely recognized
radar indicator. It is a maplike display of panoramic sweeps of the radar beam.
The intensity of the cathode-ray beam is modulated to show the presence of a target
in the same manner that a television image results from beam modulation. The
azimuthal direction of the outward trace of the spot follows the orientation of the
antenna. The example is of Palermo harbor made with a 3 cm SU-2 surface-search
shipborne radar. Rings for range are shown at 2 and 4 nautical miles. The
dark region to the lower right indicates no reflected signal because of the signal’s
reflection from water; the dark splotches over the land indicate regions hidden from
the line of sight.
produces a spot of light on a phosphorescent screen that can be moved

according to the amplitude of electric signals applied to electrodes of the
tube. In the simplest radar application, a voltage that increases linearly
with time (a time base) deflects the spot horizontally and the output of the
radar receiver deflects the spot vertically. The screen shows a horizontal
line trace with little vertical blips when the receiver picks up reflections.
The spot begins its motion across the screen at the time the transmitter
pulse is emitted. By knowing the speed of the time base, which can be
determined by electronic means, the range to any of the blips can be deter-
mined. The size of the blips depends on the reflecting power of the target
and its range. This form of indicator is called the A-scope.
Various combinations came into use for displaying a radar set’s out-
put. Some used cathode-ray tubes with horizontal and vertical inputs,
others with circular traces in which one signal moves the spot along a ra-
dius from the center and the other moves it in a circular path around the
477
center. The form of display that is most readily identified by the layman
is the plan position indicator (PPI) in which the screen gives a maplike
representation of the region interrogated by the radar. In it the time base
moves the spot outward from the center of the tube face and a circular mo-
tion follows the rotation of the radar antenna. The tube’s electron beam is
suppressed when there is no signal from the receiver and enhanced when
there is. This results in a glowing phosphor when there is a target and
darkness otherwise.
478
NOTES AND SOURCES
Citations given in capital letters are from the list of abbreviated references;
those in upper and lower case are from the general references.
Chapter 1.2 Electromagnetic waves
[1] James Clerk Maxwell A Treatise on Electricity and Magnetism article 771. Oxford:
The Clarendon Press, 1892
[2] C W F Everitt James Clerk Maxwell: Physical and Natural Philosopher p 99. New
York: Charles Scribner’s Sons, 1975
[3] GUERLAC 2, pp 290–291
Chapter 1.3 Perceptions of air power, 1919–1939

[1] JONES H, Vol 5, pp 26–32
[2] Ibid., Vol 6, pp 1–27; appendices, pp 8–14
[3] WEBSTER & FRANKLIN 1, p 45
[4] JONES H, Vol 5, pp 153–154
[5] Beverley Nichols Cry Havoc! p 63. London: Jonathan Cape, 1933
[6] Douhet op. cit. R Ernest Dupuy and George Fielding Eliot introduced Douhet’s
ideas, which they encountered in a French military journal, to their English
speaking readers in their book If War Comes pp 53–60, New York: The Macmil-
lan Co., 1937. This book presents a calm evaluation of military technology on
the obviously expected war
[7] James S Corum The Roots of the Blitzkrieg: Hans von Seekt and German Military
Reform pp 144–146, 155. Lawrence, Kansas: University Press of Kansas, 1992
[8] Much uncertainty surrounds the pre-war German decisions concerning the four-
motor bomber. The outcome probably turned on a combination of personal
preferences, economics, experience in Spain and a flawed strategic outlook.
See SUCHENWIRTH 1, pp 55–59
[9] Feldmarshall Albert Kesselring Soldat bis zum letzten Tag pp 32–33. Bonn:
Athenäum Verlag, 1953
[10] Williamson Murray, Force Strategy, Blitzkrieg Strategy and the Economic Dif-
ficulties: Nazi Grand Strategy in the 1930s Journal of the Royal United Services
Institute in Defence Studies Vol 128, pp 39–43, March 1983
[11] Saundby op. cit., p 28
[12] Peter C Smith The History of Dive Bombing Annapolis: The Nautical and Aviation
Publishing Co., 1981
[13] SUCHENWIRTH 2, pp 7, 28, 79
479
[14] Colonel Roy M Stanley Prelude to Pearl Harbor pp 133–148. New York: Charles
Scribner’s Sons, 1982
[15] Schaffer op. cit., pp 107–108
[16] Greer op. cit., pp 67–69, 91–92
[17] Spaight op. cit.
[18] Ibid., p 35
[19] JONES H, Vol 6, pp 439–477, 465
[20] J C Slessor Air Power and Armies pp 163–164. London: Oxford University Press,
1936
[21] B H Liddell Hart Paris and the Future of War. New York: E P Dutton and
Company, 1925
[22] Heinz Guderian Panzer Leader p 25. New York: E P Dutton and Co., 1952
[23] Roscoe op. cit., pp 141–174
[24] Mitchell op. cit. p 206
[25] Ibid., pp 125–126
[26] ROSKILL 5, pp 256–399
[27] ROSKILL 6, pp 392–405
Chapter 1.4 Navigation in 1939
[1] For a readable telling of the chronometer story see Dava Sobel Longitude: the
True Story of a Lone Genius Who Solved the Greatest Scientific Problem of His Time
New York: Walker Publishing Co., 1995
[2] Commander A E Fanning, Astronomical Navigation Since 1884 Royal Institute
of Navigation Vol 38, pp 209–215, 1985
[3] TERRAINE 1, p 85
[4] William E Jackson The Federal Airways System pp 219–227. Institute of Electrical
and Electronic Engineers, 1970; Robert I Colin, Otto Scheller: the Radio Range
Principle AES Vol 2, pp 481–487, 1966
[5] JONES H, Vol 5, pp 8–18
[6] TUSKA 1
[7] SIG CORPS 1, pp 185–188
Chapter 1.5 Anti-aircraft artillery, 1914–1939
[1] KOCH, p 9
[2] PILE, p 173
[3] KOCH, pp 14–15. Anti-aircraft units generally follow the rule of about four
heavy guns per battery with two or more batteries forming a battalion and
two or more battalions forming a regiment
[4] Ibid., pp 10–15
[5] Ibid., p 19
[6] Ibid., pp 20–21
[7] Ibid., p 28
[8] Major Charles Edward Kirkpatrick Archie in the AEF: the Creation of the Antiaircraft
Service of the United States Army, 1917–1918 pp 5–11. Fort Bliss, Texas: US Army
Air Defense School, 1984; James A Sawicki Antiaircraft Artillery Battalions of the
US Army Vol 1, pp 1–2. Dumfries, Virginia: Wyvern Publications, 1991
480
Notes and Sources
[9] Coast Artillery units on the west coast and in the Philippines were not drawn
for service in France until 1918, presumably through a mistrust of Japanese
intentions
[10] Kirkpatrick [8], pp 20–24
[11] Ibid., pp 181–182
[12] Sawicki [8], pp 6–8
[13] John C Reilly Jr United States Navy Destroyers of World War II pp 68–79. Poole:
Blandford Press, 1983
[14] PILE, pp 43–51
[15] Ibid., pp 52–60
[16] Liddell Hart Memoirs op. cit., Vol 1, p 127
[17] PILE, pp 41–42
[18] ROSKILL 6, p 420
Chapter 2.1 Electronic Component Development
[1] KURYLO & SUSSKIND

[2] Peter A Keller The Cathode-Ray Tube: Technology, History and Applications pp
45–54. New York: Palisades Institute for Research Service, 1991
[3] J B Johnson, A Low Voltage Cathode Ray Oscillograph Journal of the Optical
Society of America and Review of Scientific Instruments Vol 6, pp 701–712, 1922
[4] V K Zworykin, Description of an Experimental Television System and the Ki-
nescope Proc. IRE Vol 21, pp 1655–1673, 1933
[5] Manfred von Ardenne, Die Braunsche Röhre als Fernsehemfänger Fernsehen Vol
1, pp 193–202, 1930; Ueber neue Fernsehsender und Fernsehempfänger mit
Kathodenstrahlröhren Fernsehen Vol 2, pp 65–80, 1931
[6] Manfred von Ardenne Ein glückliches Leben für Technik und Forschung pp 75–77,
82–90. Berlin: Verlag der Nation, 1973
[7] Ryoka Sawada History of Electron Tubes Sogo Okamura, editor, pp 23–24. Tokyo:
Ohmsha Ltd, 1994
[8] Albert W Hull and N H Williams, Characteristics of Shielded-grid Pliotrons
Phys. Rev. Vol 27, pp 432–438, 1926
[9] Albert W Hull, Measurements of High-frequency Amplification with Shielded-
grid Pliotrons Phys. Rev. Vol 27, pp 439–454, 1926
[10] STOKES, pp 34–39. James Brittain in a letter to the author dated 24 January 1994
cautions that the entire tetrode–pentode invention may be more complicated
than it appears here because of the intense activity in vacuum tube design
during those years and the rather long time that four-element tubes had been
available for experiment
[11] Ibid., pp 54–56; see also W I G Page, The Working Principles of a New Screened
Grid Power Output Valve Explained Wireless World July, pp 7–9, 1928
[12] Zworykin, [4]
[13] H Barkhausen and K Kurz, Die kürzesten, mit Vakuumröhren herstellbaren
Wellen Phys. Zeit. Vol 21, pp 1–6, 1920. This is a beautiful little paper. It re-
ports an encounter with an unsuspected form of oscillation, describes the
experiments that revealed its nature and derives a formula that gives good
approximation for the wavelength as a function of the one adjustable param-
eter
481
[14] For a review of early magnetron work see James E Brittain, The Magnetron
and the Beginnings of the Microwave Age Phys. Today Vol 38, pp 60–67, 1985
[15] August Zacek, Über eine Methode zur Erzeugung von sehr kurzen elektro-
magnetischen Wellen Zeit. Hf. Vol 32, p 172, 1928
[16] STOKES, p 137
[17] CALLICK, pp 41–42
[18] ALLISON, p 102
[19] Ed Simmonds and Norm Smith Radar Yarns: Being Memories and Stories Col-
lected from RAAF Personnel Who Served in Ground Based Radar During World
War II p 219. Published privately by E W & E Simmonds, 15 Blair Street, Port
Macquarie, NSW 2444, Australia, 1991
[20] HOWSE, p 8
[22] Raymond R Myers, Plastics and Resins Encylopaedia Britannica Vol 21, p 339,
1993
[23] Dipl-Phys. Hans Ulrich Widdel in a letter to the author dated 18 November
1994
[24] J C Swallow, The History of Polythene Polythene: the Technology and Uses of
Ethylene Polymers A Renfrew and Phillip Morgan, editors, pp 1–10. London:
Iliffe and Sons Ltd, 1960
[25] Paul Kokulis of Washington, patent attorney and friend of Fawcett, reports the
disappointment on finding a solid
[26] Maurice V Wilkes Memoirs of a Computer Pioneer pp 45–46. Cambridge, Mas-
sachusetts: The MIT Press, 1985
[27] John Harry DuBois and Frederick W John Plastics p 47. New York: Reinhold
Publishing Company, 1967
[28] ROTTERDAM, 23 February 1943, p 3 and 17 March 1943, p 8
[29] Gordon M Kline, Plastics in Germany, 1939–1945 Modern Plastics Vol 23, pp
152a–152p, October 1945
[30] Gordon M Kline, interviewed by Jeffrey L Meikle at Lake Worth, Florida on 15
and 16 May 1987
Chapter 2.2 Beginnings, 1902–1934
[1] KERN, pp 34–40

[2] The superheterodyne principle was discovered independently in Germany with
a patent having been filed on 18 June 1918, predating Armstrong’s by six
months. See Walter Schottky, On the Origin of the Super-heterodyne Method
Proc. IRE Vol 14, pp 695–698, 1926
[3] Guglielmo Marconi, Radio Telegraphy Proc. IRE Vol 10, pp 215–238, 1922
[4] WATT, p 62
[5] L S Alder, Provisional Patent Specification No 6433/28, 1 March 1928
[6] TUSKA 2, pp 13–20, 95–115, identifies nine, and KERN, pp 44–63, ten patents
that present the idea of radar
[7] Sean Swords, The significance of radio wave propagation studies in the evolu-
tion of radar, BLUMTRITT, pp 185–197. This experiment became transformed
in Popular Science Vol 127, November 1935, p 18, as ‘a way of halting airplane
motors at a distance’
482
Notes and Sources
[8] Leo C Young in the Stanford Caldwell Hooper audio collection History of Radio–
Radar–Sonar at the Library of Congress, Reel 23
[9] Gregory Breit and Merle A Tuve, A Radio Method of Estimating the Height of
the Conducting Layer Nature Vol 116, p 357, 1925; Gregory Breit and Merle
A Tuve, A Test of the Existence of the Conducting Layer Phys. Rev. Vol 28,
pp 554–575, 1926; Merle A Tuve, Early Days of Pulse Radio at the Carnegie
Institution Journal of Atmospheric and Terrestrial Physics Vol 36, pp 2079–2083,
1974
[10] MCKINNEY, pp 84–87
[11] Ibid., p 132
[12] SOUTHWORTH, pp 79–81
[13] L F Jones, A Study of the Propagation of Wavelengths between Three and Eight
Meters Proc. IRE Vol 21, pp 349–386, 1933
[14] William H Wenstrom, Notes on Television Definition Proc. IRE Vol 21, pp 1317–
1327, 1933
[15] Carl R Englund, Arthur B Crawford and William W Mumford, Some Results
of a Study of Ultra-short-wave Transmission Phenomena Proc. IRE Vol 21, pp
464–492, 1933
[16] TUSKA 2, p 13
[17] E Giboin, L’Évolution de la Détection Électromagnétique dans la Marine Na-
tionale Onde Vol 31, pp 53–64, 1951
[18] J Bion, Le Radar La Revue Maritime July–August, pp 330–346, 456–471, 1946;
R B Molyneux-Berry, Dr Henri Gutton, French radar pioneer, BURNS, pp 45–
52
[19] J B Johnson, A Low Voltage Cathode Ray Oscillograph Journal of the Optical
Society America and Review of Scientific Instruments Vol 6, pp 701–712, 1922
[20] Heinrich Löwy, Die Fizeausche Methode zur Erforschung des Erdinnern Phys.
Zeit. Vol 12, pp 1001–1004, 1911
[21] SWORDS, pp 46–47, 59. For detail about Löwy see Ulrich Kern, Review con-
cerning the history of German radar technology up to 1945, BLUMTRITT, pp
171–183
[22] KROGE, p 16; TRENKLE 1, pp 23–24
[23] Ibid., pp 12–18; REUTER, pp 15–22
[24] LOBANOV, pp 25–33
[25] ERICKSON, pp 247–252; LOBANOV, pp 101–109
[26] LOBANOV, p 103
[27] Cited in SWORDS, pp 45–46
[28] Harold A Zahl, From an Early Radar Diary Coast Artillery Journal Vol 91, no
4 (March–April), pp 8–15, 1948. A rather accurate description of the incident
was found in ‘Mystery Ray Locates "Enemy": U.S. Army Tests Detector for
Hostile Ships and Planes’ Popular Science Vol 127, October 1935, p 25
[29] News item Elektrisches Nachrichtenwesen Vol 10, pp 24–25, 1932
Chapter 2.3 Britain builds an air defense system
[1] For a survey of the politics of the major powers up to the outbreak of war see
Richard Overy and Andrew Wheatcroft The Road to War London: Macmillan
Co., 1989
483
[2] For a glimpse into some of the ideas that had been studied see Russell Burns,
Aspects of UK Air Defence from 1914 to 1935 Proc. IEE Vol 136, pp 267–278,
1989
[3] Ronald W Clark Sir Edward Appleton pp x, 90–91, 96–102. Oxford: Pergamon
Press, 1971. The radio section of the Post Office (British) had noticed reflections
from aircraft the year before. A H Mumford and R F J Jarvis, Radio Report No
273, 18 July 1934
[4] Wright op. cit., pp 26–36
[5] Ibid., pp 53–56
[6] William Roy Piggott, who worked closely with both Appleton and Watt, in a
letter to the author dated 13 March 1997
[7] Gordon Kinsey Orfordness: Secret Site. Lavenham: Terence Dalton Ltd, 1981. A
book about the place, primarily dealing with its World War I functions. Two
chapters about the radar work there were written by A F Wilkens
[8] WATT, pp 113–115
[10] SWORDS, pp 186–236; B T Neale, CH—The First Operational Radar GEC
Journal of Research Vol 3, pp 73–83, 1985
[11] ROWE, pp 25–26
[12] Ibid., p 23
[13] WOOD & DEMPSTER, p 83
[14] WOOD & DEMPSTER, p 88; WATT, p 208
[15] The Germans made a few copies of CH, called Elefant. This was equipment
that originated in a limited excursion of the Reichspost Zentrale into radar.
I have not been able to determine whether this was to test the CH design
or, as suggested by Trenkle, experiment with over-the-horizon capability. The
transmitter array was mounted on a tower, electrically very similar to CH,
although much more compact. The receiver was improved by using a steerable
high-gain dipole array. TRENKLE 1, pp 101–102. The Japanese Army used a
design, Tachi-6, with broadcast transmission and multiple receivers having
dipole arrays for direction (see Chapter 9.2), and the Russians made a similar
broadcast-transmission radar, RUS-2 (see Chapter 5.6).
[16] Radar Bulletin (published by RAF 60 Group beginning April 1941) October 1945,
p9
[17] C P Snow Science and Government. Cambridge, Massachusetts: Harvard Uni-
versity Press, 1961, appendix 1962
[18] The Earl of Birkenhead The Professor and the Prime Minister: The Official Life of
Professor F A Lindemann, Viscount Cherwell. Boston: Houghton Mifflin and Co.,
1962
[19] ROWE, p 22
[20] BROWN, p 17
[21] Ibid., p 13
[22] R W Burns, A D Blumlein—engineer extraordinary Engineering Science & Edu-
cation Journal February, pp 19–33, 1992
[23] JAY, part 1, para 5
[24] Sidney Jefferson, LATHAM & STOBBS, p 215. Young readers may be puzzled
by the meaning of ‘dress for dinner’. They can look for instruction in Noël
Coward’s plays.
[25] SAYER, pp 20–26
484
Notes and Sources
[26] Ibid., p 117; BOWEN, p 50

[27] Michael Pearson, Coast Defence Radar Fort Vol 19, pp 93–105, 1991
[28] Ibid., p 96; SAYER, pp 120–125
[29] Alfred Price Cossor Radar: The First Fifty Years pp 10–11. Harlow: Cossor Elec-
tronics, 1985; SAYER, p 46–50
[30] SAYER, p 319
[31] P M S Blackett, Operational Research: Recollections of Problems Studied, 1940–
45 Brassey’s Annual: The Armed Forces Yearbook 1953 pp 88–106
[32] JAY, part 2, paras 312–313
[33] Russell Burns, Early History of the Proximity Fuze (1937–1940) Proc. IEE Vol
140A, pp 224–236, 1993
[34] JAY, part 1, paras 186–188
[35] A straight-vision receiver carried only the video portion of the broadcast and
in amplitude, not frequency modulation; the audio was in a separate radio-
frequency channel. This allowed the wide-band design characteristics to be
dealt with without the complication of an intermediate-frequency amplifier
or the need to separate the audio and video portions of the signal in a common
carrier
[36] BOWEN, pp 30–64
[37] J F Coales, The Origins and Development of Radar in the Royal Navy, 1935–
1945, KINGSLEY 1, pp 11–29
[38] HOWSE, p 12
[39] Ibid., pp 18–22
[40] J F Coales and J D S Rawlinson, The Development of UK Naval Radar, BURNS,
pp 53–59
[41] ROWE, p 27
[42] Ed Simmonds and Norm Smith Radar Yarns Ed Simmonds editor, p 35. Pub-
lished privately by E W & E Simmonds, 15 Blair Street, Port Macquarie NSW
2444, Australia, 1991
[43] WATT, p 46
Chapter 2.4 Americans and Germans build prototypes

[1] Carl R Englund, Arthur B Crawford and William W Mumford, Some Results
of a Study of Ultra-short-wave Transmission Phenomena Proc. IRE Vol 21, pp
464–492, 1933
[2] ALLISON, pp 68–78
[3] L A Hyland, A personal reminiscence: the beginnings of radar 1930–1934,
BURNS, pp 29–34
[4] PAGE, pp 64–66
[5] René Mesny, Constantes de Temps, Durées d’Établissement, Décréments Onde
Vol 13, pp 237–243, 1934. This approach to the problem differs entirely from
that taken by electrical engineers of the time or subsequently, but it must have
looked familiar to Page as the analysis of a galvanometer widely employed in
undergraduate laboratories
[6] G L Beers, Description of Experimental Television Receivers, Proc. IRE Vol 21,
pp 1692–1706, 1933
[7] ALLISON, pp 120–124
[8] TAYLOR, p 192
485
[9] ALLISON, p 124; BELL, pp 24–25

[10] A L Samuel, A Negative Grid Triode Oscillator and Amplifier for Ultra-High
Frequencies Proc. IRE Vol 25, pp 1243–1252, 1937
[11] W Tinus and W H C Higgins, Early Fire Control Radars for Naval Vessels Bell
Tech. J Vol 25, pp 1–47, 1946
[12] FRIEDMAN, p 172
[13] SIG CORPS 2, pp 256–257
[14] RADAR SURVEY 2„ pp 33–38
[15] Interview with Blair and other sources in MCKINNEY, pp 64–68
[17] Interview with Colton in MCKINNEY, pp 130–131
[19] GETTING, p 113
[20] SIG CORPS 1, pp 43–45; MCKINNEY, pp 132–136
[21] Arthur L Vieweger, Radar in the Signal Corps Institute of Radio Engineers Trans-
actions Vol 1-MIL, pp 555–561, 1960
[22] Interviews with Blair and Colton in MCKINNEY, pp 157–158
[23] Interview with Colton in MCKINNEY, pp 165–166
[24] SIG CORPS 1, p 123
[25] Colton op. cit., p 746; MCKINNEY, pp 168–172; SIG CORPS 1, pp 125–127;
Harold A Zahl, From an Early Radar Diary Coast Artillery Journal Vol 91, no.4
(March–April), pp 8–15, 1948
[26] Vieweger [21], p 556
[27] Ibid., pp 558–559
[28] Frank Voltaggio, The SCR-270 in Japan AES Vol 3, pp 7–14, 1988
[29] REUTER, pp 18–19
[30] KROGE, pp 17–26
[31] Ibid., pp 24–27
[32] Ibid., pp 22–23
[33] HOLLMANN
[34] KROGE, p 44; TRENKLE 1, pp 26–30; REUTER, p 23
[35] KROGE, pp 55–56; P-G Erbslöh and H-K Freiherr von Willisen, patent decla-
ration dated 27 April 1935
[36] KROGE, p 77
[37] REUTER, pp 24–25
[38] KROGE, p 136; TRENKLE 1, pp 151–152
[39] F A Kingsley, who interrogated German naval personnel after the surrender,
in a letter to the author dated 20 November 1995
[40] KROGE, p 110; REUTER, p 43(n)
[41] KROGE, p 94
[42] Ibid., p 118
[43] Ibid., pp 93–94
[44] WATT, p 405
[45] RUNGE, p 30. It should be noted that Marconi was excluded from early radar
work because of their Italian connection
[46] HOFFMANN 2, pp 329–440
[47] Iris Runge Carl Runge und sein wissenschaftliches Werk p 156. Göttigen: Vanden-
hoeck und Ruprecht, 1948
[48] Ibid., p 171
486
Notes and Sources
[49] Kurt Fränz, Wer war Wilhelm Tolmé Runge, address to the Direktionsbereich
der Forschungs-Institut-Telefunken, November 1990
[50] Electronics Vol 8, pp 284–286 (September, pp 18–19), 1935; Popular Science Vol
127, October 1935, p 25. These articles describe transmitters, receivers and
their antennas of equipment supposedly for 10 and 15 cm waves but the pho-
tographs indicate that they are almost certainly the 50 cm equipment that
Runge employed. The drawing of an array of transmitters and receivers would
seem to show the first idea for a radar system that occurred after the 50 cm
reflections
[51] RUNGE, pp 42–44
[52] Dr-Ing Gotthard Müller, Funkmessgeräte-Entwicklung bei der C Lorenz
AG, 1935–1945, pp 2–4. Stuttgart: Standard Elektrik Lorenz AG (Technisch-
wissenschaftliches Schriftum), 1983
[53] Robert I Colin, Otto Scheller: the Radio Range Principle IEEE Transactions on
Aerospace and Electronic Systems Vol AES-2, pp 481–487, 1966
[54] KROGE, p 50
[55] TRENKLE 1, pp 30–31; PRITCHARD, pp 42–47
[56] HOFFMANN-HEYDEN, pp 31–34
[57] REUTER, p 47
[58] Office of the Chief of Naval Operations, order dated 18 November 1940 (Op-
20-E/AB, (SC)A6/A1, serial 069120)
[59] HOWSE, p 162
Chapter 2.5 Five other nations

[1] NAKAGAWA, p 7
[2] TAYLOR, p 188
[3] Shigeru Nakajima on receiving the Barkhausen-Medaille der Akademie der
Wissenschaft der DDR, Wissenschaft: eine Herausforschung an den erfind-
erischen Geist, reprinted in Japan Radio Company Review No 16, 1981
[4] NAKAGAWA, pp 6–7
[5] Ibid., pp 7–8
[6] Ibid., p 11
[7] Ibid., pp 15–16, 82
[8] PRICE 2, p 291
[9] NAKAJIMA, pp 243–246
[10] ERICKSON, p 248
[11] Ibid., pp 255–257
[12] LOBANOV, pp 107–108
[13] Ibid., p 57
[14] Ibid., p 59
[15] ERICKSON, p 255
[16] Ibid., pp 256–258
[17] Steven J Zaloga, Soviet Air Defense Radar in the Second World War The Journal
of Soviet Military Studies Vol 2, pp 104–116, 1989
[18] LOBANOV, p 139
[19] Ibid., p 129
[20] Ibid., pp 240–241
[21] Ibid., pp 92–94
487
[22] ERICKSON, pp 258–259

[23] N F Alekseev and D D Malairov, Generation of High-Power Oscillations with a
Magnetron in the Centimeter Band Zhurnal Tekhnicheskoi Fiziki Vol 10, pp 1297–
1300, 1940
[24] E Giboin, L’Évolution de la Détection Électromagnétique dans la Marine Na-
tionale Onde Vol 31, pp 53–64, 1951
[25] J Oger, Pré-histoire du Radar La Revue Maritime Vol 108, pp 433–469, 1955
[26] Giboin, [24]
[27] Maurice Ponte, Sur des Apports Français a la Technique de la Détection
Électromagnétique Revue Technique Thomson-CSF Vol 1, pp 171–180, 1946
[28] M Staal and J L C Weiller, Radar Development in the Netherlands before the
War, BURNS, pp 235–242
[29] M Calamia and R Palandri, The History of the Italian Radio Detector Telemetro,
BURNS, pp 97–105
Chapter 3.1 War in Europe

[1] Manfred Bauer and John Duggan LZ 130 ‘Graf Zeppelin’ und das Ende
der Verkehrsluftschiffahrt pp 136–140, 153–166. Friedrichshafen: Zeppelin-
Museum, 1994
[2] These positions are based on the ship’s log. There is reason to believe they were
inaccurate between fixes. A photograph that purports to be of the airship
straying over British terra firma is one of LZ-127 during a pre-Nazi flight, as
the shape of the tail fins and the absence of swastikas on them attest
[3] Sir Edward Fennessy, The Zeppelin Incident and the Battle of Britain, unpub-
lished manuscript, February 1994. The map shown in Bauer and Duggan’s
book does not show the ship straying over land. Political pressures may have
intervened to require the navigator to cook his log
[4] Len Dobson, LATHAM & STOBBS, pp 1–3
[5] Letter to the authors (Bauer and Duggan [1], p 156.) from Keith Wood, a scientist
then employed at Bawdsey
[6] This interpretation of the Zeppelin incident is based on Colin Latham’s, I see
the cat but he can’t see me! News and Views (Newspaper of Marconi Radar
and Control Systems Limited), July 1992, pp 8–9; also LATHAM & STOBBS,
pp 3–6. Latham retired as Chief Engineer, Airspace Control Division, Marconi
Radar Systems
[7] Robert Watson Watt, memorandum to the Committee for the Scientific Survey of
Air Defence dated 27 February 1935, Detection of Aircraft by Radio Methods,
paragraph 9, cited in WATT, p 431
[8] Fennessy, [3]
[9] Interview with Dr Ernst Breuning on 21 February 1969 by Alfred Price
[10] WOOD & DEMPSTER, pp 98–99
[11] BOWEN, pp 83–91; GUERLAC 1, p 153
[12] Sir Edward Fennessy in a letter to the author dated 7 September 1995
[13] BOWEN, pp 83–91; BROWN, pp 45–46
[14] LOVELL, p 21
[15] BOWEN, p 84
[16] LOVELL, p 25; CALLICK, pp 38–40
[17] BOWEN, pp 117–119
488
Notes and Sources
[18] Ibid., pp 58–59

