Case Study

Design study for the Space-Web Bridge

• Innovative cable stay suspension bridge
• Fundamental frequency analysis
• Comparison with traditional suspension bridges

To address the matter of significantly increasing the spans of future suspension bridges, Gordon Rose of Rose Associates has proposed the Space-Web Bridge concept. In this design, the aerodynamic and seismic stability of a suspension bridge is enhanced by the addition of a horizontal system and network of cables incorporating a 3-Dimensional tower design. To help assess the potential benefits a design study was performed using LUSAS to calculate the theoretical natural frequencies for comparison with an equivalent traditional suspension bridge without a cable network.

As Gordon Rose explains : "The aerodynamic stability of a long span bridge depends partly on the shape of the cross section and partly on the global dynamic characteristics of the structure. There are various mechanisms of aerodynamic excitation that can lead to resonance of the bridge deck - including vortex shedding, turbulence and flutter.  However, the critical wind speed at which resonance occurs is higher for a structure with an inherently higher fundamental natural frequency."

He continues, "There is no fixed value of natural frequency to aim for in the design of long span bridges. In general, the aerodynamic stability is improved by an increase in the fundamental vertical bending frequency. It is also important to avoid the coupled oscillations associated with classical flutter by maintaining a significant difference between the frequencies at which fundamental vertical bending and torsional modes occur."

The traditional suspension bridge was formed by main cables with hangers supporting a spine beam representing the deck longitudinal and transverse members. Span configurations of 425-850-425 metres and 850-1700-850 metres were investigated with a tower height in each case of 150 metres. In both cases all materials were assumed to be steel.

The additional cable network was composed of three extra pairs of longitudinal cables together with their associated "hangers" and a lattice of secondary cables. The cables were fixed in position (but not direction) at the ends of the structure and at the towers. The deck was guided (fixed in position vertically and horizontally) at the ends of the structure and at one tower. It was fixed vertically, horizontally and longitudinally at the other tower. No account was taken of the Tower stiffness in the analysis. In each case the model was solved with and without the additional cable network in place to give four sets of results in total.

Using LUSAS, natural frequencies were obtained for vertical bending, lateral bending, and the first torsional mode for each of the four study models. From the results obtained the natural frequencies of the bridge structure were shown to be increased with the addition of the cable network although further investigation will be carried out to establish just how much the separation between the first torsional frequency and the fundamental frequency of vertical bending is enhanced.