FoMRHI Comm.

2069 Jan Bouterse

Making woodwind instruments

9.9 Frans Brüggen, Hans Coolsma, J.S. Bach and a traverso by Stanesby Junior

This article was originally intended as an appendix to my communications about making a

baroque traverso. But as it happened, I got the idea during the preparations that several
aspects of instrument research seem to fit better into an introduction to the forthcoming
chapters.

One of the problems in my first years (around 1980) as woodwind maker was finding mea-
surements of historical specimens. I had to go to museums or private collections to take
measurements of historical instruments myself and I also investigated copies wherever I
could find them. But there was an additional problem: which instruments to choose for
making a copy? That was especially difficult for the baroque traverso. I was interested in
that instrument but, coming from the recorders, I had almost no experience in playing it.
But then I stumbled on a particular traverso which looked (and sounded) very promising.
The story: Frans Brüggen, who started his career as recorder player and ended as conductor,
has also played baroque traverso (and sometimes modern flute as well). He was also the
proud owner of some fine historical flutes, such as a pair by Godefroid-Adrien Rottenburgh
(and not by I.H. Rottenburgh, which I wrote in 2014 in an obituary to the death of Frans
Brüggen, see FoMRHI Comm. 2012). But he had another interesting traverso, made by
Thomas Stanesby Junior (1692-1754). This instrument in tropical black wood with ivory
rings and a silver key was played intensively by Frans Brüggen and - as I wrote in the afore-
mentioned obituary - it was quite possible that it got one or more cracks during the
recordings (in 1971/72) of the Bach cantata No. 8: ‘Liebster Gott, wann werd ich sterben’
(‘Dearest God, when shall I die’). This cantata has among the most virtuoso and difficult
parts which Bach wrote for the traverso, in the keys of A and E (three and four sharps),
using the full compass of the instrument up to a3. The flute imitates in the opening choir of
this cantata the sound of the death bells. Alfred Dürr wonders in his book Die Kantaten von
Johann Sebastian Bach (Deutscher Taschenbuch Verlag, 1971) that this part might
originally have been written for a high recorder (which then could have been a sixth flute
in d - jan b.). At www.youtube.com/ watch?v=aRmqjW0rbq0 you can hear the original
soundtrack with the Leonhardt Consort, the Choir of King's College Cambridge, conductor
Gustav Leonhardt, and of course Frans Brüggen on traverso (Teldec 1971). And I think that
he played the flute part in this cantata in the most brilliant way, especially in the bass aria.
If Frans Brüggen may have shown any technical faults (nowadays traverso purists will
surely recognise some), then these are blown away in this performance by his great musi-
cality: Brüggen is here in my opinion unequalled by other and more experienced traverso
players on later recordings.
There is, however, a twist to this story: the booklet accompanying the gramophone record
mentions that Brüggen played a traverso of Stanesby Junior for the recording of this cantata.
But there is also a photo of a recording session which can only be that of the opening chorus
of cantata No. 8 (with two oboi d’amore and violins playing pizzicato, see photo next page).
On this photo is Frans Brüggen clearly playing another instrument: a traverso in early
French style. But which one, as most of these French traversos have a very low pitch, not
the a-415 Hz from the recordings with Harnoncourt and Leonhardt?
And I had some more doubts, which occurred to me when I tried to play a copy of this
Stanesby traverso. I will discuss that later on in this article.

Frans Brüggen and the other musicians during the recordings (or more likely: at a rehear-
sal) of Bach’s cantata No. 8.

Hans Coolsma
Hans Coolsma, well known for his recorders, made (c.1970) a series of copies of Brüg-
gen’s Stanesby traverso. It was a one-off production; he never made traversos again. That
was probably the reason he had no objections to giving me (in 1981) the bore
measurements of the original instrument and to investigate a copy of the flute. Many years
later I acquired one of these copies myself, giving me the chance to do more research.

The Stanesby copy by Hans Coolsma (head by Jan Bouterse)

With the list of measurements, I got a copy of the prospectus which came with the traverso.
It gives some detailed information about the instrument and how the copy was made:
The original for this copy, owned by Frans Brüggen, Amsterdam, and used daily by him,
was chosen for the following reasons:
-a): it is in its original state, has not been altered in any way, and is in excellent
condition;
-b): it is a beautiful instrument with a sonourous tone quality; its most noteworthy
characteristics are a powerful e1 and a remarkably pure intonation;
-c): the original was easily and often available for measuring with the result that
throughout the whole process of reconstruction the copy could be compared with the
original at frequent intervals.
The copy only differs from the original - which is tuned at a1=415-419 Hz - in one
respect: the embouchure is widened by 0.5 mm in order to sound the third register more
readily. Frequent testing has proved that when the Coolsma copy of the head joint, incor-
porating the small improvement aforementioned, is used in connection with the original
three joints by Stanesby the combination is even more satisfying.
- The original instrument is made of grenadilla with ivory mounts. The copy is of course
also of grenadilla (Dalbergia melanoxylon). The ivory has not been bleached: it will
gradually yellow in the course of time.
- The original silver key is not ornamented or engraved; this example has been respected.
- The acoustical proportions of the original were copied with extreme precision. It was
with considerable pride that we found that both the timbre and the intonation of every
individual tone and half-tone corresponds exactly to the original.
- This instrument is unquestionably the finest copy of a flauto traverso made since the
reconstruction of baroque instruments was begun again in the present century.
- Each flauto traverso is sold complete with case, cork-grease and chamois leather
cleaner. A fingering chart is added.
Additional information in the prospectus: length of the instrument: 620 mm, four pieces,
cork joints, ivory fittings and silver key; the cork in the mouthpiece has been exactly
adjusted. Base tone: d1. Old pitch: a1 = 415-419 Hz. All measures are original. Tested by
Frans Brüggen and Hans Coolsma.