[19] R W Burns, A D Blumlein—Engineer Extraordinary Engineering and Science
Education Journal February 1992, pp 19–33
[20] Derek Martin THORN EMI: 50 Years of Radar pp 4–5. Hayes, Middlesex:
THORN EMI Electronics Ltd, 1986
[21] BROWN, pp 48–51; BOWEN, p 99
[22] PRICE 4, pp 54–55
[23] LOVELL, p 25
[24] Ibid., pp 55–56
[25] TERRAINE 1, pp 95–106
[26] HOFFMANN 1, p 11
[27] Ibid., pp 12–13
[28] TERRAINE 1, pp 103–104
[29] HOFFMANN 1, pp 95–96
[30] TRENKLE 1, pp 76–78
[31] BRANDT, p 19
[33] GIESSLER, p 66
[34] GIESSLER, pp 16–17
[35] JONES 1, pp 190–192
[36] Ibid., pp 67–71
[37] JONES 2, pp 265–332
[38] TERRAINE 1, pp 95–115
[39] The Germans designated these vessels as battleships despite their 11 inch ar-
tillery. The British terminology is followed here to distinguish them from the
15 inch Bismark and Tirpitz
[40] REUTER, pp 44, 52
[41] ROSKILL 1, pp 82–87
[42] REUTER, pp 51–52
[43] Ibid., pp 111–121
[44] JONES 1, p 93
[45] Mr F A Kingsley in a letter to the author dated 16 January 1996
[46] Liddell Hart History op. cit., pp 51–53
[47] KROGE, pp 108–109
[48] HEZLET, p 193
[49] REUTER, pp 53–54
[50] HEZLET, p 193
[51] RAF SIGNALS 5, pp 56–58
Chapter 3.2 The Battle of Britain and the Blitz

[1] I am indebted to Harry von Kroge for locating the specific directive, which is
mentioned without citation in German papers; it was given by Göring on 3
February 1940. Der Führer legt entscheidenen Wert darauf, dass die Rüstung
im Jahre 1940 zur grösstmöglichen Höhe gebracht wird. Es müssen daher mit
allen Mitteln die Vorhaben gefördert werden, die im Jahre 1940 bezw. 1941 zur
Auswirkung kommen können. Alle anderen Programme, die sich erst später
auswirken, müssen, falls es die Belebung der Wirtschaft erfordert, zugunsten
der obigen Vorhaben zurückgestellt werden, Percy E Schramm Kriegstagebuch
489
des Oberkommandos der Wehrmacht Vol 1, p 962. Bonn: Bernard & Graefe Verlag.
For other discussions see GIESSLER, p 74; TRENKLE 1, pp 41–42. For the
Allied understanding see OSRD 5, p 117. For a complete discussion of this
question, which is more involved than the Göring quotation seems to imply,
see SUCHENWIRTH 2, pp 49–54
[2] TRENKLE 2, p 16
[3] WOOD & DEMPSTER, pp 66–67
[5] Alfred Price has grippingly described this day through official records and from
interviews with pilots from both sides who were present at an RAF–Luftwaffe
reunion. Read his Battle of Britain: The Hardest Day, 18 August 1940. New York:
Charles Scribner’s Sons, 1979
[6] Details about the front-line stations of Chain Home are covered by Mike Dean
Radar on the Isle of Wight. Scampton: Historical Radar Archive, 1994. This
attack was the cause of the first Military Medal to be awarded to a woman,
Avis Parsons (née Hearn) who remained at her post relaying messages while
Stukas dropped 90 bombs on the station. LATHAM & STOBBS, pp 25–29
[7] WEBSTER & FRANKLIN Vol 1, p 152
[8] NIEHAUS, pp 37–40
[9] D V Pritchard, The Battle of the Beams Ham Radio (published by Communica-
tions Technology, Greenville, New Hampshire) pp 29–39, June 1989; pp 20–29,
August 1989; pp 53–61, October 1989
[10] W D Hershberger, Seventy-five Centimeter Radio Communication Test Proc.
IRE Vol 22, pp 870–877, 1933
[11] TRENKLE 3, pp 119–120
[12] Pritchard Ham Radio [9]
[13] TRENKLE 3, pp 137–144
[14] NIEHAUS, pp 39–40
[15] Ibid., p 49
[16] Winston S Churchill The Second World War: Vol 2, Their Finest Hour pp 384–385.
Houghton Mifflin Company, Boston, 1949
[17] JONES 1, p 169
[18] Members of KGr 100 recalled in interviews with Dr Alfred Price that they were
generally more aware of the deficiencies of the navigation systems than of
British interference. Specifically they said the clear, moonlit night plus the ini-
tial fires contributed most to the accuracy of the Coventry attack. TRENKLE 3,
p 126, reports that KGr 100 was not troubled by interference until May 1941
[19] LOVELL, p 14
[20] KEMP, pp 20–22, 28
[21] Ernest Putley, Ground Control Interception, BURNS, pp 162–176
[22] BROWN, p 63
[23] SIG CORPS 2, pp 86–87, 96
[24] PILE, p 172
[25] Maurice V Wilkes Memoirs of a Computer Pioneer pp 64–65. The MIT Press:
Cambridge, Massachusetts, 1985
45 Brassey’s Annual: The Armed Forces Yearbook, 1953 pp 88–106
[27] PILE, p 173
[28] Ibid., p 287
490
Notes and Sources
[29] Viktor Reimann Goebbels p 245. Garden City, New York: Doubleday and Com-
pany, 1976. NIEHAUS, p 54
[30] ROWE, pp 73–74
[31] Churchill [16], p 333
[32] Ibid., pp 383–391
[33] Editor, Radiolocators J. Ap. Phys. Vol 12, p 511, 1941
[34] For example, Larry Wolters, Radio Locator Called a Child of Television Chicago
Sunday Tribune 13 July 1941
Chapter 3.3 The Atlantic, 1941

[1] ROSKILL 1, pp 279–282, 604–605
[2] REUTER, pp 55–57; ROSKILL 1, pp 291–292
[5] REUTER, pp 65–68; ROSKILL 1, pp 287–291, 367–372
[6] HOWSE, p 91
[7] TRENKLE 1, pp 116–120
[8] Burkard Baron von Müllenheim-Rechberg Battleship Bismarck: A Survivor’s Story
pp 130–131. Annapolis: Naval Institute Press, 1990
[9] GIESSLER, p 72
[10] ROSKILL 1, pp 403–404
[11] HOWSE, pp 95–96
[12] Müllenheim-Rechberg [8], 130–131
[13] Ibid., pp 177–178
[14] GIESSLER, p 73
[15] HEZLET, pp 207–208
[16] Ibid., p 207
[17] Müllenheim-Rechberg [8]
[18] Robert D Ballard The Discovery of the Bismarck. New York: Warner Books, 1990
[19] REUTER, p 47
[20] GIESSLER, p 75
[21] HOWSE, pp 341–343
[22] Ibid., pp 84–86
Chapter 3.4 Friend, foe or home?

[1] Robert Watson Watt, memorandum to the Committee for the Scientific Survey of
Air Defence dated 27 February 1935, Detection of Aircraft by Radio Methods,
paragraph 19, cited in WATT, p 434
[2] WATT, p 138
[3] GEBHARD, pp 251–256
[4] Lord Bowden of Chesterfield, The story of IFF (identification friend or foe) Proc.
IEE Vol A132, 1985, pp 435–437
[6] Bowden [4], p 435
[7] SIG CORPS 2, pp 242–243
[8] Harold Alden Wheeler Hazeltine Corporation in World War II Ventura, California:
Pathfinder Publishing Company, 1993
491
[9] R M Trim, The development of IFF in the Period up to 1945, BURNS, pp 436–457
[11] SIG CORPS 1, pp 264–266
[13] For examples of radar-directed fire on friendly ships see MORISON 6, pp 188,
243, 317, 354; 7, p 154; 12, pp 222, 227
[14] REUTER, pp 34–35
[15] TRENKLE 1, p 172
[16] Ibid., p 173
[17] Harry von Kroge, letter dated 7 March 1995
[19] Ibid., pp 133–135
[20] REUTER, pp 35–36
[21] Trim [9], pp 442–445
[22] Wheeler [6], pp 215–238
[23] TRENKLE 1, p 178
[24] Bowden [4], p 436
[25] RADAR, No 2, May 1944, pp 21–27
[26] ROWE, p 126
[27] TRENKLE 2, p 16
[28] WATT, p 147
[29] Ibid.
Chapter 3.5 The Japanese realize they are behind

[1] NAKAGAWA, pp 21–22. Nakagawa’s treatment of history differs from that of
Wilkinson (Roger I Wilkinson, Short Survey of Japanese Radar Transactions of
the American Institute of Electrical Engineers Vol 65, pp 370–377, 455–463, 1946),
although not with his technical descriptions of the equipment. This difference
is repeated in appendix G of PRICE 2. Consultation with a number of Japanese
sources bears out Nakagawa’s presentation, which draws on a wider base, was
researched at a more leisurely pace and was not veiled by greatly dissimilar
tongues
[2] Ibid., pp 22–24
[3] Ibid., pp 25–26
[4] Ibid., p 25
[5] Ibid., p 26
[6] Ibid., pp 26–27
[7] Ibid., pp 27–28
[8] Ibid., pp 28–29, 87
[9] Ibid., pp 32–33
[10] Ibid., p 83; PRICE 2, p 289
[11] NAKAGAWA, pp 34, 84–85
[12] PRICE 2, p 293
[14] Genzo Sato, The Secret Story of the Yagi Antenna in World War II The Radiosci-
entist Vol 2, No 4, pp 71–74, 1991. Colin MacKinnon in a letter to the author
dated 31 May 1995
[15] PRICE 2, pp 290–291. Price attributes the manufacture of Tachi-1 to Sumitomo
and Tachi-2 to Shibaura. Sumitomo was the name NEC adopted briefly during
492
Notes and Sources
the war, and Shibaura was Tokyo Shibaura Electric, later Toshiba. Japanese
Wartime Military Electronics and Communications, Section VI, Japanese
Army Radar, Technical Liaison and Investigation Division, Office of the Chief
Signal Officer, GHQ, US Army Forces Pacific found at US National Archives,
SCAP Box 7428
[17] PRICE 2, p 291. Wilkinson [1], p 375
[18] Air Technical Intelligence Group, Advanced Echelon FEAF, Report No 261,
12 December 1945
[19] SCIENTIFIC, Vol 1, pp 12–14
[20] NAKAJIMA, p 255
[21] Ibid., p 254–255
Chapter 4.1 Microwaves
[1] GIESSLER, p 63
[2] Ben R Rich and Leo Janos Skunk Works: A Personal Memoir of My Years at Lockheed
pp 19–27. Boston: Little, Brown and Co., 1994
[3] W W Hansen, A Type of Electric Resonator J. Ap. Phys. Vol 9, pp 654–663, 1938
[4] Arthur L Norberg and Robert W Seidel, The Contexts for the Development of
Radar, BLUMTRITT, pp 199–216
[5] The term ‘ionosphere’ seems to have been coined independently by Robert Wat-
son Watt and E V Appleton in 1926. C S Gillmor, The history of the term ‘iono-
sphere’ Nature Vol 262, pp 347–348, 1976. The term Ionosphäre was used by
Hans Plendl and was apparently introduced to American researchers through
his papers. Wilbert F Snyder and Charles L Bragaw Achievement in Radio: Sev-
enty Years of Radio Science, Technology, Standards, and Measurement of the National
Bureau of Standards footnote, p 172. Washington: US Government Printing Of-
fice, 1986
[6] Lord Rayleigh, On the passage of electric waves through tubes or the vibrations
of dielectric cylinders Phil. Mag. Vol 43, pp 125–132, 1897
[8] O Schriever, Elektromagnetishe Wellen an dielektrischen Drähten Ann. Phys.
Vol 63, pp 645–673, 1920
[9] Karle S Packard, The Origin of Waveguides: a Case of Multiple Rediscovery
IEEE Transactions on Microwave Theory and Techniques Vol MTT-32, pp 961–969,
1984
[10] Ibid., pp 966–968
[11] A Arsenjewa-Heil and O Heil, Eine neue Methode zur Erzeugung kurzer,
ungedämpfter, elektromagnetischer Wellen grösser Intensität Zeit. Phy. Vol 95,
pp 752–762, 1935
[12] Karl R Spangenberg Vacuum Tubes p 616. New York: McGraw-Hill Book Com-
pany, 1948
[13] GUERLAC 1, p 194
[14] Russell H Varian and Sigurd F Varian, A High Frequency Oscillator and Am-
plifier J. Ap. Phys. Vol 10, pp 321–327, 1939; W W Hansen and R D Richtmyer,
On Resonators Suitable for Klystron Oscillators J. Ap. Phys. Vol 10, pp 189–199,
1939
493
[15] Edward L Ginzton, The $100 idea: How Russell and Sigurd Varian with the
help of William Hansen and a $100 appropriation, invented the klystron IEEE
Spectrum Vol 12, pp 30–39, 1975
[16] GUERLAC 1, p 213
[17] Luis W Alvarez, Alfred Lee Loomis—last great amateur of science Phys. Today
Vol 36, 1983, pp 25–34. Adapted from Biographical Memoirs Vol 51, National
Academy of Sciences, 1980
[18] GUERLAC 1, p 221
[19] Ibid., pp 247–250
[20] David H Sloan and Lauritsen C Marshall, Ultra-High Frequency Power, (ab-
stract) Phys. Rev. Vol 58, p 193, 1940 Winfield W Salisbury, The Resnatron
Electronics February 1946, pp 92–97
[21] BOWEN, p 143
[22] Russell W Burns, The Early History of Centimetric Radar: the Contributions
of the General Electric Company, manuscript intended for publication, 1998
[23] GUERLAC 1, pp 225–226
[24] BATT, p 43
[25] The space-charge limitation of klystrons was overcome after the war by impos-
ing an axial magnetic field on the electron beam, thereby allowing much larger
currents. These were the generators that made possible the high-energy elec-
tron accelerators at Stanford. The excellent frequency stability of the klystron
was the determining factor in choosing klystron over magnetron
[26] J T Randall, The Cavity Magnetron Proc. Phys. Soc. Vol 58, pp 247–252, 1946
[28] GUERLAC 1, pp 228–231
[29] Burns, [22]
[31] HOLLMANN, Vol 2, pp 2–4
[33] Reg Batt in MAGNETRON, p 34
[34] LOVELL, pp 1–43
[35] Ibid., pp 41–42
[36] C A Cochrane, Development of Naval Warning and Tactical Radar, KINGS-
LEY 1, pp 189–203
[37] HOWSE, pp 83–84
[38] James Sayers, MAGNETRON, pp 12–14
[39] Nakajima, BURNS, pp 243–258. In Chapter 2.5 the 1938 visit of Professor
Barkhausen to Japan is described. Although it was a highly technical visit
lasting two months during which he went to the principal laboratories and
industries, he did not learn about the new centimeter-wave generator
[40] Sogo Okamura, editor History of Electron Tubes p 29. Washington: IOS Press
(republished for Ohmsha, Tokyo), 1994. The cavity magnetron was refered to
at Bell Labs for a while as the ‘Samuel oscillator’, letter from J R Wilson to L
A DuBridge, 30 April 1940
[41] Arthur L Samuel, Electron Discharge Device, US Patent No 2 063 342, 6 Decem-
ber 1936
[42] Kinjiro Okabe Magnetron-Oscillations of Ultra-Short Wavelengths and Electron
Oscillations in General pp 30–31. Tokyo: Shokendo, 1937
[43] NAKAJIMA; Marvin Hobbs, Japanese Magnetrons Electronics May, 1946, pp
494
Notes and Sources
114–115
[45] Ibid., pp 32, 89
[46] NAKAJIMA
[47] James Phinney Baxter Scientists Against Time p 142. Boston: Little, Brown and
Company, 1946
[48] N F Alekseev and D D Malairov, Generation of High-Power Oscillations with
a Magnetron in the Centimeter Band Zhurnal Tekhnicheskoi Fiziki Vol 10, pp
1297–1300, 1940; transliteration and translation by I B Benson for Proc. IRE Vol
32, pp 136–139, 1944
[50] Ibid., pp 92–94
[51] Ibid., pp 67–69
[52] F Fischer and F Lüdi, Die Posthumus-Schwingungen im Magnetron Schweiz-
erischer Elektrotechnischer Verein Bulletin Vol 28, pp 277–283, 1937
[53] F Lüdi, Zur Theorie der geschlitzten Magnetfeldröhre Helvetica Physica Acta
Vol 16, pp 59–82, 1942
[54] Hans Paul, Neuere Entwicklungen auf dem Gebiet der Zentimeterwellen Elek-
trotechnische Zeitschrift Vol 77, pp 849–854, 1956
[55] Hans H Jucker, who is preparing a study of Swiss radar history, in a letter to
the author dated 21 April 1995
Chapter 4.2 The Tizard Mission

[1] Walter Millis Road to War: America 1914–1917. Boston: Houghton Mifflin Co.,
1935
[2] Clark op. cit., pp 251–252
[3] Churchill vacillated in his support, withdrawing it even after preparations were
well under way. Lindemann favored it, despite it being Tizard’s idea. ZIM-
MERMAN, pp, 61–89
[4] BOWEN, p 151
[5] Ibid., pp 157–158
[6] Wallace is reported to have said that GL mark I was responsible for having
brought down 400 German planes with the British Expeditionary Force in
France and Belgium. GUERLAC 1, p 170. Unknown to those participating
in these discussions at the time, the Battle of Britain was demonstrating the
fantasy of these claims, as GL mark I at the time of the Tizard meetings had
not been instrumental in bringing down a single plane. PILE, p 173
[7] Interview with A E Cassevant and John J Slattery cited in MCKINNEY, p 213
[8] SIG CORPS 1, p 193
[9] Ibid., p 200
[10] Interview with Karl T Compton on 20 October 1943 by Henry Guerlac
[11] BOWEN, p 180
[12] SIG CORPS 1, pp 198–199
[13] BOWEN, p 158
[14] SIG CORPS 1, pp 288–291. Ruth F Sadler with Lt Col Herbert H Butler, His-
tory of the Electronics Training Group in the United Kingdom, March 1944,
manuscript at US Army Center for Military History, Washington, DC
[15] PILE, p 114
495
[16] Ibid., p 166

[17] Ibid., pp 171–174
[18] Ibid., p 215
[19] GUERLAC 1, p 258
[20] BELL, pp 25–26. The Bureau of Ordnance assumed responsibility for fire-
control radar and replaced the Bureau of Engineering’s FA designation with
mark 1 and so on
[21] Wilfred Eggleston Scientists at War pp 28–40. London: Oxford University Press,
1950
[22] OSRD 6, p 3
Chapter 4.3 The Radiation Laboratory

[1] GUERLAC 1, pp 259–261; RABI, p 134
[2] ALVAREZ, p 87
[3] GUERLAC 1, pp 262–263
[4] ALVAREZ, pp 88–92
[6] BELL, pp 95–100. The comparison of British and American microwave AI sets
is made nicely in BUDERI, pp 116–119
[7] GUERLAC 1, p 324
[8] GETTING, pp 3–80
[9] Lee L Davenport, IEEE, pp 61–62
[10] GETTING, pp 108–110
[11] Monopulse radar later removed this jitter much more effectively than the
smoothing circuits, but such accuracy was not required for the fire-control
problems of World War II
[12] GETTING, p 113; GUERLAC 1, p 278
[13] GETTING, pp 120–121. Davenport remembers that the mechanical M-7 director
was used for this test, IEEE, p 66. The Bell Labs history does not mention this
test but states that the T-10 prototype was functioning in December 1941.
BELL, pp 145–146
[14] GUERLAC 1, p 481
[15] SIG CORPS 2, pp 268–274
[16] GETTING, pp 133–142
[17] Perhaps the best judgment on the value of SCR-584 comes from the 1995 cata-
logue of Radio-Research Instrument Co., Waterbury, Connecticut: ‘Designed
at MIT Radiation Labs and still considered one of the finest automatic tracking
radars ever built, it is now being used in hundreds of installations in its orig-
inal form and in various modifications. We have them in stock for immediate
delivery complete in their own 20 ft trailer van containing the entire system’
[18] MORISON 10, p 154
[19] POLLARD, pp 53–60
[20] GUERLAC 1, pp 273–274, 399–400
[21] The administrative history is entirely from GUERLAC 1, pp 647–692
[22] Britton Chance, IEEE, p 50
[23] Ernest C Pollard, IEEE, pp 214–215
[24] Louis N Ridenour, Editor in Chief Radiation Laboratory Series New York:
McGraw-Hill Book Company, 1947
496
Notes and Sources
[25] RABI, pp 164–165
Chapter 4.4 The proximity fuze—the smallest radar

[1] The Luftwaffe tried without notable success the ‘bomb the bombers’ tech-
nique when the air defense of Germany began to call for desperate measures.
BEKKER 3, pp 404–405
[2] Russell Burns, Early History of the Proximity Fuze (1937–1940) Proc. IEE Vol
140A, pp 224–236, 1993
[3] Dennis Newton, The Remarkable Work of Alan Butement Despatch: Journal of
the New South Wales Military Historical Society Vol 27, pp 70–82, 1992; Betty
Williams Dr W A S Butement, the First Chief Scientist for Defence Canberra:
Australian Government Publishing Service, 1991
[4] JONES 2, pp 307–313
[5] The Proximity Fuze US Naval Administrative Histories of World War II part II, Vol
II, pp 192–193. Washington: Bureau of Ordnance
[6] Section T Final Summary Report, 19 April 1942, Archives of the Department of
Terrestrial Magnetism. Applied Physics Laboratory, The ‘VT’ or Radio Prox-
imity Fuze, with confidential supplement, 20 September 1945; Merle A Tuve
and Richard B Roberts, US Patent 3 166 015, Radio Frequency Proximity Fuze,
6 January 1943; M A Tuve, Affidavit on behalf of Butement et al v. Varian before
the Board of Patent Interferences, No 86 648, 4 September 1964
[7] R B Roberts, extract from a manuscript autobiography and reminiscences,
Archives of the Department of Terrestrial Magnetism, Carnegie Institution
of Washington, 1978
[8] Statement to the author circa 1965
[9] Roberts, [7]
[10] Either Roberts’s memory was in error or he was misinformed because the
Cleveland was soon to take part in the invasion of North Africa
[11] Administrative Histories [5], p 225
[12] Ibid., pp 233–236
[13] MORISON 5, p 107
[14] BALDWIN, pp 181–189
[15] Administrative Histories [5], pp 253–255
[16] SIG CORPS 3, pp 297–298
[17] Botho Stüve Peenemünde West: Die Erprobungsstelle der Luftwaffe für geheime
Fernlenkwaffen und deren Entwichlungsgeschichte pp 761–777. Munich: Bechtle
Verlag, 1995
[18] R W Burns, Factors Affecting the Development of the Radio Proximity Fuze
1940–1944 IEE Proc. Sci. Meas. Technol. Vol 143, pp 1–9, 1996
Chapter 4.5 Greater and lesser microwave sets

[1] LOVELL, p 45
[2] RAWNSLEY, pp 185–189
[3] James Sayers, MAGNETRON, pp 12–14
[4] LOVELL, p 52
[5] Ibid., pp 65–84
[6] BROWN, p 33
497
[7] Sir Bernard Lovell, H2S/ASV, MAGNETRON, pp 38–40

[8] Scope Distortion, RADAR, No 11, pp 29–32, 1945
[9] In allocating electron tubes in April 1942 Cherwell favored elimination of GL
mark III as the best means of reconciling production with need. JAY, part 3,
para 69
[10] BELL, pp 83–84
[11] D H Tomlin, The Origins and Development of UK Army Radar to 1946, BURNS,
pp 284–295; D H Tomlin, Army Radar 1939–1945, MAGNETRON, pp 53–58
[12] ALVAREZ
[13] RADAR, No 10, pp 28–33, 1945
[14] ALVAREZ, pp 101–103
[15] GUERLAC 1, pp 386–394
[16] RADAR, [13]
[17] RADAR SURVEY 1, pp 127–132
[18] RADAR, [13], p 33
[19] For a description of some of these systems and the problems of implementing
them see William L Leary, The Search for an Instrument Landing System,
1918–48 Innovation and the Development of Flight Roger D Saunius, editor, pp
80–99. College Station, Texas: Texas A&M University Press, 1999
[20] GUERLAC 1, pp 497–505
[21] Luis W Alvarez, Alfred Lee Loomis—Last Great Amateur of Science Phys. Today
Vol 36, No 1, pp 25–34, 1983
[22] Captain C W Watson, Ground-Controlled Approach for Aircraft Electronics pp
112–115, 1945; Charles Fowler, IEEE, pp 101–114
[23] ALVAREZ, pp 98–101
[24] RADAR, No 7, pp 19–25, 1945. Details of the operation of GCA draw in part
from a letter to the author from Squadron Leader (ret.) T Winchcombe dated
31 December 1997. The trials of mark I are the basis of an historically and
technically correct autobiographical novel by Sir Arthur C Clarke Glide Path
London: Sidgwick and Jackson, Ltd, 1963. Sir Edward Fennessy wrote that he
had been the Wing Commander who selected Clarke as RAF Technical Officer
to join the team working with Alvarez. The decision was influenced by the
marginal comment on Clarke’s résumé: ‘brilliant but mad’.
[25] GUERLAC 1, pp 448–459
[27] Milton A Chaffee, IEEE, pp 42–46
[28] KEMP, pp 120–121
[29] GUERLAC 1, pp 1025–1030
[30] Ibid., pp 537–550
Chapter 5.1 The Mediterranean, 1940–1942

[1] HOWSE, pp 20–21
[2] Ibid., pp 63–64
[3] RAF SIGNALS 4, p 70
[4] HEZLET, p 194
[5] Macintyre The Battle for the Mediterranean, p 36
[6] Charles Lamb War in a Stringbag pp 105–113. London: Cassell and Company,
1977; Vice-Admiral Brian Betham Schofield The Attack on Taranto pp 40–52.
498
Notes and Sources
Annapolis: US Naval Institute, 1973. Imitation is often said to be the highest

form of flattery. The Kriegsmarine built a number of multi-purpose aircraft for
their carrier that was never finished, the Graf Zeppelin that had a design very
similar to the Swordfish, the Fiessler Fi 167. It had a more powerful engine
than the Swordfish, which provided higher speed and greater range
[7] Lamb [6], pp 40–44
[8] Tony Spooner, Goofingtons and Malta’s War Aeroplane Monthly July, pp 409–412,
1988
[9] S W C Pack Night Action off Cape Matapan Annapolis: US Naval Institute, 1972
[10] Lutton op. cit., p 84
[11] Peter Brain, Sheilah Lloyd and F J Hewitt South African Radar in World War II
Cape Town: The SSS Radar Book Group, 1993; B AAustin, Radar in World War
II: The South African Contribution Engineering Science and Education Journal
June, pp 121–130, 1992
[12] Ralph Bennett Ultra and the Mediterranean Strategy pp 72–74. New York: William
Morrow and Company, 1986
[13] Ibid., pp 38–39
[15] Playfair op. cit., Vol 2, p 45
[17] Flying Officer John W Findlay, 202 Group, RAF, in a letter to the author dated
11 February 1994
[18] KEMP, pp 43, 52, 93, 109, 121
[19] HOFFMANN 1, pp 201–204
[20] Ibid., pp 206–207
[22] Flying Officer George A Emery, 230 Squadron, RAF, in a letter to the author
dated 16 December 1994
[23] Macintyre The Battle for the Mediterranean, p 141
[24] JONES 1, p 256
[25] C Powell, A personal reminiscence: GL radar, an elementary ECCM technique,
BURNS, pp 503–505
[26] German aerial photographs in the US National Archives, Record Group 242
[27] TEDDER, p 245
[28] Adolf Galland Die Ersten und die Letzten: Die Jagdflieger im zweiten Weltkrieg
p 185. Darmstadt: Franz Schneekluth, 1953
[29] HOWSE, p 154
[30] PRICE 1, p 94; REUTER, p 109
[32] BOWEN, pp 113–114. This brings up again the question of when Germany got
an ASV According to Leo Brandt it was May 1941. For further discussion see
Chapter 5.3, The Channel, 1942
Chapter 5.2 War in the Pacific
[1] Harold A Zahl Radar Spelled Backwards p 74. New York: Vantage Press, 1972
[2] Myron J Smith Pearl Harbor, 1941: a Bibliography New York: Greenwood Press,
1991
499
[3] US Congress Pearl Harbor Attack: Hearings Before the Joint Committee on the Investi-
gation of the Pearl Harbor Attack part 26, pp 367–375. Washington: Government
Printing Office, 1946
[4] The Army Air Corps changed into the Army Air Forces on 20 June 1941
[5] CRAVEN & CATE 1, pp 289–291
[6] Congress [3], part 26, pp 379–386; part 27, pp 615–632
[7] Ibid., part 10, pp 5027–5080; part 26, pp 517–536; part 32, pp 341–351
[8] SIG CORPS 2, pp 10–15
[9] HOWSE, pp 122–126
[10] Sir John Rupert Colville The Churchillians p 140. London: Weidenfeld and
Nicholson, 1981
[11] Ed Simmonds, Historic Background More Radar Yarns Ed Simmonds, editor,
pp 2–29. Published privately by E W & E Simmonds, 15 Blair Street, Port
Macquarie NSW 2444, Australia, 1992
[12] Ed Simmonds in a letter to the author dated 19 November 1994
[13] Ed Simmonds in a letter to the author dated 6 August 1994
[14] WATT, pp 308–310
[15] Elting E Morison Turmoil and Tradition: a Study of the Life and Times of Henry L
Stimson p 562. Boston: Houghton Mifflin Company, 1960
[16] SIG CORPS 2, pp 95–102
[17] Ibid., p 97
[18] CRAVEN & CATE 1, pp 291–293
[19] Elting Morison [15], 563
Chapter 5.3 The Channel, 1942

