Picture a spot at the middle of the deck of a 6km suspension bridge. Not where you'd want to be when a stiff breeze gets up. Images of America's Tacoma Narrows Bridge pitching itself to pieces one windy November day in 1940 have haunted generations of suspension bridge designers.
The stiffest, most aerodynamic designs have, to date, daringly spanned just over 2km. An extensively tested proposal for a link between Sicily and mainland Italy across the Straits of Messina, should it come to fruition, would deliver a span of 3.3km. That takes conventional design wisdom to its utmost limit. But bridges of 6km and above are entirely feasible, says independent consultant and Knight Piesold non-executive chairman David Piesold. Following six years' research the firm has patented an idea it believes can double the length of the world's longest suspension bridge spans and is now talking to prospective clients.
Knight Piesold's big idea is active aerofoil stabilisation, a responsive system of aerodynamic flaps on either side of a bridge deck that would enable it to 'fly' through a high wind.
Wind causes long, thin bridge decks to twist from side to side and undulate along their length - movement described more properly as torsional and flexural oscillation. Though undesirable, individually neither is catastrophic. Problems start to occur when the frequency of torsional and flexural oscillations coincides. A critical wind speed can cause the harmonic oscillations to diverge, introducing a potentially destructive condition known as 'flutter'. It was flutter that demolished the Tacoma Narrows Bridge and it is one of the bridge designer's greatest fears.
Flutter is also experienced in aeroplanes. Yet, Piesold noted as he bounced through turbulence aboard a passenger jet, planes are able to control the twisting and bending experienced in their wings in flight with the aid of aerofoils. Knight Piesold linked up with flight control experts GEC Marconi and Imperial College to look at active aerofoil stabilisation of cable supported bridge decks in 1993.
Piesold believes the attraction of active aerofoil stabilisation lies in the opportunities it offers for improving safety and reducing cost. Though the aerofoils and control system represent a new cost, they allow lighter construction for the deck itself. They also reduce wind-induced stress on the deck and the bridge as a whole. Combined, these factors allow significant materials and cost savings in the suspension structure and supporting pylons.
Piesold calculates that savings through reduction of dead load would become appreciable in bridges of over 700m. And by spanning larger distances with lighter, more stable bridges it should be possible to cut the major cost of intermediate pylons in multi-span structures.
The research focused on hollow box section decks which were generally lighter, less expensive and more aerodynamically evolved than trussed designs. Tests showed that box girders were longitudinally stiffer than trussed structures. And with a section comparable to that of an up-side- down wing, the foil shape of, say, the Humber bridge deck meant wind passing across it generated a stabilising down force. However, box girders lacked torsional rigidity. Aerofoils could be most useful in the control of twisting.
While aircraft control air flow with aerofoils on only one edge of the wing it was clear that a bridge deck would need foils on both edges to deal with varying wind directions. Modelling showed that, while planes gained optimum stability with foils on the trailing edge, control of the leading edge foil achieved best results in a bridge. The trailing edge foil lent added stability.
Knight Piesold/GEC Marconi found torsional oscillation could be best limited by deflecting both foils counter and equal to the rotation of a twisting deck. In other words, if the deck were to rotate about its centre dropping the leading edge by 5degrees, the leading edge foil would deflect upwards by 10degrees. The trailing edge foil, meanwhile, would deflect downwards by 10degrees.
Piesold explains: 'The purpose of aerofoils is not to increase downward pressure on the deck but to reduce oscillation.' This is achieved by the foils disrupting the flow of wind across the deck. Though aerofoils could not eliminate torsional oscillation they would prevent it developing a regular frequency. Piesold says aerofoils placed at intervals along the length of the deck would be effective. He points out that 'on aeroplanes you don't have stabilisers over whole length of wing'.
Having established the principles by which flutter could be prevented, the research group tackled the mechanisms by which active aerofoil stabilisation could be delivered. Effective foils needed to be supremely responsive.
Capitalising on GEC Marconi's experience of supplying flight control systems for civil and military aircraft, the research team started to look at computer-controlled aerofoils. Exhaustive investigation went into working through the different configurations of motion sensor-computer- mechanical actuator-aerofoil.
Questions quickly arose over how to power a fast enough response. Depending on the mass of a deck, aerofoils would need to be 2-4m or more wide to gain adequate purchase in the wind, says Piesold. High wind loads, meanwhile, would place aerofoils under large amounts of stress calling in turn for large amounts of power.
The best means of delivering power, worm gears and hydraulic rams, were relatively slow and very power hungry. And the preferred computer-controlled solutions, in which each aerofoil would be served by a dedicated sensor, computer and actuator, were wholly reliant on consistent power supply. This would be most at risk when most needed - in foul weather. It would also be most difficult to install and maintain in long span structures - those where dependable active aerofoil stabilisation was most essential. Maintenance and updating of components in a dedicated computer controlled and mechanically actuated system was also seen as complex. Though not ruling it out, Knight Piesold/GEC Marconi went on look for a simpler solution.
A purely mechanical system makes the aerofoils deflect, exploiting the change in angle between the bridge deck and hangers when the deck rotates. The system uses the mass of the deck itself to generate the force required to control the aerofoil.
Piesold envisages perhaps two pairs of hangers for each twinned set of aerofoils. A 4m rigid suspension member is incorporated into each hanger which, except for minor movement, remains in a set plane. Aerofoils, linked via crank arms to the suspension members, are activated by rotation in the deck element. Depending on the gearing of the simple crank mechanism, movement of the foils will be directly proportional to torsion in the deck itself. No computers and no electricity are needed.
Knight Piesold/GEC Marconi has calculated the novel loading aerofoils would place on a bridge and is revising the design. What is required next, says Piesold, are detailed studies and wind tunnel modelling of aerofoils for a real project - unlike the aeronautics industry, Piesold notes, there is rarely scope for prototyping in engineering.
Aerofoils could be applied to the Messina Straits bridge but, considering the design is so far advanced, this would require too fundamental a re- evaluation of principles, he thinks. If current confidential discussions go well though, active aerofoil stabilisation could be deployed on suspension bridges in China early next decade.