Thus far Hans Coolsma gives this jubilant description of his traverso project. He also told
me in 1981 that he had copied the bores very precisely (with an accuracy of 0.01 mm),
using a copy-reaming machine that was exclusively designed for this project.
Quite a number of players must have ordered one of these Stanesby copies - I am told that
200 have been sold by Coolsma - but I do not know how expensive the instrument was.
There must, however, have been an issue with the instruments: I do not know any profes-
sional player who has played or used these copies for performances and recordings. And
Brüggen sold his original Stanesby to Mashiro Arita in Japan. It is time to have a closer
look at the copy by Hans Coolsma.

The Stanesby traverso

First impression: the copy is a sturdy looking and weighty instrument, luxuriously made in
black tropical wood with thick ivory rings. Is the original flute indeed made from grenadilla
(African blackwood)? I have always some doubts when other people are so sure to see the
difference with the other option, which is ebony wood. During my study at the Wageningen
University I have done some wood identification and I have seen samples of both species
which
showed a considerable overlap in colour, structure and weight. Oiling the wood intensively
makes it even more difficult to recognise its characteristics. You have to carry out a micro-
scopic test to be sure which is the wood species in question. But such tests are destructive,
because you have to cut off a sample of the wood.

The silver key looks in comparison small and is actually rather short, only 2 mm protruding
over the socket ring. Noticeably, but found on most traversos by Stanesby Junior and some
other English makers, is that the head joint has no socket, but a tenon which fits into the
socket of the upper middle joint. Contemporary flute makers on the continent worked the
other way around. The disadvantage of this design is that it is not ideal to make corps de
rechange, extra joints for playing in other pitches. But as far as I know Stanesby didn’t
make such extra joints.
Quite a lot of traversos by Stanesby did survive, Phillip T. Young in his 4900 Historical
Woodwind Instruments (London, 1993) has listed about 40, of which 25 are made of ivory.
I know of a drawing (by Fred Morgan) of one Stanesby traverso which was made in the
other way, thus with a socket in the head joint. This flute (from an unknown private collect-
ion) is also made of a black tropical wood, has a much shorter lower centre joint and foot
than the ex-Brüggen traverso (230.4 against 239 mm). The total sounding length (from the
centre of the mouth hole to the lower end of the flute) is 569 mm, which is about 12 mm
longer than the same distance of the Coolsma copy, and 17 mm longer than the original ex-
Brüggen Stanesby traverso. This means that the acoustical design is different, the pitch is
also lower, about 10 Cents under a-415 Hz.

The toneholes and their undercuttings

The toneholes on the Stanesby copy by Coolsma are all almost perfectly circular in diame-
ter, with differences up to only 0.2 mm between minimum and maximum diameter. This
seems to be also the case with the toneholes of other traversos by Stanesby Junior. It can be
seen as a sign that Stanesby drilled the holes directly in the desired size, only leaving him
with some undercutting. The sockets and tenons on the copy are all almost perfectly cylin-
drical, the internal diameters of the sockets don’t change from the open end to the shoulder
rim, I have not seen that often on historical woodwind instruments.
More weird, however, is one of the fingerholes on the Coolsma copy. On most baroque
traversos we see the that hole 1 and 2 on the upper middle joint have about the same size,
hole 3 is somewhat smaller; similarly hole 4 and 5 on the lower middle joint are again
about equally sized with hole 6 being clearly smaller. But hole 4 on the Coolsma copy with
a diameter of 5.0 mm is much smaller than usual and has about the same size as hole 6; the
diameter of hole 5 is 6.2 mm.
This has repercussions for the pitch of some notes. A disadvantage on my copy is that a#2
(with 1. 3) is very flat and hardly usable, which is a problem for playing music with three or
more sharps (as on the aforementioned Bach cantata!). This a#2 is in baroque temperament
always a bit flatter than the b-flat2 (fingered with 1 2 . 3 4 5 6 7), but surely not so much as
on this instrument.
What was going on here? I was also wondering about the undercuttings of some of the
toneholes, which looked to me only slightly done (holes 3, 4, 6 and 7). When I saw these
features, I became more interested in the original instrument. I asked around and one of the
members of the ‘Bouwerskontakt’ was so kind to send me a drawing of the Stanesby traver-
so, made by Jean-François Beaudin (after measurements and a sketch by Fred Morgan). And
it became immediately clear to me that Coolsma had not only made changes to the mouth
hole, but also to some other important elements of the flute. So was the upper middle joint
about 5 mm longer and even more important: hole 4 was much bigger: Ø 5.9 instead of the
5.0 mm of the copy. The other toneholes on the Stanesby had about the same dimensions as
on the copy, and were also almost perfectly circular in diameter.

The bore profile

It is time to look at the bore measurements of both instruments. The graph on the next page
presents the bore profiles of the flute, as presented to me by Hans Coolsma. Only the bore
of the head joint is taken from his copy, and so are the positions of the toneholes.
The bore profiles have a common pattern: the bore is in the head (I) cylindrical (Ø 19.2 to
19.3 mm), in both middle joints (II and III) narrowing from 18.8 to 15.5 and from 15.4 to
13.7 mm, in the foot (IV) narrowing from 13.7 to 13.2 mm and at the lower end widening
again to 13.7 mm. There are at the connections between the joints no, or only minor, steps
in the bore diameters. That can be conceived as a sign that there has not been fiddled with
the instrument, such as shortening joints.
Remark: I have seen some very fine baroque traversos which for some reasons have
rather wide steps in the bore between the parts, for instance resulting in a profile where
the bore diameter at hole 4 is wider than at hole 3. I will discuss this phenomenon in a
forthcoming article.
Characteristic for the Stanesby traverso is that the bore in the lower part of joint III is hardly
tapering, it is almost cylindrical over a length af about 40 mm. And also that Stanesby
probably used the same reamer from both ends of the foot (mirrored profile). The narrowest
point of the bore of the flute is thus halfway along the foot, and not as I have seen on other
instruments at hole 7 or even higher, at the connection of the lower middle joint and the
foot.
The walls of the flute are fairly thick: at hole 2 about 4.1 mm, at hole 5 about. 5.3 mm
(this difference between the wall thicknesses between the two joints can be seen on many
more baroque traversos). For a traverso with the supposed pitch of a-415 Hz the bore is
rather wide. But it looks to me as if this design of the bore profiles is well thought-out by
Stanesby. There are some differences from the measurements (taken by Fred Morgan) of
the original instrument:
- the bore of the head is with Ø 19.55 mm a bit wider (on the copy: Ø 19.2 to 19.3 mm)
- the bore of the upper middle joint is 0.2 to 0.4 mm wider
- the bore of the lower middle joint is only in some places a bit wider (0.2 mm)
- the bore of the foot is only in the upper section a bit wider (0.2 mm)
These differences are probably caused by another way of measuring. Morgan might have
given the maximum diameters where the bore has come slightly oval in cross section. But
the general pattern of the profiles is not so much different. However, I suspect that Coolsma
has made the bore of the head joint on his copies deliberately narrower. I have no reason to
think that, for instance, the bore of the copy did shrink so much.
 ....-

