Monitoring for research at Yetminster rail underbridge
Monitoring at on site at Yetminster and later reconstructed at the University of Surrey overcame tight timescales, adaptable scope and even flooding conditions.
Data collection for the Yetminster rail underbridge will inform research studies into the behaviour of offset U-frame stiffening and fatigue in metallic structure connections. The Yetminster Bridge in Dorset was scheduled for reconstruction, providing the ideal opportunity for load testing.
Network Rail have taken a proactive approach to the management of their ageing Victorian assets in a plan to lead the industry through handling the stewardship responsibilities in this and other areas. Bringing together the expertise of academia, SMEs and specialists, together we aspire to improve the effectiveness of asset management. Short and long term monitoring is being used to collect vital data and target the prioritisation of efforts and investment.
Research and investigations
The bridge consists of two girders in a half-through arrangement supporting a single rail track. The 129 year-old superstructure is formed of early-steel plates fixed with wrought iron angles and rivet connections. Cross-beams of steel joist section connect to the main girders through riveted angles and provide a U-frame restraint against lateral torsional buckling in main girders.
Differing from current rules of assessment calculations, the cross-beams are offset from the vertical stiffener angles and rely on a torsional resistance in the section of unstiffened girder between the two. This is not abnormal in structures of this age but there is no guidance for numerical assessment in today’s standards. The opportunity was taken for a research programme involving monitoring on site and reconstructed full-scale off-site.
The research planned to transport the whole superstructure, unaltered, to the University of Surrey facilities, but identified that measurements in-service would be vital to understand the real loading behaviour. Consultant Mott Macdonald, worked for Network Rail and created a specification. Testing on site was earmarked for a framework monitoring contractor, but at short-notice the supplier were unable to provide the services required. James Fisher Testing Services (JFTS) were contacted to collaborate with less than 2 weeks’ notice before the decommissioning.
Adaptation and collaboration
From our wealth of experience working previously with Mott Macdonald on similar scopes of rail underbridge monitoring schemes, JFTS were able to collaborate immediately. With a stock of monitoring equipment ready for implementation, we could serve the monitoring requirements, and by understanding in detail the research needs we could suggest alternative methods to achieve the objectives.
The scope as presented to us, assumed the installation of structural arms for measurement of displacement of the operational bridge girders relative to adjacent unused girders and the river bed. However, the preparation works for the reconstruction had already installed river narrowing temporary dams, which obstructed such methodology.
Monitoring and testing
JTFS are able to conduct monitoring by digital image correlation techniques. From a static camera position, sub-millimetre measurements are taken at many points within the field of view, replacing the need for physical displacement gauges. Augmented with accelerometers, additional data was taken correlating with the digital imagery and adding locations which were out of view. These techniques required no arms for measurement relative to other features and could be installed in less time on site. Soon after, the bridge was off-loaded at the University of Surrey where JFTS continued testing, by jacking controlled loads into the bridge and measuring at higher accuracy and more intensely the effects of the abnormal U-frame stiffening. Measurements included displacements and strains, which could be directly comparable to finite element modelling.
In the days immediately preceding the deconstruction, JFTS had mobilised to site and began installation of gauges, when Storm Dennis hit the south of England. River levels rose to above the girder bottom flange and access became unsafe. The already limited working time was reduced by a further 2 days before river level subsided and it was safe to continue, but JFTS were able to reset the monitoring and continue collecting data for train movements in the final day that the bridge was in-service.
For testing off-site, we acknowledged that the scope would need to respond to the findings and therefore were willing to make adjustments as testing commenced. As this began, we found a need for 50% more strain gauges to be added, which were quickly applied. Jacking was originally planned for only one or two load application positions at any time, but a desire to closer represent actual rail bogey loads was determined and required increase to four load points. JFTS were capable to upscale at short notice and continue.
JFTS collected data on the bridge live load effects in this demanding scheme. Having proven expertise and flexibility, JFTS where selected as partner organisation of choice and continue to maintain great working relationships with all parties involved.