On rail breaks

Sir - I was very interested to read the Behind the News item on rail breaks in RG 6.99 p346, which contains fascinating figures regarding rail breaks and rail replacements on the Railtrack network.

You report that Railtrack expects to replace only about 0·8% of its 32000 km of rail in each of the next 10 years, compared to a European ’norm’ of about 2%. Another report 1 suggests that in the 1980s British Rail’s rail replacement rate was similar to that of other European railways, but has declined fairly continuously since then. Clearly if Railtrack could achieve such a reduction in rail replacement, it would represent enormous savings. However, it would also imply that our European colleagues, and perhaps also British Rail in the past, have been missing something fairly fundamental.

I am not aware of Railtrack’s reasons for believing that a reduction of rail renewals to 0·8% can be achieved without heartache and the problems mentioned in your report. However, there is a very well documented example of where such a reduction has been achieved, and others which are not so well documented. One freight railway in North America with around 17000 track-km used to renew 2·4% of its rail, or 410 track-km per annum. Between 1982 and 1994, this rate was cut, so that in 1994 only 0·75% or 130 track-km was replaced. Renewals have continued at approximately this level since. Rather ironically, this almost directly reflects Railtrack’s reduction in rail replacements over the same period.

However, contrary to the experience in Britain, significant reductions have also been obtained in rail defects, including breaks. Clearly the conditions are not identical to those on Railtrack. While speeds are lower, and traffic is largely freight rather than passenger, the network, climate and vehicles are such that conditions for fatigue failure of rails are much more severe. So what Railtrack would like to achieve is possible, in principle, and since this ’experiment’ in North America has now been running for more than 15 years, its success would appear not to be a stroke of luck. Similar results are being obtained in a European experiment with which I have been involved for the last few years. From this it would appear that one does not need to wait for 15 years to obtain the benefits, and that the track does not realise it is in a different part of the world and behave differently accordingly.

The North American railway credits its achievement largely to its steady adoption of a preventive maintenance policy, the main component of which has been a radical change in its application of rail grinding. In 1982, the railway ground 17% of its track, or 2900 track-km, and on average about 75 MGT of traffic was carried between grinding. But by 1994 this had increased to 15000 track-km or 88%, and the average tonnage carried between grinding was a mere 15 MGT.

Much of the track is ground several times per annum, while lightly trafficked routes may not be ground at all. Almost all grinding is undertaken in a single pass. Moreover, this helps to keep the cost of grinding per finished km to a small fraction of the typical cost in Europe.

For the sake of Railtrack, its customers, and the travelling public, I hope that its experiment is a success. However, one must assume that Railtrack believes in an alternative way of reducing the cost of rail maintenance and renewals in the long term than that adopted successfully elsewhere, since significantly less than 10% of the track in Britain is ground every year. Your report of an ’alarming rise in rail breaks’ suggests that this alternative strategy is still under development.

Dr Stuart Grassie

Bearsden, Glasgow, Great Britain

Reference.1. Railway Finances. Fourth Report of the Transport Committee of the House of Commons, HMSO, London, 1995. Vol 1, para 98.