-
i. ...........
1

15

10
L:O 50 100 150 200
Relation between size and position of the toneholes
There is a general rule for the size and position of toneholes on a traverso. Because on these
instruments all primary tones (the first harmonics) can be overblown into their octaves (the
second harmonics), the position of the holes is established for a given bore profile.
For instance: placing a hole too low (in the direction of the foot) means that the octave
interval which is tuned at that hole has the tendency to be too sharp. It means also: when
you change the bore profile, you must inevitably adapt the position and/or size and under-
cuttings of the toneholes (and vice versa). This means also that it is very strange when on a
copy a tonehole is much bigger or smaller than on the original instrument.

This brings me to the deviating size of hole 4 on the Stanesby copy. Hole 4 is important for
tuning g1 (1 2 3), g2 (1 2 3), g3 (1 . 3) and d3 (as the third harmonic of the g1, with the
fingering . 2 3 . . . 7, thus with the key pressed, which means that hole 7 is opened).
Making this hole too big (or undercutting too much) destroys the balance between these
tones, the intervals becoming too wide. The balance on the copy for these tones and inter-
vals is all right, so hole 4 seems to have the right size and shape for this instrument. But
making this hole bigger for a better a#2 (1 . 3), things will become worse for the other
tones.
But what could Coolsma have brought to change the size of hole 4? He claimed that the
original flute had ‘a remarkably pure intonation’. I already mentioned that Coolsma altered
not only the size of the mouth hole, but made the second joint c. 5 mm longer.* The only
reason for this can be to lower the pitch of the flute, I suppose to bring the pitch closer to
a1=415 Hz. But that change in length can’t explain the alteration to hole 4, nor does that the
difference in diameter of the head joint. I have made several corps de rechange for other
traversos and have never met the need to make adaptations to the holes on the right hand
joint. * Coolsma added these 5 mm in the upper section of the second joint: the
distances between the toneholes (from 1 to 6, or from 3 to 4) stayed almost the
same. It was maybe better to move holes 1, 2 and 3 a little bit upwards. But
even that can’t explain the events with hole 4.

Conclusion: Coolsma must have had another reason for to make the copy as it is. I had a
discussion with Tanja Obalski - she is a professional soprano singer - who did a course in
Stapleford in making a copy of a Stanesby traverso. Robert Bigio, who was the teacher at
that course, told her that on several original Stanesby traversos he had played the notes g1
and g2 are very sharp. I asked him about that and he answered me:
I have played a number of Stanesby flutes. They are remarkably consistent, and they all
have the same flaws (but I use the word 'flaws' advisedly): the G is indeed far too sharp,
and the D is far too flat. My friend Helen Crown wrote her PhD thesis on Lewis Granom,
an 18th-century London flute player who is known to have played a Stanesby flute. Helen
tells me that Granom wrote some low C sharps in his music, which can only be played if
the D is already very flat. One effect of the sharp G is that the only way to get C natural
in tune is to finger it with the addition of two fingers in the right hand. Making it smaller
allows you to play C natural with the standard fingering. Most players have found these
aspects of the Stanesby to be impossible. I have cheated by making the G hole smaller
and have shortened the footjoint to raise the D.
It is quite possible that exactly these problems were the reason for Coolsma to make altera-
tions to the copy: to lower the g in all registers. But doing so he had to cope with the pro-
blems to restore the balance with the other tones.
We also discussed the other qualities of the Stanesby copy. Has it indeed what Coolsma
said, ‘a sonourous tone quality and a powerful e1'? And what exactly is ‘sonorous’? The
dictionary says: ‘a deep and rich sound’. But my personal view on the instrument is that
you must put a lot of energy into it to come a bit close to that type of sound. I feel quite
much resistance, the response being rather slow. The sound is less free and more delicate
than on several of my boxwood traversos, for me not at all corresponding with the weight
and appearance of the flute. The e1 is not particulary powerful as well. I know very well
that other (and much more experienced) players may have other opinions about the
Stanesby copies by Coolsma. But my conclusion is - and Simon Polak, who is a
professional Dutch traverso maker, agrees with me - is that there is something wrong with
these copies. They are surely not exact copies of the original Stanesby traverso, and I doubt
some of the statements by Coolsma about their qualities as well.
Now I asked Mashiro Arita about his opinion about the ex-Brüggen Stanesby,now his. He
answered me (and made excuses for his very poor English):
So, I could to say that it is possible to play the instrument very good in tune from bottom
(1st octave) of d' to the top of register with 18-19 and ca.1940's. without any trouble and
we can make very deep sound, colorful and rich sound!!!
Also to say it is not necessary to arrange for tone -holes sizes, bore and shortened foot
joint.etc..if we can have good condition's original old instruments!
I hope that you could imagine what's a problem on the flute playing is, on the old
flute I think the pitch of original Stanesby black one is c.415 with on my playing and
no retuning all tone-holes. ... And I believe that Brüggen played on original Stanesby
for Cantata BWV 8 by Bach, that he told me in that time. He played super excellent!!!!

Three Stanesby traversos from the collection of Mashiro Arita.

