Wind Wizard: Alan G. Davenport and the Art of Wind Engineering

WIND WIZARD

WIND WIZARD

Alan G. Davenport and the Art of Wind Engineering

SIOBHAN ROBERTS

PRINCETON UNIVERSITY PRESS

PRINCETON AND OXFORD

Copyright © 2013 by Siobhan Roberts

Requests for permission to reproduce material from this work should be sent to Permissions, Princeton University Press

Published by Princeton University Press, 41 William Street, Princeton, New Jersey 08540

In the United Kingdom: Princeton University Press, 6 Oxford Street, Woodstock, Oxfordshire OX20 1TW

press.princeton.edu

Jacket photograph by Ron Nelson. Courtesy of the estate of Ron Nelson.

All Rights Reserved

Library of Congress Cataloging-in-Publication Data

Roberts, Siobhan.

Wind wizard : Alan G. Davenport and the art of wind engineering / Siobhan Roberts.

p. cm.

Includes bibliographical references and index.

ISBN 978-0-691-15153-3 (hardcover : alk. paper) 1. Wind-pressure. 2. Davenport, Alan G. 3. Buildings—Aerodynamics. 4. Bridges—Aerodynamics. I. Title.

TA654.5.R636 2012

624.1′75—dc23           2012015170

British Library Cataloging-in-Publication Data is available

This book has been composed in Minion Pro

Printed on acid-free paper. ∞

Printed in the United States of America

1 3 5 7 9 10 8 6 4 2

In memoriam

Alan Davenport

1932–2009

CONTENTS

I

Sowing Wind Science    1

II

Tall and Taller Towers    32

III

Long and Longer Bridges    129

IV

Project Storm Shelter    183

Acknowledgments    227

Notes    229

Interview Sources    243

Glossary    245

Bibliography    251

Index    267

WIND WIZARD

I

Sowing Wind Science

No sooner did the Tacoma Narrows Bridge—the world’s third longest suspension bridge, and the pride of Washington State—open in July 1940 than it earned its epitaphic nickname, Galloping Gertie. The 4,000-foot structure, its main span reaching 2,800 feet, twisted and bucked in the wind. The pronounced heave, or more technically speaking the longitudinal undulation, caused some automobile passengers to complain of seasickness during crossings. Others observed oncoming cars disappearing from sight as if traveling a hilly country road. By November 7, amid 39-mile-an-hour winds, the $6,400,000 bridge wobbled and flailed, then rippled and rolled, then twisted like a roller coaster, until in its final throes it plunged, with a beastly roar, 190 feet into the waters of Puget Sound.

See Glossary for definitions

Speaking to a New York Times reporter the day after the collapse, Leon S. Moisseff, the bridge’s designer and engineer, was at a loss to explain the cause, placing blame on a peculiar wind condition.

Wind engineer Alan G. Davenport, founder of the Boundary Layer Wind Tunnel Laboratory at the University of Western Ontario, often summoned the memory of the Tacoma Narrows Bridge disaster as a cautionary tale. Most features of this disaster are too familiar to bear repeating, he told his audiences, whether assembled at technical lectures or at popular talks. Both occasions always included screenings of grainy film footage capturing the bridge misbehaving as though fashioned from rubber—footage now preserved, owing to its cultural, historical, and aesthetic import, in the United States National Film Registry, as well as on YouTube, with numerous clips garnering more than six million cumulative views. Nonetheless, Davenport noted, as familiar as this disaster may be to the collective consciousness, the consequences bear continued consideration. What’s past is prologue, Shakespeare observes in The Tempest (and said the jester Trinculo, Here’s neither bush nor shrub to bear off any weather at all, and another storm brewing; I hear it sing i’ th’ wind). In broadening the moral of the Tacoma disaster and applying it to the behavior of all structures—bridges, buildings, and beyond—Davenport made of this cautionary tale a professional credo that governed his lifelong fascination with the wind, and with balancing the wind’s fickle forces.

Figure 1A. The Tacoma Narrows Bridge displayed torsional oscillation and longitudinal undulation even before it opened on July 1, 1940. University of Washington Libraries, Special Collections, UW21413.

