London’s Southbank University was the venue for a fascinating seminar on Practical Surveying and Gauging, organised by the London Section of the Permanent Way Institution. The panel of speakers were all experts in their own fields so it promised to be an excellent event. The session was chaired by the Institution President, Steve Whitmore, and the keynote address was given
by Network Rail’s senior survey engineer Chris Preston. Chris developed the theme that he had first raised in the January issue of The Rail Engineer (issue 99) and challenged his audience with the title of his address, “So you like taking risks?” The essential message of this was the seriousness of the risks that are implicit in conducting infrastructure projects without the right surveying.
Good surveying need only take up two or three percent of a project’s budget, but the costs incurred by proceeding without the right information can far exceed this. Chris referred to Network Rail’s published standards for surveying, and briefly described how projects should work from these to ensure that they delivered surveys and survey data that were compliant. Finally he described some of the innovative survey systems the company is now using.
These included vehicle-mounted and helicopter- mounted LiDAR systems used for such purposes as assessing level crossings for vehicle grounding risk and looking for trees that are at risk of falling onto the line from adjoining land. He also outlined a system called RILA, a combination of laser scanner and GNSS receiver which can be mounted in minutes on the rear of a service train to survey track geometry.
Another Network Rail speaker, Barry Gleeson, survey systems engineer for Thameslink, described how that project is delivering the good practice that Chris had previously outlined in its Key Output 2 (KO2) phase. In his talk on “Survey for Track Alignment”, Barry spoke of the lessons learned from the mistakes of earlier phases of the project. He described the challenges of surveying and mapping infrastructure on a curved planet, the methods used to manage this within the tight tolerances required for railway track works and the management processes and procedures being applied by the project team collaboratively in order to deliver what is required.
Attwell Mlilo, Balfour Beatty Rail’s survey manager for Thameslink KO2, went into some detail in discussing “Setting Out”. He gave a description of this process, described its purpose, and covered the standards and principles, good practices, methods and equipment that are involved. He described how setting out for track renewal today is undertaken using 3D modelling and laser control of machines in 3D, with 2D modelling available as a means of checking this process. Done well, setting out reduces installation costs, improves margins and cuts down the follow-up requirements whilst resulting in better quality track and lower maintenance costs in the future.
Track surveying and gauging
Peter Roberts of Costain spoke on “Track Survey and Tamping” and covered his subject in a similar fashion. He looked back at the history of tamping, described its objectives and the elements of good practice. The current state of the art was defined and the methods of recording and verifying the quality of the results were outlined. Finally Peter discussed recent and likely future innovation in the field.
“The Network Rail Structure Gauging Train (SGT)” was covered by Kevin Hope, principal engineer plant and T&RS in Network Rail’s Technical Services department. He described the limitations of the old SGT inherited from BR Research that had led to the decision to introduce the current system, SGT1, in 2009. The new system, LaserFlex from SGT, uses a clever arrangement of lasers and cameras, in which a combination of tight control of laser frequency, narrow band pass filters and pixel masking on the cameras virtually eliminates interference from ambient lights such as platform lighting on stations. It is mounted on the end of a vehicle, in between vehicles in the train.
The system has been so successful that its software is to be modified to allow it to gauge platforms, something that would have been unthinkable with its predecessor. A second train is to be procured too, SGT2. Together these trains will allow Network Rail to fully implement a risk based structure gauging programme where those structures with tight clearances are re-checked more frequently, particularly those on high speed and busy routes. Previously the standards required re-gauging on a fixed frequency irrespective of risk.
Kevin completed his presentation with a brief review of how Network Rail records and analyses gauging data, and looked at some current and future innovations. These include the TiCleD system that actually reviews the clearances on a route by vehicle type for each type cleared to run on that route.
The final speaker was the well known Dr David Johnson, formerly of LaserRail and now with DGauge. David looked to the future of gauging and the probable introduction of overtly probabilistic methods of route clearance. He showed the need for such an approach by describing how Platform 12 at Liverpool Street Station is frequently used every day by Class 321 trains without incident despite the fact that current route clearance procedures show that there should be a clash of about 75mm between train and platform!
David showed how this arises as a result of current methodology, which sums the worst possible combinations of the variables to examine the clearance that remains. Variables include train loading, vehicle movement (sway), track variance from design position and so on. It is intuitively obvious that in reality it is unlikely that the worst values of all of these will occur at once, and so it is in practice for the Class 321 at Liverpool Street. Fortunately, these trains were running there before anyone applied the gauging rules and they acquired “grandfather rights” before they were stopped!
How can this be avoided in future? Well, David was able to suggest an answer. This was the application of probabilistic analysis that would allow the derivation of numeric probabilities to given amounts of clearance (or clash) at the point under consideration. This would permit the application of engineering judgement in a numerate and auditable manner. Thus it might be agreed that a clash of 5mm was allowable if its probability was less than, say 0.1%, or that a clearance of only 5mm was acceptable if it was 99.99% certain to be no lower than that.
David’s presentation brought the proceedings to a close. Together, the speakers had given a fascinating insight into various aspects of surveying and gauging which kept the audience enthralled.
Next month’s issue of The Rail Engineer will include a Focus feature on Surveying in which these and other topics will be covered in more detail. Make sure you see your own personal copy by contacting subscribe@rail- media.com today!