[1] ROSKILL 2, pp 147–150
[2] REUTER, p 76
[3] Apparently the ships were able to obtain bearings on shore Seetakt stations by
using the directional capability of their receivers, but the wording of the only
source for this, REUTER, is ambiguous
[4] John Deane Potter Fiasco: the Break-out of the German Battleships pp 62–63. New
York: Stein and Day, 1970
[5] JONES 1, pp 233–234
[6] HOWSE, pp 129–131
[7] ROSKILL 2, p 154
[8] Captain H J Reinicke, The German Side of the Channel Dash US Naval Institute
Proceedings Vol 81, pp 636–646, 1955
[9] BRANDT, pp 39–40. Brandt is quite specific in giving the time of a German
examination of an ASV set: ‘Am Tage der Versenkung der "Bismarck" befand
sich ein "ASV"-Gerät im Laboratorium der Firma Telefunken’. This together
with Brandt’s extraordinary competence and high position would seem to
exclude the possibility of the date being a typographical error, yet Metox
radar receivers to counter ASV and designed from a knowledge of it were not
issued to U-boats until August 1942. REUTER, p 153
[10] WATT, pp 358–378
[11] Sir Robert Cockburn, The Radio War BURNS, p 337
[12] F A Kingsley, Electronic Countermeasures in the Royal Navy KINGSLEY 2, pp
196–199
500
Notes and Sources
[13] JONES 1, pp 121–125; for a detailed description of the raid see Millar, op. cit.
[14] Cockburn [11], pp 337–338
[15] JONES 1, pp 130–134; Millar, op. cit., pp 109–133
[16] Ibid., p 192
[17] PRICE 1, p 78
[18] Cockburn, loc. cit.
[19] JONES 1, pp 233–249
[20] Ibid., pp 244–246
[21] ROWE, pp 128–134
[22] Terence Robertson Dieppe: the Shame and the Glory Boston: Little, Brown and
Company, 1962
[23] John P Campbell Dieppe Revisited: a Documentary Investigation pp 130–132.
London: Frank Cass and Co., 1993
[24] Ibid., p 137
[25] ROSKILL 2, p 241
[26] Campbell [23], p 131
[27] NIEHAUS, pp 93–94; HOFFMANN 1, pp 268–269; Campbell, [23], p 141
[28] J R Robinson, Radar Intelligence and the Dieppe Raid Canadian Defence Quar-
terly Vol 20, pp 37–43, 1991
[29] Jack Nissen and A W Cockerill Winning the Radar War: a Memoir pp 162–191.
New York: St. Martin’s Press, 1987
[30] Robinson [28], p 41
[31] JONES 1, pp 195–198; Robinson [28], p 42
[32] JONES 1, p 402
[33] Campbell [23], pp 147, 168
[34] ROSKILL 2, pp 246–247
[35] Campbell [23], p 148
[36] Derek Howse, Type Number of Radar Sets, Operational or Designed, KINGS-
LEY 1, pp 372–373
[37] Hugh G Henry III, draft of a dissertation intended for submission to St John’s
College, University of Cambridge
[38] Robinson [28], p 42
Chapter 5.4 Carrier warfare defined

[1] FRIEDMAN, p 149
[2] LUNDSTROM 1, p 63
[3] Ibid., pp 91–93
[4] Ibid., pp 96–98, 101–102
[5] Ibid., pp 118–119
[6] Ibid., pp 148–149
[7] Ibid., p 169
[8] Ibid., p 196
[9] Ibid., pp 210–211
[10] Ibid., pp 243–248
[11] Ibid., p 293
[12] MORISON 4, pp 141–159
[13] Walter Lord Incredible Victory New York: Harper and Row, 1967
[14] Gordon Prange Miracle at Midway New York: McGraw-Hill, 1982
501
[15] LUNDSTROM 1, pp 309–449

[16] MORISON 4, p 104
[17] Mitsuo Fuchida and Masatake Okumiya Midway: the Battle That Doomed Japan:
The Japanese Story pp 243–244. Annapolis, Maryland: US Naval Institute, 1955
[18] NAKAGAWA, pp 32, 89
Chapter 5.5 The South Pacific, 1942

[1] The determination of when vessels received radar was made from the collection
of photographs in the Still Picture Division of the US National Archives. Pho-
tographs were taken on completion of construction, repair or modification
and generally allowed the radar antennas to be recognized
[2] Letter dated 16 January 1995 to the author by Captain Russell S Crenshaw, Jr,
who was Gunnery Officer aboard USS Maury during these actions
[3] FRIEDMAN, pp 147–148. The important information, that the San Juan carried
an SG radar at the time of the Battle of Savo Island on 8/9 August 1942, was
confirmed at the National Archives by examination of photographs taken on
31 May 1942 in Boston Harbor. She also had an SC and two FDs
[4] MORISON 5, p 154
[5] MCNALLY, pp 13a–13c
[7] PRICE 2, pp 47–48, 290
[9] Lieutenant Lewis C Mattison, USNR, and Master Sergeant Dermott H Mac-
Donnell, USMC, Report on Fighter Direction at Cactus, October 8, 1942 to
January 1, 1943. This document obtained through the courtesy of Dr John B
Lundstrom, Milwaukee Public Museum
[10] Frank op. cit., p 111
[11] Ibid., p 117
[13] Frank op. cit. p 178
[15] Jennings B Dow, Navy Radio and Electronics During World War II Proc. IRE
Vol 34, pp 284–287, 1946
[16] Robert C Rasmussen, It Helped Sink Six Jap Warships Bell Laboratories Record
Vol 24, pp 201–202, 1946
[18] LUNDSTROM 2, pp 467–471
[19] Frank op. cit., pp 402–403
[20] Ibid., p 379
[21] LUNDSTROM 2, pp 339–340
[22] Ibid., pp 384, 458
[23] Ibid., pp 454–455
[24] C W Kilpatrick The Night Naval Battles in the Solomons p 81. Pompano Beach,
Florida: Exposition Press of Florida, Inc., 1987
[25] The gun flashes were only on American ships, as the Japanese used flashless
powder
[26] LUNDSTROM 2, pp 478–483
502
Notes and Sources
[27] ROSKILL 2, p 232

[28] Undersea exploration by Dr Robert D Ballard, Woods Hole Oceanographic
Institution presented at the conference ‘World War II in the Pacific’, 10–12
August 1994, Crystal City, Virginia
[29] Frank op. cit., pp 486–487
[30] Ibid., p 207
[31] SIMMONDS & SMITH, pp 97–108; see also [32]
[32] Samuel Milner United States Army in World War II The War in the Pacific: Victory
in Papua Washington: Office of the Chief of Military History, 1957
[33] SIG CORPS 2, pp 261–265. Lt Col Harold A Zahl and Maj John W Marchetti,
Radar on 50 Centimeters Electronics January 1946, pp 98–104; February 1946,
pp 98–103
[34] SIMMONDS & SMITH, pp 39–53, 97–108
Chapter 5.6 The Eastern Front

[1] Steven J Zaloga, Soviet Air Defense Radar in the Second World War Journal of
Soviet Military Studies Vol 2, No 1 (March), pp 104–116, 1989
[2] LOBANOV, pp 238, 245, 80
[3] Zaloga [1], p 105
[4] Generalleutnant a.D Klaus Uebe Russian Reactions to German Airpower in World
War II p 10. New York: US Air Force Historical Division, Aerospace Studies
Institute, Air University, Arno Press, 1964
[5] PLOCHER 1, pp 239, 156
[6] LOBANOV, pp 184–185
[7] Ibid., pp 228–235
[8] Pekka Eskelinen, View of Military Radio Systems and Electronic Warfare in
Finland During WWII AES Vol 11, pp 3–7, 1996
[9] Kurt Petsch Nachtjagdleithschiff ¡Togo’ pp 58–100. Reutlingen: Preussischer
Militär-Verlag, 1988
[11] Ibid., pp 191–197
[12] PRICE 3, p 338
[13] LOBANOV, p 199
[14] Zaloga [1], p 111
[15] Ibid., pp 111–112
[17] PLOCHER 1, p 163
[19] Walter Schwabedissen The Russian Air Force in the Eyes of the German Comman-
ders p 360. New York: US Air Force Historical Division, Aerospace Studies
[20] HOFFMANN 1, pp 169–170
[21] Ibid., pp 303–305
[22] PLOCHER 3, p 64
[23] Walter Bartig, Geschichte einer Funkkompanie und ihrer Männer, p 34.
Manuscript, Berlin, September 1987
[24] Schwabedissen [19], pp 318–323
[25] PLOCHER 3, p 136
503
[26] HOFFMANN 1, pp 163–165

[27] Schwabedissen op. cit., p 376
[28] HOFFMANN 1, pp 184–188
Chapter 6.1 The destruction of German cities initiated

[1] WEBSTER & FRANKLIN 1, pp 155–166
[2] Ibid., p 459
[3] The term ‘AN-Verfahren’ is traced back to the Lorenz beam-navigation system
in which the pilot heard dots, if he were on one side of the correct course (in
one of the two radiation lobes), dashes if he were on the other side (in the
second lobe) and long dashes if on course (equal amplitudes from both lobes).
Straying a little to the dot side generated from the unequal amplitudes an
audible signal that fooled the ear into recognizing a Morse letter A (dit–dah);
straying to the dash side a Morse letter N (dah–dit)
[4] KROGE, p 121
[5] REUTER, p 244; TRENKLE 1, pp 91–93
[7] One version of Wassermann shows a cabin half way up the structure that is
frequently taken to be the operating room. The dipoles had to be fed in phase,
which was done by using the same lengths of cable to connect each dipole
with the generator. The cabin stored excess cable on reels
[8] FRIEDMAN, p 173
[9] See Chapter 4.5
[11] BRANDT, pp 32–33
[12] RUNGE, p 47
[13] REUTER, p 34
[14] Ibid., p 86
[15] Udet’s genial nature is demonstrated by his long friendship with Carl Zuck-
mayer, a strong anti-Nazi playwright, who was forced to flee Europe minutes
ahead of the Gestapo with each new advance of the dictator’s power. His play
‘Des Teufels General’ attempts to salvage the memory of his friend
[16] PRICE 1, pp 64–65
[18] STREETLY, pp 214–215
[19] TRENKLE 3, pp 182–192
[20] ADERS, pp 76–77
[21] R Cockburn, The Radio War, BURNS, pp 330–356
[22] HOFFMANN 1, pp 330–332; TRENKLE 1, pp 102–104
[23] R V Jones in a letter to the author dated 18 April 1995
[24] TRENKLE 1, pp 104–106; Theodor Schultes, Funkmess- Übersichtsverfahren
Die Funkortung der deutschen Flugsicherung pp 56–94. Dortmund: Verkehrs-
und Wirtschafts-Verlag, GmbH, 1953
[25] STREETLY, p 215
[26] TRENKLE 1, pp 51–54
[27] A E Hoffman-Heyden, German World War II Anti-jamming Techniques,
BURNS, pp 374–396
[28] RUNGE, pp 54–55
504
Notes and Sources
[29] REUTER, pp 87–88

[30] PRICE 1, pp 69–70
[31] RUNGE, p 49
[32] ADERS, pp 77–79
[33] HOFFMANN 1, pp 32–34
[34] PRICE 1, p 70
[35] RUNGE, pp 55–57, 76–82
[36] Manfred von Ardenne Ein glückliches Leben für Technik und Forschung pp 129–
130. Berlin: Verlag der Nation, 1973
[37] V Ardenne [36], pp 105, 101, 124–128, 67–68
[38] Radar attracted others who had little stomach for some kinds of war research.
George Valley, who pushed for the construction and designed in the late 1940s
the radar system to guard the North American continent from Soviet attack,
was a Rad Lab veteran. His graduate research had been in nuclear physics,
and when demands for radar work slacked DuBridge asked him to transfer
to Los Alamos for bomb work. Valley refused, saying ‘No, I think that’s filthy,
I won’t do it’. BUDERI, p 358
[39] ROWE, p 108
[40] There are numerous descriptions of Gee. Colin Latham provided this one
[41] ROWE, p 109
[42] The frequencies used were the same used by Knickebein, as these were those
of the pre-war Lorenz blind approach system that the RAF used under license
and for which their planes were equipped
[43] PRICE 1, pp 98–104
Chapter 6.2 Countermeasures

[1] PRICE 2, p 11
[2] STREETLY, pp 17–18, 160; RAF SIGNALS 7, pp 191–194
[3] STREETLY, p 18–19; RAF SIGNALS 7, pp 76–78
[4] R Cockburn, The radio war, BURNS, pp 330–356
[5] REUTER, p 142
[6] OSRD 5, pp 12–13
[7] Ibid., p 18
[8] PRICE 2, pp 11–33
[9] Ibid., p 59
[10] OSRD 5, p 28
[12] Ibid., p 41
[13] PRICE 2, pp 86–89
[14] STEPP, p 72
[15] J S Hey and G S Stewart, Radar Observations of Meteors Proc. Phys. Soc. Vol
59, pp 858–883, 1947
[16] J S Hey, Solar Radiations in the 4–6 Metre Radio Wave-Length Band Nature Vol
157, pp 47–48, 1946
[17] M J B Scanlan, Chain Home Radar—a Personal Reminiscence GEC Review Vol 8,
No 3, pp 171–183, 1993. Scanlan points out that the CH signal/noise ratio was
established at the antenna, which made imperfections in the long transmission
lines to the receivers less troublesome
505
[18] PRICE 1, pp 112–114

[19] WATT, p 395
[20] PRICE 1, p 114
[21] HOFFMANN 1, pp 201–202
[22] RAF SIGNALS 5, pp 149–150, 162
[23] The Würzburg’s rotating polarization allowed a blanking circuit to suppress
the plane-polarized jammer, Carpet, which then had to be provided with a
rotating field antenna. OSRD 5, p 77
[24] REUTER, pp 106–107
[25] There are reports that the name comes from the Danish village Dybböl (German
Düppel) on the Baltic Sea where tests were presumably made
[26] PRICE 2, pp 31, 62–63
[27] OSRD 5, p 25
[28] The Würzburg transmitter had to be altered for these techniques. The pass
band of the receiver was much larger than typical Doppler shifts, and filters
designed to pass the shifted radio frequency would have failed to generate an
observable signal, if the phase of the radio frequency had not been coherent
from pulse to pulse, as it was not in the self-exciting oscillator used. Coherence
was attained by using a continuously-running, low-power auxiliary oscillator
to stimulate the onset of oscillation (demanded by the modulator pulse) to be
in phase with this reference signal, a process called phase locking or injection
locking
[29] PRICE 2, p 284
[30] HOFFMANN-HEYDEN, p 382
[31] Ibid., pp 194–202
[32] Ibid., pp 386–387
[33] DAVIS, p 527
[34] Memorandum of the Reichsminister der Luftfahrt und Oberbefehlshaber der
Luftwaffe, Berlin, 5 January 1944
[35] PRICE 1, p 142
[36] KINGSLEY 2, pp 197–201
[37] JONES 1, pp 284–286
[38] PETERSEN, pp 18–20
[39] PRICE 2, pp 47–56
[40] E H Cooke-Yarborough, Countermeasures Receiver Techniques, BURNS,
pp 365–373,
Chapter 6.3 An air war of attrition

[2] JONES 1, pp 258–259. For a technical description that illustrates the extent of
Allied interest, see J H Buck and J A Pierce, Nonradar Navigational Methods
Radar Aids to Navigation John S Hall, editor Radiation Laboratory Series Vol 2,
pp 47–50. New York: McGraw-Hill Book Company, 1947
[3] F E Jones, OBOE—a precision ground controlled blind bombing system, BURNS,
pp 319–329
[4] ROWE, pp 144–147
[5] F E Jones [3], pp 324–325
[6] Sir Edward Fennessy in letters to the author dated 7 September and 23 November
1995
506
Notes and Sources
[7] DAHL, p 10
[9] REUTER, pp 96–97
[10] Ibid., p 123
[12] REUTER, p 250
[16] This belief came full blown from the pages of the air prophet. Giulio Douhet
The Command of the Air pp 371–389. New York: Coward-McCann, Inc., 1942
[17] GUERLAC 1, p 773
[18] Ibid., pp 766–772
[19] The name H2S, the origin of which has various traditions, caused 10 cm mi-
crowave equipment to be referred to as S-band. H2X, whose nominal origin
is obscure, resulted in 3 cm microwaves being referred to as X-band
[20] GUERLAC 1, pp 776–783
[21] Just How Accurate is H2X Bombing? RADAR, No 9, pp 43–44, 1945
[22] GUERLAC 1, pp 788, 800
[23] Ibid., pp 746–753
[24] CRAVEN & CATE 2, pp 681–706, 848–850
[25] REUTER, p 134
Chapter 6.4 Arbeitsgemeinschaft Rotterdam

[1] The number has been recorded ranging from 1 500 to 9 000
[2] TRENKLE 1, pp 62–63
[3] REUTER, pp 48–49, 113–114. It is useful to note the German names for the various
bureaus and offices just given. Air Ministry = Reichs-Luftfahrts-Ministerium;
Air Signals = Luftnachrichtentruppe; Chief of Air Signals (Martini) = Chef
des Nachrichtenverbindungswesens der Luftwaffe; Special Commissioner
for Radar (Martini) = Sonderbeauftragter für Funkmessgeräte; Supervisor of
Technical Communications (Fellgiebel) = Generalbevollmächtiger für technis-
che Nachrichtenmittel; Plenipotentiary of High-frequency Research (Plendl,
Esau) = Bevollmächtiger für die Hochfrequenzforschung
[4] Arbeitsgemeinschaft does not translate into committee, although it functioned
as such. Working group is perhaps better
[5] HOLLMANN Denis M Robinson cites a German book by Thoma in an inter-
view published in IEEE Frederick Seitz has shown this to have certainly been
Hollmann’s book
[6] ROTTERDAM, 23 February and 17 March 1943
[7] A technical report based on a meeting about antennas held 24–26 March 1943
and issued by Zentrale für wissenschaftliches Berichtswesen der Luftfahrt-
forschung, pp 153–161
[8] ROTTERDAM, 23 February, p 3; 17 March 1943, p 8
[9] Ibid., 8 April 1943, p 1
[10] Ibid., 1 and 22 June and 14 December 1943
[11] Ibid., 14 December 1943, p 6
507
[12] Asuccinct statement about German radio amateurs can be found in Die geheimen
Konferenzen des Generalluftzeugmeisters Georg Hentschel, editor, pp 124–125.
Koblenz: Bernard und Graefe Verlag, 1989. The following dialogue discloses
not only the difficulties faced by radio amateurs but the extraordinary mistrust
by the Nazis for a large part of the German population. Pasewaldt: ‘I think
the progress of this field in England and America is essentially the result of
the unheard of importance of their radio amateurs, while in Germany they
have been unyieldingly suppressed . . .’ Feldmarschall Milch: ‘It was done by
the offices responsible for security. The whole German radio group before the
war had been trained for pure communist espionage. The amateurs were up
to 99% Moscow boys, so we said that now we are going to cut the wires of
those fellows, which was just as it should have been. . . . After the war we must
introduce amateur radio into the Hitler Youth and let it bloom.. . .’ (In answer
to the problem that some were still communicating illegally with radio, Milch
discloses a less endearing side of his personality.) ‘Those scoundrels are lucky
that I am not Chief of the Gestapo for there would be far more executions.
They are far too mild and humane; they cannot compare themselves with the
Russian GPU’
[13] STEGSKOPFER The search for qualified technicians continued by other means.
Public notices calling for soldiers engaged in non-electronic duties to apply
for transfer, even recommending that family members apply for those at the
front who might miss the announcements. There was no mention of radar.
News item, Hochfrequenz-Fachkräfte für die Luftwaffe Funkschau p 295, Vol
16, October–December, 1943
[14] JONES 2, pp 325–326
[15] Letter from Staatsrat Dr-Ing H Plendl to Reichsführer SS Himmler dated 7
January 1944
[16] JONES 2, loc. cit.
[17] Letter from Hans Plendl, Jr., dated 14 March 1995
[18] Fritz Trenkle, Zum 90. Geburtstag von Hans Plendl Funkgeschichte No 78, pp
3–5, 1991. See Chapter 10.3, Secrecy and the Technical Imperative.
[19] H Frühauf, H E Hollmann zum 60. Geburtstag Hochfrequenztechnik und Elek-
troakustik Vol 68, pp 141–143, 1959
[20] REUTER, pp 198–200
[21] Personal communication dated 10 April 1996 from Frau Anna Maria Elstner,
Runge’s daughter
[22] Professor Kurt Fränz, Als Student und Doktorand in der Weimarer Republik
und im Dritten Reich, unpublished manuscript communicated in a letter to
the author dated 15 July 1996
[23] Professor Hans Plendl, Jr, in a statement to the author, 26 August 1996
[24] DAHL, pp 16–24; KAUFMANN, pp 43–46
[25] REUTER, pp 135, 141
[27] Ibid., pp 248–264, 293
[28] REUTER, pp 147–148
[29] Ibid., p 195
508
Notes and Sources
Chapter 6.5 The destruction of German cities completed

[1] For a study of how Germany coped with the air attacks see Earl R Beck Un-
der the Bombs: the German Home Front, 1942–1945 Lexington, Kentucky: The
University Press of Kentucky, 1986
[2] PRICE 1, p 225
[3] Ibid., pp 213–214
[4] Ibid., pp 230–233
[5] Ibid., pp 221–222
[6] HOFFMANN 1, p 329; PETERSEN, pp 60–61. These books provide the most
detailed account of how the Luftwaffe night fighters used radar and radar
control in their ever changing struggle with Bomber Command
[7] PRICE 2, pp 280–288
[8] KAUFMANN, pp 101–114
[9] KOCH, p 75
[10] Sir Edward Fennessy in a letter to the author dated 23 November 1995
[11] PRICE 2, pp 189–190
[12] G Förster, German experiments in Jamming H2S airborne radar, BURNS, pp
397–404
[13] TRENKLE 2, pp 138–139; ROTTERDAM, 29 September 1943
[14] OSRD 5, p 107
[15] CRAVEN & CATE 3, p 723
[16] Ibid., pp 666–669
[17] GUERLAC 1, pp 772, 1076–1079
[18] RADAR SURVEY 1, pp 145–149, 213–216
[19] Ibid., p 189
[20] ROWE, p 117
Chapter 7.1 The Battle of the Atlantic, 1939–1945

[1] For a brief description of the code war see Brian Johnson The Secret War pp
305–349. New York: Methuen Inc., 1978
[2] HOWSE, pp 58, 100
[3] David Zimmerman, Technology and Tactics The Battle of the Atlantic, 1939–
1945: the 50th Anniversary International Naval Conference pp 476–489, Stephen
Howarth and Derek Law, editors. London: Greenhill Books, 1994
[4] Captain Donald Macintyre U-Boat Killer p 50. Annapolis: Naval Institute Press,
1976
[5] HOWSE, p 100
[6] Ibid., pp 108–110; C A Cochrane, Development of Naval Warning and Tactical
Radar Operating in the 10-cm Band, 1940–5, KINGSLEY 1, pp 185–276
[7] Macintyre [4], p 63
[8] BOWEN, pp 102–103
[9] PRICE 4, pp 78–79
[10] Ibid., pp 60–65; Axel Niestlé in a letter to the author dated 15 April 1995
[11] PRICE 4, pp 87–91
[12] Blair, 1996 op. cit.
[13] TRENKLE 2, p 44. See also the historical survey presented by Dr Bode at a
meeting of the Rotterdam Committee. ROTTERDAM, 26 April 1944, pp 30–40
[14] BRANDT, pp 39–40
509
[15] Axel Niestlé, German Technical and Electronic Development The Battle of the At-
lantic pp 430–451, Stephen Howarth and Derek Law, editors. London: Green-
hill Books, 1994.
[16] PRICE 4, pp 94–95
45 Brassey’s Annual: the Armed Forces Yearbook, 1953 pp 88–106
[18] Marshal of the Royal Air Force Sir John Slessor The Central Blue: Recollections
and Reflections pp 524–525. London: Cassell and Company Ltd, 1956
[19] The equipping of these planes with microwave radar had taken place with the
active participation of Rad Lab personnel, even to the attacking of some of the
raiders
[20] Max Schoenfeld Stalking the U-Boat: USAAF Offensive Antisubmarine Operations
in World War II pp 3–6. Washington: Smithsonian Institution Press, 1995
[21] ROTTERDAM, 17 March 1943
[22] PRICE 4, p 168
[23] Niestlé [15], p 443
[24] This was the case for SCR-517A and C-SCR-517B had a PPI indicator
[25] Schoenfeld [20], pp 37–53. In letter to the author dated 15 April 1995 Niestlé
asserts this one sinking was not justified
[26] Russell W Burns, Impact of Technology on the Defeat of the U-boat, September
1939–May 1943 IEE Proceedings: Science, Measurement and Technology Vol 141,
pp 343–355, 1994
[27] LOVELL, p 161
[28] As late as April 1944 Brandt opened a meeting of the Rotterdam Committee
with an admonition not to discuss matters with anyone not officially involved,
clear evidence of a demand to tighten security. ROTTERDAM, 5 April 1944,
p1
[29] Based on a search of applicable records by Axel Niestlé reported in letters to
the author dated 12 March and 16 April 1995
[30] REUTER, p 161
[31] Schoenfeld [20], pp 80–83
[32] Brian McCue U-Boats in the Bay of Biscay: an Essay in Operations Analysis p 65.
Washington: National Defense University Press, 1990
[33] REUTER, p 161
[34] PRICE 4, pp 165–171; REUTER, pp 162–166
[35] PRICE 4, pp 165–171
[36] ROTTERDAM, 26 April 1944, pp 12–22
[37] RUNGE, pp 50–51
[38] ROSKILL 3, pp 365–366
[40] Richard Natkiel, Maps Battle of the Atlantic see [15], p 23
[41] Slessor [18], p 518
[42] TERRAINE 2, pp 767–768
[43] R W Burns, The Background to the Development of the Cavity Magnetron,
BURNS, pp 259–283
[44] The Allies lost 2 353 ships of which only 19 were from convoys with air cover.
Macintyre, 1976 [4], p 173
[45] Air Commodore Henry A Probert, Head of the RAF Air Historical Branch, in a
letter to the author dated 2 February 1995; Axel Niestlé in a letter to the author
510
Notes and Sources
dated 15 April 1995