Well, since everything Robert Bigio and Mashiro Arita assumed about the Stanesby
traverso is true, what Frans Brüggen achieved playing this instrument in the Bach cantata is
even more remarkable. He had, however, one advantage: because the three and four sharps
he had mainly to play g# and not g, and the d1is completely absent in his parts.
At the end of this quest I should do one job: to make an exact copy of the Stanesby flute.
Two new middle joints should be enough to prove what Bigio told me. But I am a bit
reluctant to make them; what to do with an instrument which is of no use to me?
Drawings and measurements of traverso’s by Stanesby Junior

I have made a drawing of the copy by Hans Coolsma of the Stanesby Junior traverso (ex-
Brüggen, now Arita). I added measurements to scans and photos of the instrument parts, see
example below. I can send members of the FoMRHI a digital copy in colour of the full
drawings (email me: info -at- mcjbouterse.nl).

There are various plans with measurements of several other traversos by Stanesby Junior in
circulation. Some of them are published by the collections, it is even possible to download
some of the drawings from the corresponding websites.
See http://www.flute-beaudin.com/Plans.htm for the list of drawings you can order from
Jean-François Beaudin. No. 15 on this list is the ex-Brüggen traverso, now in the collection
of Masahiro Arita in Japan. No. 104 is in the Dayton C. Miller Collection in the Library of
Congress in Washington DC (USA), inv. no. DCM 1125. There is a plan of this instrument
(also by Jean-François Beaudin), and downloadable on line at www.loc.gov/collections/
dayton-c-miller-collection/?q=stanesby+flutes. There are several more Stanesby traversos in
this collection (some of them damaged, or with later added keys).
Another traverso by Stanesby Junior is in the Musée de la Musique in Paris, a drawing
(again by Beaudin) is available, see http://collectionsdumusee.philharmoniedeparis.fr/doc/
MUSEE/0162306 for pictures and more information.

Epilogue
A last word about Frans Brüggen and the flute he played on the photo. I did some research
and found that Friedrich von Huene made a traverso after an instrument by Chevalier (in the
Museum of Fine Arts in Boston), which he sold in two versions, one at the original pitch of
a1=410 Hz, and the other in a1-415 Hz. See picture below, from a prospectus by Friedrich
von Huene (I am not sure if the Von Huene workshop still makes this flute). See www.mfa.
org/collections/object/flute-50386 for a picture and more information about the original
traverso by Chavalier.

I am pretty sure that Brüggen played this Chavalier copy during the session on the photo
(see second page of this article), but as I have to believe Mr. Arita, he did not so at the
definitive recording. But if he had played that instrument, it would actually not have been a
bad choice. It is said that a French flutist (Buffardin) has inspired Bach to write some of
finest works for the flauto traverso when he was visiting Germany.
 hole 7 at L 37.5, Ø 5.7x5.7, undercut to L 6

Stanesby Junior - traverso, copy by Hans Coolsma

from instrument in collection Frans Brüggen (now Arita)
measurements in millimeters, by Jan Bouterse
FoMRHI Comm. 2068 Jan Bouterse

Making woodwind instruments

9: Flutes (traversos)

9.1 What is a flute?

About the name: the term ‘flute’ is in English language generally used for the transversely
blown instrument with a mouth hole, also called the embouchure hole. The term
‘embouchure’ is sometimes used for the mouth hole on itself, but means actually the
position and shape of the hole between the lips of the player in relation to the mouth hole.
It is likely that that the name ‘flute’ comes from the Italian flauto. The same applied to
other European languages: Flöte in German, flûte in French, fluit in Dutch. But in the 17th
and 18th centuries these terms were used in these countries, in the first place and for a long
period, for the instruments which are called ‘recorders’. And ‘flute’ in English can still (or
reviving historic useage) sometimes mean a recorder as well. With the reintroduction of
historical instruments in the 20th century some original names also reappeared, such as for
the sixth flute (soprano recorder in d2) and the voice flute (tenor recorder in d1).

This is a small flute (piccolo) for children. It is made from plastic and has - as a recorder -
seven fingerholes on the front and a thumb hole at the back. The mouth hole is about the
same shape as that on a modern Böhm flute.
There is even the possibility to play this flute with a short
windway device which directs the breath at precisely the
right angle. But with this tool you can’t vary that angle and
so you cannot so easily influence the pitch and other aspects
of the sound as well. That is probably why there are also
double holes on this flute, to avoid fork fingering for some
notes.

The term flauto traverso was often used in the 18th century for the wooden or ivory flute
with one key or, later in that century, more keys: four or six. Today many players use the
abbreviated term ‘traverso’ for a baroque-type instrument, and so will I on these pages.
It is perhaps not quite correct to use the term ‘traverso’ for older types of flutes, the instru-
ments with a cylindrical bore and without a key, which were played well into the 17th
century. Similar types of flutes with a cylindrical bore and no keys are played in other
cultures, such as bamboo flutes, for instance in Japan. Some flutes are not, or not entirely,
held in a transverse position, for instance some modern bass flutes.
Modern concert flutes (Böhm flutes) are classified as woodwind instruments, despite the
fact that most of them are made of metal. Other materials were used as well for flutes and
traversos: ivory, crystal glass, and plastic. Comprehensive information about the history
and use of flutes can be found in several encyclopaedia and on internet.
9.2 Some thoughts about making traversos
In this series on making woodwind instruments, I will give inform-
ation about making the flutes as they were played in the 16th to 18th
centuries - thus renaissance and baroque traversos, made of wood or
(artificial) ivory. I must begin with a warning: I have never played an
original old renaissance flute (I even don’t know if any specimens did
survive which can be played). And I am very not an experienced
player on this type of flute. I only have made some copies and done
some experiments. So I can tell you that traversos are at first glance
simple instruments to make. You have not - as on a recorder - to cut a
windway and make a block, nor you have the difficult job of learning
- as for an oboe or bassoon - how to make a reed and staple. It all
seems simple: there is the inside (the bore) which is made by drilling
and reaming; and there is the exterior which is shaped on a lathe. And
then you have to drill a mouth hole and some fingerholes and put a
cork in the bore not far from the mouth hole. But that simplicity is
deceptive. For a start, it is not so easy for everyone to produce a
sound by blowing over the edge of a mouth hole. It took me several
days to get it more or less right when, long ago, in 1976, with only
some experience as recorder player, |I bought a baroque traverso, a
cheap instrument made from fruit wood which came from the former
German Democratic Republic. And I had a problem when I gave a
course in traverso making , ten years later, when none of the five
students were able to produce a sound either. I then had to do all the
voicing and tuning myself. My advice, if you have no experience in
flute playing: take a bottle (Coca Cola or what ever), close your lips
leaving a small hole in the centre and blow over the rim of the bottle.
Adjust the direction of the air stream if there is no direct sound. For
playing a traverso you need a very similar technique.
One step further is making a flute out of a plastic tube. My first one
was made of PVC tube bought in a DIY shop. I could find only one
size, with an internal diameter of about 25.5 mm, external diameter
32 mm. That is too wide for a flute with the fundamental d1, but with
a length of 635 mm you will get the note b (= a minor third lower
than d1) which makes it a flauto d’amore. The position of the
fingerholes was derived from a photo in a book. Important: the
mouth hole must be kept small: 9 mm circular. It is one of the
mistakes of some people who are experimenting with making flutes
from plastic tubes, that they have the much wider and slightly square
size of the mouth hole of the modern concert flute in their mind. It is
much easier to get a nice sound from a smaller circular hole! The
hole must also be undercut a little bit, the rim must be finished
properly with a clean edge.