Figure 1B. Photographer Howard Clifford of the Tacoma News Tribune snapped a few shots and ran. University of Washington Libraries, Special Collections, UW20731.

Figure 1C. The bridge’s main span collapsed into the waters of the Tacoma Narrows on November 7, 1940. University of Washington Libraries, Special Collections, UW21422.

Figure 1D. The new and improved Tacoma Narrows Bridge, 1950. Courtesy of R. A. Dorton.

It was in 1965 that Davenport established the world’s first dedicated boundary layer wind tunnel designed to test civil engineering structures. The planetary boundary layer is the region of the atmosphere extending from the earth’s surface upward about 3,000 feet, the wind churning the air into turbulent eddies, average velocity increasing with height. A boundary layer wind tunnel mimics these marbled striations of air—it mimics wind energy—in order to test designs for buildings and bridges that will face the wind when built. Making its debut in the 1960s, Davenport’s wind tunnel arrived the same decade as the laser, the computer mouse and the Internet, handheld calculators and the ATM, Apollo 8, string theory, and Rachel Carson’s Silent Spring. In the years to come, Davenport’s revolutionary lab would investigate the windworthiness of some of the world’s most innovative structures: many of the tallest buildings, including New York City’s World Trade Center and Citicorp Tower, Chicago’s Sears Tower, Boston’s John Hancock Tower, Shanghai’s World Financial Center, and Toronto’s CN Tower (which, strictly speaking, is not a building but rather a freestanding structure), and many of the longest bridges, among them Florida’s Sunshine Skyway, the proposed Straits of Messina span in Italy, France’s Millau Viaduct and the Pont de Normandie, as well as the iconic Golden Gate Bridge and New York’s Bronx-Whitestone Bridge. The Bronx-Whitestone Bridge, being very similar in design to the Tacoma Narrows crossing, has over the years required an extensive rehabilitation regime that continues to this day.

In addition to the buildings and bridges that came through the lab, there were also a few exceptions and eccentricities. Legend has it that in the early days the lab conducted tests on portable toilets, and later on Arctic tents to be deployed by the Canadian military. NASA commissioned a study on the ground wind loads for the Jupiter launch vehicle (occasionally Davenport said he wished he’d been an astronaut). The 2,421-foot illuminated Glorious Cross of Dozulé had its day in the tunnel, though it has yet to grace the countryside of Normandy. Sports Illustrated splurged on an investigation of Augusta National’s twelfth hole, the lynchpin of the Masters’ famed Amen Corner, said to be among the toughest holes in golf, in part because of the seemingly indecipherable winds (see sidebar, Driving into the Wind, below).

Driving into the Wind

Sports Illustrated turned to the lab to decipher the maddening winds at what’s been called the meanest hole in golf, the 12th hole at Augusta National. Augusta National is a one-of-a-kind golf course, the article reported, but all it takes to reproduce it (albeit at a scale of 1 to 200) is high-density foam sculpted with drywall compound, more than 600 trees made of sponge and wire, an acrylic Rae’s Creek (complete with tiny silicone waves) and, for good measure, foam golfers that are nearly as stiff as the real thing.… The shot’s path was represented on the model by a fixed piece of copper tubing 5/16th of an inch in diameter. Meteorological data from 1949 through ’99 (collected at Augusta Regional Airport, about 10 miles south of Augusta National) was then analyzed by computer to create a simulation of the typical April winds that blow through Amen Corner. Smoke was used to give these breezes visual paths. To illustrate the turbulence at higher elevations, a wire coated with oil was fixed upwind from the model. An electrical current was sent through the wire until the oil burned, producing yellowish smoke. To depict the wind’s effects along the trajectory of the shot, 13 evenly spaced holes were drilled along the copper tubing. Inside, titanium tetrachloride was introduced, producing bright white smoke. Results showed that one wind took two directions: On the tee the wind is in the golfer’s face, quartering slightly to the left (east), in the direction of the 11th fairway. About 25 yards into its flight the ball encounters a crosswind blowing to the east. Another 40 yards toward the green, as the shot is approaching its apex, the ball is slammed by a wind shear, with gusts blowing to the west toward the 13th fairway. This wind dissolves into low-speed swirling 20 yards from the green, as the ball is passing over Rae’s Creek. The lab’s project leader, Greg Kopp, concluded, The challenge used to be trying to figure out the wind. Now the players have all the information, but they may wish they didn’t. It’s still a frightening shot into a very difficult wind.

