The Goons sang about walking backwards for Christmas!

That's how I feel about my progress with the Crab conversion; not exactly backward but pretty well stationary over this last few days.

I have it running very near as well as I find acceptable - in forward gear! - but swing it in to reverse and the thing's as lumpy as mother-in-laws gravy.

Normally I would blame poor relation between wheel centres and coupling rod holes, and/or, bad quartering of the wheels.

It can be none of those. It worked OK as OO with the same chassis, the same wheels and the same coupling rods, and I have checked and re-checked the quartering. The original Bachmann axles were 3mm diameter and were replaced by EM length 3mm ones but I did notice that the rearmost axle was a bit sloppy in its slot, with too much fore and aft movement. Obviously I had found the problem.

"Obviously"- nothing!

I checked that a 1/8" axle sat the slot comfortably and it did, so I made a new axle from 1/8" silver steel and re-assembled the chassis. Absolutely no change in the running. I have now used every swear word I know and made up another half dozen, but the thing will not run smoothly in reverse. For the time being and particularly for my sanity, the model is currently in the "Try again later Dave" box.

So I will this week tell you about my method of EM-ing Bachmann driving wheels. On all such drivers that I have inspected, and certainly on the K3 and the Crab, each driving wheel has a short boss around the axle hole on the wheel's rear face. I tried chucking a wheel on this small boss and found that my three-jaw chuck would just not hold it. On the wheels I have, these bosses are about 1/8"in projection and ¼" in diameter and while the three-jaw would just not grip them, a ¼"collet would.

Even then the grip was minimal but with a newly sharpened tool and very shallow cuts, I was able to skim off the required amount and get a wheel that was 0.090" thick and with the spokes, balance weight (part of the wheel casting) and the front axle bosses, skimmed by the same amount, the wheel becomes a good profile for EM.

The job has to be approached with rather more gentleness than one associates with a machine like the Myford ML7 and I admit that on more than one occasion I succeeded in flicking the casting out of the collet. However I quickly realized that those occasions had all been achieved when I tried to quicken up the progress by taking deeper cuts. Keeping the cuts to no greater than 0.002" allowed the job to be completed to fruition.

I decided that to re-use the Bachmann plastic, top-hat bushes that insulate the all-metal wheels from the chassis, and matching the turned down ends of the original axles, was a better proposition than making new insulated bushes on a plain axle. This was very possibly an error because I have come to think that the lumpy running problem may be due to one of the wheels on the motorised axle not being truly square to its axle and the very slight wobble is producing a virtual length change of that axle's centre line.

When I have had a few hours of more successful modelling on some other project I'll come back to the Crab and solve its problems. Mean while I'm going to build a few wagon kits.