Left: measurements of my PVC-flute. The actual instrument was

made by me in two parts, connected with a coupling (also available
in the DIY-shop). Advice if you want to make a copy: take a slightly
longer piece of tube and drill all holes at first a bit smaller. Adjust
everything when you are voicing and tuning the flute.
PVC is perhaps not an attractive material, but easy to work, for instance to drill holes in.
Formerly you could only buy tubes in a dull grey colour, nowadays also in a more
interesting white, but only in a restricted range in diameters. A tube with an internal
diameter of 17 mm or 18 mm and a wall thickness of 2.5- 3.0 mm should come closer to
historical instruments, but I am afraid you must order that from a specialised company,
which might be expensive. I have tried also acryl glass, available in more diameters, but
these tubes are generally thinner and acryl glass is also much more brittle.

A different approach to music

There is another difficulty in making the traverso: you have to learn the appropriate finger-
ings. There are similarities, but also important differences from the fingerings for recorders.
Traversos have no thumb holes; overblowing the notes of the lower register must be done
by blowing harder and slightly changing the direction of the breath. You can change the
volume of the sound of the traverso as well, but then you have to make pitch corrections:
turning the flute inwards makes the opening of the mouth hole smaller, which makes the
note flatter; turning it outwards makes the note sharper. Such corrections are also needed
for some notes which are (surely for our modern ears) a bit out of tune.
Playing the traverso means that you have to learn about the differences in character of the
keys (modes) in which the music is written. That means that you must develop a new way
to look at music, hearing in a different way. We are so lucky now that you can find instruct-
ions and fingering tables from old sources in reprints, such as the books of Hotteterre (Prin-
cipes de la Flûte) and Johann Joachim Quantz (Versuch einer Anweisung die Flöte traver-
siere zu spielen) and these are also available in English.

9.3 Historical renaissance flutes

Flutes with a cylindrical bore, with six fingerholes and without a key had a long-lasting
tradition in Europe. The instrument were played in intimate settings as well as in armey
bands. Renaissance traversos were made in different lengths (consorts). At
http://www.flutehistory.com/Resources/Lists/Renaissance_flutes.php3 you can find a
comprehensive list of early flutes. There are two famous collections of these instruments,
both in Verona in Italy: in the Bibliotheca Capitolare and the Accademia Filarmonica.
See for more information Filadelfio Puglisi, 'A Survey of Renaissance Flutes', in the
Galpin Society Journal 41 (1988), p. 67-82. See also his book I flauti traversi
rinascimentali in Italia (Studio per Edizioni Scelte, 1995). Puglisi published in his book full
data of the flutes in Verona - but please be aware that he used for all instruments the same
schematical drawing, to which he added the measurements (see example on the next page).
There are roughly two sizes of flutes in the Verona collections: tenors with a sounding
length (= the distance from the centre of the mouth hole to the lower end of the flute) of
around 540-575 mm, and bass flutes with a sounding length of around 810- 860 mm. He
gives no information about pitches and the other playing aspects of the instruments.
Drawing on the following page: renaissance flutes often have mouth holes which are
slightly oval, with the longest diameter slightly diagonal to the axis of the instrument.
Puglisi doesn’t give measurements of the undercutting of the mouth hole and the
fingerholes. The holes are generally drilled in a straight line, which allowed the
players to hold the flute to the left or to the right. Yet it creates no acoustical
problems if you drill some of the holes slightly off-line.
Drawing from Filadelfio Puglisi, I flauti traversi rinascimentali in Italia (Studio per
Edizioni Scelte, 1995).
 The Nova Zembla flute

This is a picture of another historical renaissance flute.

It was found on Nova Zembla (Novaya Zemlya), an
island in the Arctic Ocean in northern Russia and the
extreme northeast of Europe. In 1596-7 Dutch sailors
tried to find a route with their ships to the East Indies,
but were trapped by the ice and had to stay there for the
whole cold winter. Several artefacts have been found
where the sailors made a house. One of these was this
flute which - apart from a missing piece of wood at the
lower end - has survived rather well. The owner put his
mark with a knife at the front of the flute, which is
probably made of fruit wood.
The instrument is now in the Rijksmuseum in Amster-
dam (see https://www.rijksmuseum.nl/nl/zoeken
/objecten?q=Nova+Zembla++fluit&p=1&ps=12&st=O
BJECTS&ii=0#/NG-NM-7692,0 for a colour photo.