[46] For a summary of such matters see Burns, [43]
[47] I base this statement on the extensive descriptions of the use of both by Don-
ald Macintyre, an outstanding escort commander. See Macintyre [4]. For an
excellent account of the HF/DF technique stripped of its secrecy, see P G
Redgment, High-Frequency Direction Finding in the Royal Navy, KINGSLEY
2, pp 229–266. For the German perspective, see J Rohwer, Die Funkführung
der deutschen U-Boote im zweiten Weltkrieg Wehrtechnik pp 324–328, 360–
364, 1969. For an extensive review of American HF/DF see Kathleen Broome
Williams Secret Weapon: US High-Frequency Direction Finding in the Battle of
the Atlantic Washington: Naval Institute Press, 1996. For a very complete de-
scription of all aspects of U-boat wireless communication and British surface
vessel HF/DF see Arthur O Bauer Funkpeilung als aliierte Waffe gegen deutsche
U-Boote 1939–1945 Rheinberg (D-47486), Germany: Herausgeber & Vertrieb
(Postfach 301 217), 1997
[48] Francis Harry Hinsley British Intelligence in the Second World War: its Influence on
Strategy and Operations Vol 2, p 177. New York: Cambridge University Press,
1981
[49] P G Redgment [47], p 255
[50] Axel Niestlé in a letter to the author dated 15 April 1995
[51] Y’Blood op. cit., pp 273–274
[52] Ibid. op. cit.
[53] MORISON 10, pp 171–177
[54] TRENKLE 1, pp 125–126
[55] Niestlé [15], pp 441–442
[56] TRENKLE 1, pp 139–142; REUTER, pp 167–168; Niestlé, [15], p 441
[57] GUERLAC 1, p 726
[58] TERRAINE 2, pp 626–627
[59] Niestlé [15], pp 445–446
[60] Y’Blood op. cit.
Chapter 7.2 Radar in arctic waters

[1] PLOCHER 2, pp 35–49
[2] Ibid., p 38
[3] ROSKILL 2, pp 138–144
[4] Ibid., pp 279–285
[5] Ibid., p 291
[6] HOWSE, pp 158–159; ROSKILL 2, loc. cit., reports that only two escort vessels
had radar, of which the minesweeper was one. Given Howse’s sources, this is
clearly in error. Dudley Pope 73 North: the Battle of the Barents Sea Annapolis:
Naval Institute Press, 1989, gives detail of radar use by all British destroyers
[7] HOWSE, pp 158–159
[8] For details of the engagement, skillfully told, see Pope [6]
[9] Pope [6], p 185
[10] Walter Bartig, Geschichte einer Funkkompanie und ihrer Männer, p 38.
Manuscript, Berlin, September 1987
[11] PLOCHER 3, p 209
[12] G Muller and R Bosse, German Primary Radar for Airborne and Ground-Based
Surveillance, BURNS, pp 200–208
511
[13] BEKKER 2, p 240

[14] ROSKILL 3, pp 78–88
[15] The disposition with time of installation of radar in the Royal Navy is given in
appendix 3 of KINGSLEY 1, pp 387–400
[16] There is a report (HOWSE, pp 187–188) of a naval adaptation of the 55 cm
air-search set Hohentwiel on the mainmast, but none of the reports suggest
that it made any contribution to the battle; it may have been knocked out by
the same shot that disabled the forward Seetakt
[17] For details of all these attempts and those that follow see Ludovic Kennedy
The Death of the Tirpitz Boston: Little, Brown and Company, 1979
[18] PRICE 1, pp 244–246
Chapter 7.3 The Mediterranean, 1943–1945

[1] CRAVEN & CATE 2, p 3
[2] SIG CORPS 2, pp 374–375
[4] War Diary, 62nd Coast Artillery, 12 November 1942 to 26 April 1946. Copy
furnished by Colonel John M Godfrey, Adjutant
[5] John Manning, commander of Battery C, 62nd CA (AA), in a letter to the author
dated 23 March 1994
[6] War Diary, 62nd Coast Artillery, 30 September 1943
[10] SIG CORPS 2, pp 257–260, 377–378
[11] GUERLAC 1, pp 699–706
[12] HOFFMANN 1, pp 212–221
[13] TERRAINE 1, pp 390–393
[14] E Giboin, L’Evolution de la Detection Electromagnetique dans la Marine Na-
tionale Onde Vol 31, pp 53–64, 1951; Maurice Ponte, Sur des Apports Français a
la Technique de la Détection Électromagnétique Revue technique Thomson-CSF
Vol 1, pp 171–180, 1946
[15] PRICE 2, pp 71–79
[17] Ibid., p 325
[18] WATT, pp 146–147, 332–334
[19] JAY, part 2, para. 30
[20] HOFFMANN 1, p 224; PRICE 2, p 301
[21] CRAVEN & CATE 2, pp 520–545
[23] SIG CORPS 3, p 303
[24] CRAVEN & CATE 3, pp 346–352
[26] RADAR, No 5, 30 September 1944, pp 10–12
Chapter 7.4 Japanese shipping destroyed

[1] FRIEDMAN, pp 146–147
512
Notes and Sources
[2] BELL, pp 69, 75–81; RADAR SURVEY 4, pp 135–136

[3] Blair, 1975 op. cit., p 113
[4] Ibid., pp 321–322
[5] Roscoe, 1949 op. cit., pp 170–172
[6] Blair, 1975 op. cit., p 530
[7] Vice Admiral Charles A Lockwood, Electronics in Submarine Warfare Proc. IRE
Vol 35, pp 712–715, 1947
[8] Mochitsura Hashimoto (translated by Commander E H M Colegrave) Sunk: the
Story of the Japanese Submarine Fleet, 1941–1945 pp 200–205. New York: Henry
Holt and Company, 1954
[9] See Chapter 4.2 for details
[10] Richard C Knott Black Cat Raiders of WWII Annapolis: The Nautical and Avia-
tion Publishing Company, 1981
[11] GUERLAC 1, pp 997–1001
[12] BELL, pp 71, 100–102; RADAR SURVEY 1, pp 117–120
[13] Low Altitude, High Precision, RADAR, No 1, April 1944, pp 19–22
[14] LAB vs. Jap Shipping, RADAR, No 5, November 1944, pp 3- 9
[15] Ibid.
[16] GUERLAC 1, pp 1006–1013
Chapter 7.5 The wide Pacific

[1] FRIEDMAN, p 149; RADAR SURVEY 2, pp 137–138
[2] FRIEDMAN, p 150; RADAR SURVEY 2, pp 143–144; GUERLAC 1, pp 439–442
[3] FRIEDMAN, pp 151–152; RADAR SURVEY 2, pp 163–164; GUERLAC 1, pp
442–443
[4] MONSARRAT, p 46
[5] Ibid., pp 53–57
[6] MORISON 6, p 108
[7] MONSARRAT, pp 61–62
[8] PAGE, p 148
[9] MCNALLY
[10] A E Fanning, The Action Information Organization, KINGSLEY 2, p 168
[11] FRIEDMAN, pp 147–148; RADAR SURVEY 2, pp 131–132
[12] FRIEDMAN, p 173; RADAR SURVEY 2, pp 39–44
[14] H T Friis and W D Lewis, Radar Antennas Bell Tech. J. Vol 26, pp 219–317, 1947
[15] TRENKLE 1, pp 147–151
[16] Lloyd V Berkner, Naval Airborne Radar Proc. IRE Vol 34, pp 671–706, 1946
[17] BELOTE, pp 204–205. There is evidence that O’Hare was shot down by one
of the Japanese aircraft; John Lundstrom and Steve Ewing Fateful Rendezvous
Annapolis: Naval Institute Press, 1997
[18] Berkner [16], pp 692–695
[19] NAKAGAWA, p 49
[20] Ibid., pp 57–58
[21] NAKAGAWA, p 66
[22] Japanese Monograph No 118, Operational History of Naval Communications,
December 1941—August 1945, pp 110–111. Washington: Office of the Chief of
Military History, Department of the Army, 26 May 1953
513
[23] NAKAJIMA, p 255

[24] NAKAGAWA, p 36
[25] BELOTE, p 208; PRICE 2, p 293; NAKAGAWA, p 91
[27] PRICE 2, p 143
[29] Charles A Lockwood and Hans Christian Adamson Battles of the Philippine Sea
pp 14, 57. New York: Thomas Y Crowell Company, 1967
[30] MORISON 6, p 394
[31] Lockwood and Adamson [29], pp 19–23
[32] MONSARRAT, p 77
[33] Brian Garfield The Thousand-Mile War: World War II in Alaska and the Aleutians
pp 24–32. Garden City, New York: Doubleday and Company, 1969
[34] PRICE 2, pp 53–55
[35] Garfield [33], pp 717–179
[36] Guadalcanal preceded Attu but the landing was unopposed; the tough part
came later
[37] NAKAGAWA, p 46
[38] Action Report of USS Mississippi 26 July 1943, Office of Naval Records and
Library
[39] Garfield [33], pp 271–282; PAGE, p 151; MORISON 7, p 59–61, erroneously
includes the Idaho and the Monaghan in this action
[40] Mississippi Action Report, loc. cit.
[41] MORISON 6, p 104
[42] John Miller, Jr United States Army in World War II The War in the Pacific. Cartwheel:
the Reduction of Rabaul p 94. Washington: Office of the Chief of Military History,
1959
[43] GUERLAC 1, pp 994, 1020
[44] MORISON 8, pp 233, 321
[45] William A Klingaman APL—Fifty Years of Service to the Nation: a History of the
Johns Hopkins University Applied Physics Laboratory pp 14–15. Laurel, Maryland:
The Johns Hopkins University Press, 1993
[46] The details of the Battle of the Philippine Sea, including its radar component
are well told in the general references given
[47] John Hightower, Most Secret Weapon, Radar The Evening Star (Washington),
21 June 1943, p A5; 22 June, p A6; 23 June, p A6. Radar Stories Are Released
by US and Great Britain Electronics June, pp 274–282, 1943
[48] US Naval Historical Center, Washington Navy Yard, photograph NH72278-KN
Chapter 8.1 Invasion

[1] Mark A Stoler D-Day 1944 Theodore A Wilson, editor, pp 298–317. Lawrence,
Kansas: University of Kansas Press, 1994
[2] JONES 1, pp 400–401
[3] Ibid., pp 116–118
[5] Friedrich Ruge D-Day [1], p 127
[6] PRICE 2, p 128
[7] STREETLY, pp 53–54; RAF SIGNALS 7, pp 232–233
514
Notes and Sources
[8] Ibid., loc. cit.

[9] F A Kingsley, Electronic Countermeasures in the Royal Navy, KINGSLEY 2, pp
214–221
[10] Professor Sir Martin Ryle, D-13: Some Personal Memories of 24–28th May 1944
Proc. IEE Vol 132A, pp 438–440, 1985
[12] Ron Colledge, LATHAM & STOBBS, pp 137–139
[13] PRICE 2, pp 123–127
[14] Kingsley [9], p 219
[15] PRICE 2, pp 128–130
[16] HOFFMANN, p 278
[17] JONES 1, pp 410–411
[18] HOWSE, pp 212–213
[19] GUERLAC 1, pp 837–838
[20] HOWSE, pp 213–217
[21] WATT, pp 331–334; RADAR SURVEY 4, pp 5–8, 99–101
[22] TERRAINE 1, pp 630–631
[23] It is possible that advanced forms of the light-weight air-warning equipment,
AN/TPS-1 and TPS-3, were also on the beaches. Portable Radar Gets Better,
RADAR, No 3, June, pp 22–23, 1944
[24] KEMP, pp 54–57
[25] HOWSE, pp 217–219
[27] Sixty-six Tons of MEW, RADAR, No 3, June, pp 6–7, 1944
[28] HOFFMANN 1, pp 278–285. The British record of this minor affair is noted in
the situation map of 10 June. Major L F Ellis Victory in the West Vol 1 The Battle
of Normandy map opp. p 248. London: Her Majesty’s Stationery Office, 1962
[29] GUERLAC 1, pp 838, 842–844
[30] Gerhard L Weinberg D-Day [1], pp 324–326
Chapter 8.2 Flying bombs

[1] Lieutenant Colonel F M Rickard, Ordnance Encyclopaedia Britannica (12th edi-
tion), Vol 31, pp 1202–1203, 1922
[2] Michael J Neufeld The Rocket and the Reich: Peenemünde and the Coming of the
Ballistic Missile Era p 50. New York: The Free Press, 1995
[3] JONES 1, pp 359–360
[4] Ibid., pp 414–415
[5] Neufeld [2], pp 197–200
[6] PILE, pp 329–330
[7] Ibid., p 303
[8] Air Chief Marshal Sir Roderic Hill, Air Operations by Air Defence of Great Britain
and Fighter Command in Connection with the German Flying Bomb and
Rocket Offensives, 1944–1945 The London Gazette (supplement), 19 October
1948, pp 5585–5617
[9] Ibid., pp 5596–5598
[10] PILE, pp 339–340
[11] Henry B Abajian and Lee Davenport, IEEE, pp 1–19, 72–73
[12] BALDWIN, pp 260–262
515
[13] Hill [8], p 5599

[14] BUDERI, pp 226–228
[15] GUERLAC 1, pp 448–459, 857–859
[16] Sir Edward Fennessy in a letter to the author dated 18 June 1996; M S Dean, The
UK’s First Ballistic Missile Early Warning System, unpublished manuscript,
August 1996
[17] TRENKLE 1, pp 101–102
[18] Neufeld [2], p 247
[19] SIG CORPS 3, pp 319–321; Pat Hawker, Aspi 5, Task Z & Operation ‘Silent
Minute’: the Secret Story of the Vain Attempt to Jam the V2 Radio Bygones
August/September, pp 16–19, 1944, October/November, pp 9–13, 1944
[20] Neufeld [2], p 273
Chapter 8.3 The battlefield transformed

[1] The Mobile MEW, RADAR, No 5, September, pp 31–33, 1944
[2] Radar to Fighter to Target, RADAR, No 7, January 1945, pp 36–37
[3] GETTING, pp 136–137
[4] Radar at the Front, RADAR, No 8, February 1945, pp 9–15
[5] GUERLAC 1, pp 906–907
[6] KEMP, pp 109–121
[8] KEMP, p 108
[9] GUERLAC 1, pp 891–892
[10] KEMP, p 118
[11] GUERLAC 1, p 488
[12] GETTING, p 138
[13] Countermeasure for the Mortar Menace, RADAR, No 10, June, pp 3–8, 1945
[14] GUERLAC 1, pp 883–885
[15] Ibid., pp 885–886
[16] DAVIS, pp 453–482
[17] BALDWIN, p 279
Chapter 8.4 Post Mortem

[1] HOFFMANN 1, pp 351–352
[2] Exercise ‘Post Mortem’: Report on an Investigation of a Portion of the German
Raid Reporting and Control System, pp 3–5. Air Ministry Report, 1945
[3] Generalmajor (a.D.) Alfred Boner, Ein Kapitel Nachkreigsgeschichte, p 10. Un-
published manuscript dated 23 September 1985 provided the author by Harry
von Kroge
[4] STREETLY, p 122
[5] Ibid., pp 15–20
[6] Ibid., pp 7–10
[7] Ibid., pp 43, appendix pp 1–2
[8] Post Mortem [2], appendix p 19
[9] Ibid., p 5; STREETLY, p 122
[10] PETERSEN, p 52
[11] STEGSKOPFER
516
Notes and Sources
[12] PRICE 1, p 242

[13] Post Mortem [2], p 43
[14] Boner [3], p 15
[15] Gunnar Krogsøfe as told to Niels Chr Bahnson and related in a letter to the
author dated 12 August 1996
[16] Dipl-Phys. Hans Ulrich Widdel in a letter to the author dated 4 September
1994. Mr Widdel has been active for many years at the Max-Planck-Institut
für Aeronomie, where his professional activities caused him to know and
work with many of the top German radar engineers
[17] BEKKER 1, chap 1. The third edition of this book, BEKKER 2, omits the incident
[18] ROTTERDAM
[19] PRITCHARD, p 219
[20] Dr-Ing Gotthard Müller, Funkmessgeräte-Entwicklung bei der C Lorenz
AG, 1935–1945, p 37. Stuttgart: Standard Elektrik Lorenz AG (Technisch-
[21] RUNGE, p 87
[22] KROGE, pp 191–192
[23] Fritz Trenkle, Zum 90. Geburtstag von Hans Plendl Funkgeschichte No 78, pp
3–5, 1991
[24] Kurt Mauel, Leo Brandt Männer der Funktechnik pp 25–27, Sigfrid von Weiher,
editor. Berlin: VDE-Verlag, GmbH, 1983
Chapter 9.1 The Philippines and Okinawa

[1] Captain Rikihei Inoguchi and Commander Tadashi Nakajima The Divine Wind
pp 90–99. Annapolis: US Naval Institute, 1958
[2] Inoguchi and Nakajima [1], p 64
[3] FRIEDMAN, p 150
[4] MORISON 12, p 206
[5] Ibid., p 191
[6] Ibid., p 224
[7] The guns of Yamato fired projectiles about 50 times heavier than those of the
destroyers
[8] GUERLAC 1, pp 1020–1021
[10] Unit histories of 583rd and 597th Signal Air Warning Battalions, US National
Archives
[11] GUERLAC 1, p 1022
[12] Initially five planes flew together: three kamikazes with two escorting fighters.
The fighters had experienced pilots who were to report the effectiveness and
recommend alterations in tactics. Inoguchi and Nakajima [1], p 62
[14] GETTING, pp 173–185
[15] MONSARRAT, pp 102–107
[16] Ibid., pp 112–113, 133
[17] Ibid., pp 118–119
[18] Commander R H Maynard, USN, Radar and Weather Journal of Meteorology Vol
2, pp 214–225, 1945
[19] The Li’l Abners, RADAR, No 11, September 1945, pp 22–25
517
[20] BELOTE, p 39
[21] Colonel Hiromichi Yahara The Battle for Okinawa p xiii. New York: John Wiley
and Sons, 1995
[22] BELOTE, p 267
[23] MORISON 14, pp 234–235
[24] Roy E Appleman, James M Burns, Russell A Gugeler and John Stevens United
States Army in World War II Okinawa: the Last Battle p 102. Washington: Office
of the Chief of Military History, 1948
[25] BELOTE, p 307
[26] Ibid., p 267
Chapter 9.2 The destruction of Japanese cities

[1] Barbara W Tuchman Stilwell and the American Experience in China, 1911–45 pp
349–509. New York: The Macmillan Company, 1970; Eric Larrabee Commander
in Chief: Franklin Roosevelt, his Lieutenants, and Their War pp 509–578, 606–614.
New York: Harper and Row, 1987.
[2] Report of Division 15 Vol 1 Radio Countermeasures pp 311–313. Washington: Na-
tional Defense Research Committee, 1945. For a table of Japanese radars see
NAKAGAWA, pp 83–91
[3] PRICE 2, pp 151–154, 292.
[4] CRAVEN & CATE 5, pp 94–102.
[5] PRICE 2, p 160.
[6] Ibid., p 226.
[7] This is based on the interrogation of a Japanese radar officer specializing in IFF at
the Tama Laboratories, who produced a document outlining their investiga-
tion of a captured American IFF Air Technical Intelligence Group, Advanced
Echelon FEAF, Report No 275, 14 December 1945
[8] PRICE 2, pp 227–231
[9] Gee and Loran, RADAR, No 9, April 1945, pp 14–23
[10] CRAVEN & CATE 5, pp 656–657
[11] GUERLAC 1, pp 1050–1053
[12] NAKAGAWA, pp 50, 67–68; SCIENTIFIC, Vol 1, p.6
[13] PRICE 2, pp 232–233
[14] For details of the destruction of these two cities see Richard Rhodes The Making
of the Atomic Bomb New York: Simon and Schuster, 1986
[15] Edward S Miller War Plan Orange: the US Strategy to Defeat Japan, 1897–1945
Annapolis: Naval Institute Press, 1991
Chapter 10.1 Navigation transformed

[1] A E Fanning, The Action Information Organization, KINGSLEY 2, pp 170–171
[2] The author was a passenger on an American-flag freighter in 1958 that had
neither radar nor loran
[3] F J Wylie The Use of Radar at Sea pp 175–181. London: Hollis and Carter, 1952
[4] Philip Van Horn Weems Air Navigation pp 153–158. Annapolis: Weems System
of Navigation, 1955
[5] R A Sheppard and M C Stevens, The Development of IFF and SSR in the Post
War Years, BURNS, pp 458–461
518
Notes and Sources
[6] Jack Gough Watching the Skies: a History of Ground Radar pp K1-K7. London: Her
Majesty’s Stationery Office, 1993
[7] Richard M Trim, Secondary Surveillance Radar—Past, Present and Future,
BLUMTRITT, pp 93–120
[8] Hall op. cit., pp 252–255; GUERLAC 1, pp 615–617
[9] Hall op. cit., pp 131–142
[10] TRENKLE 1, pp 136–142
[11] GUERLAC 1, pp 525–529
[12] Pierce et al op. cit., pp 1–34
[13] Ibid., p 97
[14] Ibid., p ix
[15] Elbert S Maloney Dutton’s Navigation and Piloting pp 701–732. Annapolis: Naval
Institute Press, 1978
[16] Hall op. cit., pp 76–77
[17] GUERLAC 1, pp 1082–1083
[18] Carl I Aslakson, Velocity of Electromagnetic Waves Nature Vol 164, pp 711–712,
1949
[19] A A Michelson, F G Pease and F Pearson, Measurement of the Velocity of Light
in a Partial Vacuum Ap. J. Vol 82, pp 26–61, 1935
[20] Wilmer C Anderson Rev. Sci. Inst. Vol 8, pp 239–247, 1937
[21] L Essen, Velocity of Electromagnetic Waves Nature Vol 159, pp 611–612, 1947
Chapter 10.2 Science and the electronic age

[1] Karl G Jansky, Electrical Disturbances Apparently of Extraterrestrial Origin Proc.
IRE Vol 21, pp 1387–1398, 1937
[2] Grote Reber, Cosmic Static Ap. J. Vol 100, pp 279–287, 1944
[3] J S Hey op. cit., p 20
[4] STEPP, pp 75–76
[5] Sir Edward Appleton and J S Hey, Solar Radio Noise Phil. Mag. Vol 37, p 73, 1946
[6] General Pile, chief of AA Command, writes that the Germans never attempted
to jam GL mark II PILE, pp 288, 330
[7] Hey op. cit., p 15
[8] Hey op. cit., pp 19–23. J S Hey and G S Stewart, Radar Observations of Meteors
Proc. Phy. Soc. Vol 59, pp 858–883, 1946
[9] Sir Bernard Lovell, Impact of World War II on Radio Astronomy Serendipitous
Discoveries in Radio Astronomy K Kellermann and B Sheets, editors. Green Bank,
West Virginia: National Radio Astronomy Observatory, 1983
[10] P M S Blackett and A C B Lovell, Radio Echoes and Cosmic Ray Showers Proc.
Roy. Soc. Vol 177, pp 183–186, 1941
[11] Hey op. cit., p 124
[12] BOWEN, pp 196–199, 205–207
[13] J L Pawsey, R Payne-Scott and L L McCready, Radio-frequency Energy from
the Sun Nature Vol 157, pp 158–159, 1946
[14] M Ryle and D D Vonberg, Solar Radiation on 175 Mc./s. Nature Vol 158, pp
339–340, 1946
[15] John H DeWitt and E K Stodola, Detection of Radio Signals Reflected from the
Moon Proc. IRE Vol 37, pp 229–242, 1949
[16] Zoltán Bay, Reflections of Microwaves from the Moon Hungarica Acta Physica
Vol 1, pp 1–22, 1947 (Note the 2.4 m wavelength described as microwaves.)
519
[17] Andrew J Butrica To See the Unseen: a History of Planetary Radar Astronomy
Washington: National Aeronautics and Space Administration, 1996
[18] TRENKLE 4, pp 112–119
[20] STEPP, p 95. Hans Ulrich Widdel has pointed out the curious manner in which
this sentence is incorporated into the typescript. It occupies the bottom of
the page in space that is left blank on all other pages. Taking into account the
circumstances of a military occupation, one cannot exclude the possibility that
it was added after the thesis had been accepted by a nervous faculty committee
[21] This lecture was a tutorial on radar sensitivity in which Stepp discussed the
Moon observations as an example. W Stepp, Die Reichweite von Funkmess-
geräten Hochfrequenztechnik und Weltraumfahrt Dr-Ing R Merten, editor, pp
36–43. Stuttgart: S Hirzel Verlag, 1951
[22] Dr Hans J Albrecht in a letter to the author, dated 8 June 1995
[23] Mr Widdel has made a study of the propagation of various German radars to
see what was possible and concluded that the Würzmann could have seen the
Moon as enhanced noise with real-time signals. He also concluded that a very
alert operator might have seen it with Wassermann or Mammut, but there are
no reports of such
[24] Wilhelm Stepp, Ueber die erste Erfassung des Monds mit einem Funkmess-
gerät (Radar-Gerät) in Deutschland Der Seewart Vol 35, No 2, p 71, 1974. This
paper was the basis of another, more often cited paper. Hans Mogk, Die Mon-
dentfernung 1943 funktechnisch vermessen Funkgeschichte No 87, pp 323–324,
1992
[25] G C Southworth, Microwave Radiation from the Sun J. Frank. Inst. Vol 239, pp
285–297, 1945
[26] Hey op. cit., pp 23–25. For details of the work by Purcell see BUDERI, pp
291–307
[27] H I Ewen and E M Purcell, Observation of a Line in the Galactic Radio Spectrum
Nature Vol 168, p 356, 1951. C A Muller and J H Oort, The Interstellar Hydrogen
Line at 1,420 Mc/sec., and an Estimate of Galactic Rotation Nature Vol 168, pp
357–358, 1951
[28] Robert H Dicke, The Measurement of Thermal Radiation at Microwave Fre-
quencies Rev. Sci. Inst. Vol 17, pp 268–275, 1946
[29] Antenna temperature is the standard way radio astronomers report the inten-
sity of received radiation. If one has a perfect receiver and connects a resistor
across the input, the output will result from the amplification of the random
motion of the free electrons in that resistor. The learned have shown that black-
body radiation of temperature T incident on an antenna will generate the same
noise signal as a resistor input of temperature T . Black-body radiation is rare
in radio astronomy, but the Dicke comparison method makes temperature the
natural unit for a given frequency band
[30] R E Burgess, Noise in Receiving Aerial Systems Proc. Phys. Soc. Vol 53, pp
293–304, 1941
[31] Robert H Dicke and Robert Beringer, Microwave Radiation from the Sun and
Moon Ap. J. Vol 103, pp 375–376, 1946
[32] For references to this early work see Forman op. cit., pp 416- 419, and J C Slater,
The Design of Linear Accelerators Rev. Mod. Phys. Vol 20, pp 473–518, 1948
[33] Luis W Alvarez, The Design of a Proton Linear Accelerator Phys. Rev. Vol 70,
520
Notes and Sources
pp 799–800, 1946; ALVAREZ, pp 153–160

