| doi:10.3850/978-981-08-6218-3_SHM-We028 |
 Final Paper PDF
|
CONTROL OF EXCESSIVE FOOTBRIDGE VIBRATION USING TUNED MASS DAMPERS
S. Browna, JP. Hwangb and A. Parkerc
Structural Dynamics, Heggies Pty Ltd, Lane Cove, NSW.
asteve.brown@heggies.com
bjoon-pil.hwang@heggies.com
candrew.parker@heggies.com
EXTENDED ABSTRACT
Many bridge structures have suffered from greatly amplified dynamic responses to a range of excitation sources at or near their resonant frequencies. This in turn can cause undesirably high accelerations or displacements of the structure, which can lead to elevated stresses and potential fatigue problems as well as discomfort to bridge users. Tuned mass dampers (TMDs) are often used in large structures such as tall towers to mitigate their dynamic responses to external excitations such as wind and ground vibration or internal excitations such as footfalls. This paper presents a case study of the measurement, modelling and analysis techniques implemented to design a TMD for reducing footfall vibration on a pedestrian/cycle bridge recently opened in Sydney.
1. INTRODUCTION
The Falcon Street Pedestrian and Cyclist Bridge was recently constructed and commissioned over the Warringah Freeway in Sydney, Australia. During the final stages of fit-out and commissioning, construction workers noted that they were concerned with the level of uncontrolled vertical vibration on the bridge whilst people were walking over it. Heggies was engaged by the Roads and Traffic Authority of NSW (RTA) to investigate the excessive vertical vibrations resulting from footfalls on the bridge. This paper provides a detailed overview of the investigation including the specification and testing of the Tuned Mass Damper (TMD) system that was implemented in order to mitigate the excessive levels of vibration on the bridge.
2. FOOTBRIDGE VIBRATION SEVERITY
Vibration measurements were recorded at selected mid span and quarter span locations. A range of activities were carried out on the footbridge in order to quantify the severity of the vibration. The result of a 3 person in phase point excitation activity (notionally a “vandal scenario”) was a very disturbing 1.9m/s/s and was not by any means the limit of response for 3 participants. It was noted that as the bridge response increased during this activity, it became easier for the participants to synchronise and “lock in” the excitation with the response.
3. MODAL ANALYSIS AND FINITE ELEMENT MODELLING
The measurement and definition of the natural frequencies and mode shapes of a structure is referred to as modal analysis. A modal analysis was undertaken in order for the vibration of the bridge to be characterised and to determine the target frequencies for mitigation design. The modal analysis indicated that the first two vertical bending modes of the bridge were at 1.3Hz and 1.99Hz.
A Finite Element model of the pedestrian bridge was modified and updated by Heggies based on the measured results of the modal analysis. This model was then used to predict the effect of the addition of the proposed tuned mass dampers. A sensitivity analysis was undertaken in order to determine the optimal number of TMDs that would be required to provide the required reduction in vibration levels. The results indicated that only a small benefit was gained by the use of three TMD’s over two TMD’s.
The optimal target frequency and damping of the TMD’s was calculated from the well established formulas for a simple two degree of freedom system. Based on these initial calculations the nominal design parameters were used in the FE model to predict the structures response. The final TMD design adopted for the bridge required two TMD’s 3,700 kg each, supported by 4 coil springs to provide a bounce mode of 1.85Hz and bridged by four viscous dampers each with damping rates of 4000 N/m/s.
4. FACTORY ACCEPTANCE TESTING
In order to minimise the installation time on site and to avoid major modifications to the TMD support system, one TMD was assembled in the workshop and pre tuned to 1.85Hz. The frequency tuning was achieved by adjustment of the total mass of the TMD. The TMD was first assembled and supported by a stiff frame, and then excited while the frequency of the bounce mode was measured. Approximately 90kg was removed from the total TMD mass to achieve the required frequency. Subsequently the viscous dampers were attached and the TMD was again excited to measure the modal damping, the resulting TMD modal damping was measured at 18%, just lower than the target 19%.
5. FIELD TESTING
The TMD’s were installed into the mid span cavities with the pre-tuned weights and viscous dampers connected. Diving point Frequency Response Functions were measured at mid span locations on span 2 and span 3, (the main spans) where the TMD’s were installed. Further frequency tuning was not required as the resulting bridge Frequency Response Functions showed that the target TMD frequencies and bridge damping had been achieved. The maximum vibration level for walking or jogging was now measured at 0.2m/s/s and 0.3m/s/s for vandal excitation. This result represents almost an order of magnitude reduction in the bridge vibration response to pedestrian related excitation, and brings the response comfortably below the recommendations of AS5100 and European standards.