The bore of the flute is cylindrical, but because of some

cracks and distortion not easy to measure, Ø 17.7 mm at
hole 5. The wall is very thin; about 1.8 mm at the upper
end, 2.4 mm at the lower end.
The mouth hole is 7.2 x 8.2 mm. The fingerholes are
not much undercut. Some measurements:
total length: 639; SL (sounding length, from centre of
mouth hole to lower end): 535

hole A B C
mouth hole 0 23.0x21.8 7.9x8.2
hole 1 235 23.2x22.1 6.3x6.7
hole 2 272 22.2x21.0 6.5x6.9
hole 3 306 23.0x21.5 6.4x7.0
hole 4 355 nm 6.4x6.8
hole 5 390 23.2x21.8 6.5x6.9
hole 6 427 nm 6.7 (damaged)
A: position of the hole - distance from centre hole to
center of mouth hole
B: diameter (min/max) of the flute
C: diameter (min/max) of the hole
nm: not measured
I have made a copy of this flute in palissander wood, with a bore
diameter of 17.5 mm. It plays at a pitch of a-435 to 440 Hz (depend-
ing on how much you cover the mouth hole), the lowest note is d1.
Maybe that with a slightly narrower bore the pitch will be higher.
The cork is set by me about 13 mm from the centre of the mouth hole.
My first wooden renaissance flute
Before this copy of the Nova Zembla flute, I had made a few instruments with thicker walls
(exterior diameter about 27 mm), based on a copy of unknown origin. I made these also in
two joints, with a socket on the head and a tenon on the lower joint. That allowed me to
make the bore in the top of the lower joint a bit wider over some distance: that made it
easier to tune the a2 and b2 (becoming a bit sharper). This thicker flute plays (for me)
m o r e easily than the Nova Zembla copy, the pitch is also a bit higher, a-440 Hz. Some
measurements in mm: Sounding length (SL): 519; head L 255, SL 186; socket: L 29.5, Ø
25. Bore: Ø 17.5, but 18.0 over about 50 mm in the lower section the head. Lower joint L
without tenon: 333, tenon L 29. Ø bore: 17.5, but up to 18.0 over about 100 mm in the top
section and up to 18.0 over about 100 mm in the lower section. Mouth hole: 8.5x8.8 mm,
Ø wood 26.6; tone holes (L to centre mouth hole, Ø hole, Ø exterior) hole 1: 222, 7.0x7.4,
27.2; hole 2: 258, 7.0x7.2. 27.0; hole 3: 293, 6.5x6.6, 26.9; hole 4: 346, 6.5x6.7, 26.8; hole
5: 381, 7.7x72, 26.6; hole 6: 415, 5.7x5.8, 26.3. All holes are moderately undercut.

9.4 Some practical tips for drilling and turning a renaissance flute

Drilling a hole with a length of over 600 mm is not easy. See part 7 of this series (comm.
2060 in FoMRHI Q 136) for some methods and tools. It is important to take a piece of
wood that is thick enough: if the drill goes slightly off line, you may still have enough milli-
meters leeway in the wall to rescue yourself.
For my Nova Zembla copy I had a piece of Santos palissander (surely not authentic for this
flute, but very practical) that was about 40x40 mm thick. I bored a pilot hole with a diame-
ter of 10 mm, but the drill was far too short, I had to put it on an extension piece for flat
bits. But before I could use that I had to
widen the bore as far as possible, with a
17 mm flat bit on which I had put an end
piece with a pilot (photo, right), again
mounted on an extension piece. There was
the danger that parts of this combination
might come loose and be stuck in the
wood. It would have been much easier to
buy a steel rod and grind it to a d-bit drill
(other photo). After this drilling
adventure I had to ream out the bore to 17.5 mm. That must be done before turning the flute
on the lathe: the wall will become so thin that reaming is then very dangerous.

Turning such long and thin pieces of wood is another issue. It is inevitable that the wood
will vibrate, which will result in ripple marks that destroy the surface. You must somehow
support the wood, giving it counter pressure close to where the chisel does its work. See
Comm. 2056 in FoMRHI Q. 135 for some solutions to this problem. I can tell you that I had
to use a lot of sand paper to get a clean surface.
It is of course not always necessary to use a lathe to get a round flute. As most renaissance
flutes have a simple smooth surface, without decorative patterns (rings, rims, grooves or
whatsoever), you can plane the wood into its round shape. It is not unlikely that this techni-
que was used in the past, also for instance for making crumhorns (which are made from
straight pieces of wood; the bending is done in a later phase, after the finishing of the bore
and the surface).
9.5 Fingerings for the renaissance flute in d1

d1 1 2 3 4 5 6
d2 . 2 3 4 5 6
e1 and e2 1 2 3 4 5
f1 and f2 1 2 3 4 . 6 - turn flute inwards or lower breath pressure to make this
note a bit flatter
f#1 1 2 3 4 - turn flute outwards to make this note a bit sharper
or: 1 2 3 . 5 6 - turn flute a bit inwards
f#2 1 2 3 4 - turn flute a bit inwards
or: 1 2 3 . 5 6 - turn flute a bit inwards
g1 and g2 1 2 3
g#1 1 2 . 4 5 6 - turn flute a bit inwards, if necessary
g#2 1 2 . 4 - turn flute a bit inwards
a1 1 2 (6) - hole 6 might be covered for a better stability
a2 1 2
or: 1 2 . 4 5 6 - this fingering is a bit sharper than the other one, and
sounds louder; it is actually the third harmonic, an
octave + fifth above the d1
b-flat1 1 . 3 4 (5) - covering hole 5 lowers this note a bit
b-flat2 1 . 3 - is often rather flat
or: 1 . 3 4 5 6 - is often rather sharp, turn flute inwards
b1 1 . . . . (6) - hole 6 might be covered for a better stability
b2 1 . . 4 5 6 - covering 4, 5 and 6 makes this note a bit sharper
(otherwise the b2 is a bit flat)
c2 . 2 3 . . (6)
or: . 2 . 4 5 6 - may give a clearer sound
c3 . 2 . 4 5 - other fingerings possible
c#2 . . . (4 5 6)
c#3 . 2 3 4
d3 . 2 3 4 . 6
e3 1 2 . . 5 6 - higher notes (third regsiter) are possible, for instance
g3 with 1 . 3

