Wheels ~2-inch farther out on commuter railcar axle derail and then rerail at switch frog.

[Robert Gift]

Reminds me of the Atlas HO track black plastic RR crossing sections which rerail derailed cars.
Never knew of a track switch rerailing any cars.

 

[808Beachbum]

Pretty interesting, especially since I worked for at least a couple years as a Journeyman Machinist in the Wheel Shop at UP Omaha. This is where the old wheelset assemblies were disassembled, each component re-worked, reassembled, gauged and inspected prior to shipping to other points or being installed on freight cars or locos in our complex.

Once disassembled, the axles would be inspected electronically for internal defects within the journal area, prior to being either cut down, or reground, depending on condition and final diameter that could be obtained within specs. Wheels also were re-profiled if sufficient material remained, both loco and freight wheels, and re-bored to fit a specific new or reworked axle. Roller bearing boxes for the locos, and roller bearing assemblies for freight wheels were also stripped, inspected, reworked and rebuilt to spec in the Wheel Shop "clean room".

The interference fit for mounting wheels was 1/10,000 per 1 inch journal diameter, and it took somewhere between 60 and 120 TONS of pressure per square inch to press them onto the axles. The actual tonnage varied based on journal diameter and depth of the wheel or gear seat. (we also re-profiled gears for the drive axles of locos that mounted into the traction motors. They were press fit in a similar manner prior to mounting wheels) It was amazing to see how far "most" of those cast steel wheels would flex as the pressure built. I say most, because every once in a while, both during the dismounting OR mounting operations, one would fragment. Press operators always stood at the outside end of the press main beam until they reached the proper pressure, with no indicated signs on the charts created live of any "galling" between the wheel and axle. That would be when friction wins, and one or the other component starts peeling deep gouges out of the other component. The Press pressure would spike up, and the operator would stop and release all pressure. The wheels, only partially on the journal, would need to be dismounted, and either manually ground smooth for another try, or sent back for re-machining and mounting another day.

One point they don't make mention of, at 2 inches out of gauge, and depending on the actual profile of the wheels on the axle that derailed, they would be riding VERY high on the rail, with the wheel radius lifting the tread completely off the rail I would think. The wheelsets certainly never went out the door being so far out of gauge, although perhaps they were damaged in some other incident and not taken out of service. That seems doubtful, since they tested and found so many out of gauge.

It's hard for me to imagine a wheel being able to work itself off that way. Perhaps they use an inferior metal for either the axles or wheels compared to what UP did. My Dad was actually the head of the MVP&M Lab, and they were always doing destructive testing on a wide variety of components including wheels and axles. I had some good access to observe some of their testing, very interesting processes.