I have not had an empty workbench since I last posted something here. Here's something completely different in prototype, scale, the nature of the work and the type of model. This time the job is to improve the look of the wheels of a ready-to-run H0 loco. It's the Fleischmann ex-Prussian T16 tank loco from the 1970s or 1980s, bought in a rather battered condition for purposes of experimentation. The standard of fidelity and the workmanship Fleischmann achieved with these models were very good for their day, though of course expectations have risen since then. But having a solid basis suggests the models will be good candidates for some modification. In particular, reducing the huge flange depth.
Here's a composite view to show the difference in flange height before and after. Judged on the shaping of the wheel boss, spokes and tyre rim this is a nice casting and I think that this adds a lot to the positive overall impression given by the model. Newer mass-market models in H0 scale (which probably have a common origin in Asia) tend not to have such nice looking wheels. It really helps that this is a single metal casting.
Here's the first trial, after I'd worked on the six rearmost wheels. The front two wheels are less easy to see, but the enormous flange still shows up clearly.
And here's the whole lot done.
Working on a lathe is still new territory for me and most jobs require trial and error. In this case I had a set of wheels from another loco to test my ideas. What worked best was to put each wheel face-first in a 15mm collet and then cut straight through the wheel back towards the chuck. I set the tool position to take off an amount that brings the flange height fairly close to the relevant NMRA RP-25 dimension (which is based on the full width of the tyre, not as manufactured but after thinning the wheel). The amount I removed from the flange height is about 0.7 mm, the original height being about 1.2 mm.
I haven't thinned the flange yet because I wanted to reassemble the whole lot and see whether it could cope with real trackwork first. I'd already added some washers to the axles, which had masses of sideplay (the most on the middle axle as expected). More adjustment is still needed now and I might fiddle with the front to back depth of the flanges on the centre drivers -- which would be replicating what was (usually?) done on prototype locos with lots of axles in a rigid frame.
The prototype locos had very closely spaced axles,
Here is a prototype view of a similar loco, apparently missing as many parts as my model. Fleischmann's flanges meant that the wheelbase has to be overscale by about 4 mm, but RP-25 flanges on correctly scaled wheels would still not allow the correct wheelbase. The Fleischmann wheel diameter is 0.5mm underscale so there is a little leeway here. What I would like to do in the end is make new frames for the model and re-use these wheels. This would mean I could get rid of the 'pancake' motor and bring the wheelbase in closer to scale. As noted in my earlier posts this is another area where I have not yet got much experience, so taking time first to tweak the sideplay and find out how to minimise the chances of the loco ending up on the ballast seems to offer a sensible chance of progression. In the past two different aftermarket chassis for the model were available with full compensation and these are still sought after.
Leaving aside the details that were missing when I bought the loco, as well as the ones that I either pulled off deliberately or may have broken, it is not very difficult to pick out features such as the sketchy valve gear that are pretty crude. But if the model can 'take' the new wheels then that's an encouragement to re-detail because it lifts the overall impression so much. The later releases of this model have much more faithful valve gear, available as a spare part if necessary, and there is a cast metal front buffer beam available from a third party manufacturer to replace Fleischmann's one which has the buffers set the wrong distance apart. In my view this is one of those models that seems to look right somehow, a trick that doesn't seem always to correspond exactly with fidelity to the prototype, so I hope to give mine some more pushes in the right general direction.
Ben