THERE IS now a far greater understanding of rolling contact fatigue in the UK than there was on October 17 2000 when the Hatfield derailment occurred. In part this is due to the Wheel-Rail Interface System Authority, set up in 2001 to co-ordinate the response to RCF by a shattered and demoralised industry struggling to cope with the aftermath of Hatfield. But WRISA’s members voted on December 9 to wind up the organisation, and plans to replace it have yet to be firmed up. With 12 individuals and two companies facing trial for manslaughter and other charges arising from Hatfield, it proved impossibly expensive to get personal liability insurance.Addressing in November a conference in London on wheel-rail interface management, WRISA’s Chairman Brian Clementson said of the RCF pandemic ‘similar problems are occurring on railways around the world’. But while the basic causes are better understood, there is always risk in changing wheel profiles or rail steels. So progress must involve carefully-monitored trials on the live railway.Hence the frustration expressed by delegates at the apparent inability to change anything at all in the UK. Stuart Grassie, who has contributed several articles on wheel-rail interaction to this journal, complained that ‘there doesn’t seem to be anyone who can make a decision about anything’, and this applied ‘no matter how much evidence there is’. London Underground and the rest of Europe all use BN260 rail – indeed ‘every railway in the world uses head-hardened rail’ (p42), but ‘because Hatfield occurred on head-hardened rail, you cannot use anything except 220 rail’.Rob Haskel of ProRail, reported ‘a huge rise in rail defects’ in the Netherlands, which doubled from 654 in 2000 to 1585 in 2001. But the research programme reported on pp38-40 appears to be getting results. In the UK, RCF research is being hobbled by the difficulty of working on live track, according to Mark Dembosky, Network Rail’s Dynamic Systems Specialist. Access to high speed lines in Germany and France was far simpler than ‘the amount of rigmarole you have to go through to get on the track here’, he said. Tangential forces that result from what Dembosky called ‘a dancing contact patch’ often cause RCF ‘clusters’ when the track is subject to short wavelength misalignments, as in switches and crossings, or with wheel treads that have worn hollow. A shifting contact patch causes the rolling radius to change so quickly that the wheels skid momentarily, generating large shear forces. But RCF cracks only become dangerous when they grow faster than the rail wears. Dembosky explained why the high rail in curves of around 1500m radius is particularly vulnerable. In sharper curves, the leading wheelset in each bogie wears away the gauge corner. In shallower curves, the self-steering quality of coned wheelsets keeps flanges away from the rail. The Hatfield curve was just under 1500m.