[34] Forman op. cit., p 417. Luis W Alvarez et al , Berkeley Proton Linear Accelerator
Rev. Sci. Inst. Vol 26, pp 111–133, 1955
[35] TM 11–1106. Technical Manual, Radio Set SCR-268 pp 101–102. Washington: War
Department, 26 August 1942
[36] For references to these first experiments see D W Fry and W Walkinshaw,
Linear Accelerators Reports on Progress in Physics Vol 12, pp 102–132, 1948–49
[37] E G Bowen, O O Pulley and J S Gooden, Application of Pulse Technique to
the Acceleration of Elementary Particles Nature Vol 157, p 840, 1946. W D
Allen and J L Symonds, Experiments in Multiple-Gap Linear Acceleration of
Electrons Proc. Phys. Soc. Vol 59, pp 622–628, 1947
[38] D W Fry, R B R-S Harvie, L B Mullett and W Walkinshaw, Travelling-Wave
Linear Accelerator for Electrons Nature Vol 160, pp 351–353; A Travelling-
Wave Linear Accelerator for 4 MeV Electrons Nature Vol 162, pp 859–861, 1948
[39] E L Ginzton, W W Hansen and W R Kennedy, A Linear Electron Accelerator
Rev. Sci. Inst. Vol 19, pp 89–108, 1948
[40] Felix Bloch, William Webster Hansen Biographical Memoirs Vol 27, pp 121–137.
Washington: National Academy of Sciences, 1952
[41] J E Nafe, E B Nelson and I I Rabi, The Hyperfine Structure of Atomic Hydrogen
and Deuterium Phys. Rev. Vol 71, pp 914–915, 1947. J E Nafe and E B Nelson,
The Hyperfine Structure of Hydrogen and Deuterium Phys. Rev. Vol 73, pp
718–728, 1948
[42] Willis E Lamb, Jr and Robert C Retherford, Fine Structure of the Hydrogen
Atom by a Microwave Method Phys. Rev. Vol 72, pp 241–243, 1947
[43] N Bloembergen, E M Purcell and R V Pound, Relaxation Effects in Nuclear
Magnetic Resonance Absorption Phys. Rev. Vol 73, pp 679–712, 1948
[44] Felix Bloch, W W Hansen and Martin Packard Nuclear Induction Phys. Rev. Vol
69, p 127, 1946; Felix Bloch, W W Hansen and Martin Packard The Nuclear
Induction Experiment Phys. Rev. Vol 70, pp 474–485, 1946
[45] GUERLAC 1, pp 515–522
[46] Robert H Dicke, Robert Beringer, Robert L Kyhl and A B Vane, Atmospheric
Absorption Measurements with a Microwave Radiometer Phys. Rev. Vol 70,
pp 340–348, 1946
[47] John H Van Vleck and Victor F Weisskopf, On the Shape of Collision-broadened
Lines Rev. Mod. Phys. Vol 17, pp 227–236, 1945
[48] B Bleaney and R P Penrose, Ammonia Spectrum in the 1 cm Wave-length
Region Nature Vol 157, pp 339–340, 1946
[49] J P Gordon, H J Zeiger and C H Townes, Molecular Microwave Oscillator and
New Hyperfine Structure in the Microwave Spectrum of Ammonia Phys. Rev.
Vol 95, pp 282–284, 1954
[50] For a detailed report of this early work see Forman [32], pp 407–410
[51] Commander R H Maynard, USN, Radar and Weather Journal of Meteorology Vol
2, pp 214–225, 1945
[52] Arthur E Bent, Radar Detection of Precipitation Journal of Meteorology Vol 3, pp
78–84, 1946. GUERLAC 1, pp 641–642
[53] Horace E Byers and Richard D Coons, The ‘Bright Line’ in Radar Cloud Echoes
and its Probable Explanation Journal of Meteorology Vol 4, pp 75–81, 1947
[54] J S Marshall, R C Langille and W McK Palmer, Measurement of Rainfall by
Radar Journal of Meteorology Vol 4, pp 186–192, 1947
521
[55] E G Bowen, Radar Observations of Rain and Their Relation to Mechanisms of

Rain Formation Journal of Atmospheric and Terrestrial Physics Vol 1, pp 125–140,
1950
[56] SIG CORPS 3, pp 465–466
[57] HOLLMANN
[58] G C Southworth and A P King, Metal Horns as Directive Receivers of Ultra-
short Waves Proc. IRE Vol 27, pp 95–102, 1939
[59] CALLICK, p 92
[60] Henry C Torrey and Charles A Whitman Crystal Rectifiers. Radiation Laboratory
Series Vol 15, pp 5–11. New York: McGraw-Hill Book Company, 1948
[61] Frederick Seitz, Research on Silicon and Germanium in World War II Phys.
Today January, pp 22–27, 1995, The Tangled Preclude to the Age of Silicon
Electronics Proceedings of the American Philosophical Society Vol 140, pp 289–
337, 1996, and with Norman G Einspruch Electronic Genie: the Tangled History
of Silicon Urbana, Illinois: University of Illinois Press, 1998. For details of diode
development during the war see J H Scaff and R S Ohl, Development of Silicon
Crystal Rectifiers for Microwave Radar Receivers Bell Tech. J. Vol 26, pp 1–30,
1947
Chapter 10.3 Secrecy and the Technical Imperative

[1] Fuehrer Directives and Other Top-level Directives of the German Armed Forces, 1939–
1941 p 81. Washington: Office of the Judge Advocate General, ca. 1946
[2] Letter from Hans Ulrich Widdel to the author dated 11 August 1994
[3] Dr-Ing Gotthard Müller, Funkmessgeräte-Entwicklung bei der C Lorenz AG,
1935–1945, pp 35–36. Stuttgart: Standard Elektrik Lorenz AG (Technisch-
[4] KROGE, pp 174–175
[5] SIG CORPS 2, pp 268–274
[9] Annalee Saxenian, Lessons from Silicon Valley Technology Review July 1994, pp
42–51
[10] Akio Morita Made in Japan: Akio Morita and Sony pp 211–213. New York: E P
Dutton, 1986
[11] KEMP, p 93
[13] Anne Stobbs, LATHAM & STOBBS, p 198
[15] HOWSE, p 54
[16] Technical Data on Japanese Radio and Radar Equipment, Naval Research Lab-
oratory, Washington, July 1944
[17] GUERLAC 1, p 249
[18] MCKINNEY, p 130
[19] Ibid., p 156
[20] TM 11–1106. Technical Manual, Radio Set SCR-268 p 103. Washington: War
Department, 26 August 1942
[21] PAGE, p 129
522
Notes and Sources
[22] BELL, p 51
[23] KERN, pp 126–127
[24] Weyers Taschenbuch der Kriegsflotten 1939 p 220. München/Berlin: J F Lehmanns
Verlag, 1939
[25] R V Jones in a letter to the author dated 26 October 1995
[26] JONES 1, p 93
[27] Ibid., p 224
[28] STREETLY, p 46
[29] Ibid., pp 26–31
[30] Richard Rhodes Dark Sun: the Making of the Hydrogen Bomb pp 83–120. New
York: Simon and Schuster, 1995
[31] Ronald Radosh and Joyce Milton The Rosenberg File: a Search for the Truth p 72.
New York: Holt, Rinehart and Winston, 1983
[32] PRICE 3, pp 338–340
[33] To contrast this with the strangled attempts at exchange when attempted dur-
ing the preceding few months see ZIMMERMAN, pp 25–48
[34] Britton Chance, IEEE, p 57
Chapter 10.4 An evaluation

[1] TUSKA 2; KERN, pp 44–63
[2] Frederick Emmons Terman Radio Engineering pp 663–671. New York: McGraw-
Hill Book Co., 1937
[3] KERN, p 160
[4] David P Billington The Tower and the Bridge: the New Art of Structural Engineering
pp 90–91. New York: Basic Books, Inc., 1983
[5] BROWN, p 13
[6] RABI, p 164
[7] TAYLOR, p 247
523
BIBLIOGRAPHY
General references
Hugh G Aitken Syntony and Spark: the Origins of Radio Princeton, New Jersey:
Princeton University Press, 1979
Hugh G Aitken The Continuous Wave: Technology and American Radio, 1900–1932
Princeton, New Jersey: Princeton University Press, 1985
Park Benjamin A History of Electricity (The Intellectual Rise in Electricity) From Antiq-
uity to the Days of Benjamin Franklin New York: John Wiley and Sons, 1898
Clay Blair Hitler’s U-Boat War: the Hunters 1939–1942 New York: Random House,
1996
Clay Blair Jr Silent Victory: the US Submarine War Against Japan Philadelphia: J B
Lippincott Company, 1975
John H Bryant, The First Century of Microwaves—1886 to 1986 IEEE Transactions
on Microwave Theory and Techniques Vol 36, pp 830–58, 1988
Russell W Burns Television: an International History of the Formative Years London:
The Institution of Electrical Engineers, 1998
Ronald William Clark Tizard Cambridge, Massachusetts: The MIT Press, 1965
Roger B Colton, Radar in the United States Army Proc. IRE Vol 33, pp 740–753, 1945
Giulio Douhet The Command of the Air New York: Coward-McCann, Inc., 1942
C W F Everitt James Clerk Maxwell: Physical and Natural Philosopher New York:
Charles Scribner’s Sons, 1975
John A Fleming, Electricity Encyclopaedia Britannica (11th edition) Vol 9, pp 179–193,
1911
John A Fleming, Telegraph. Part II—Wireless Telegraphy Encyclopaedia Britannica
(11th edition) Vol 26, pp 529–541, 1911
Paul Forman, ‘Swords into ploughshares’: breaking new ground with radar hard-
ware and technique in physical research after World War II Rev. Mod. Phys.
Vol 67, pp 397–455, 1995
Richard B Frank Guadalcanal: the Definitive Account of the Landmark Battle New York:
Random House, Inc., 1990
Noble Franklin The Bombing Offensive Against Germany: Outlines and Perspectives
London: Faber and Faber, 1965
Thomas H Greer The Development of Air Doctrine in the Army Air Corps, 1917–1941
Washington: Office of Air Force History, 1955
John S Hall, editor Radar Aids to Navigation: Radiation Laboratory Series Vol 2. New
York: McGraw-Hill Book Company, 1947
Max Hastings Bomber Command New York: Dial Press, 1979
524
Bibliography
J S Hey The Evolution of Radio Astronomy New York: Science History Publications,
1973
Basil Henry Liddell Hart History of the Second World War New York: G P Putnam’s
Sons, 1970
B H Liddell Hart The Liddell Hart Memoirs Vol 1, 1895–1938; Vol 2 The Later Years
New York: G P Putnam’s Sons, 1965
John B Lundstrom The First South Pacific Campaign: Pacific Fleet Strategy December
1941–June 1942 Annapolis: Naval Institute Press, 1976
Wayne Charles Lutton Malta and the Mediterranean: a Study in Allied and Axis Strat-
egy, Planning and Intelligence During the Second World War Dissertation, South-
ern Illinois University, 1983
Donald Macintyre The Battle of the Atlantic London: B T Batsford Ltd, 1961
Donald Macintyre The Battle for the Mediterranean New York: W W Norton and Co.,
1964
George Millar The Bruneval Raid: Flashpoint of the Radar War Garden City, New York:
Doubleday and Co., 1975
William Mitchell Winged Defense: the Development and Possibilities of Modern Air
Power—Economic and Military New York: G P Putnam’s Sons, 1925
Williamson Murray Luftwaffe Baltimore: The Nautical and Aviation Publishing Co.,
1985
Paul J Nahin Oliver Heaviside: Sage in Solitude: the Life, Work, and Times of an Electrical
Genius of the Victorian Age New York: The Institute of Electrical and Electronic
Engineers, 1987
Vincent Orange A Biography of Air Chief Marshal Sir Keith Park London: Methuen &
Co., Ltd, 1984
J A Pierce, A A McKenzie and R H Woodward, editors Loran: Radiation Laboratory
Series Vol 4. New York: McGraw-Hill Book Company, 1948
Major General I S O Playfair The Mediterranean and Middle East Vols 1–3. London:
Her Majesty’s Stationery Office, 1954, 1956, 1960
Gordon W Prange with Donald M Goldstein and Katherine V Dillon Pearl Harbor:
the Verdict of History New York: McGraw-Hill Company, 1986
R A F Air Historical Branch The Rise and Fall of the German Air Force, 1933–1945 New
York: St Martin’s Press, 1983 (a reprint of the classified, 1948, Her Majesty’s
Stationery Office edition)
Theodore Roscoe On the Seas and in the Skies: a History of the US Navy’s Air Power
New York: Hawthorn Books, 1970
Theodore Roscoe United States Submarine Operations in World War II Annapolis: US
Naval Institute, 1949
Friedrich Ruge (translated by M G Saunders, RN) Der Seekrieg: the German Navy’s
Story, 1939–1945 Annapolis: US Naval Institute, 1957
Dudley Saward Bomber Harris: the Story of Marshal of the Royal Air Force Sir Arthur
Harris Garden City, New York: Doubleday and Co., 1985
Air Marshal Sir Robert Saundby Air Bombardment: the Story of Its Development New
York: Harper and Brothers, 1961
Ronald Schaffer Wings of Judgment: American Bombing in World War II New York:
Oxford University Press, 1985
George C Southworth, Survey and History of the Progress of the Microwave Arts
Proc. IRE Vol 50, pp 1199–1206, 1962
J M Spaight Air Power and the Cities New York: Longmans, Green and Co., 1930
525
Gerald F Tyne Saga of the Vacuum Tube Indianapolis: Howard W Sams, 1977
E T Whittaker A History of the Theories of Aether and Electricity from the Age of Descartes
to the Close of the Nineteenth Century New York: Longmans, Green, and Co.,
1910
Robert Wright The Man Who Won the Battle of Britain New York: Charles Scribner’s
Sons, 1969
William T Y’Blood Hunter–Killer: US Escort Carriers in the Battle of the Atlantic An-
napolis: Naval Institute Press, 1983
William T Y’Blood Red Sun Setting: the Battle of the Philippine Sea Annapolis: Naval
Institute Press, 1981
References cited through abbreviation. General references are cited as

op. cit. in Notes.
ADERS = Gebhard Aders History of the German Night Fighter Force, 1917–1945 (trans-
lation of Geschichte der deutschen Nachtjagd 1917–1945) London: Jane’s Publish-
ing Company, 1979
ALLISON = David Kite Allison New Eye for the Navy: the Origin of Radar at the Naval
Research Laboratory NRL Report 8466. Washington: US Government Printing
Office, 1981
ALVAREZ = Luis W Alvarez Alvarez: Adventures of a Physicist New York: Basic
Books, Inc., 1987
BALDWIN = Ralph B Baldwin The Deadly Fuze: Secret Weapon of World War II San
Rafael, California: Presidio Press, 1980
BATT = Reg Batt The Radar Army: Winning the War of the Airwaves London: Robert
Hale, 1991
BEKKER 1 = Cajus Bekker (pseudonym of Hans Dieter Berenbrok) Radar: Duell im
Dunkel Oldenburg/Hamburg: Gerhard Stalling Verlag, 1958
BEKKER 2 = Cajus Bekker Augen durch Nacht und Nebel: Die Radar Story Olden-
burg/Hamburg: Gerhard Stalling Verlag, 1964
BEKKER 3 = Cajus Bekker The Luftwaffe War Diaries: the German Air Force in World
War II (translation of Angriffshöhe 4000) pp 404–405. Garden City, New York:
Doubleday and Co., 1968
BELL = A History of Engineering and Science in the Bell System: National Service
in War and Peace (1925–1975) M D Fagen, editor. New York: Bell Telephone
Laboratories, 1978
BELOTE = James H Belote and William M Belote Titans of the Seas: the Development
and Operations of Japanese and American Carrier Task Forces During World War II
New York: Harper and Row, Publishers, 1975
BLUMTRITT = Tracking the History of Radar Oskar Blumtritt, Hartmus Petzold and
William Aspray, editors. Piscataway, New Jersey: IEEE Press, 1994
BOWEN = E G Bowen Radar Days Bristol: Adam Hilger, 1987
BRANDT = Leo Brandt Zur Geschichte der Radartechnik in Deutschland und Gross Bri-
tannien Genoa: Pubblicazioni dell’Istituto Internazionale dell Comunicazioni,
1967
BROWN = R Hanbury Brown Boffin: a Personal Story of the Early Days of Radar, Radio
Astronomy and Quantum Optics Bristol: Adam Hilger, 1991
BUDERI = Robert Buderi The Invention that Changed the World New York: Simon
and Schuster, 1996
526
Bibliography
BURNS = Radar Development to 1945 Russell Burns, editor. London: Peter Peregri-
nus Ltd (for the IEE) 1988
CALLICK = E B Callick Metres to Microwaves: British Development of Active Com-
ponents for Radar Systems 1937 to 1944 London: Peter Peregrinus Ltd (for the
IEE) 1990
CRAVEN & CATE 1 = Wesley Frank Craven and James Lea Cate, editors The Army
Air Forces in World War II Vol 1 Plans and Early Operations, January 1939 to
August 1942 Chicago: University of Chicago Press, 1948
CRAVEN & CATE 2 = Vol 2 Europe: Torch to Pointblank, August 1942 to December
1943 1949
CRAVEN & CATE 3 = Vol 3 Europe: Argument to V–E Day, January 1944 to May 1945
1951
CRAVEN & CATE 5 = Vol 5 The Pacific: Matterhorn to Nagasaki, June 1944 to August
1945 1953
DAHL = Alexander Dahl Bumerang: Ein Beitrag zum Hochfrequenzkrieg München:
J F Lehmanns Verlag, 1973
DAVIS = Richard G Davis Carl A. Spaatz and the Air War in Europe Washington:
Center for Air Force History, 1993
ERICKSON = John Erickson, Radio-location and the Air Defence Problem: the
Design and Development of Soviet Radar Science Studies Vol 2, pp 241–63,
1972
FRIEDMAN = Norman Friedman Naval Radar Annapolis: Naval Institute Press,
1981
GEBHARD = Louis A Gebhard Evolution of Naval Radio-Electronics and Contributions
of the Naval Research Laboratory NRL Report 8300. Washington: US Government
Printing Office, 1979
GETTING = Ivan Alexander Getting All in a Lifetime: Science in the Defense of Democ-
racy New York: Vantage Press, 1989
GIESSLER = Helmuth Giessler Der Marine–Nachrichten–und–Ortungsdienst Mu-
nich: J F Lehmanns Verlag, 1971
GUERLAC 1 = Henry E Guerlac Radar in World War II New York: Tomash–
American Institute of Physics Publishers, 1987
GUERLAC 2 = Henry Guerlac, The Radio Background of Radar J. Frank. Inst. Vol
250, pp 285–308, 1950
HEZLET = Vice-admiral Sir Arthur Richard Hezlet Electronics and Sea Power New
York: Stein and Day Publishers, 1975
HOFFMANN 1 = Karl Otto Hoffmann Ln—Die Geschichte der Luft–nachrichtentruppe
Vol 2, part 1 Der Weltkrieg: Der Flugmelde–und Jägerleitdienst 1939–1945
Neckargemünd: Kurt Vowinckel Verlag, 1968
HOFFMANN 2 = Vol 2, part 2 Der Weltkrieg: Drahtnachrichtenverbindungen Richt-
funkverbindungen 1939–1945 1973
HOFFMANN-HEYDEN = Adolf-Eckard Hoffmann-Heyden Die Funkmessgeräte
der deutschen Flakartillerie (1938–1945) Dortmund: Verkehrs–und Wirtschafts–
Verlag GmbH, 1953
HOLLMANN = Hans E Hollmann Physik und Technik der ultrakurzen Wellen Berlin:
Springer Verlag, 1936
HOWSE = Derek Howse Radar at Sea: the Royal Navy in World War 2 Annapolis:
Naval Institute Press, 1993
IEEE = IEEE Center for the History of Electrical Engineering RAD Lab: Oral Histories
527
Documenting World War II Activities at the MIT Radiation Laboratory Piscataway,

New Jersey: IEEE Press, 1993
JONES 1 = R V Jones The Wizard War: British Scientific Intelligence, 1939–1945 (Amer-
ican edition of Most Secret War) New York: Coward, McCann and Geoghegan,
Inc., 1978
JONES 2 = R V Jones Reflections on Intelligence London: William Heinemann, 1989
JONES H = H A Jones The War in the Air Oxford: The Clarendon Press, 1935
KEMP = John Kemp Off to War with ¡054’ Braunton: Merlin Books, Ltd, 1989
KINGSLEY 1 = The Development of Radar Equipments for the Royal Navy, 1935–45 F
A Kingsley, editor. London: Macmillan Press, Ltd (for the Naval Radar Trust)
1995
KINGSLEY 2 = The Application of Radar and Other Electronic Systems in the Royal
Navy in World War 2 F A Kingsley, editor. London: Macmillan Press, Ltd (for
the Naval Radar Trust) 1995
KOCH = Horst-Adalbert Koch Flak: Die Geschichte der deutschen Flakartillerie Bad
Nauheim: Verlag Hans-Henning Podzun, 1954
KURYLO & SUSSKIND = Friedrich Kurylo and Charles Susskind Ferdinand Braun:
a Life of the Nobel Prizewinner and Inventor of the Cathode-Ray Oscilloscope Cam-
bridge, Massachusetts: The MIT Press, 1981
LATHAM & STOBBS = Colin Latham and Anne Stobbs with Foreword by Sir Ed-
ward Fennessy, CBE Radar: a Wartime Miracle Stroud: Alan Sutton Publishing
Ltd, 1996
LOBANOV = M M Lobanov Nachalo Sovetskoy Radiolokatsii (The Beginnings of Soviet
Radar) Moscow: Sovetskoye Radio, 1975
LOVELL = Sir Bernard Lovell Echoes of War: the Story of H2S Radar Bristol: Adam
Hilger, 1991
LUNDSTROM 1 = John B Lundstrom The First Team: Pacific Naval Air Combat from
Pearl Harbor to Midway Annapolis: Naval Institute Press, 1990
LUNDSTROM 2 = John B Lundstrom The First Team and the Guadalcanal Campaign:
Naval Fighter Combat from August to November 1942 Annapolis: Naval Institute
Press, 1994
MAGNETRON = Fifty Years of the Cavity Magnetron P M Rolph, editor. Birmingham:
School of Physics and Space Research, University of Birmingham, 1991
MONSARRAT = John Monsarrat Angel on the Yardarm: the Beginnings of Fleet Radar
Defense and the Kamikaze Threat Newport, Rhode Island: Naval War College
Press, 1985
MORISON 1 = Samuel Eliot Morison History of United States Naval Operations in
World War Two Vol 1 The Battle of the Atlantic, September 1939–May 1943 Boston:
Little, Brown and Company, 1947
MORISON 3 = Vol 3 The Rising Sun in the Pacific, 1931–April 1942 1948
MORISON 4 = Vol 4 Coral Sea, Midway and Submarine Actions, May 1942–August
1942 1949
MORISON 5 = Vol 5 The Struggle for Guadalcanal, August 1942–February 1943 1948
MORISON 6 = Vol 6 Breaking the Bismarcks Barrier, 22 July 1942–1 May 1944 1950
MORISON 7 = Vol 7 Aleutians, Gilberts and Marshalls, June 1942–April 1944 1951
MORISON 8 = Vol 8 New Guinea and the Marianas, March 1944–August 1944 1953
MORISON 10 = Vol 10 The Atlantic Battle Won, May 1943–May 1945 1956
MORISON 12 = Vol 12 Leyte, June 1944–January 1945 1958
MORISON 13 = Vol 13 The Liberation of the Philippines, Luzon, Mindanao, the Visayas,
528
Bibliography
1944–August 1945 1959

MORISON 14 = Vol 14 Victory in the Pacific, 1945 1960
NAKAGAWA = Yasuzo Nakagawa Japanese Radar and Related Weapons of World War
II Laguna Hills, California: Aegean Park Press, 1997; an English translation
edited by Louis Brown, John Bryant and Naohiko Koizumi of relevant parts of
the author’s Document: Kaigun Gijutsu Kenkyu-sho (The Naval Technical Research
Department) and Jishu Gijutsu de Ute: Nippon Denki ni Miru Erekutoronikusu Hat-
ten no Kiseki (Electronics Development at Nippon Electric) Tokyo: Nihon Keizai
Shinbunsha Inc., 1987
NAKAJIMA = Shigeru Nakajima, The history of Japanese radar development to
1945, BURNS, pp 243–258
NIEHAUS = Werner Niehaus Die Radarschlacht 1939–1945: Die Geschichte des
Hochfrequenzkriges Stuttgart: Motorbuch Verlag, 1977
OSRD 5 = Science in World War II Vol 5 Electronics: a History of Divisions 13 and 15
and the Committee on Propagation C Guy Suits, editor. Boston: Little, Brown
and Company (for Office of Scientific Research and Development) 1948
OSRD 6 = Vol 6 The Administrative History of the Office of Scientific Research and
Development Irvin Stewart, editor, 1948
PAGE = Robert Morris Page The Origin of Radar New York: Doubleday and Com-
pany, 1962
PETERSEN = Carsten Petersen Luftkrig over Danmark Vol 3 Jagerkontrol 1943–1945,
Karup 1940–1945 Ringkøing, Denmark: Bollerrup Boghandels Forlag, 1988
PILE = General Sir Frederick Pile Ack-Ack: Britain’s Defence Against Air Attack Dur-
ing the Second World War London: George G Harrap, 1949
PLOCHER 1 = Generalleutnant Hermann Plocher, edited by Harry R Fletcher The
German Air Force Versus Russia Vol 1 1941 New York: US Air Force Historical
Division, Aerospace Studies Institute, Air University, Arno Press, 1965
PLOCHER 2 = Vol 2 1942 1966
PLOCHER 3 = Vol 3 1943 1967
POLLARD = Ernest C Pollard Radiation: One Story of the MIT Radiation Laboratory
Durham, North Carolina: The Woodburn Press, 1982
PRICE 1 = Alfred Price Instruments of Darkness: the History of Electronic Warfare
London: Macdonald and Jane’s Publishers, 1977
PRICE 2 = Alfred Price The History of US Electronic Warfare: the Years of Innovation-
Beginnings to 1946 Washington: Association of Old Crows, 1984
PRICE 3 = Alfred Price The History of US Electronic Warfare: the Renaissance Years,
1946 to 1964 Washington: Association of Old Crows, 1989
PRICE 4 = Alfred Price Aircraft versus Submarine: the Evolution of the Antisubmarine
Aircraft, 1912–1980 London: Jane’s Publishing Co., Ltd, 1980
PRITCHARD = David Pritchard The Radar War: Germany’s Pioneering Achievement
1904-45 Wellingborough: Patrick Stephens Limited, 1989
RABI = John S Rigden Rabi: Scientist and Citizen New York: Basic Books, Inc., 1987
RADAR = Radar (magazine). Washington: Headquarters, Army Air Forces, 1944–
1945 (a Ferranti limited edition reprint as a book)
RADAR SURVEY 1 = US Radar Survey Vol 1 Airborne Radar Washington: National
Defense Research Committee, 1945
RADAR SURVEY 2 = Vol 2 Shipborne Radar
RADAR SURVEY 3 = Vol 3 Ground Radar
RADAR SURVEY 4 = Vol 4 Navigational Radar
529
RADAR SURVEY 5 = Vol 5 Radar Definitions

RADAR SURVEY 6 = Vol 6 Test Equipment
RADAR SURVEY 7 = Vol 7 Nomenclature Index
RAF SIGNALS 4 = The Second World War, 1939–1945: Royal Air Force Signals Vol 4
Radar in Raid Reporting London: The Air Ministry, 1952
RAF SIGNALS 5 = Vol 5 Fighter Control and Interception
RAF SIGNALS 7 = Vol 7 Countermeasures
RAWNSLEY = C F Rawnsley and Robert Wright Night Fighter New York: Henry
Holt and Company, 1957
REUTER = Frank Reuter Funkmess: Die Entwicklung und Einsatz des RADAR-
Verfahrens in Deutschland bis zum Ende des Zweiten Weltkrieges Opladen: West-
deutscher Verlag GmbH, 1971
ROSKILL 1 = Captain Stephen Wentworth Roskill The War at Sea, 1939–1945 Vol 1
The Defensive London: Her Majesty’s Stationery Office, 1954
ROSKILL 2 = Vol 2 The Period of Balance 1956
ROSKILL 3 = Vol 3, part 1 The Offensive 1960
ROSKILL 4 = Vol 3, part 2 The Offensive 1961
ROSKILL 5 = Stephen Roskill Naval Policy Between the Wars Vol 1 The Period of
Anglo–American Antagonism, 1919–1929 London: Collins, 1968
ROSKILL 6 = Policy Vol 2 The Period of Reluctant Rearmament 1976
ROTTERDAM = Sitzungsprotokolle der Arbeitsgemeinschaft Rotterdam (Minutes of
the Rotterdam Committee) Leo Brandt, editor. Düsseldorf: Sonderbücherei für
Funkortung, 1953
ROWE = Albert Percival Rowe One Story of Radar Cambridge: The University Press,
1948
SAYER = Brigadier A P Sayer The Second World War, 1939–1945: army Radar London:
The War Office, 1950
SCIENTIFIC = Report on the Scientific Intelligence Survey in Japan Scientific and Tech-
nical Advisory Section, General Headquarters United States Army Forces,
Pacific, 1945. US National Archives
SIG CORPS 1 = Dulany Terrett United States Army in World War II. The Signal Corps:
the Emergency (to December 1941) Washington: Office of the Chief of Military
History, 1956
SIG CORPS 2 = George Raynor Thompson, Dixie R Harris, Pauline M Oakes and
Dulany Terrett United States Army in World War II. The Signal Corps: the Test
(December 1941 to July 1943) Washington: Office of the Chief of Military History,
1957
SIG CORPS 3 = George Raynor Thompson and Dixie R Harris, United States Army
in World War II. The Signal Corps: the Outcome (Mid-1943 through 1945) Wash-
ington: Office of the Chief of Military History, 1966
SIMMONDS & SMITH = Ed Simmonds and Norm Smith Echoes Over the Pacific:
an Overview of Allied Air Warning Radar in the Pacific from Pearl Harbor to the
Philippines Campaign Murwillumbah: Murwillumbah Print Shop, 1995
SOUTHWORTH = George C Southworth Forty Years of Radio Research: a Reportorial
New York: Gordon and Breach Science Publishers, 1962
STOKES = John W Stokes 70 Years of Radio Tubes and Valves Vestal, New York: Vestal
Press, 1982
STREETLY = Martin Streetly Confound and Destroy: 100 Group and the Bomber Support
Campaign London: Jane’s Publishing Company, 1978
530
Bibliography
SUCHENWIRTH 1 = Richard Suchenwirth The Development of the German Air Force,

