Sarod...:
No, I can’t explain.... Your question isn’t something that lends itself easily to a few simple formulas or a number of steps out of a M.E. or Dynamics text book. This has been something that has been studied and developed over the years. Whole companies exist around this subject. The railroads, the AAR (in the U.S.A. and Canada, Mexico), the RR truck and bearing manufactures, have reams of data and testing on this subject, it has just evolved from a basic understanding of journal bearings and then to roller bearings. These groups are constantly discussing, testing and refining these components. It is truly an area of study unto itself, and the bearing manufacture’s literature and AAR Specs. would be a good place to start your study. Obviously, the general theory on bearings remains the same, but this is a very tough environment, and the load classes vary from 50ton, 70, 100, to 125ton in the U.S.; also the railcar types vary (C.G., length, flexibility, etc.) which lead to different dynamic actions such as rock-n-roll, truck hunting, etc. Then the running environment also varies; unit trains tend to run under full load or empty, constantly, many miles per year; while general service railcars run few miles per year and under varying loadings, not always at max. cap’y.; finally maintenance-of-way and jointed vs. welded rail plays a big part in car dynamics, and this is generally improving while miles per year is increasing. Basically, one bearing design serves the worst of all of these conditions. Bearings have become an incredibly refined and dependable railcar component, not to say they don’t fail, but they aren’t the maintenance headache they once were, with journal bearings. Press them on a proper axle to the manufacturer’s specs. and put them under a properly fitted truck and they perform quite well.