SECTION: Vol. 86 ; Pg. 60; ISSN: 0040-1692
LENGTH: 3759 words
HEADLINE: The Brooklyn Bridge at 100; centennial of John Roebling's Brooklyn Bridge
BYLINE: Birdsall, Blair
The Brooklyn Bridge at 100
JOHN Roebling, the creator of the Brooklyn Bridge, is unique among bridge builders. He was the only one in the early nineteenth century, when modern suspension bridges first appeared, who never let fashion or untested theories eclipse his basic common sense. While others around him built suspension bridges that became famous for blowing down in the wind, his bridges have gone out of service only because of obsolescence. Roebling's largest and most famous design is as viable and sound today-- except for repairs to restore the materials to their original condition--as the day it was inaugurated in May 1883.
To appreciate the dimensions of his genius, Roebling's achievements must be put in the context of what was going on elsewhere in the field of suspension bridges. It was a time of experimentation. Many fine, intelligent engineers applied themselves to this field. But many, taking steps that they believed were based on experience or elementary attempts at new theory, found that their steps were too large or even in the wrong direction. Roebling's hand was singularly sure.
Rope-Borne Catwalks to Steel-Borne Highways
During Roebling's active career, roughly from 1830 to 1870, suspension bridges in the modern sense-- structures wth a relatively horizontal travel-way supported from overhead cables--were just beginning to gain wide attention. In contrast, in previous suspension bridges the platform had been directly connected to and followed the curve of the cables. (Roebling died just as construction of the Brooklyn Bridge was beginning, leaving supervision to his son.)
It was clear to many designers that the suspension bridge was then the only means available to span distances required for crossings such as New York's East River. But while Roebling was drawing his conservative plans for the Brooklyn Bridge with a record span of 1,600 feet, other designers were increasing span lengths without Roebling's intuitive sense for stability, despite the fact that many suspension bridges in the United Kingdom and the United States had been destroyed by wind. Roebling also uniquely recognized the value--in terms of strength, stability, and economy--of a single cable mass. While others called for large cables composed of several small ones laid side by side, Roebling's cables were designed to become a single, compact bundle.
By the end of the nineteenth century, most engineers had come to agree with Roebling's sense of the need for stiffening structures on a suspension bridge. In response, a mathematical theory was developed by the Austrian bridge designer Joseph Melan to analyze a stiffened suspension bridge. The first to be designed on the basis of this theory was the Manhattan Bridge, built during the first decade of the twentieth century. The theory was successful, and stiffness has been recognized ever since as essential in suspension bridge design. Indeed, many bridges were overdesigned in this respect, with the result that the structures were heavier and more expensive than necessary. But this fault can be excused in the first application of the new theory.
Other problems, however, have made the other East River suspension bridges less distinguished in maturity than Roebling's. These problems include the selection of materials, erection methods, and the balancing of dead loads (the weights of the bridge structures). One example is the use of nongalvanized wire on the Williamsburg Bridge. Another is the failure, in the design of the Manhattan Bridge, to account for the fact that two of the four identical cables were located so that they should carry more dead load than the other two. Because the cables were designed for equal loads, the actual load had to be distributed to the four cables by means of the floorbeams, which were not really designed for that purpose.
By the early 1920s, when engineers had gained experience and confidence in using the new theory, stiffness was achieved with somewhat greater finesse. Examples include the Benjamin Franklin Bridge in Philadelphia, the bridges across the Hudson River at Bear Mountain and Poughkeepsie, the bridge across the Ohio River at Maysville, Ky., the several suspension bridges between West Virginia and Ohio, and the Waldo Hancock Bridge in Maine.
From Economy to Disaster
But then came the 1930s and the Great Depression. Bridge design engineers found themselves under the greatest imaginable pressure to economize, leading them to extend the Melan design theory to its limits. Thus many bridges built during this period--the Deer Isle Bridge in Maine, the Thousand Islands Bridge between New York and Canada, the Whitestone Bridge at New York City over the Long Island Sound, and the Tacoma Narrows Bridge at Tacoma, Wash., over an arm of Puget Sound--were designed with a relatively shallow plate girder instead of the heavier, conventional truss adjacent to the platform. The goal was to provide the minimal stiffness necessary to prevent excessive movement of the roadway under heavy loads while maximizing the economy of construction and the aesthetics of a slender, even gossamer structure.