These fingerings are based on my experiences with the copy of the Nova Zembla flute.
Some notes tend to be sharp or flat, but sometimes it is supposed that they are sharp or flat
in meantone tuning, when compared with equal temperament. Fork fingered notes often
sound a bit muffled, turning the flute inwards or outwards changes not only the pitch but
also the quality of the sound. See for further fingerings on internet:
http://www.oldflutes.com/charts/ren/

On Youtube there is a film (http://Youtube.com/watch?v=-sGJiNXl_9g) with a superb

demonstration of the possibilities of the renaissance flute: Hans-Joachim Fuss playing a
recercata by Aurelio Virgiliano on a flute after Bassano (made by Philippe Allain-Dupré
from Paris). There are no muffled sounds at all, and especially surprising is the third register
of this instrument.
9.6 Tuning the renaissance flute

a - some theory
The d1 (the first harmonic) cannot be tuned on one of the fingerholes: the only way to
change its pitch is by widening the bore at the lower end of the flute, or shortening the flute
at that point. For d2 (as the first overtone, or second harmonic), and even more the d3 (the
fourth harmonic), we need to push the cork into the right position. Moving the cork closer
to the mouth hole makes the octave interval d1 to d2 wider, and does that even more so
with the interval d1-d3. For playing d2 you have to open hole 1, and for d3 hole 5 as well.
Opening these holes (and blowing harder) makes that in the bore of the flute new antinodes
can be formed, which correspond with the harmonics.

This picture with the approximate positions of the nodes (N) and antinodes (A) of the
fundamental d1 and its overtones makes it clear.
d1: fundamental (first harmonic) d2:
first overtone (second harmonic)
a2: second overtone (third harmonic)
d3: third overtone (fourth harmonic)
The upper (most left) antinodes in the picture are placed several millimetres above the
mouth hole. Are they really there? In traversos the mouth hole is much smaller than
the diameter of the bore at that place. That means that at that spot there is no equali-
sation of pressure. This antinode is actually a fictive antinode, which is supposed to
be (if you want to do calculations etc.) around 40 mm outside the mouthhole (see
Otto Steinkopf, Zur Akustik der Blasinstrumente, Moeck Verlag 1983). At the lower
end the soundwave is also a bit longer than the length of the tube (about 6 mm).
We see the importance of opening hole 1 for producing d2 and d3. Opening hole 4 just
helps to produce the a2, played as second overtone of the d1, fingered 1 2 3 . 5 6. Opening
hole 1 destroys however this a2, because it needs there a node of the sound wave. When
these holes have the function for overblowing the note in one of the upper harmonics, the
size of it has no influence of the pitch of these notes.
The other (and easier) a2 is played as 1 2, as the first overtone (and second harmonic) of
the a1. But the nodes and antinodes of the a2 with this fingering is quite different com-
pared with the third harmonic of the d1!

b - about tuning holes

Tuning the following notes implies that the holes on which they are tuned must be drilled in
the right place and have the right size; unless you are very sure and accurate: drill all holes
initially 0.5 to 1.0 mm too small. Doing that, it is better to put a piece of wood in the flute
that fills up most of the bore, this to prevent that the drill causes splinters when it breaks
through the wall.
Another problem is to get the fingerholes precisely at their positions on the flute,
preferably on a straight line. What I normally do is to put a piece of writable adhesive tape
over the
length of the flute, draw a straight line on the tape and indicate on that line the positions of
the holes. After checking the distances, I pierce small points on the hole positions. Another
method is to make a template (in metal or plastic) in which the hole positions are drilled,
and attach that to the flute.

These are the tuning holes:

- d1, d2 and d3: no tuning hole (or the open end of the bore is to be regarded as a hole)
- e1 and e2: hole 6
- f1 and f2: hole 5, together with and f#1 and f#2
These notes have to be corrected with your embouchure: f1 and f2 (1 2 3 4 . 6) are
always too sharp, and f#1 and f#2 (1 2 3 4) too flat with the usual fingerings. Both
groups of notes are tuned on hole 5. Making that hole bigger means that these notes
all rise in pitch, with hardly any effect on the difference between f and f#.
- g1 and g2: hole 4
- a1 and a2: hole 3, together with g#1 and g#2
- b1 and b2: hole 2, together with a#1/b-flat1 and a#2/b-flat2
- c2: hole 1, together with c#2
- higher notes can rarely be tuned at the fingerholes; you must know to which lower harmo-
nics these notes are related.
On all holes of the flute you can overblow the related note into the corresponding second
harmonic, that is an interval of an octave. The same applies to the baroque traverso.
But there might be problems on longer types of renaissance flutes to overblow all the
notes of the first register into a useful octave interval. On these instruments you must then
use alternative fingerings.

c - tuning rules
Tuning means enlarging the fingerholes which have been initially drilled a bit (for instance
1 mm) too small. But the cork position must be absolutely correct before you do that: the
notes d1, d2 and d3 must have the correct mutual relations. When tuning a baroque traver-
so, you can make some adjustments to the bore profile; this is not, or only in a very resticted
way, possible for a renaissance flute with its cylindrical bore.
As we tune the flute from hole 6 to hole 1, we have to be especially careful to the get the
first notes right, after the d1/d2/d3 that e2/e3, f and f#, and so on. We saw in the list above
that on each hole more than one note has to be tuned. Most important is that the octave
intervals (e1 to e2, f1 to f2) are correct. We already knew to begin with the right position of
the cork. Before you start with the work on the tuning holes for the other notes. But playing
octave intervals correctly depends also on your embouchure and breathing technique. Be
aware that you can change the pitch of most tones with your embouchure up to about a
quarter of tone (20 cents or more).