1919–1939 New York: US Air Force Historical Division, Aerospace Studies
SUCHENWIRTH 2 = Richard Suchenwirth Historical Turning Points in the German
Air Force War Effort New York: US Air Force Historical Division, Aerospace
Studies Institute, Air University, Arno Press, 1968
SWORDS = Sean S Swords Technical History of the Beginnings of Radar London: Peter
Peregrinus (for IEE) 1986
TAYLOR = Albert Hoyt Taylor Radio Reminiscences: a Half Century Washington: US
Naval Research Laboratory, 1948, 1960
TECHNICAL = Reports of the US Naval Technical Mission to Japan 1945, 1946. Oper-
ational Archives, US Naval Historical Center
TEDDER = Lord Arthur William Tedder With Prejudice: the War Memoirs of Marshal
of the RAF Lord Tedder Boston: Little, Brown and Company, 1966
TERRAINE 1 = John Terraine A Time for Courage: the Royal Air Force in the Euro-
pean War, 1939–1945 (American edition of The Right of the Line). New York:
Macmillan Publishing Co., 1985
TERRAINE 2 = John Terraine The U-boat Wars, 1916–1945 (American edition of
Business in Great Waters: the U–Boat Wars, 1916–1945) New York: G P Putnam’s
Sons, 1989
TRENKLE 1 = Fritz Trenkle Die deutschen Funkmessverfahren bis 1945 Heidelberg:
Alfred Hüthig Verlag, 1986
TRENKLE 2 = Fritz Trenkle Die deutschen Funkstö rverfahren bis 1945 Ulm: AEG–
Telefunken Aktiengesellschaft, 1982
TRENKLE 3 = Fritz Trenkle Die deutschen Funkführungsverfahren bis 1945 Heidel-
berg: Alfred Hüthig Verlag, 1987
TRENKLE 4 = Fritz Trenkle Die deutschen Funknachrichtenanlagen bis 1945 Vol 2 Der
Zweite Weltkrieg Heidelberg: Alfred Hüthig Verlag, 1990
TUSKA 1 = C D Tuska, Historical Notes on the Determination of Distance by Timed
Radio Waves J. Frank. Inst. Vol 237, pp 1–20, 83–102, 1944
TUSKA 2 = C D Tuska, Pictorial Radio J. Frank. Inst. Vol 253, pp 1–20, 95–124, 1952
WATT = Sir Robert Watson-Watt The Pulse of Radar: the Autobiography of Robert
Watson-Watt (American edition of Three Steps to Victory) New York: Dial Press,
1959
WEBSTER & FRANKLIN 1 = Sir Charles Webster and Noble Franklin The Strate-
gic Air Offensive Against Germany, 1939–1945 Vol 1 Preparation London: Her
Majesty’s Stationery Office, 1961
WEBSTER & FRANKLIN 2 = Vol 2 Endeavor 1961
WEBSTER & FRANKLIN 3 = Vol 3 Victory 1961
WEBSTER & FRANKLIN 4 = Vol 4 Annexes and Appendices 1961
WOOD & DEMPSTER = Derek Wood and Derek Dempster The Narrow Margin: the
Battle of Britain and the Rise of Air Power, 1930–1940 Washington: Smithsonian
Institution Press, 1990
ZIMMERMAN = David Zimmerman Top Secret Exchange: the Tizard Mission and the
Scientific War Montreal: McGill–Queen’s University Press, 1996
Principal dissertations, unpublished reports and private publications
cited through abbreviation.
JAY = K E B Jay, History of the Development of Radio and Radar. Unpublished
531
manuscript prepared for contribution to the Official History of the Second World
War 1946, located at Library of the Defence Research Agency Malvern
KAUFMANN = Robert Kaufmann Funk–Radar–Bumerang aktiv dabei Florstadt: pri-
vate publication, 1986, author’s address Lauterbacher Strasse 36, D–61197
Florstadt/Paderborn
KERN = Ulrich Kern Die Entstehung des Radarverfahrens: Zur Geschichte der
Radartechnik bis 1945 (Dissertation, Universität Stuttgart) Historisches Insti-
tut der Universität Stuttgart, 1984
KROGE = Harry von Kroge GEMA—Berlin: Geburtsstätte der deutschen aktiven
Wasseschall–und Funkortungstechnik 1998. Private printing by Lühmanndruck,
Hamburg. Author’s address: Sinstorfer Kirchweg 68, D–21077 Hamburg
MCKINNEY = Colonel John B McKinney, Radar: a Reluctant Miracle, a paper
written for the Research Seminar in Technological Innovation at the Harvard
Business School, 1960, Professor J R Bright, Seminar Director. Located in the
archives of the Historical Electronics Museum, Baltimore
MCNALLY = Commander I L McNally, Radar Reflections, Manuscript memoirs
contributed to the Vice Admiral Edwin B Hooper collection, 1 July 1975. Navy
Department Library, Washington, DC
RUNGE = Wilhelm Runge, Ich und Telefunken. Unpublished manuscript de-
posited in Telefunken Library, Ulm, 1971
STEGSKOPFER = Friedrich Janssen et al Wir Stegskopfer: Die Funkmess–Einheiten
Prinz Eugen und Tegetthoff, 1943–1945 Published privately by Hans–Joachim
Menzel, Benniger Weg 11, D-7141 Murr, Germany, 1989
STEPP = Wilhelm Stepp, Ueber die Reichweite drahtloser Anlagen im Wellengebiet von
1 cm bis 20 m (Dissertation, Technische–Hochschule Darmstadt) Darmstadt:
Hessische Landes– und Hochschul–Bibliothek, 1946
Principal journals
AES = IEEE Transactions on Aerospace and Electronic Systems

Ann. Phys. = Annalen der Physik
Ap. J. = Astrophysical Journal
Bell Tech. J. = Bell System Technical Journal
J. Ap. Phys. = Journal of Applied Physics
J. Frank. Inst. = Journal of the Franklin Institute
Onde = L’Onde Électrique
Phil. Mag. = Philosphical Magazine
Phys. Rev. = Physical Review
Phys. Today = Physics Today
Phys. Zeit. = Physikalische Zeitschrift
Proc. IEE = Institution of Electrical Engineering Proceedings
Proc. IRE = Proceedings of the Institute of Radio Engineers
Proc. Phys. Soc. = Proceedings of the Physical Society
Proc. Roy. Soc. = Proceedings of the Royal Society (London)
Rev. Mod. Phys. = Reviews of Modern Physics
Rev. Sci. Inst. = Reviews of Scientific Instruments
Zeit. Hf. = Zeitschrift für Hochfrequenztechnik
Zeit. Phys. = Zeitschrift für Physik
532
NAME INDEX
Note: Entries in the end notes are Barkhausen, Heinrich

indicated by a suffix ‘n’. Japan, 84
Positive-grid tube, 37
Abajian, Hurach (later Henry) Visits Japan, 494n
Microwaves in Pacific, 447 Barrow, Wilmer L, 147
V-1, 396 Bay, Zoltán, 436–7
Alekseev, N F, 158 Bayler, Walter L, 249
Allison, James B, 70 Beaverbrook, Lord, 120
Alvarez, Luis Becker, Ludwig, 287
Airborne, 167–8 Bekker, Cajus, 407
Antenna, 191 Bergquist, Kenneth P, 217
Atomic bomb, 425 Beria, Lavrenty P, 87
Countermeasures, 292 Billington, David, 460
Eagle, 192–3, 366, 424 Birkenhead, Earl of, 56
GCA, 171, 193–4, 416, 429 Blackett, P M S
Linear accelerator, 439 AA Command, 118
Los Alamos, 174 Meteors, 435–6
MEW, 195 Operational analysis, 339
Phased array, 281, 458 Proximity fuze, 175
‘Radar scientist’, 461 Rutherford, 187
Anderson, S H, 69 Blair, William, 43–4, 69
Ando, Hiroshi, 35 Bloch, Felix, 441
Appleton, Edward V, 51 Blumlein, Alan Dower
Arco, Graf von Death of, 189
Anti-Nazi, 287 Minimum range, 101
Telefunken, 315 Böhm, Otto
Ardenne, Manfred von Leaves Germany, 79
Radar proposal, 287 Photo incident, 315
TV, 35 Bohr, Niels, 177
Arisaka, Iwao, 136 Bonch-Bruevich, Mikhail
Armstrong, Edwin H, 10 Alexandrovich
Early location attempts, 41 Death, 263
Regenerative receiver, 130 Lenin’s favor, 88
Arnold, Henry, 71 Resonant magnetron, 158
Aschenbrenner, Friedrick, 113–4 Boner, Alfred, 404
Bainbridge, Kenneth T, 167 Boot, H A H
Los Alamos, 174 Physicist’s design, 461
Bainbridge-Bell, L H, 105, 451 Resonant magnetron, 151, 451
533
Bowen, E G Bush, Vannevar

Airborne radar, 59–61, 168 British Branch Rad Lab, 308
ASV mark II, 102 Critical orders, 168
Joins Rad Lab, 162–3 Dispute over fuze, 396
Meteorology, 443 NDRC, 149–50
Microwaves, 151 Proximity fuze, 176
Relieved airborne, 153 Rad Lab, 164
Solar radiation, 436 Tizard Mission, 160
Strife with Rowe, 100 Butement, W A S
Tizard Mission, 160 CD/CHL, 58–9
Towns seen with radar, 189 Proximity fuze, 60, 175, 179
Bowen, Harold G, 65 Byrne, John, 293
Gives radar priority, 66 Callaghan, Daniel J, 256–7
Tizard Mission, 160 Calpine, H C, 63
Bowles, Edward L Cassevant, A F, 71
Blind landing, 149, 193 Chennault, Claire, 420
North Africa, 355 Chernishev, A A, 86
Rad Lab, 163 Cherwell, Lord
Stimson’s radar man, 223 Appointed to Tizard Committee,
Brandt, Leo 55
Düppel, 296 Convinced of German radar, 104
Examines ASV, 340, 501n
Dipole jamming, 295
Kindersarg, 407
Germany versus Atlantic, 339
Postwar, 408
Navigational beams, 116
Replaces Runge, 311
Opinion of AA, 27, 396
Rotterdam, 311–2, 315
Opposes AA radar, 119, 191
Würzburg, 283
Pushes H2S, 189
Brandt, Walter, 74
Chiang Kai-shek, 420
Braun, Ferdinand, 33
Chu, L J, 296
Breit, Gregory
Fission, 177 Churchill, Winston, 83
Ionosphere, 42, 176 AA fire, 119
Breuning, Ernst, 97 Air war, 317
Brinker, Emil, 139 Casablanca, 305
Brooke-Popham, Sir Robert, 27 Demands H2S, 189
Brown, John Lawrence (The Great Dieppe, 233
Brown), 400 Discounts radar, 120
Brown, Robert Hanbury Invasion worries, 386
ASV, 61 Malta, 206
ASV mark II, 102 Pledges aid to Soviets, 348
EMI, 460 SCR-584, 395
Fourth-best dictum, 101 Tizard Mission, 160, 454
Microwave airborne, 117 US Army planes for Atlantic, 339
Recalls Bawdsey, 58 Clarke, Sir Arthur C, 498n
Brown, Wilson, 239 Coales, J F, 63
Browning, John, 24 Cockburn, Robert, 116
Buckner, Simon Bolivar, 419 Experiments with Window, 296
Burgess, R E, 439 Terman, 293
Burns, Russell, 343 Invasion, 388
534
Name Index
Cockcroft, John D Fighter control 103, 282, 287

AI, 151 Dippy, R J,
Proximity fuze, 179 Gee, 288
Rutherford, 187 Loran, 430
Tizard Mission, 160 Dominik, Hans, 46
Coke, Charles, 207 Dönitz, Karl
Colton, Roger B, 69–70 Ardenne-Hollmann, 287
Orders SCR-584 and 545, 171 Blames radar, 346
Secrecy, 450 Commands Navy, 350
Sees microwave radar, 170 Compared to Harris, 334–5
Compton, Karl T, 150 Dismantles Hipper, 352
Conant, James Bryant Meets Japanese, 136
Electronics Training Group, 163 Metox, 341
NDRC, 149–50 Passage of Biscay, 342–3
Condon, Edward U, 167 Resists U boats for
Cooke, A H, 168, 188 Mediterranean, 212
Cornford, Clifford, 302 U-boats effective, 169
Cotton, F Sidney, 229 Views on submarines, 19, 104,
Courant, Richard, 79 121, 128, 334
Crane, H R, 181 Doorman, F M, 236
Crerar, Findlay, 98 Douhet, Giulio, 13
Cunningham, Sir Andrew, 208 Dowding, Hugh
Curran, Joan, 295 Aerial photos, 230
Dahl, Alexander Battle of Britain, 108, 111–2
Naxburg, 315–6 Organizes Fighter Command,
Oboe, 303 51–2
Davenport, Lee Relieved, 119–20
GCA, 194 Downing, A C, 188
SCR-584, 169–70 Draper, Stark, 255
V-1, 396 DuBridge, Lee
David, Pierre, 44, 89–90 British Branch, 308
Davis, N E, 228 Heads Rad Lab, 167
Dee, P I, 153 North Africa, 355
Microwaves, 187 ‘Radar won the war’, 465
Sees towns with AIS, 189 Steering Committee, 173
De Forest, Lee Eaker, Ira C
Regenerative receiver, 130 8th Air Force, 306–7
Triode, 10 Daylight attacks, 309
Dennison, David M, 181 Eastman, Melville, 430
Devyatkov, N D, 158 Edwards, G C, 444
Dewey, George, 411 Eggert, Joseph, 380
Dewhurst, H, 99 Eichelberger, Robert L, 261
DeWitt, John H, 437 Eisenhower, Dwight D
Diamond, Harry, 180 Invasion strategy, 386
Dicke, Robert H Night fighters, 354
Atmospheric absorption, 438 SCR-584, 395
Radiometer, 438–9 Elliott, George E, 218
Diehl, Hermann Erbslöh, Paul-Günther
Freya-Fahrstuhl, 280 Co-founds Tonographie, 47
535
Lobe switching, 280 Crete, 211

NVA, 73 No fighters for Tirpitz, 352
Esau, Abraham No Würzburg for Japanese, 136
Countermeasures, 322 Saves Martini, 315
Replaces Plendl, 299 Sees radar, 78
Ewing, Douglas H, 308 Technically naive, 82–3
Falck, Wolfgang, 282, 287 Window, 299, 445
Faraday, Michael, 8 Gorcke, P, 321
Faulkner, H W, 160 Goto, Arimoto, 254
Fawcett, E W, 39 Green, G K, 179
Fegen, E S F, 124 Greenglass, David, 452
Fellgiebel, Erich, 311 Griggs, David T, 307
Fennessy, Edward (later Sir Edward) Guthrie, Robert, 66
CH installation, 99 Gutton, Camille
Clarke, A C, 498n Collision avoidance, 45, 90
Fermi, Enrico, 177 Lack of range, 458
Fletcher, Jack Gutton, Henri, 45
Carrier raids, 238 Oxide cathode, 152
Coral Sea, 240 Hafstad, Lawrence, 179
Midway, 242–4 Halsey, William
South Pacific, 249, 252 Carrier raids, 238
Foders, Heinrich, 139 Coral Sea, 242
Fowler, R H, 179 Leyte, 414
Franklin, Benjamin, 7 Proximity fuze, 182
Fraser, Sir Bruce, 418 South Pacific commander, 254–5,
Fuller, J F C, 17 257
George, Lloyd, 11 Hamada, Shigenori, 138
George V, King, 42 Hamazaki, Ryo, 137
George VI, King, 100 Hansell, Haywood, 422
Getting, Ivan Hansen, W W
Criticizes Navy radar, 416 Cavity resonance, 146
Rad Lab, 167 Linear accelerator, 440
SCR-584, 169, 191 Rhumbatron, 148
Style, 191, 461 Velocity of light, 432
V-1, 396 Hara, Chuichi, 241
Ghormley, Robert L Hashimoto, Chuji, 137
Esperance, 253 Hashimoto
Relieved, 254 Submarine commander, 363
Solomon Islands, 246 Harris, Arthur
Gibson, R O, 39 Battle of Berlin, 309, 317
Gilbert, C W, 187 Compared to Dönitz, 334–5
Gilbert, William, 7 Day–night controversy, 307
Goebbels, Josef Effect on postwar, 426
Coventry, 119 Gee, 289
V-1 and V-2, 393 Lübeck-Cologne, 300–1
Göring, Hermann, 25 Ruhr, 303
Air defense, 289, 304 Harrison, John, 20
Ardenne-Hollmann, 287 Hart, Raymond, 55
Attempt to destroy CH, 387 Heaviside, Oliver, 9
536
Name Index
Helmholtz, H L F von, 146 Reflection observations, 43

Henderson, John T, 165 Iachino, Angelo, 208
Hershberger, William D Inoue, Kyoshiro, 140
Information exchange, 450 Irifune, Naosaburo, 136
Microwaves, 69–70 Ito, Yoji, 84
Hertz, Heinrich Death ray, 424
Experiments suggest radar, 40, Magnetron, 156
48, 69 Mission to Germany, 135, 454
Inspires resonant magnetron, 151 Ito, Tsuneo, 156
Produces EM radiation 9–10 Jansky, Karl, 434
Pupil of Helmholtz, 146 Jaumann, Professor, 347
Hey, J S, 294, 434–5 Jewett, Frank B
Heydenburg, Norman, 177 NDRC, 150
Hill, A V, 159, 454 Joffe, A F
Hill, Roderic, 395 Helps Oshchepkov in prison, 88
Himmler, Heinrich, 315 Opposes microwaves, 158
Hiraga, Yuzuru, 83 Supports radar, 85
Hitler, Adolf Johnson, J B, 34, 46
Air defense, 304 Jones, R V
Attacks Russia, 279 Cabinet meeting, 116
Avoids U-boat incidents with US, Channel dash, 226
337 Dipole interference, 295
Belief in V-2, 394 Gee, 288
Channel dash, 225 Identifies German radar, 229, 290
Declares war on US, 318 Invasion, 387
Fears loss of capital ships, 349 Klein-Heidelberg, 285
Invasion fears, 234 Malta radar, 213
Night fighters, 285 Oboe, 302
Secrecy, 445 Oslo Report, 115
Sees radar, 78 V-1 and V-2, 393, 396
Stop research order, 489–490n Würzburg, 230
U-boats to Mediterranean, 212 Joubert de la Ferté, Sir Philip
Hodgkin, Alan, 188 Bombers for Coastal Command,
Hoffmann-Heyden, A-E, 317 338
Hollmann, Hans E GL radar, 187, 190
Crystal diode, 154, 443 Leigh light, 337
GEMA, 73–4 Relieved of CC, 340
Leyden, 314 Kammhuber, Josef
Radar proposal, 287 Air defense by radar, 324
Rotterdam, 311 Bypassed, 304
Hopkins, Harry, 339 NATO, 408
Hosogaya, Boshiro, 375 Night fighters 282–3, 289
Hull, Albert, 36–7 Wants airborne radar, 286
Hülsmeyer, Christian Kanner, Morton H, 195
Collision avoidance, 40, 46, 456 Kaufmann, Robert
Lack of range, 458 Jamming, 321
Hyatt, John, 39 Naxburg, 315–6
Hyland, Lawrence Kesselring, Albert
Leaves NRL, 65 Battle of Britain, 111
537
Holds Italy, 357–8 Leigh, Humphrey de Verde, 336–7

Malta radar, 213 Leigh-Mallory, Trafford
Transferred to air service, 13 Battle of Britain, 111
Kimmel, Husband E, 216 Dieppe, 233
Kimura, Masatomi, 376 Dispute with Park, 120
King, Ernest LeMay, Curtis
Anti-sub warfare, 338, 344 Bombing Japan, 422
Naval AA, 26 Effect on postwar, 426
Opposes Philippine landing, 412 Warns cities of attacks, 424
Pacific offensive, 246 Lewis, W B, 187
Proximity fuze, 396 Liddell Hart, B H
Kinkaid, Thomas High opinion of Pile, 27–8
Guadalcanal, 258 Military scholar 16
Leyte landing, 412, 414 Lindemann, Frederick see Cherwell,
Knox, James, 83, 222 Lord
Kobayashi, Masatsugu, 84 Lobanov, M M, 88
Examines British radar, 138–9 Lockard, Joseph, 218
Kohlrausch, Friedrich Wilhelm, 8 Loomis, Alfred Lee
Kondo, Nobutake, 257 GCA, 167–8, 416
Konoe, Fumimaro, 140 Inspects with Watt, 222
Korovin, Yu K, 48, 86 Loran, 430
Kramar, E, 113 Microwaves, 149
Kühnhold, Rudolph Rad Lab, 163
Credit for radar, 456 Shows Colton roof-top unit, 170
Demonstration, 128 Loomis, Wheeler, 167
GEMA, 76 Lothian, Lord, 160
Microwaves, 310, 455 Lovell, Bernard
Navigational beams, 113 CH, 154
Postwar, 408 Heads H2S, 190
Rotterdam, 311 Meteors, 435–6
Seeks radio location, 47, 459 Lovett, Robert A, 307
Telefunken, 73 Löwy, Heinrich, 22, 46
Kummetz, Oskar, 350 Ludendorff, Erich, 323
Kunkel, ‘The Mysterious Lüdi, F, 158
Lieutenant’, 284–5 Luebke, E A, 193
Kurita, Takeo, 413–4 Lugg, Sydney
Kusaka, Ryunosuke, 84 Leigh light, 336
Lamb, Willis, E, 440 Radar beacon, 102
Langmuir, Irving, 39 Lütjens, Günther
Lark-Horovitz, Karl, 444 Bismarck, 124–5
Laue, Max von, 287 Breaks radio silence, 126
Lauritsen, C C, 176 Dependence on radar, 260
Lawrence, Ernest Raiding, 122–3
Cyclotron, 439 MacArthur, Douglas
Rad Lab, defense interests, 150, Liberation of Philippines, 411–2
163, 167 Papua, 261
Tuve, 177 Strategy, 377
Lawson, James, 168 Macintyre, Donald, 511n
Lee, Wilis A, 257–8 McMillan, Edwin M, 168
538
Name Index
McMorris, Charles, 375 Megaw, E C S, 152

McNair, Leslie, 183 Michelson, A A
Madsen, Sir John, 220 Velocity of light, 432
Maertens, Erhard Wm Blair, 69
Navy signals chief, 341 Mikawa, Gunichi, 251
Treason suspected, 315 Milch, Erhard
Mahan, Thayer, 224 Competence, 289
Malairov, D D, 158 On amateurs, 508n
Marconi, Guglielmo Mitchel, John P, 217
Decimeter radar, 84 Mitchell, William
Predicts radar, 41 AA, 27
Wireless telegraphy 10 Planes versus ships, 17, 224, 245,
Marlborough, John Churchill Duke 465
of, 124 Mitscher, M A, 374
Marsden, Ernest, 209, 222 Moffett, William, 17
Marshall, George C Montgomery, Bernard L, 449
Compared to Sprunace, 379 Moore, James, 71
Invasion strategy, 386 Moran, Edward J, 253–4
U-boats, 339 Morita, Akio, 447
Marshall, Lauritsen C Morita, Masanori, GL mark II, 139
British Branch, 308 Mountbatten, Lord Louis, 230, 449
Sloan-Marshall tube, 294 Müller, Gtthard, Lorenz, 81
Martini, Wolfgang Hohentwiel, 351
Air defense by radar, 324 SEL, 408
Anzio, 359 Murray, A J L, 62
Aware of arrests, 314 Mussolini, Benito
CH, 290 Fall, 357
Channel dash, 226–7, 293 Takes Italy into war, 205
Examines British radar, 110 Muth, Hans, 286
Heads radar, 311 Nagaoka, Hantaro, 373
IFF, 133–4 Nagumo, Chuichi
Kindersarg, 407 Death of, 381
LZ-130, 99 Indian Ocean, 235
Malta radar, 213 Midway, 243–4
Microwaves, 310 South Pacific, 252
NATO, 408 Nakajima, Shigeru, 84, 140
Selects radar, 81 Resonant magnetron, 156
Supports Diehl, 103 Staff reduced, 373
Suspected of treason, 315 Nawa, Takeshi, 372
Watson-Watt’s friend, 78 Nimitz, Chester
Marx, E, 321 Carrier raids, 238
Mattison, Lewis C, 249 Coral Sea, 242
Mauborgne, Joseph O, 160 Fast carriers, 367–9
Mayer, Emil, 79, 315 Favors Philippine landing, 412
Mayer, Hans Ferdinand Midway, 243
Dachau, 314 Okinawa, 418
Oslo Report, 104 South Pacific, 254
Proximity fuze, 176 Strategy, 377
Maxwell, James Clerk, 8 Understanding of radar, 248
539
Nishimura, Shoji, 413 IFF, 132–3

Nissenthal, Jack, 234 Microwaves, 311
Norden, Carl L, 192 Navigational beams, 113–4
Oberlin, Edgar, 65 Relieved from HF research, 299,
O’Connor, R N, 210, 212, 220 322
Oersted, Hans Christian, 7 Rotterdam, 311
O’Hara, E H, 371 Son, 407
Ohnishi, Takijiro, 412 Technicians, 313
Okabe, Kinjiro, 37, 85 Plücker, Julius, 34
Magnetron, 156 Pöhlmann, H, 299
O’Kane, B J, 444 Pollard, Ernest
Oliphant, M L SG, 172
Microwaves, 151 Steering Committee, 174
Rutherford, 187 Pollard, P E
Oppenheimer, Robert, 174 Designs GL, 59
Oshchepkov, Pavel K Early radar, 175
Early experiments, 48 Portal, Sir Charles, 307
Gulag, 88 Pretty, Walter P G
Rapid, 87, 263 LZ-130, 98
Ozawa, Jisaburo Post Mortem, 403–4
Leyte, 413 Price, Alfred, 490n
Philippine Sea, 378–80 Pryor, W L, 172
Page, Robert M Purcell, E M
Assigned to radar, 65 21 cm radiation, 438
Credit for radar, 456 Nuclear magnetic resonance, 441
Duplexer, 67, 168 Rabi, I I
Ring oscillator, 450 Hydrogen, 440
Paris, E T, 58 Magnetron, 167
Park, Keith Rodney Rad Lab report, 174, 454
Battle of Britain, 111–2 Raeder, Erich
Dispute with Leigh-Mallory, 120 Inspects radar, 78
Parsons (née Hearn), Avis, 490n Prefers surface raiders, 120
Patterson, C C, 152 Radar for surface raiders, 128
Patton, George, 403 Resigns, 350
Pearce, F L, 160 Ramsey, Norman, 167
Pears, Charles, 352 Randall, J T
Peregrinus, Peter, 7 Resonant magnetron, 151, 451
Pershing, John J, 17 Physicist’s design, 461
Phillips, Sir Tom, 220 Reber, Grote
Pile, Sir Frederick Influence on Dutch, 438
AA radar, 118 Radio astronomy, 434
Commands AA Division, 27–8 Reeves, A H, 302
Solar flares, 435 Rice, C W, 70
V-1, 395 Ridenour, Louis N
Pither, A G, 220 Edits Rad Lab report, 174, 454
Planck, Max, 287 North Africa, 355
Plendl, Hans Roberts, G A, 54
Ardenne-Hollmann, 287 Roberts, Richard Brooke
Hans Mayer, 314 Fuze tests, 182
540
Name Index
Shock tests tubes, 177 Postwar, 408

