One of things I've found (and that many already know) is that jigs and fixtures to make parts can take far longer to make than the parts themselves.
The 3D printer has helped with this a lot in recent years. Often jigs do not need to be that that strong or durable, or even that accurate. Reproducible is often the major criterion.
With a preference for split frames and a general absence of accurate and quality wheels to a profile that I'm happy with I'm more or less constrained to cast driver centres. I also like the idea of turning the crankpin on the centre (whether cast or as a fitted steel blank). Doing this is not particularly difficult, but it does take time to make fixtures and each wheel design needs something slightly different.
For crankpins it's essential that they all have the same throw and cosmetically ideal if they are centred on the boss. Precise crankthrow is not that important. I've made fixtures from brass etc to do this in a more or less kosher way. But it takes time. Big lumps of metal in the Unimat are not that much fun either.
So previously I've drawn a fixture with a male/female relationship to the centre (a casting from digital print) and printed it. Held in the 3 jaw chuck this was effective for turning crankpins on S scale drivers. Any exact centre issues could be overcome by doing all wheels in a batch without removing the fixture from the chuck.
Recently I've found that printed threads of M4 and up (I haven't tried smaller), seem to be able to hold reasonable loads. The resins I use don't tap all that well as they tend to chip rather than cut, although a tap (if available) may be useful in touching things up. For 375 these observations had me wondering if I could print a fixture that would fit directly to the Unimat headstock (M14x1 - no tap on hand). So here it is (M4 screw with a reduced head retaining the casting in a printed thread). Images are after the required wheels have all been done.
The casting has previously been faced off on the rear.
I'm sure there are caveats on the loads such fixtures can sustain, but the loads from crankpin turning will be relatively small. There was a bit of learning and fiddling to get the threads and fixture right, but this is largely one off. I could probably have used a larger OD and eased the fit with the casting a whisker. Future fittings will be cut and paste.
Is it accurate? It seems to be. I don't have a DTI, but visually things looked good and the cast crankpin blank was pretty true before turning. The wheel was a press fit into the fixture with no slop so they should all be the same at least, and visually they are nicely centred in the boss.
I guess if it's possible to print driver centres, then this sort of thing will be no less accurate.
I'll report back when the chassis is wheeled up as to how well it turns out. In the meantime this was a pretty painless exercise. It is not a universal solution, but there are plenty of cases where this approach might save some time.