The main rules:

- a flute without fingerholes is pitched several cents higher than when the holes are drilled
and closed by your fingers (this because the empty hole spaces in the wall of the flute add to
the volume of the bore of the instrument);
- making a hole wider means that the related notes become sharper; but the note of the
second register (the first overtone) responds more to the size of the hole than the note of the
first register; and it does that even more so when you enlarge a hole in the direction of the
lower end of the flute.
- formulated in a different way: when a hole influences a note and its octave, the position
will have more influence on the lower note, while the diameter will have more influence on
the upper note;
- that means that you have a problem when as at the beginning the e2 is already correct, but
the e1 still too flat; in that case the position of hole 6 is wrong: it should have been drilled
higher on the flute (more close to the mouth hole);
- sometimes a correction is possible by filling the hole in (with wax, nail polish etc.) at one
end, and filing out at the other side;
- enlarging a hole has a quicker (sharpening) effect on the fork-fingered note than on the
simple fingered one (for instance: the fork-fingered f with 1 2 3 4 . 6 and f# with 1 2 3 4)
- undercutting a hole has more or less the same effect as giving the whole a bigger diameter,
but has a bit stronger effect on the fork-fingered notes;
- enlarging a hole has a strong effect on its related notes (for instance hole 4 for g1 and g2),
but much less so for higher notes (for instance hole 4 for a1/a2 and b1/b2);
- enlarging and undercutting a hole means that the bore profile of the flute in that place is
changed: that might have some influence on notes of the second and higher registers; to find
these relations, you must make a complete scheme of the nodes and antinodes of all tones of
the instrument (see part 3a of this series: Practical acoustics for woodwinds: sound waves
and tuning, Comm. 2040 in FoMRHI Q 132).

d - tuning techniques
The French recorder maker Philippe Bolton gives on its website interesting information
(also in English http://www.flute-a-bec.com/accordgb.html) about the tools and techniques
of enlarging and undercutting fingerholes. He recommands as the safest tool for enlarging
holes a coarse round file, because fine files are not efficient in wood. Finding the right type
of files (which are not too coarse and have the right diameters) is, however, not easy. I am
always nosing around for these tools.
An alternative is working with a piece of sandpaper (or better, emery cloth) wrapped
around a small stick. You can buy that in several qualities, from fine to coarse. I work often
with emery cloth with a grid of 150, for finishing a higher number is needed.
Most professional woodwind makers have one or another type of undercutting knife.
Bolton writes about that tool: this technique needs learning. A 3 mm double-bevelled blade
is perfect for this. The top and bottom areas of the hole must be cut separately, beginning on
one side and working around to the other, stopping before the blade gets parallel to the
fibres, when it will start splitting the wood instead of cutting it.

Undercutting knife of
Philippe Bolton; with a
cross section of the blade
of one of my own under-
cutting tools.

Always start cutting at the top or bottom end of a hole, working around to the side. That
means that you begins with cutting straight through the fibers and stop before the blade gets
parallel to the fibres, where it will start splitting the wood instead of cutting it.
Some instrument makers use a Dremel mini-drill, with a small milling fraise. But you
have to be very experienced to make a regularly shaped undercutting with such tools.
9.7 The Schweizer Pfeif
‘Schweizer Pfeif’ (Swiss pipe) is in German language the name for the piccolo flute as it
was played in the bands of Swiss soldiers who fought as hirelings in foreign armies.
I do not know about finds of early small flutes from the 17th century and earlier, there are
some baroque types with one key from the 18th century.
This piccolo flute is still used today in bands and had also found his way to the modern
orchestra, after it had undergone the same development as the longer flutes: with a bigger
mouth hole and a key system.

I have made a simple Schweizer Pfeif, the upper one

in boxwood, the lower one in elder wood. The funda-
mental tone is d2, the pitch a-440 Hz, the bore is cyl-
indrical (Ø 12 mm), the cork is set at about 12 mm of
the centre of the mouth hole. All holes are widely un-
dercut. Hole sizes (max/min): mouth hole: 7.2x7.2;
hole 1: 6.3/6.6; hole 2: 6.6/7.3; hole 3: 6.4/6.5; hole 4:
6.2/6.4; hole 5: 6.2/6.3; hole 6: 5.5/5.6.
The double holes make the instrument more practical
for modern music (and for modern ears): you can play
now d#/e-flat with 1 2 3 4 5 6a. The f2 and f3 are
played 1 2 3 4 . 6 (fork fingered), whereas for f# you
have to cover only the bigger of the two holes at posi-
tion 4 (1 2 3 4a), making it a bit sharper than the flat
mean tone f# with hole 4 fully covered. There is also
an easier g#/a-flat with 1 2 3a. The b-flat3 sounds best
with 1 2 . 4 5 6a.
Elder wood is surprisingly useful for smaller flutes
as you can drill a hole through the soft pith in the
centre of the stem (or branches). The problem is to
find pieces which are long and straight enough for
this flute. I oiled the wood heavily with linseed
oil; the flute sounded nicely.
Some measurements of the double holes (in fact:
the diameter of the drills which I used, before
tuning with files): hole 3a: 4.5 mm; hole 3b: 3.2
mm; hole 4a: 4.5 mm; hole 4b: 2 mm; hole 6a: 4
mm, hole 6 b: 3 mm.
One warning: playing the piccolo too much can harm your ears!
9.8 Some conclusions

The renaissance type of flute with a

cylindrical bore and 6 fingerholes is
in interesting instrument for the
amateur woodwind maker. It is not
too difficult to make and you have
not to buy expensive tools.
However, playing these flutes is a
different matter, for instance when
you are accustomed to the modern
Böhm flute. It is not so much about
the new fingerings which you have
to master: the embouchure you have
to use is really different. You must
not fill the tube with energy as on
the modern flute. With traversos
(from renaissance and baroque)
you have to develop another, more
intimate type of relation: it is more
a question of listening to the instru-
ment, to what it gives back to you.

I certainly believe that the

renaissance type of flute has great
possibilities for modern music.
You can play the instrument in
equal temperament with some
simple adaptations to the
fingerholes. See the instructions in
par. 9.7 where I applied these ideas
to the piccolo flute; see also the
middle instrument on the photo,
left, which is a tenor in d1 with the
same type of double holes.

From left to right:

- modern Böhm flute (white metal)
- renaissance flute (acryl glass)
- ‘modern renaissance flute’ in two
parts in santos palissander, with
double holes
- renaissance flute in two parts in
santos palissander, single holes
- baroque traverso in four parts in
boxwood
All flutes tuned at a-440 Hz.