U fission, 177 Press release, 487n
Rochefort, Joseph, 243 Rotterdam, 311
Röhrl, Anton Telefunken, 73
Microwaves, 310 Rutherford, Lord, 57
Rotterdam, 311 Ryle, Martin
Rommel, Erwin, 211 Invasion, 388
Malta, 214 Solar radiation, 436
Roosevelt, Franklin D Sakura, Yoshio, 372
Casablanca, 305 Samuel, Arthur L
Defend Australia, 240 Credit by Japanese, 156
Electronics Training Group, 163 Precursor of magnetron, 458
NDRC, 150 VHF triode, 68
Neutrality Zone, 337 Sasaki, Kiyoyasu, 137
Releases Army planes for Satake, Kinji
Atlantic, 339 In Germany, 137
Tizard Mission, 160, 454 Würzburg, 139
Rosenberg, Julius, 452 Saville, Gordon P, 223
Ross, Alfred W, 62 Sayers, James, 151, 155
Rottgardt, Karl Scheller, Otto, 21, 80
Commissioner for Radar, 317 Schonland, Basil, 209
Microwaves, 310 Schottky, Walter, 36
Runge’s dislike for, 287 Schriever, O, 147
Soviet prison, 408 Schultes, Theodor
Telefunken, 315 GEMA, 73–4
Rowe, A P Scott, Norman
AI, 153 SG, 256–7, 449
Bawdsey director, 58 South Pacific, 253
Drops Bowen, 162 Scrugham, James, 66
Gee, 288 Sedyakin, A I
IFF and the poor, 134 Gulag, 88
Moves lab, 98-99 Seitz, Frederick, 444
Moves TRE to Malvern, 231 Shaw, Bernard, 232
Oboe, 302 Shea, Michael, 363
Pushes H2S, 324 Shearing, G, 62
RDF, 83 Shembel, B K, 87–8
Strife with Bowen, 100 Sherbrooke, R St V, 350
Tizard Committee, 52 Sheridan, Philip, 15
Sunday Soviets, 174, 446 Shire, E S, 175, 179
Rozhanski, D A, 86 Short, Walter C, 216
Rumyanysev, D N, 86 Shrapnel, Henry, 183
Runge, Carl, 79 Sims, Charles, 380
Runge, Wilhelm Sims, William, 17
Biscay, 342 Sinclair, Don, 292
Darmstadt, 438 Skinner, H W B, 154, 444
Decimeter waves, 78–9 Slater, John C, 167
Inactive, 311 Slessor, John
Leaves Telefunken, 315 Coastal Command, 339–40
Lichtenstein, 286–7 Tanks, 16
541
Slim, William, 421 ‘Radar fought the war’, 465

Sliozberg, M L, 265 Son with Barkhausen, 84
Sloan, David Taylor, William E G, 216
Linear accelerator, 146, 439 Tedder, Arthur William, 212
Tuba, 294 Teller, Edward, 177
Slutskin, A A, 86 Terman, Frederick
Smith, Holland, 377 Countermeasures, 293
Smuts, Jan Christian, 11 Ferrets, 356
Snow, C P, 56 Window, 296
Sommerfeld, Arnold, 147 Tesla, Nikola, 41
Sommerville, James, 160 Thiel, Walter, 394
Sorrell, Bob, 293 Thiel, Willi, 437–8
Southworth, George C Thomson, A F H, 175, 179
Crystal diode, 154, 443 Thurston, Sir George, 257
Dielectrics, 370 Tinus, W C, 68
Solar radiation, 438–9 Tizard, Henry
Waveguide 146 Airborne radar, 60
Spaatz, Carl, 307 Leaves PM council, 120
Spaight, J M, 15 Mission to US, 159, 454
Speer, Albert, 317
Scientific committee, 50, 52
Sperrle, Hugo, 111
U-boats, 339
Sprague, Thomas L, 414
Tojo, Hideki, 359
Spruance, Raymond
Torrey, Henry, 444
Commander of crucial battles,
Touch, A Gerald, ASV, 102, 153
243–4
Townes, Charles, 442
Philippine Sea, 379–80
Trenchard, Hugh, 51
Staal, C H J A, 91
Trump, John G, 167
Stalin, Josef, 348
Stepp, Wilhelm Tucker, S M, 83
Long-wave noise, 294, 434 Tukhachevskii, M N, Gulag, 88
Moon, 437–8 Turner, Cobden, 176
Würzburg, 80 Tuttle, Geoffrey W, 229
Stepushkin, A B, 86 Tuve, Merle
Stilwell, Joseph W, 420–1 Directs entire fuze project, 184
Stimson, Henry L Fuze safety, 181
Bowles, 223, 355 Ionosphere experiment, 42
Calls for Watt, 222 Proximity fuze, 176
Secretary of War, 83 Tyler, Kermit A, 218
Stodola, E K, 437 Udet, Ernst
Stratton, J A, 149 Suicide, 283
Sutton, Robert W, 153 Zuckmayer play, 504n
Takagi, Takeo, 236, 241 Ushijima, Mitsuru, 419
Takahashi, Kanjiro, 156 Valley, George, 505n
Takayanagi, Kenjiro, 137 Van de Hulst, H C, 438
Tanaka, Raizo, 252, 257, 259 Van der Bijl, H J, 34
Taylor, Hoyt Van Fleck, J H, 442
Credit for radar, 456 Van Keuren, Arthur H, 162
Heads NRL radio, 64 Varian, Russell H, 148
Observes radar-like effects, 42 Varian, Sigurd F, 148
542
Name Index
Vivian, J G P, 449 Williams, N H, 36

Vvedenskiy, B A, 86 Willisen, Hans-Karl Freiherr von
Wallace, B E, 226 Co-founds Tonographie, 47
Wallace, F C Lobe switching, 280
Extravagant claims, 495n Loses GEMA, 408
Tizard Mission, 160–1 NVA, 73
Watanabe, Yasushi, 373 Wilson, Carrol, 163
Watson, Paul, 70 Wimer, C J, 219
Watt, Robert Watson (later Sir Wimperis, Harry
Robert Watson-Watt), 45–6 Death ray, 50, 424
Against Tizard Mission, 160 Tizard Mission 159
CH disclosed to French, 90 Winn, Roger, 344
Channel dash, 228 Wolff, Irving, 44
Credit for radar, 456 Wood, R W, 149, 458
Daventry experiment, 52 Woodnutt, Bill, 221
Did not want EEs, 460 Woodward-Nutt, A E, 160
Dundee move, 99 Wright, Carlton H, 258
HMSS, 62 Wright, Charles, 62
Leader of Bawdsey, 58 Microwaves, 151
Noise on CH, 295 Wright, Stuart P, 366
Radar memorandum, 50, 424 Yagi, Hidetsugu
RDF 1.5, 61 Antenna, 37, 84
Third-best dictum, 64 Resources committee, 373
US inspection, 222–3 Yamamoto, Isoroku
Wavell, Archibald Percival, 211 Forewarning by, 367
Wavell, Claude, 387 Midway, 242-243
Weber, Wilhelm Eduard, 8 Tokyo raid, 240
Weiler, J L W C von, 91 South Pacific, 252–4
Whipple, Fred, 296 Y’Blood, William T, 345–6, 348
Wiesner, J B, 197 Yeo, R A, 62
Wilkins, Arnold F Young, Leo
CH disclosed to French, 90 Amateur, 65
Credit for radar, 460 Credit for radar, 456
Daventry experiment, 52 Observes radar-like effects, 42
IFF, 129 Zacek, August, 37
Noise on CH, 295 Zahl, Harold, 71
Radar memorandum, 50 Pearl Harbor, 215
Williams, F C VT-158, 262
IFF mark III, 131 Zworykin, Vladimir Kosma, 35
543
SUBJECT INDEX
Note: Entries in the end notes are Organizations and companies,

indicated by a suffix ‘n’. American
Aircraft Radio Laboratory (Signal
Corps)
This index is divided into the Altimeter 22
following sections: IFF, 131
APL (Applied Physics Lab), 184
Bell Labs (BTL), Bell Telephone
Organizations and companies, Laboratories
American AA director, 170
Organizations and companies, Introduced to radar, 67
British Magnetron, 163
Organizations and companies, Radio astronomy, 434
German SJ, 361
Tube designs, 34, 46, 186
Organizations and companies,
Bureau of Standards see NBS
Japanese
Carnegie Institution
Organizations, Soviet
APL, 183–4
Organizations and companies, Ionosphere 42
other nations NDRC, 163
Radars and related electronics, Proximity fuze, 176
American Coast and Geodetic Survey, 432
Radars, Australian Dahlgren see Naval Proving Ground
Radars and related electronics, Dupont, 40
British Eimac
Tube designers, 37
Radars, Canadian
Zahl tube, 262
Radars and related electronics, Eitel-McCullough see Eimac
German
Field Artillery Board, 183
Radars, Japanese GE (General Electric)
Radars, South African AN/TPS-2, 414
Military and naval units, American IFF, 131
Military and naval units, British Magnetron, 151, 154
Rice’s microwaves, 70
Military and naval units, German
SCR-584, 188
Ordnance Vacuum tubes, 36–7
Vacuum tubes (specific) General Radio Company, 292
General Gilfillan Brothers Company, 194
544
Subject Index
Hazeltine Corporation, 131 Union Carbide, 39–40

Microwave Committee, 150 Western Electric
Naval Proving Ground (Dahlgren), FH, 370
179–80 LAB, 366
NBS (National Bureau of Standards) Signal Corps, 71
Proximity fuze, 180 Westinghouse
Radar proposal, 70 NDRC, 150
NDRC (National Defense Research Power supplies, 163
Committee) Rad Lab, 167
Organized, 150 Signal Corps, 71
Proximity fuze, 176
Service labs, 162 Organizations and companies,
Tizard Mission, 160 British
NRL (Naval Research Laboratory), ADEE (Air Defence Experimental
22, 186 Establishment), from Army
Early work, 42–3 Cell, 99, 186
IFF, 131 ADRDE (Air Defence Research and
Radar suggestions, 42–4 Development
Tizard Mission, 160 Establishment), 99, 186
XAF, 64007 Air Council, 51
Philco AMRE (Air Ministry Research
ASV mark II, 162 Establishment)
IFF mark II, 131 Change to TRE, 186
Radio Research Laboratory Dundee, 99
(Harvard), 293 Swanage, 153
AN/APR-7, 323 Army Cell at Bawdsey, 58
Rad Lab (MIT Radiation Becomes ADRDE, 99, 186
Laboratory) ASE (Admiralty Signal
British Branch, 173, 398 Establishment), 186
North Africa, 355 Bawdsey Research Station, 56–8
Organized, 163 Evacuation, 98–9
Recruiting, 167 Birmingham University
Steering Committee, 173 Crystal diode, 154
Raytheon Corporation Microwaves, 151
SG, 248 BTH (British Thompson-Houston),
SO, 413 187
RCA (Radio Corporation of Crystal diode, 444
America), 22–3 Central Interpretations Units, 387
CXZ, 67 Committee for the Scientific Survey
Early radar, 43–4 of Air Defense see Tizard
Signal Corps Committee
Air defense, 27 Decca Radio Ltd, 431
Early work, 43–4 EK Cole Ltd, 102
IFF, 131 EMI (Electrical and Musical
NDRC, 150, 160 Industries)
Visit Britain, 118 Propose radar, 101
War outbreak, 217–23 TV, 58, 61
Sperry Gyroscope Company, 149 GEC (General Electric Company,
Sylvania, 180 Ltd)
545
AI, 151–3 Improving Freya, 282–3

Crystal diode, 444 Incorporated, 74
Resonant magnetron, 152 Lorenz, 80
HMSS (His Majesty’s Signal Navy wants exclusive use, 78
School), 37, 62, 186 Not at Rotterdam meetings, 311
Reflex klystron, 153 Heereswaffenamt (Ordnance
Type 271, 154 Department), 81
Imperial Chemical Industries, 39 IG Farben, 40, 312
Ministry of Economic Warfare, 280 Julius Pintsch, 47
National Physical Laboratories, 432 Leybold-von Ardenne, 35
Noise Investigation Bureau, 228 Lorenz, AG
Pye Radio, 102 Beam navigation, 21–2, 113
Radio Countermeasures Board, 291 Collaboration with GEMA, 80
Radio Research Board’s Propagation Hohentwiel, 351
of Waves Committee, 51 Rotterdam, 311
Radio Research Station at Slough, 50 TV, 35
RAE (Royal Aircraft Establishment), Marine-Nachrichtendienst (Navy
102 Signals Service), 128
Gets Freya serial 1, 407 Max-Wien-Institut, 297
RRDE (Radar Research and NVA (Nachrichtenmittel-Versuchs-
Development Anstalt)
Establishment), 186 Early radar, 73
Tizard Committee Established, 46
Organized 50, 55 NVK (Nachrichtenmittel-Versuchs-
Proximity fuze, 175 Kommando), NVA renamed,
Tizard Mission (British Technical 123
and Scientific Mission) Physikalisch-technische
Canada, 164 Reichsanstalt, 311
IFF, 131 Reichsluftfahrtsministerium
Proximity fuze, 179 (National Air Ministry), 82
US, 159–61 IFF, 133
TRE (Telecommunications Research Reichspostzentralamt,
Establishment) Elefant-Rüssel, 397
From AMRE, 186 Siemens & Halske, 104
Oboe, 302 Jagdschloss, 285
To Malvern, 231–2 Proximity fuze, 185
Window, 295–6 Scientist imprisoned, 314
Sonderkommision für
Organizations and companies, Funkmesstechnik (Special
German Commissioner for Radar),
Arbeitsgemeinschaft Rotterdam 317
(Rotterdam Committee), Telefunken
310–3 Jewish director, 78–9
Extensive listing, 507n Kühnhold, 73
GEMA (Gesellschaft für Lichtenstein, 286–7
elektro-akustische und Mannheim, 285
mechanische Apparate) Postwar, 408
IFF Erstling, 132 Rotterdam, 311
Improves AW sets, 103 Stops microwave research, 310
546
Subject Index
Tonographie, 47 Radar for Moscow, 264

Early radar, 73 Resonant magnetron, 157
Zeppelin Company, 283 NKVD (People’s Commissariat of
Internal Affairs), 87
Organizations and companies,
PVO (Air Defense Forces), 87
Japanese
QRL (Central Radio Laboratory), 48,
Electrical Engineering Research
87
Department (Denki Kenkyu
UFTI (Ukrainian Physical-technical
Bu), 137
Institute), 89
Fuji Electrical Apparatus
Evacuation, 263
Manufacturing Co (Fuji
Denki Seizo), 138 Organizations and companies,
Japan Radio (Nihon Musen), 85 other nations
Japan Victor (Nihon Victor), 138 Bernard Price Institute (South
Naval Air Technical Depot, Africa), 209
Oppama, 373 Brown, Boveri & Cie (Switzerland),
Naval General Staff (Kaigun Kansei 158
Hombu), 137 Constructions Navales (France), 355
NEC (Nippon Electric, Nippon Kammerling-Onnes-Institute
Denki), 84, 137–8 (Netherlands), 314
NEC Ikuta Research Office Branch New South Wales Government
(Ikuta Kenkyu Bunsho), 138 Railways (Australia), 221
NHK Japan Broadcasting Philips Physics Laboratory
Corporation, 137 (Netherlands), 91
NTRD (Naval Technical Research Radiophysics Laboratory
Department) (Australia), 220
Death ray, 424 Linear accelerator, 440
Germany, 135 Radio astronomy, 436
Origins, 83–4 RIEC (Regio Instituto Elettrotecnico
Wartime problems, 372–3 e della Communicazioni
Toshiba Research Institute (Toshiba della Marina, Italy), 91
Kenkyusho), 138 SADIR (Societe Anonyme des
Francaise
Organizations, Soviet
Radiotelegraphiques,
Extensive listing, 86–7
France), 90
GAU (Main Artillery
Société LMT (France), 355
Administration), 47
United Incandescent Lamp and
AA radar, 87–8
Electrical Co (Hungary), 436
LEFI (Leningrad Physical-technical
Institute), 87 Radars and related electronics,
LFTI (Leningrad Electro-physical American
Institute), evacuation, 263 AN/APN-2 (US Rebecca), 391
NIIIS-KA (Scientific Research AN/APQ-5B (LAB)
Institute), 48 Japanese shipping, 366–7
AA radar, 87–8 Not used in Atlantic, 338
NII-9 (Scientific Research Institute AN/APQ-7 (Eagle)
9), 88–9 Development, 192–3
AA unit, 265 Precision, 424
Evacuation, 263 AN/APR-4 (intercept receiver), 422
Klystron, 158 AN/APR-7, 323
547
AN/APS-3, 345 Carrier raids, 239

AN/APS-4 (ASH) Comparison, 82
3 cm AI, 371 Coral Sea, 241–2
Photograph, 382 Deficient for fighter control, 245,
AN/APS-6 (AIA), single-seat AI, 368
371 Düppel, 357
AN/APS-13, 323 Eastern Solomons, 253
Atomic bomb trigger, 425 Lack of A-scope, 237
AN/APS-15 (H2X) Pearl Harbor, 218
Accuracy measured, 308 Poor performance at Santa Cruz,
Japan, 423 255
Micro-H, 323 Secure line of sight
Suspected of Nagasaki bombing, communication, 369
425 Status in 1942, 247
AN/CPS-1 (MEW) CXAS
Army support, 398–400 Bell Labs prototype, 68
Development, 195–6 Redesigned for magnetron, 164
Invasion, 391 CXBA, prototype FH, 370
Operations room, photograph, CXBL, prototype SM, 368
204 CXZ, RCA prototype, 67
Photograph, 202 Eagle see AN/APQ-7
V-1s, 396 Eureka see AN/PPN-1
AN/CPS-4, 443 GCA see AN/MPN-1
AN/MPN-1 (GCA) H2K, water absorption, 442
Berlin Blockade, 194 H2S (American use), 308
Development 193–5 H2X see AN/APS-15
Rejected by civil aviation, 429 IFF mark II (SCR-535)
AN/PPN-1 (US Eureka), 391 Few admirers, 252
AN/TPQ-3, mortars, 401 Philco, 131
AN/TPS-2, guerrillas, 414–5 IFF mark III (American), Coral Sea
AN/TPS-3, 400 and Midway, 243
AN/TPS-10 (Little Abner) IFF mark IV (American), 132
Iwo Jima, 418 IFF mark V (American), 133
Photograph, 203 LAB see AN/APQ-5B
AIA see AN/APS-6 Little Abner see AN/TPS-10
ASB (60 cm ASV), NRL, 371 Loran, 430–1
ASE (US ASV mark II) Mandrel (US version), 293–4
7000 ordered from Philco, 162 Mediterranean, 356
Aleutians, 375 Mark 3 (FC), 68
Black Cats, 364–5 Adapted to magnetron, 164
Santa Cruz, 255 Battle of Pips, 376
ASG (George, S-band ASV), Rad Comparison, 82
Lab, 169, 172 Difficulties at Surigao, 413
ASH see AN/APS-4 Offset shooting, 377
ASV mark II see ASE Photograph, 270
Carpet (US version), 293–4 Status in 1942, 247
Mediterranean, 356 Weakness, 370
Chaff see Window Mark 4 (FD), 68, 237
CXAM, 67 Absence noted, 239
548
Subject Index
Adapted to magnetron, 164 Photograph, 142

Comparison, 82 SCR-296, 68
First use of proximity fuze, 259 SCR-515, 131
Kamikazes, 415–6 SCR-516 (MAWD), 223
Mitscher makes radar night SCR-517, 169
defense, 374 Against U-boats, 339, 340
Offset shooting, 377 Japanese freighters, 365
Photograph, 271 SCR-520, 169
Status in 1942, 247–8 SCR-527 (GCI), 118
Weakness, 370 Panama, 223
Mark 8 (FH) SCR-540 (AI mark IV), 118
Introduced, 247 SCR-545 (Bell Labs GL)
Photograph, 273 Anzio, 359
Replaces FC, 370 Back up for 584, 191
Surigao, 413 Bell Labs GL radar, 191
Mark 12 (improved FD), 247 SCR-582, 354
Luzon, 416 SCR-584 (GL mark 3A)
Photograph, 272 AA Battle of January 1st, 402
Mark 22 (3 cm height finder), Anzio, 358–9
photograph, 272 Army support, 398–401
Mark 35, fire direction radar, 416 Automatic tracking, 169–70
MEW see AN/CPS-1 Colton, 191
Rebecca see AN/APN-2 Copied by Soviets, 452
SCR-268 Current use, 496n
Anzio, 359 Development, 165
Attempted airborne, 162 Flying bombs, 185
Blind fire, 354 M-9 director, 171–2
Compare SCR-584, 170 Photograph, 328
Comparison, 82 Soviets, 265
Demonstration of, 71 Surveying, 414–5
Düppel, 357 Tests H2S and H2X, 308
Guadalcanal, 249–50 V-1s, 395–6
Japanese copy, 139 SCR-587 (intercept receiver), Kiska,
Japanese Window, 299 375
Linear accelerator, 439–40 SCR-602
Philippines, 219 Düppel, 357
Photograph, 326 Expendible, 378
Record shoot at Rendova, 378 Invasion, 391
Soviets, 265 North Africa, 353
SCR-270 Pacific landings, 262
Comparison, 82 Photograph, 143
Guadalcanal, 249–50 SCR-720, 169
Midway, 243 Replaces other AIs, 188
Milne Bay, 262 SC (small CXAM)
Pearl Harbor, 215–8 Düppel, 357
Philippines, 219 Dutch Harbor, 375
Used in Korean War, 72 Luzon fighter direction, 417
SCR-271, 72 Photograph, 30
Observes Moon, 436 Status in 1942, 247–8
549
Savo Island, 251 Radars and related electronics,

SD (submarine AW), 361–2 British
SF (small SG), 172 AI mark I
SG Calls for, 150
Appreciated, 248 Initiated, 61
Battle of Pips, 376 Production too early, 101
Cape Esperance, 253–4 AI mark II, 357
Common knowledge, 256 AI mark III, 101
Development, 171–2 AI mark IV
Luzon fighter direction, 417 Demanded for North Africa, 354
Pacific island navigation, 370 Düppel, 357
Problems in use, 257 First interception, 117–8
Prototype photograph, 30 Night fighters over Germany, 319
Savo Island, 251 Satisfactory, 101, 186
SG-1, removes faults, 370 AI marks VII, VIII and IX, 188
Fears about Window, 296
SJ (submarine periscope radar)
AI mark X (SCR-720)
Bell Labs, 361
Night fighters over Germany, 319
Extensive use, 362–3
Photograph, 385
Wolf packs, 369
Unaffected by Window, 296
SK
AIS (10 cm AI, general designation),
Improved CXAM, 67
186–8
Fleet standard, 368
AMES (Air Ministry Experimental
Kamikazes, 416 Station), 53
Philippine Sea, 380 AMES type 1 (CH), 53–7
Photograph, 272 Appraisal, 460
PPI, 237 Attempts to jam, 290
SM Blitz, 117
Fighter control, 368 Germans learn details from IFF,
Photograph, 272 134
SO (torpedo boat), 413 Klein-Heidelberg, 284–5
SP (light SM) 368 Photograph, 92
Tuba, 294 Proof against jamming and
XAF, 67, 82 Window, 295
XT-1 (see also SCR-584) fails for Rejected by Signal Corps, 223
GCA, 193–4 V-2s, 397
Follows shells, 401 AMES type 2 (CD/CHL)
YE (homing radio), 239 Blitz, 117
Photograph, 272 Channel dash, 226
Comparison, 82
Radars, Australian Covers CH, 59
AW, 221 Photograph, 199
LW/AW, 140 AMES type 5 (COL, overseas
Development, 221 version of CHL)
New Guinea, 378 Düppel, 357
No landing secure without, 262 Malta, 211
Photograph, 143 North Africa, 353
MAWD (modified SCR-268), 221 AMES type 9 (Mobile Radio Unit,
Milne Bay, 262 MRU), 59
550
Subject Index
Dunkerque, 110, 136 CHL see AMES type 2

Immune to Düppel, 358 COL see AMES type 5
Malta, 207, 211 Decca Navigator
North Africa, 353 Invasion, 390
Rejected by Signal Corps, 223 Postwar, 431
Tracks airship, 98 Eureka, 135
Used in France, 107 Sicily, 356–7
AMES type 11 (50 cm warning), 400 GCI see AMES types 16, 17 and 26
AMES type 13 (height finder), 399 Gee
AMES type 15 (mobile GCI), 400 Description, 288
AMES types 16 and 17, GCI), 59 Evaluation, 324
Effective, 119 Feared captured, 289
Fears about Window, 296 Invasion, 390
First interception, 117–8 Weaknesses, 301
Invasion, 391 Gee-H, 322
Photograph, 198 Evaluation, 324
AMES type 22 (50 cm light Invasion, 389
warning), 400 Velocity of light, 432
AMES type 26 (GCI), 400 GL mark I
ASV mark I Blitz, 118
Examined by Telefunken, 500n Dunkerque, 110, 136
Initiated, 61 GL mark II, 59
Production too early, 101 Blitz, 118
Tizard Mission, 160 Photograph, 327
ASV mark II Radio astronomy, 434–5
Bismarck, 127 Singapore, 138–9
Channel dash, 227 GL mark III (GL mark 3B)
Completed, 153 Low priority, 60, 190
Effects on U-boats, 336–8 Photograph, 328
Gains priority, 187 Ground Grocer, 294
Germans capture, 214 HF/DF (high-frequency direction
Milne Bay, 262 finder)
North-African convoys, 208 Bismarck, 126–7
Photographs, 382–4 Ship borne, 345, 348
Tizard Mission, 160 Shore-based, 345
ASV mark III Huff-Duff see HF/DF
Atlantic use decided at Cabinet, H2S
339 Accuracy measured, 308
Destroys confidence in Metox, Competition with ASV mark III,
340 339
Development, 188–9 Development, 188–90
Boozer (passive receiver), 292 Evaluation, 324
Carpet (Würzburg jammer), 286 Examines ASV, 338
CD mark IV Inaccuracy, 301
Channel dash, 226 Rotterdam, 310
Dieppe Raid, 234 Telefunken duplicates, 313
Not detected by Martini, 310 US wants, 308
CD/CHL see AMES type 2 H2X
CH see AMES type 1 Captured, 316
551
Carpet bombing, 322 Fails from gunfire, 350

IFF mark I, 130 Photograph, 275
IFF mark II Type 273Q, 351
J-switch, 291 Type 279, 61, 125
Radar beacon, 102 Immune to Düppel, 358
Tizard Mission, 130–1 Mediterranean, 206, 214
IFF mark III, Allied standard, 132 Type 279M, 126
J (cover name), 289 Type 281, 351
LW Type 284, 63
Invasion, 391 Arctic, 351
North Africa, 353 Compared to FC, 68
Photograph, 142 Holds Bismarck, 125
Mandrel (Freya jammer) Hood, 126
Introduced, 292 Photograph, 329
Invasion, 387–8 Type 285, 63
Monica (tail radar), 292 Arctic, 349
Moonshine Compared to FD, 68
Helps 8th Air Force, 291 Type 286, 63
Revived for Invasion, 389 Arctic, 349
MRU see AMES type 9 Düppel, 357
NT-271 see CD mark IV Mediterranean, 206
Oboe Type 286M
Development and use, 301–2 Early introduction, 125
Evaluation, 324 First kill, 336
Superior to optical, 303 Type 286P, 336
US wants, 307
Radars, Canadian
Pip Squeak, 110
GL mark 3C, 165, 191
RDF 1.5, 61
Night Watchman, 165
Rebecca, 135
Sicily, 356–7 Radars and related electronics,
SCL, Singapore, 138 German
Serate, 306 A-1, A-2 and A-3 designations, 81–2
Shoran see Gee-H Darmstadt, 80
Type 79X, 61 DeTe-I, sea search, 75, 77
Type 79Z, 125 DeTe-II
Type 91 (jammer) Air warning, 75, 77
Dover, 228 Demonstrated to Gëring, 78
Invasion cacophony, 388 Dreh-Freya (panoramic Freya), Post
Type 271 Mortem, 404–5
Arctic, 349 Egerland, 317
Beam shape, 190 Elefant-Rüssel (copy of CH), 397
Compare SG, 171–2 Over-the-horizon, 484n
Lack of PPI limits, 336 Post Mortem, 404–6
Late as sea search, 129 Tracks V-2, 397
Photograph, 275 Elektra Sonne (postwar Consol)
Rushed to completion, 154, 187 Postwar Omnirange, 428
Type 272, 234 RAF use, 301
Type 273 Erstling (IFF), 132
Arctic, 349 Unloved, 284
552
Subject Index
Flakleit g, photograph, 327 Unaffected by Window, 305–6