The most extreme design in this quest for economy was the Tacoma Narrows Bridge. Even during construction, workers on this bridge noticed that the structure was undulating in the breeze, and motorists using the bridge when it was completed were often treated to the odd spectacle of seeing cars ahead disappear behind a wave in the floor. Though the engineers recognized that this was not a desirable condition, they did not recognize it as catastrophic. They therefore decided to let the bridge stand as built until careful model studies could be completed to determine the best means of reducing the instability and deformation. However, nature chose not to wait for the perfect solution, and in November 1940 the bridge was destroyed by the harmonics induced by a broadside wind of only modest velocity--about 45 miles per hour.
The George Washington Bridge between Manhattan and New Jersey barely escaped a similar fate. It was designed in the late 1920s for ultimate service as a double-deck bridge, with ample stiffness. But only the upper deck was installed at first, with the lower deck and the main stiffening members omitted until heavier traffic should require more capacity. This strategy actually placed the bridge in a potentially precarious situation. However, it had one saving advantage: the main cables as built were adequate to support the proposed double-deck bridge. The dead weight of these cables--very heavy with respect to the initial construction--provided adequate stiffening to prevent catastrophe.
The Thousand Islands Bridge, designed by the founder of the author's firm, David B. Steinman, was the first of these relatively unstiffened bridges. During construction, Steinman received word that the bridge was undulating in the breeze, and he immediately acted to restrain the bridge and break up the harmonics. His solution was to add stays radiating from the roadway level at the tower to the main cable in the adjacent spans. The building of Steinman's Deer Isle Bridge in Maine was at that time just beginning, and he carefully designed a system of stays of the same type to be incorporated during construction. Roebling himself used stays between tower and bridge deck for stability in all his suspension bridges. Indeed, these very visible diagonals are Roebling's trademark--evidence of his early sensitivity of a problem that eluded some of his successors for 60 years.
Solving the Stability Problem
Steinman's experiences, together with the Tacoma disaster, brought engineers face to face with a force of nature that until then had gone unrecognized. A perfectly horizonal broadside wind could lift the deck like an airplane wing, generating an upward force of such intensity that the stiffening members acted as little more than cloth ribbons. The engineers found they had to deal with aerodynamics and not simply the static forces applied by wind. Indeed, the Tacoma Bridge was perfectly designed, on the basis of existing criteria, to withstand a wind of up to 125 miles an hour considered as a simple horizontal force against the bridge.
Why did engineers, who were surely aware that in the previous century suspension bridges had frequently been destroyed by wind, have this blind spot that veiled the potential for trouble in their new structures? One can only surmise that they were unable to relate the light wooden platforms of earlier days to 6,000 tons of steel and concrete.
Unfortunately, a catastrophe of this type is sometimes the only way to discover natural forces hitherto unrecognized. But once the problem became clear, studies mushroomed overnight. There was never any possibility of a subsequent disaster from the same cause. The only problem was that of identifying the best and most economical methods of solution. Two schools of thought developed. One solution--a "brute-force' engineering approach--was to increase the stiffness of the deck, returning to designs that had proven resistant to whatever forces the wind might bring to bear. The other solution, based on aerodynamics, was to develop a bridge cross-section with a shape less affected by wind, so that catastrophic aerodynamic forces would not be generated.
The first school of thought is represented by countless modern bridges provided with increased resistance to both vertical and rotational movement. The "brute force' approach soon mellowed to a more sophisticated philosophy that includes adequate stiffness and a due regard for aerodynamics. Most of the designs therefore include features such as open strips in the deck floor to reduce wind lift. Examples include the second Tacoma Narrows Bridge, the Mackinac Bridge in northern Michigan, the Delaware Memorial and Walt Whitman Bridges over the Delaware River, the Chesapeake Bay Bridge at Annapolis, and the Throgs Neck and Verrazano Narrows Bridges in New York. Almost all designers now take advantage of the opportunity to test their designs by means of sectional or full models in wind tunnels.
The other school of thought led to a new type of structure pioneered by the British, the first example of which was the suspension bridge over the Severn River between Bristol, England, and Wales. The suspended structure, comparatively shallow in depth, has a cross-section like the streamlined hull of a ship that discourages the generation of vertical aerodynamic forces, assuring freedom from catastrophic lift.
