Increasing the sizes of rail freight wagons causes a big problem - what to do with overbridges. The most common approach to providing enough clearance is to lower the track - in theory a more cost effective solution than replacing the overbridge. But track lowering can adversely affect track quality and ultimately line speed, if the consequences are not fully understood and a robust trackbed design adopted.
This was highlighted in a recent investigation undertaken on a rural railway line by Scott Wilson Pavement Engineering.
The line had experienced substantial freight growth over the last few years, and site investigation undertaken pre-privatisation of the rail network had been insufficient to predict deterioration in track quality. This led to an under-engineered specification that had since required further remedial works.
Dr Phil Sharpe, principal railway engineer with Scott Wilson Pavement Engineering, explains:
'Traditionally, a trackbed investigation consists of a series of hand dug trial pits. The work is often performed in short time periods between trains, or at night, which are far from ideal conditions.
The trial pits are shallow and do little more than scratch at the surface.
And according to Sharpe, this is where the problems lay. When the track was lowered to increase the clearance under the existing bridge, the original sand blanket layer, acting as a filter between the ballast and clay subgrade, was destroyed.
Although new ballast was placed on a geomembrane, the reduction in overburden pressure and creation of a 'sump' encouraged moisture to enter the trackbed and soften the clay subgrade. Clay slurry accumulated and track quality deteriorated.
'Attempts to rectify the matter only made things worse and line speeds were reduced to 30km/h, ' says Sharpe. At this stage nobody really knew what had gone wrong so Railtrack called in SWPE to perform a detailed site investigation.
The techniques used were fairly new to the rail industry.
Ground probing radar was employed first, producing a longitudinal profile which clearly showed a reduction in ballast depth as the line went beneath the bridge - it also indicated that clay slurry was pumping to the surface.
GPR was useful for an overview, but physical sampling was required to properly identify the cause of the problems.
Rather than trial pitting, SWPE used its automatic ballast sampler, a device specifically developed for trackbed sampling.
'The design has been perfected over many years, ' says Sharpe.
'Typically, we take 30 ABS samples, 1m deep, through ballast and into subgrade, in a four hour possession.'
Results indicated that the trackbed contained various layers of soft clay, sand blanket and slurried ballast. Sometimes fragments of geomembrane were recovered, installed with the new ballast but subsequently destroyed as attempts had been made to stabilise the track by cleaning the ballast.
There was no choice available to rectify the situation other than the installation of a deep blanket, new ballast and improved drainage to lower the water table.
Further testing on the line was conducted with a falling weight deflectometer. These tests showed the weakness of the trackbed layers where track lowering had been performed.
'Our research has clearly shown a correlation between FWD deflection, ballast contamination and track quality, ' says Sharpe. 'In fact, we can now begin to predict the likely track quality for a range of maintenance and renewal techniques like tamping, ballast cleaning or installation of a blanket, once we have identified the characteristics of a site or route.'
SWPE technical director Robert Armitage says: 'This is a classic example, which demonstrates the importance of an appropriate engineering assessment before work such as renewals or track lowering are performed. The techniques are based on those we introduced to highway pavement assessment some 15 years ago - we have now adapted them for railways and they are being acknowledged by Railtrack as a key factor to help bring about improved track quality.'
Armitage claims that, by adopting the SWPE trackbed investigation techniques along an entire route, maintenance and renewal costs could be slashed by 50% over a 10 year period - through better targeting of the resources available.