Flamme, 406 Mammut
Flensburg, 292, 319 Developed, 103, 280
Freya Evaluation, 324
Africa, 212 Phased array, 281
Air lifted to Norway, 106 Photograph, 96
Bruneval, 230 Post Mortem, 404–6
Comparison, 81–2 Mannheim
Feeder arrangement for, 77 Altered for 10 cm, 317
For blind bombing, 268 Developed from Würzburg, 81,
IFF, 133 285
Japanese see, 136 Marbach, 317
Located by Ferrets, 356 Metox (passive receiver for U-boats)
Photograph, 94 ASV mark III destroys
Photograph proves Germans confidence, 340
have radar, 104 Counters ASV-Leigh light, 338
Post Mortem, 404 Japan, 373
Romania, 269 Local oscillator emanation, 341
Tunisia, 354 Naxburg, 316
Use by Finns, 264–5 Naxos U (passive receiver for
Used in invasion of U-boat)
Czechoslovakia, 78 Failures, 340
Freya-Fahrstuhl Proposed, 312
Developed, 103, 280 Naxos Z (passive receiver for
Photograph, 93 fighter)
Heidelberg, 294 Locates bomber stream, 319
Heinrich, 301 Proposed, 312
Hohentwiel, ASV, 351 Neptun R-2, 306
Scharnhorst, 512n Neuling (IFF), 133
Jagdschloss Nürnberg, 299
Developed, 285 Seetakt
Photograph, 200 Arctic, 349
Post Mortem, 404–5 Channel dash, 227
Klein-Heidelberg, 284 Comparison, 81–2
Korfu, 312 Early operational failures, 76
Kulmbach, 317 Photograph, 270
Kurfürst, 81 Photograph published 1939, 32
Kurmark, 81 Prototype, 75
Kurpfalz, 81 Rejected by U-boats, 346
Lichtenstein B/C Scharnhorst, 351–2
Detected by Ferret, 299 With raiders, 121–9
Evaluation, 324 Stendal A (passive Würzburg
Jamming, 294 receiver), 297
Night fighters, 286 Stichling (prickly fish), 133
Photograph, 385 Tastlaus, 299
Window, 297 Wanz G-1
Lichtenstein SN-2 Replaces Metox, 341
Captured, 319 Wassermann
Effectiveness, 309, 319 Developed, 103, 280
553
Evaluation, 324 Togo, 264

Located by Ferrets, 356 Y-Verfahren
Phased array, 281 From Y-Gerät for fighter control,
Photograph, 95 284
Post Mortem, 404–6 Zwilling (IFF, twin), 133
Wismar (anti-jamming), 286
Incompatible with Würzlaus, 297 Radars, Japanese
Wotan 1 see X-Gerät Mark 1 model 1
Wotan 2 see Y-Gerät Comparison, 371
Würzburg Guadalcanal, 248–9
Africa, 212 Japan, 421
Blind fire, 282 Kiska, 300
British capture, 214 Modified for submarine AW, 363
Bruneval, 230–1 Photograph, 143
Comparison, 82 Prototype, 138
GCI, 292 Mark 2 model 1
IFF, 133 Eastern Solomons, 252
Japanese see, 136 Gunfire failures, 371–2
Locates Soviet submarine base, Photograph, 276
351 Mark 2 model 2
Photograph, 329 Anti-submarine, 363
Photograph at Flak battery, 330 Comparison, 371
Prototype, 80–1 Kiska evacuation, 376
Pulsed Doppler, 506n Midway, 157
Romania, 269 Photographs, 277
Transfer to Japan, 139 Success at Bougainville, 372
Use by Finns, 264–5 Mark 2 model 3 (copy of
Window, 297 Würzburg), 139
Würzburg-Riese (giant) Mark 3 model 2, photograph, 278
Developed for GCI, 283 Mark 4 model 2 (copy of SCR-268),
Inadequate for GCI, 304 139
Photograph, 199 Mark 4 model 3, Saipan, 421
Photograph as radio telescope, Mark 4 model 4, Peleliu, 421
200 Mark 6 airborne
Photograph from US Embassy in Introduced, 373
Berlin, 451 Photograph, 384
Post Mortem, 404 Tachi-1, 138
Radio astronomy, 438 Japan, 421
Togo, 264 Tachi-2, 138
Würzlaus, 297 Japan, 421
Würzmann, observes Moon, 437 Photograph, 331
X-Gerät, 114–5 Tachi-3, 139
X-Leitstrahlbake (direction beacon), Japan, 421
114 Receiver, photograph, 332
Y-Gerät Transmitter, photograph, 331
Beam navigator, 115 Tachi-4
Postwar DME (distance Improvements from SCL, 138
measuring equipment), 428 Japan, 421
Preferred as IFF, 284 Photograph, 332
554
Subject Index
Tachi-6, 138 Japanese shipping, 366–7

Japan, 421 15th Air Force, visual bombing rare,
Photograph, 144 322
Tachi-24 (copy of Würzburg), 139 20th Bomber Command, 420
Photograph, 333 21st Bomber Command, 424
Tachi-31 (1.5 m GL with Würzburg 62nd Coast Artillery Regiment
indicator), 421 North Africa, 353
Type A Bi-static, 85, 138 Tests SCR-268, 71
Type B see Tachi-6 109th AAA Gun Battalion, 402
164th Infantry Regiment, 253
Radars, South African
315th Wing, 424
JB
403rd AAA Gun Battalion, 328
Developed, 140
413rd AAA Gun Battalion, 402
Superior to MRU, 209
480th Group
Radars, Soviet Bay of Biscay, 340
B-2, 265 Morocco, 341
B-3, 265 482nd Pathfinder Group, H2X, 308
Neptune (GL mark III), 265 555th Signal Corps AW Battalion,
Rapid see RUS-1 399
RUS-1 (Rhubarb), 87–8 560th Signal Corps AW Battalion,
Uselessness, 263 353
RUS-2 (Redoubt), 88 561st Signal Corps AW Battalion,
Value as AW, 263 353
RUS-2S, limited production, 263 583rd Signal Corps AW Battalion,
Sleep, 265 with guerrillas, 415
SON-2 (copy of GL mark II), 265 597th Signal Corps AW Battalion,
SON-4 (copy of SCR-584), 265, 452 with guerrillas, 415
Storm, 87
Military and naval units, British
Zenith, 157
AA Command, 28
Military and naval units, American Blitz, 118–9
Coast Artillery Corps, AA assigned, AA Division, 28
26 Army Cooperation Command, 400
Backs radar, 69 Secrecy restricts, 448
Electronics Training Group, 162–3 ATS (Auxiliary Territorial Service),
1st Marine Division, 246 119
3rd Fleet Bomber Command
Leyte, 412–4 Atlantic, 344
Typhoon, 417 Dispute over H2S, 339
7th Fleet, Leyte, 412–4 Invasion, 392
7th Infantry Division, 376 Rejects day attacks, 104
8th Air Force Tirpitz, 362
Invasion, 392 Coastal Command
Uses H2S, 308–9 Bay of Biscay, 338–40
Visual bombing rare, 322 Channel dash, 227
9th Marine Defense Battalion, 378 Fighter Command, 13, 28
12th Air Force, radar ruse, 358 Openess about radar, 448
14th Air Force Organized, 52
Attack Japan, 420 Trains with radar, 55
555
Fleet Air Arm, 19 20 mm Oerlikon, 24

Pathfinder Force, 303 Britain adopts, 40
Royal Tank Corps, 16, 27 US adopts, 26
WAAF (Women’s Auxiliary Air 37 mm Browning, 24
Force), 64 37 mm Maxim
Blitz, 119 AA, 24
WRNS (Women’s Royal Naval Proximity fuze, 177, 179–0
Service), 119 40 mm Bofors, 24
11 Group, 111 Britain adopts, 40
12 Group, 111 Santa Cruz Islands, 255
80 Signals Wing, 116, 291 US adopts, 26
100 Group, 319 57 mm gun, 180
Post Mortem, 405 88 mm triple purpose gun, 25
109 Squadron, 230 90 mm antiaircraft gun, 26
148 Squadron, 295 Tests with SCR-584, 170–1
220 Squadron, 61
617 Squadron, 352 Vacuum tubes (specific)
CV43, 168
Military and naval units, German DS310, 351
Afrika Korps, 211 DS320, 81
Flak, organized, 24–5 IG-8, 264
Fliegerkorps X, 211 LG76, 313
KGr 100 (Kampfgruppe) LMS10, 313
Beam navigation, 114, 116 LMS11, 313
Evaluation, 490n LMS100, 321
Legion Condor, 25 LS180
Luftflotte 2 Comparisons, 262
Battle of Britain, 111–2 For Würzburg, 80
Mediterranean, 213 Photograph, 410
Luftflotte 3 M-60, 156–7
Battle of Britain, 111–2 M-312, 157
Luftflotte 5 Photograph, 409
Arctic, 349 NT46, 53
Battle of Britain, 112 NT99 (Micropup), 100
Luftnachrichtentruppe (Air Signal Comparisons, 262
Corps), 78
RD12Tf, 80–1
Luftnachrichten-Helferinnen (Air
Basis for Hohentwiel, 351
Signals Women), 304
Comparisons, 262
Luftnachrichten-Versuchs-Regiment
TS6, comparisons, 262
Këthen (Experimental Air
Turbator, 158
Signal Regiment Këthen),
VT90 (Micropup, US), 100
103, 114
VT158 (Zahl tube), 262
Ordnance WE-224
A-4 rocket (V-2), 393–4 Developed, 34
FZG-76 (V-1), 393–4 Thunderstorms, 45
1.1 in automatic gun, 255 WL-530, 72
5 in (dual purpose gun) 2C40, 441
Adopted, 26 2C43, 441
Proximity fuze, 180–1 100TH, 67
556
Subject Index
Linear accelerator, 440 Description, 36

Ring oscillator, 72 Velocity effects, 148
931 (RCA), 294 Barrage fire (AA)
954, 955 and 956 (RCA Acorns), 37 Blitz, 118
Defined, 24
General Batteries (proximity fuze), 179–80
‘AA Battle of January 1st’, 402 Battle of the Beams, 115–6
Admiral Graf Spee B-Dienst, 335
Seetakt, 76, 105, 451 Beacons (radar), 134–5
Raider, 121 Beaufighter
Admiral Hipper, 121–2 Bomber Command, 306
Admiral Scheer, 121, 123–4 Photograph, 385
Air attacks Bentley Priory, 63
China, 14 Berlin, Battle of, 309–10
Inaccuracy of, 21 Bismarck, 121, 124–7
World War I, 11–2 Black Cats
Air Defense Center (Berlin), South Pacific, 364–5
photograph, 201 Surigao, 413
Air Defense School (Mitchel Field), Bletchley Park, 210
217 Boise, USS, 253–4
Aircraft carrier, inter-war debate Bomber, German heavy, 479n
and construction, 17–9 Bruneval Raid, 229–31
Akagi, 244 Bumerang (boomerang), 302
Altimeter (radio) B-18, 168
Lëwy, 46 B-29, designed for radar, 420–2
RCA, 22–3 Cadillac (Project), 196–7
Amateurs (radio) Calibration, CH, 54
Eimac tubes, 37–8 Camouflage (radar)
Hallicrafter S-27, 292 Cities, 322
Nazi suppression, 313, 508n U-boats, 347
Radio astronomy, 434 Cape Esperance, 253–4
Taylor and Young, 65 Carpet bombing, 322
Anti-submarine weapons, 344–5 Catalina see PBY
Antenna temperature, 520n Cathode-ray tube
Antwerp, 396 (electrostatic-focussed), 35
Ark Royal, HMS, 127 Thought too fragile, 75
Army support Cathode-ray tube (gas-focussed),
Egypt, 211–2 34–5
In France, 108 Cavity resonator, 146
RAF, 12–4, 16–7 Chain Home, 46, 53
A-scope, problems, 207, 237, 248 Ready in 1939, 97, 110
Atlantic, Battle of the, 334–8 Channel microwave link, 30, 49
Atlantis (Raider C), 121 CIC (Combat Information Center)
Atomic bombs, 424–5 Carrier raids, 237
Attu, invasion, 376 Destroyers at Luzon, 416–7
Ballon-abwehr-kanone, 23 Kamikazes, 416
Barkhausen-Kurz tube Midway, 245
As local oscillator, 74 Photograph, 274, 276
Channel link, 49 Province of reserve officers, 369
557
Circuit (proximity fuze), 178–9 Window, 297

Cleveland, USS, 181–2 Draper sight (40 mm), 255
Cologne, 301 Dresden, 323
Conical scan Duke of York, HMS, radar victory,
Rejected for improved FD, 415 351–2
SCR-584, 170, 188 Dundee, AMRE, 99
Tachi-4, 138 Dunkle Nachtjagd (dark night
Würzburg, 80, 188 fighting), 283
Convoy JW-51B, 349–50 Duplexer (see also TR switch), Page,
Convoy JW-55B, 351 67
Convoy PQ-17, 349 Duplication of design, 461
Coral Sea, 240–2 Düppel, 296
Countermeasures Salerno, 357
Allied bomber loses, 298–9 Eastern Solomons, 252–3
Beam navigation, 115–6 Electrical scanning, 192
Channel convoys, 228–9 Electron magnetic moment, 440
Channel dash, 224–8 Electron optics, 35
Dieppe Raid, 233 English Channel, photograph of
Dover, 299 microwave link, 30
Invasion, 387–9 Enterprise, USS, carrier raids, 238–9
Malta, 213 Eastern Solomons, 252–3
Coventry, 119 Midway, 242–3
Coxial cables, 39, 147 Santa Cruz, 255–6
Crystal diode see Diode Escort carrier, photograph of radars,
Dachau, 314 273
Daventry experiment, 52 Against U-boats, 346
‘Dawn Patrol’ (movie), 52 Espionage (see also Ferrets)
Death rays Atomic bomb, 455
Dutch, 90 Radar, 452
General fear, 49, 50 Essex, USS, CIC, photograph, 274
Japanese, 424 Ferrets, 230
Derax, 83 Detect Lichtenstein, 299
Dieppe Raid, 232–5 Mediterranean, 356
Radar lessons for Invasion, 390 Pacific use, 299–300
Diode (crystal), microwave detector, Fighter Command, photograph of
154 Operations room, 93
German difficulties, 312, 317 Flak (see also military organization
Wartime research, 443–4 Flak)
Diode (vacuum), 10 Suffers from countermeasures,
Dipole array, 59 320
Dipole interference, 295 World War I, 24
Director (AA), 60 Flug-abwehr-kanone see flak
Blitz, 118 Fruit Machine, 54
M-9, 170–1 Gas regulator tubes (for jamming),
Proximity fuze, 185 293
Dive bombing, origins of, 14 ‘Ghost fleet’, radar-created, 389
‘Dog’s Breakfast’, 187 Glorious, HMS, 106
Doover, 83, 221 Gneisenau, North Sea, 104, 106
Doppler shift, CW radar, 44, 74 Channel dash, 224–8
558
Subject Index
Raider, 121–3 Infrared, 41

Ground-controlled interception (see Signal Corps, 49, 69
also British radar, GCI) Invasion, 386–92
Himmelbett, 283–4 Ionosphere, height determined, 42
Introduction by RAF, 117–8 Origins of name, 494n
SCR-268, 72 Iwo Jima, 418
Unforeseen use of SCR-268, 72 Jodrell Bank, 435
Guadalcanal, strategic importance, J-switch, 291
246 Kaga, 244
Naval Battles of, 256–8 Kamikazes
Hallicrafter (for electronic Failure, 419
espionage), 292 Justification, 412
Hamburg, 306 Success, 418
Heavy bombers, German, 14 Kammhuber line, 283
Heil tube, 148 K-band (1.25 cm radar), 187;
Helena, USS, 182 Kindersarg (Operation Child’s
First use of proximity fuze, 259 Coffin), 407–8
Helle Nachtjagd (illuminated night Kiska, radar evacuation, 376
fighting), 282 Klystron
‘Hell’s Angels’ (movie), 52 Axial magnetic field, 494n
He-111, tests H2S copy, 313 Invention, 148–9
Himmelbett stations, in Russia, 265, Oliphant, 151–2
267 Soviet invention, 158
Failures, 304 Klystron (reflex)
In Germany, 283–5 Invention, 153
Post Mortem, 405 Soviet, 158
Retained, 305 Knickebein (bent leg), 114–5
Hood, HMS, 125–6 ‘Knock out blow’, 12
Hornet, USS, Coral Sea and Komadorskis, Battle of, 375
Midway, 242–4 Laffey, USS, radar picket, 419
Hydrogen hyper-fine structure, 440 Leary, USS
Hydrogen line (21 cm), 438 Experimental radar, 67
Hyuga, 244 Photograph of, 31
IFF, early tests, 129–39 Sunk, 346
Coral Sea and Midway, 243 Leigh light
Dominates radar costs, 131 In Bay of Biscay, 336–7
Erstling, 132–3 Outside Bay, 344
Few admirers, 252 Psychological importance, 347
German lack of plane-to-plane, Leningrad, 264
134 Lexington, USS, Coral Sea, 241–2
Germans prefer Y-Verfahren, 284 Leyte Gulf, Battle, 412–4
Mark II adopted by US, 131 Linear accelerator
Mark III Allied standard, 131 Alvarez, 439
Post Mortem, 406 Sydney, 440
Illustrious, HMS, fighter direction, Lobe switching, 59
206–7 Colton demands, 71
Blitz, 211 Lübeck, 300–1
Okinawa, 418 Lupolen H, 312
Taranto, 208 Luzon, 416–7
559
LZ-127, 84, 97 Moving target, 429

LZ-130 New York, USS, 67
Espionage flight, 97–8 New Zealand, 221–2
Mistaken as LZ-127, 488n Norden bombsight, 15
Photograph, 31 Use in Eagle, 192
Magnetron (resonant) Normandie, 45, 90
Bell Labs, 163 Nuclear magnetic resonance, 441
British invention, 151–2 Offset shooting, 377
Drawing of, 152 Okinawa, 418–9
Found in wreckage, 40 Operational research
Japanese invention, 156 Blitz, 118
Japanese use, 85 Defined, 55
Secrecy, 189 Orfordness, 53
Soviet invention, 89, 157–8 Orion, HMS
Strapped, 155 Detects Italian fleet, 208
Swiss invention, 158 Radar, 206
Tizard Mission, 160 Oslo Report
Magnetron (split anode) Author imprisoned, 314
French, 44–5 Flying bombs, 393
GE, 70 Jones uses, 115–6
GEMA, 73–4 Left at British Embassy, 104
Invention, 37 Proximity fuze, 175
RCA, 44 Over-the-horizon radar, 54 484n
Malta Oxide cathode, 38–9
First Blitz, 211 GEC magnetron, 152
Second Blitz, 213 High pulse current, 101
Strategic importance, 205–8, 212 Lorenz, 80
Martlesham Heath, 61 Pan American Neutrality Zone, 337
Maxwell’s equations, 6, 9 PBY
Meddo (US H2X), 316 Aleutians, 375
Meteorology, 442–3 Black Cats, 364
Meteors, 435 Reconnaissance, 238
Me-110 Pearl Harbor, radar, 215–8
Night fighter, 286 Peenemünde, proximity fuze, 185
Photograph, 385 Pentode, 35–6
Weakness, 111 Perth, 100
Microwaves, 69 Phased array, 281
Advantages of, 145 Phase locking, 506n
Assigned to Rad and Bell Labs, Philippine Sea, radar effects, 379–80
164 Photomultiplier tubes (for
Midway, 242–5 jamming), 293
Milne Bay, 261 Pips, Battle of, 376
Moon Ploesti, 269
Radar capabilities, 520n PPI (Plan Position Indicator)
Radar observations, 436–7 ASG, 169
Moscow, 264 Convoys,336, 349
Mosquito Dieppe Raid, 234
Night fighter, 319 GCA, 194
Used with Oboe, 302–3 H2S, 189–90
560
Subject Index
Jagdschloss, 285 Newspapers, 381

MAWD, 223 Radar beacon, 102
Many inventors, 457 Radar Center, 248
Navigation, 355, 361, 370, 390 Radar pickets, 418
Radar maps, 308 Radar trains
SJ, 361 Photograph, 330
SK, 237 Use against French underground,
Used during Blitz, 117 266
XT-1, 171 Use in Russia 267
Polyethylene Radio Amateur’s Handbook, 209
First used, 100 Radiometer, 439–9
Produced, 39–40 Radio screen, NRL, 43
Rotterdam, 312 French (barrages), 44, 89
Polythene see polyethylene Japanese, 85, 138
Post Mortem (Operation) NBS suggests, 70
Photographs, 201, 203 Soviet (Rapid), 48–9
Radars examined, 404 Useless in defense of France, 107
Predictor, AA see director Watt’s approach, 52
Prince of Wales, HMS Rawalpindi, HMS, 105
Against Bismarck, 125–6 Receiver (heterodyne), 41, 482n
Radar, 219–20 Receiver (passive radar), Bismarck,
Prinz Eugen 127
Atlantic, 121, 124–9 Receiver (regenerative), 10, 443
Channel dash, 224–8 Resnatron (Sloan-Marshall tube)
Prisoner interrogations, Battle of Invention, 150
Atlantic, 341–2 Tuba, 294
Proximity fuze Rhumbatron, 148
Battle of Bulge, 403 Rodney, HMS, type 79 installed, 62
Circuit, 178–9 Rope (meter-wave Window)
Combat, 182 Invasion, 388
Early work, 60 Japan, 42
First kill, 259 Rostock, 307
Flying bombs, 185 Rotterdam (H2S), 310
German efforts, 185 Rotterdam-X (H2X), 316
Kamikazes, 415 Ruhr, Battle of the, 306
Mitscher’s use, 374 Russo-Finnish War, 88
Names, 182 Safety devices (proximity fuze), 181
Not used Antwerp, 596 St Anthan, 100
Philippine Sea, 184 Salerno, 357–8
Quality control, 181 Sandy Hook, 44
Standard, 237 Santa Cruz Islands, 254–6
Test firing, 181–2 Santee, USS, CIC, photograph, 274
V-1s, 396 Sardonyx, HMS, prototype 50 cm
Pulsed anode operation, 53 set, 63
Purges (Soviet) 87–8, 157 S-band (10 cm radar), 187
Quantum electrodynamics, 440 Scharnhorst
Radar Channel dash, 224–8
Goes public, 120 North Sea, 104, 106
Name invented, 83 Radar lost, 351
561
Raider, 121–3 Taranto, 208

Sunk, 352 Tassafaronga, Battle of, 258–9
Schnorkel, 347 Technical mission
Secondary radar, 428–9 (Japanese-German), 135–7
Secrecy Television
IFF, 134 Japanese, 84
Magnetron, 189 Relevance to radar, 46
Silicon Valley and Route 128, 447 Tetrode, 35–6
US labs avoid, 70 Texas, USS, 67
Semmes, USS ‘Things to Come’ (movie), 13
Photograph of SC and SG Tirpitz, 352–3
prototypes, 30 Togo (radar ship), 264–5
Test of SG, 172 Toksovo, 264
Sheffield, HMS, type 79 installed. 62 Tokyo
Shock tests (vacuum tubes), 177, 179 Doolittle Raid, 239–40
Shokaku Holocaust, 422
Coral Sea, 241 Torpedo problems (US), 360
Good radar, 255 Torpedo School Ship G 10
Silica tubes, 37, 62 Photograph, 32
Slotted waveguide antenna, 191–2 Secrecy breach, 450
Solar flares, 435 TR switch
Soryu, 244 Oxford, 168, 188
Specular reflection RAD Lab, 167
Argument against microwaves, Treason, German scientists accused,
145 314–5
Joffe, 157 Triode, 10
Stealth technique, 145 Tuxedo Park, microwaves, 149
Stegskopfer Tizard Mission, 160
End of war, 407 Typhoon, 417
Radar training, 314, 320 Ultra
Straight-vision receiver Atlantic, 335
Described, 485n Beam navigation, 116
Used for IF, 61 Mediterranean, 210
Strategic bombing Unconditional surrender, 306
Doctrine changes, 279–80 Uranium fission, 177
Legal and moral aspects, 15 Velocity of light, 432–3
Submarine tactics, 335 Venerable, HMS, aircraft direction
Suffolk, HMS, 125–6 room, photograph, 276
Swanage, 102 Ventnor CH station, 111
Microwaves, 153 VHF radio
Swiss, neglect radar, 158 Absence in Mediterranean, 207
Swordfish Absence in Pacific, 245, 252
ASV mark II, 208 Advantages, 99
Bismarck, 127 Victorious, HMS, 127
Channel dash, 227 V-1 (FZG-76)
German copy, 499n Evades manual track radars, 395
Rockets, 345 Oslo Report, 393
U-boat passage of Gibraltar, 336 V-2 (A-4)
Tanks, 16–7 Early design, 393
562
Subject Index
Production, 394 Poor theory, 59

Waveguide, 146–7 RUS-2, 88
Whippany Singapore, 138
Established, 68 Type 284, 63
Magnetron test, 163 Y-Dienst
Wilde Sau (wild boar), 304–5 Ability to use H2S, 405
Window Direction finders, 284
Gëring’s contest, 299 Post Mortem, 404
Introduced by Japanese, 299 Y-Service, 284, 291
Okinawa, 418–9 Yorktown, USS
Origins, 296–7 Carrier raids, 238–9
Post Mortem, 405 Coral Sea, 241–2
X-band (3 cm radar), 187 Midway, 243–4
Yagi antenna Yorktown, USS (replacement),
ASV mark II, 208, 371 photograph of radars, 272
GEMA, 74 Zahme Sau (tame boar)
Japanese submarine, 363 Fighter technique, 305
Mark 6, 373 Particularly effective, 319
NRL, 67 Zuikaku, Coral Sea, 241
563

Technical and Military Imperatives: A Radar History of World War II

Uploaded by

Copyright:

Available Formats

Technical and Military Imperatives: A Radar History of World War II

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Technical and Military Imperatives: A Radar History of World War II

Uploaded by

Copyright:

Available Formats

Technical and Military

Operation Epsilon: The Farm Hall Transcripts

Echoes of War: The Story of H2 S Radar

Bofﬁn: A Personal Story of the Early Days of

Heavy Water and the Wartime Race

Electronic Inventions and Discoveries

British Library Cataloguing-in-Publication Data

Library of Congress Cataloging-in-Publication Data are available

Production Editor: Jacky Mucklow

Published by Institute of Physics Publishing, wholly owned by The Institute of

PREFACE WITH ADVICE TO THE READER ix

5 YEARS OF ALLIED DESPAIR AND HOPE 205

A.6 The Yagi antenna 473

of radar history by judiciously skipping technical matters because most of

novel or decisive effect. Because of the lessons revealed, certain actions

From the beginning I decided against using interviews as a research tool,

Hans SORGENFREI, Rudolf STARITZ, Jane E TUCKER, Oliver VOLLAND

Information Center) of the National Institute of Standards and Technology

1.1. RADIO VISION FOR WAR

too large. Without radar, ﬁnding a tanker on the fog-covered northern

1.2. ELECTROMAGNETIC WAVES

concluded a striving to understand mysteries that had intrigued our ances-

1.3. PERCEPTIONS OF AIR POWER, 1919–1939

1933 and learned to ﬂy at the age of 48 [9]. It is hardly remarkable that he

end of his dominance by ﬁrearms. It did not require a lot of imagination

whereas sound naval doctrine emerged from the American cacophony of

1.4. NAVIGATION IN 1939

war requires a knowledge of the state of navigation at the opening of the

Fliers navigated during the day by comparing their view of the

light bombing. If clouds obscured vision, navigation suddenly became

1.5. ANTI-AIRCRAFT ARTILLERY, 1914–1939

1.5.2. United States

That anti-aircraft artillery was assigned to the Coast Artillery Corps

1.5.3. Great Britain

guns began to improve rapidly as gun-laying radar and better directors

Prototypes of the SC and SG

Failed intelligence. In August 1939 Colonel Wolfgang Martini, Chief of Luftwaffe

Failed intelligence. A German publisher issued annually a pocketbook describing

2.1. ELECTRONIC COMPONENT DEVELOPMENT

2.1.1. Cathode-ray tubes

applied to electrodes in vacuum tubes with effects dependent on the degree

2.1.2. Multielectrode tubes

tween electrodes almost entirely or substantially’ [7]. No use was made of

2.1.3. Electron velocity effects

by conductors connected to the grid. The radiating electrons spiral down

2.1.4. Transmitter tubes

High-frequency operation of transmitter triodes. Inter-electrode reactance and

signiﬁcantly to the heat load. Thoriated tungsten, the discovery of Irving

2.2. BEGINNINGS, 1902–1934

experiments: a spark-dipole transmitter with a cylindrical parabolic re-

2.2.1. The United States

2.2.3. Great Britain

the propagation of pulsed high-frequency waves [21].

2.2.5. The Soviet Union

administrative framework they saw their proposals rejected as unrealistic

2.3. BRITAIN BUILDS AN AIR DEFENSE SYSTEM

Dowding was interested but wanted something more than ‘specu-

Daventry experiment, the creation of equipment required much detailed

3 See ‘Antennas’ and ‘Vertical lobes’ in the Appendix (pp 469–72).