Still more recently, the stability problem in longspan bridges has been solved by the so-called stayed-girder design. This has the roadway and towers of a suspension bridge, but the cables are straight stays from the top of the tower to various points on the roadway. There are no massive flexible suspension cables in which harmonics can develop. This type of bridge has been found to be competitive in cost with conventional suspension bridges for spans of up to about 1,500 feet, the length of the Brooklyn Bridge.
The Gerontology of Bridges
Despite its remarkable design that makes it a crown jewel among bridges, the Brooklyn Bridge is not immune to danger as it enters its second century of service. Like most other bridges supported by public funds, it suffers from lack of proper maintenance.
The Brooklyn Bridge and its three younger sisters--the Manhattan, Williamsburg, and Queensboro Bridges, which form the four major overhead crossings of the East River between Manhattan and Brooklyn or Queens--serve as a good sample of the nationwide problem. Though some national funds are available for bridge maintenance, the program is by no means adequate to do all the work that is needed. State and local authorities establish priorities and assign available funds accordingly. These four bridges have found their place in the priority, and inspection, rating, and rehabilitaton have been in progress since 1978.
Some $ 30 million is to be spent in 1983 on the Brooklyn and Queensboro Bridges, and at least $ 100 million of work will have to be done in the next several years on each of the four East River spans. Although this is a major investment, it is far less than the cost of rebuilding any of these bridges, if only because of the disruption of traffic and property during reconstruction. And reconstruction would be unthinkable to many because these bridges are a cherished part of the metropolitan scene in New York. Indeed, the Landmarks and Art Commissions do not want to see a single change in outline.
Each of these bridges has special characteristics and needs. The Queensboro Bridge is a large truss bridge of the cantilever type spanning just under 1,200 feet, which now carries eleven lanes of vehicular traffic. The Williamsburg Bridge is a 1,600-foot suspension bridge carrying eight lanes of vehicular traffic and two lanes of rail traffic. The Manhattan Bridge is a suspension bridge with a 1,470-foot main span that carries seven lanes of vehicular traffic and four lanes of rail traffic, while the Brooklyn Bridge carries six lanes of traffic (limited because of the original design to passenger cars and light vans) and a 12-foot elevated pedestrian and cycle promenade.
Most of the deterioration on the Queensboro Bridge is simple corrosion due to the savage environment of New York City and the use of de-icing salts. In addition, there is pavement deterioration caused by heavy traffic. The problems of the Williamsburg Bridge are similar but in addition there is a special problem. Unlike those of most suspension bridges, the main cables were made of ungalvanized wire. The deterioration is not nearly as extensive as one might expect after 80 years of service, but engineers will still have to do a great deal of soul-searching to find the proper way to prevent further corrosion among the cable wires.
In addition to all the problems of the Queensboro Bridge, the Manhattan Bridge (its cables are of galvanized were) has two special problems of its own. In the first place, rail traffic has had a devastating effect on this bridge. The four tracks are located at the sides of the bridge, one pair at each edge. Whenever a train crosses the main span, there is torsional stress and movement of the structure. When two trains travel in parallel, the forces are greater; and when trains travel simultaneously on all four tracks, the forces are very large indeed. The second problem, which compounds the first, is that the suspended structure is attached to the cables in a way that makes it very difficult to determine the dead load stresses in the structure.
The design of the Brooklyn Bridge also makes it difficult to determine the distribution of dead load to the four cables. Another problem is inherent in the atypical features of this design: the short suspenders at midspan swing by several degrees away from the vertical in response to extremes of temperature. The bottoms of the suspenders move toward midspan in summer and toward the towers in winter. The suspenders have never been properly attached to the cables to allow movement of this type, and new materials and details are now being considered to provide a better solution.
Beyond these difficulties, the main problems are those caused by the environment. Corrosion is especially concentrated in the main cable anchorages and the anchorages of the diagonal stays at the tops of the towers. In the anchorages, where the cables are held to the bridge foundations, many broken wires will have to be repaired, and some bundles of wires may have to be cut off where they emerge from the main cable and tied back to the anchor heads with new material. The problem is that the anchor chambers are not waterproof; this will have to be corrected, and perhaps the chambers provided with controlled atmosphere, to arrest corrosion in the future. At the tops of the towers, all of the diagonals that pass through the tower top will have to be reanchored and made more accessible for maintenance.
Moisture is, of course, a principal culprit in all this deterioration, especially when it appears in the presence of salt used for de-icing. Indeed, the greatest boon in this field would be to find a de-icing material other than salt, for it is the greatest corrosionproducing agent in the environment.