Figure 2. Wind tunnel tests on Amen Corner, Augusta National Golf Course. Courtesy of Robert Walker for Sports Illustrated.

Figure 3. In the early days, the lab conducted tests on a design for portable toilets, and later on a ten-man Arctic tent to be deployed by the Canadian military. Courtesy of the Boundary Layer Wind Tunnel Laboratory.

Figure 4. Stabilizing the Glorious Cross of Dozulé against the wind in Normandy, France, proved difficult to finance (the clients attempted to get papal recognition of the location as the site of a miracle, which would have made raising the necessary funds easier). Courtesy of the Boundary Layer Wind Tunnel Laboratory.

Figure 5. Davenport with a model of the Jupiter launch vehicle. In 1966, he presented two papers at NASA’s Langley Research Center for the Meeting on Ground Wind Load Problems in Relation to Launch Vehicles. Courtesy of Alan Noon.

Himself a great sailor, Davenport reveled in the testing of sails for an America’s Cup vessel. There were also investigations into how better to spray fruit trees so that the mist would not be blown off course by the wind. The solution proposed by Davenport’s colleague, the electrical engineer Ion Inculet, was to make the fluid electrostatically charged, so that it would be attracted to trees and repelled from the ground. There were inquiries into how to ensure a clean airflow over surgical patients during hip replacement procedures, which are particularly susceptible to infection. And after the construction in 1984 of a second-generation wind tunnel that doubled as a wave tank, Davenport tracked the wind-induced drift of icebergs and observed the battering of BP and Exxon oil rigs in the open seas.

With such a multidisciplinary portfolio, Davenport quickly accumulated unparalleled expertise in the nascent field of wind engineering—indeed, the field emerged and evolved largely because of his work. With one pioneering example after another, he set the agenda for investigating the effects of wind on the natural and built environments.

Figure 6. In 1984, with the construction of a second-generation wind tunnel facility that doubled as a wave tank, the lab began studying wind and waves—investigating the effects of wind on oil rigs and icebergs. Courtesy of the Boundary Layer Wind Tunnel Laboratory.

Figure 7. Davenport and his daughter Clare sailed windsurfers in the wave tank at the opening of the new wind tunnel facility. Courtesy of the Boundary Layer Wind Tunnel Laboratory.

When the interaction between wind and our environs is not properly factored into structural design, the consequences can be catastrophic. A powerful lashing of wind assaulted a family viewing Christmas lights during a walk around Toronto’s City Hall in 1982. The wind lifted the plywood from the promenade, threw the family into the air, over a protective parapet, and then dropped them 20 feet onto Nathan Phillips Square below, with serious injuries resulting. Climate change is arguably exacerbating extreme weather events, such as the deadly tornado outbreak in the southern United States in 2011, Hurricane Katrina in 2005, and the North American ice storm in 1998. And our exposure and risk are only heightened in the more fringe and fragile edges of nature where people are living, in both developed and developing countries—whether those fringes are the Florida Gulf resort communities or the rural coastal villages of Sri Lanka, where Davenport attended a disaster relief conference after the Boxing Day tsunami of 2004. Wind engineering is vitally concerned with how to prevent wind-induced disasters, the most costly disasters in terms of property damage and casualties. And although most knew him as a wind expert, Davenport in his versatility was among the first to advocate that the same mindset of preparedness be applied to all shapes and forms of natural disaster, not just those powered by wind.

Figure 8A. The lab put to the test several structures by Spanish architect Santiago Calatrava, including the Valencia Opera House. Courtesy of the Boundary Layer Wind Tunnel Laboratory.

Figure 8B. Calatrava’s World Trade Center PATH Terminal at Ground Zero. Courtesy of Santiago Calatrava.

