Part of UP's GP-38-2 problem in Spokane came from them being left idling for long periods and then sent to Plummer in Run-8 from basicly cold, carbon built up from the long idle period lets go once the manifolds get good and hot and there are no turbo blades to break it up so the large pieces stay hot enough to make it to the ground and start a fire. UP also has had lots of problems with the ex-MoPac non-turbo Geeps and their liberated exhausts that MoPac liked to add --- then not maintain --- they not only carbon up but throw oil and make them good fire starters too. GP-38's, and other non-turbo units, that are run regularly and not left in idle for long periods of time don't start fires from their exhaust. GE Dash-8 and Dash-9 units and a few of the Evo series start fires for other reasons in addition to carbon build up. They have lower pressure exhaust out the stack and can build up carbon even when kept hot, especially if they are running a bit rich. They also can get into a bog down situation and produce a lot of black smoke that can turn into a ball of fire when it gets air. (Good for lighting timber tunnel linings and overhead trees on fire.) BNSF has quite a number of GP-38's, and a couple of SW-1500's in the northwest. Also, another reason that the class ones tend to isolate older Geeps moving over the road in power consists it to prevent traction motor damage to the units. The older units try to keep up with the high pulling capacity of their computerized cousins and the short time ratings of the traction motors (and the dynamic brakes) is often exceeded to the point of melt down causing expensive damage to the units. This is the main reason older Geeps are usually isolated in main line power consists led by modern units on both UP and BNSF, exhaust caused fires is a very secondary reason in most areas.