In the design of new structures, every detail is now reviewed specifically with corrosion protection in mind. Some designers use corrosion-resistant steel, but this must be used with care because it is not panacea. Expansion joints along the roadway have been a frequent cause of trouble, for it is very difficult to prevent moisture from passing through these joints to the steel below. However, joints are being developed that promise to go a long way in this direction. It is currently normal practice to study drainage problems carefully so that moisture will be carried off and away from the structure as quickly as possible. Concrete typically used for the platform slab is not impervious to moisture, so engineers require a bed of special sealing membrane over the concrete before applying a final roadway surface of asphalt. Alternatively, several concrete additives are in experimental use to reduce the permeability of the upper layer of the concrete slab.
Ending the Neglect of Public Bridges
These problems are not the fault of the people responsible for maintaining our bridges. On the contrary. That these four bridges--and many others throughout the country--are still viable for transportation is a great tribute to the skill and dedication of their original builders and of those charged with their maintenance through the years. The problem is that money has often been unavailable for regular preventive maintenance. Indeed, in the competition for funding, bridge maintenance has been pitted in a lopsided contest against many other more visible public needs. Bridge maintenance has too often been operated like a fire department--rush to the scene of a problem when one appears.
Public awareness of this problem was finally sparked by the disastrous failure owing to corrosion of the suspension bridge over the Ohio River at Point Pleasant, W. Va., at the end of 1967, in which 46 lives were lost. Extensive publicity resulted from the inspection of deteriorated bridges mandated by legislation after the West Virginia disaster, modest funds have now been made available, and some of the bridges are finally receiving the attention they need.
There is hope that bridge maintenance can be accelerated as a result of the recently enacted gasoline surtax. But even these additional funds are likely to be inadequate, and public understanding must be increased. In short, we must find some way to avoid repeating the cycle of deterioration and crash repairs.
The neglect of public bridges is in marked contrast to the careful maintenance generally accorded toll-supported bridges, which generate their own funds for maintenance. Those responsible for upkeep are able to both plan and carry out preventive maintenance programs. The result is that, in general, our toll bridges are in excellent condition.
In today's economic climate it would not be politically possible to place tolls on bridges that have been free for many years. But here is a modest proposal to encourage public understanding of the fact that "free' bridges are not really free.
For any bridge, it is possible to budget preventive maintenance for a two-year period with considerable certainty and for five years reasonably well. Given accurate traffic predictions, one can determine, approximately, the amount of maintenance money associated with each anticipated vehicular crossing. A display system could readily exhibit to drivers crossing the bridge the growing maintenance liability-- typically several cents per crossing--that their use of the bridge entails. This would be a graphic way of keeping the public continually aware of a problem that is too easily neglected.
Indeed, it would be more than fitting if celebration of the first century
of service of the Brooklyn Bridge could become the catalyst for wide and
increasing public commitment to the rehabilitation and maintenance of our
great bridges, returning them to, and then retaining them in, the condition
in which they were left by their designers and constructors.
Photo: Pioneers crossing the East River. The Brooklyn Bridge is the longest example of the unique cabling system devised by John Roebling in the 1850s. The 6,000 galvanized-steel wires were carried across the river, two at a time, by a traveling wheel. These continuous wires formed 19 skeins for each cable. The photos show cable workers and the cableway atop one tower with half of the skeins in place and an early visit of the bridge trustees venturing onto a temporary walkway used in building the cables. With cables complete came the task of suspending the platform (drawing at far right). A strap wrapped around the cable is fitted with two downward-extending ears, into which a suspender, fitted with matching ears, is to be fastened by means of a pin.
Photo: When theory is pressed beyond its limits. To economize on costly steel, engineers turned suspension bridges into gossamer webs of wire and concrete in the Depression years. But that trend ended abruptly when a 45-mph wind lifted the Tacoma Narrows Bridge to its destruction in 1940. (Note the single plate girder forming both rail and stiffener.) The rebuilt bridge on the same towers (right) shows the "brute-force' solution--a deck stiffened by a full-length box frame.
Photo: The new and the old. When engineer David Steinman learned that the platform of the Thousand Islands Bridge was moving in the wind, he added diagonal stays such as those on his later Deer Isle Bridge in Maine (left). The Verrazano Narrows Bridge is stiffened by under-platform girders. But the Brooklyn Bridge (right) remains "the most cherished and admired work of civil engineering in the Western Hemisphere,' says the American Society of Civil Engineers.