We’ve now seen that even reliable old Percy can lose the grip between axles and wheels when things get rough. I’m driven to ask what exactly is the best way to connect the wheels to the axles so they don’t slip?
There are plenty of scale locomotives running around with an interference fit between wheel and axle, and many of these have plastic wheel centres, especially in Britain. #10 was built this way, with EM Gauge Society split axles and Mike Sharman’s Millimetre Range wheels, and should still be running like its similarly-constructed brethren. Yet it isn’t. Perhaps the very air in Britain is special. Maybe that would explain how that island has produced Cornish pasties, The Two Ronnies, bagpipes, and now wheel centres that stick. So to be clear, I am looking for a method of attaching drivers to axles so that they not only maintain their quartering, but so that they maintain their quartering for me, here in North Vancouver, where the air only produces Bannock Burgers, Douglas Coupland1, Bryan Adams and wheel centres that don’t stick.
It’s time for a little science! Maybe very little. In fact, I’m not sure this would pass muster in my kids’ high school science classes, but I have only so much energy to apply to pure research.
I devised an experimental setup involving clamps, a scrap of brass that pressed over a hex-shaped test piece representing the wheel centre, some incredibly strong dental floss, and a ginger ale bottle full of coins. My son, who knows exactly what high school science experiments require blanched in horror when he saw this apparatus. However, remembering that they laughed at Newton, I pressed on and tested four methods of holding onto a slippery stainless steel axle:
- Interference: a 3D printed wheel centre pressed onto the axle with a press.
- Capillary CA: The same interference fit centre, pressed on and CA applied through capillary action. This mimics a likely scenario of rotating the wheels into quarter and gluing after becoming happy with the running.
- Printed key: A .009″ key 3D printed as part of the wheel centre that engages in a slot cut in the end of the axle. The key thickness matches the slot created by my jeweller’s saw.
- Solder: Soft solder flowed into a machined brass centre.
|Failure weights (g)||Trial 1||Trial 2||Trial 3||Trial 4||Average|
I weighed the ginger ale bottle at the point when the joint failed completely. The Capillary CA joints, in particular, collapsed slowly, rather than spilling all the coins on the floor at once. The soldered joints failed to break before running out of coins and room in the ginger ale bottle.
The lever arm is 2.5 cm long, and I suppose I could calculate the torque applied in each instance, but that is not as interesting as the relative strength. Obviously, for maximum strength, solder is the winner. However, that adds to the complexity of wheel construction. The Key also adds complexity, but it is probably manageable. It would be great if an adhesive, especially one that allows for adjustment, could exhibit strength similar to solder or even to the key. I have a few more to try.
1 Actually, I think Douglas Coupland might be from West Van, but ignoring real estate prices, it’s basically the same place.
2 The alignment on this trial was poor, giving it a slight advantage over the other three. Discarding this result, the average is 214g.