Upstream, Downstream

Pollution control efforts to cut sulfur emissions from chimneys at old coal-burning power stations resulted in a busy time for the lab’s Barry Vickery, an expert on towers, chimneys, and stacks. (Vickery is now retired and serving as a consulting director of the lab.) In 1979, the Canada-USA Sulphur Emissions Reduction Protocol mandated a 30 percent reduction in emissions by 1993. To meet this mandate, companies such as American Electric Power, in the Ohio Valley, faced constructing new chimneys at preexisting plants with no choice but to put them in proximity to both the generating unit and the old chimneys. Companies often want to avoid the considerable cost of dismantling the old chimneys, explains Vickery. But having a number of stacks clustered together results in nasty aerodynamic problems—the upstream and downstream structures strengthen the vortex shedding produced by each, creating interference effects and loads powerful enough to damage these tall stacks, sending their exterior brick walls flying, or knocking the interior linings and filters. Vickery developed the best method for predicting the response of chimneys in such scenarios, a method he subsequently applied to well over 100 chimneys. Problems resulting from interference effects can be solved in a variety of ways, but simply removing the top third of an old stack due for decommissioning usually solves the problem.

Figure 9. The close proximity of cooling towers and old and new chimneys precipitates aerodynamic problems. Mark William Richardson/Shutterstock.com

In a career spanning half a century, Alan Davenport published hundreds of technical papers and scientific reports on the wind. He also seized every opportunity to impart a more popular, cultural, even philosophical and romantic perspective. One staple talk in his repertoire, Sowing the Wind, he delivered at the 1979 commencement ceremony at Belgium’s University of Louvain upon receiving one of his numerous honorary degrees. He explained that the title was taken from the words of the Old Testament prophet Hosea, For they have sown the wind, and they shall reap the whirlwind, suggesting that one must heed the action and power of the wind or deal with the consequences. Our ancestors, and civilizations before us in the ancient world, respected the wind for both practical and spiritual reasons, he observed. The wind has the power to drive turbines and evoke emotions. It pushes ships, and sinks them. It winnows grain and turns windmills grinding grain to flour, but it also flattens crops and blows down barns. Spiritually, Davenport noted, Primitive people were no doubt awed by wind …

What exactly they thought about this mysterious invisible force we cannot be sure. Did they believe, like the popular cartoon philosopher Charlie Brown, that the clouds pushed the wind along? Or, like Ogden Nash, that the wind is caused by the trees shaking their branches? We can only guess that the unseen hand causing the rippling of the water and the rustling of the leaves was a sign that the natural world was animated by life-giving forces; that when these ripples turned to storm tossed waves, those forces were angry.

The principal god of the Aztecs, Davenport noted, was Quetzalcoatl, the plumed serpent and the god of wind. Quetzalcoatl’s equivalent in ancient Egypt was Ammon, who shared the same godly status as Ra, the sun god. And the wind, in fact, derives its power not from the clouds or trees but from the sun—from the thermal shifts of the day-night cycle, as well as from the cooling effects of bodies of water and higher altitudes. Wind is defined by Environment Canada as the horizontal movement of air relative to the earth’s surface caused by geographic variations in temperature and pressure. Solar radiation, beating down strongest at the equator and weakest at the poles, produces temperature and pressure differentials in the air. These pressure pockets are never stationary, always changing in their patterns, and altered by the earth’s rotation. Air follows an eddying motion as it moves from areas of high pressure to low, rising as it warms, bringing cold air rushing in below to take its place. This mixing creates wind—the easterly trade winds of the Northern Hemisphere, which steer tropical storms across oceans and onto continents, and the violent prevailing westerlies of the Southern Hemisphere, known as the Roaring Forties, Furious Fifties, and Shrieking Sixties.

Figure 10. A few specimens from Davenport’s collection of wind masks. Courtesy of the Boundary Layer Wind Tunnel Laboratory.

The wind was first treated as a subject of scientific inquiry in ancient Greece, and Aristotle’s treatise on the elements, Meteorologica, written in 350 BC, endured as the standard reference in Europe for nearly two thousand years. Aristotle’s pupil Theophrastus wrote two treatises on winds and weather, proposing that winds could be predicted according to the behavior of animals and the human body: A dog rolling on the ground is a sign of a violent wind.… If the feet swell there will be a change to a south wind. This also sometimes indicates a hurricane. From Sir Isaac Newton, in his Principia, published in 1687, came the discovery that the wind force on a given shape is directly proportional to the shape’s area, the air density, and the square of the wind’s velocity. In 1759, another Englishman, John Smeaton, often called the father of civil engineering, proposed a colloquial classification of the wind’s force in a paper presented to the Royal Society. He listed eleven common appellations, corresponding to the velocity of the wind by miles covered in one hour:

In 1805, the Irish admiral Sir Francis Beaufort devised his eponymous Beaufort Wind Force Scale. It provided a standardized measure of wind speed based on sea conditions and is still used today (though now calculated with an empirical formula factoring in land conditions), ranging from zero, a calm and flat sea, to twelve, hurricane force with huge waves, foam, and spray.

By the end of the nineteenth century, fluid dynamics—the scientific domain with air and wind flow in its purview, as well as water flow—had advanced considerably. Equations developed in classical hydrodynamics produced beautiful solutions. The flow of air around a cylinder, for instance, produced a symmetrical pattern framed neatly by two stagnation points, the place where the air would come to a stop at the front and again at the back of the structure. But since classical hydrodynamics assumed ideal fluids—fluids with no viscosity, no resistance or friction—these beautiful solutions seemingly did not pertain to any real-life problems. As the British chemist and Nobel laureate Sir Cyril Hinshelwood reportedly lamented, In the 19th century fluid dynamicists were divided into hydraulic engineers who observed things that could not be explained, and mathematicians who explained things that could not be observed.

In the early twentieth century, however, Ludwig Prandtl, a German scientist known for his seminal work applying mathematics to aeronautics, wedded these theoretical and practical solitudes. He proposed the concept of a boundary layer, and the notion that a viscous fluid actually possesses no velocity at the surface when it passes over an object. In dealing with ideal fluids mathematically and theoretically, as classical aerodynamicists did, fluids had been assumed to have no viscosity simply for the sake of simplicity, to make complex calculations easier. Now what Prandtl proposed was that this idealized theoretical assumption could be extended to practical applications as well. That the wind blowing over any surface doesn’t have any velocity at the very interface with the surface is counterintuitive—a property of wind that even concrete-minded engineers describe as magical, though it can be explained. If a viscous fluid such as air flows over a surface—say, the surface of a building, or the surface of the Arctic tundra—it is a fundamental property of the air that its particles have no motion when they meet the surface. Instead, on first contact a layer of air particles sticks to the surface. This is because at the microscopic level any surface is a craggy affair, and individual air particles can’t help being trapped in the crags. Following this initial layer of air particles there is a second layer that viscously slides over the first. As a result, the second layer does have some velocity when it meets the surface; it shears past at a very slow velocity, dragged down by the particles that are locked into the crags. And so it goes, up and up and up. As the air moves away from the surface, its speed rapidly increases, and eventually the fluid reaches a speed more or less independent of viscosity, driven by a pressure field somewhere else, above or below. This stratification of the air and its variegated velocity make up the boundary layer.

While Prandtl applied his insights about the boundary layer mostly in the field of aeronautics, others, such as the French engineer Gustav Eiffel, played with structures of various shapes in wind tunnels. Eiffel was one of the earliest engineers to factor in the important effects of wind on tall structures. When he designed his famous tower in 1887, he told the French newspaper Le Temps (defending himself against an artist’s protest), Now to what phenomenon did I give primary concern in designing the Tower? It was wind resistance. Well then! I hold that the curvature of the monument’s four outer edges, which is as mathematical calculation dictated it should be … will give a great impression of strength and beauty, for it will reveal to the eyes of the observer the boldness of the design as a whole. By Eiffel’s account to the Société des Ingenieurs in 1885, however, his wind calculations had been cautiously conservative: "With regard to the exposed surfaces, we have not hesitated in assuming, in spite of the apparent severity of the assumption, that on the upper half of the tower all the lattice work is replaced by solid surfaces; that in the intermediate section, where the openings become more important, the frontal area is taken as four times the actual area of iron; below this (the first stage gallery and the upper part of the arcs of the legs) we assume the frontal area is solid;