John Cutler
An update.
I was not at all happy with the slow-speed running of the converted Terrier. So I bit the bullet and followed Pete Hill’s advice to replace the wheels with 2.5mm axle versions from Alan Gibson and carefully ream out the Hornby brass bearings. The bearings protrude slightly from the frames so I remeasured the spacing washers needed and settled on 2x1mm for each of the rear (driven) wheels but only a 1mm and a 0.5mm washer for the others. I suspect you do not have to over-worry about this as the splasher clearances are so tight the wheels are not going to wander too much. One recommendation is not to glue the final gear into place. Its position on the driving axle is quite critical if you want to avoid unnecessary friction. So by relying on just abrasion of the axle to retain it, the gear can be (carefully!) slid along the axle a bit if you get the position slightly wrong at the first attempt.
I found that some of the splashers were tight to the wheels, especially near the ends of the splashers. So more careful filing was needed. I ended up using 3M Magic Tape to insulate the insides of the splashers (a bit fiddly to cut the shapes required but once you get one right you can use it as a template).
However the running was still a bit erratic. The Hornby coupling rods seemed to flop all over the place. I reduced this by inserting brass AG coupling rod insert bushes between the rods and the crankpin nuts. Thus restricted, the coupling rod (rivet) joint proceeded to foul the axle end once per revolution! So another bush was inserted behind the AG crankpin bearing on the centre axle and this gives adequate clearance. I did buy the Alan Gibson coupling rods as substitutes but they are very chunky compared to the Hornby ones and need some careful riveting work to replicate the Hornby joints. I do wonder whether the jointed rods are overengineered considering the middle axle is not sprung (or compensated). I persevered with the original Hornby rods.
I discovered the rear axle crankpin nut would occasionally and erratically (caused by the rods flailing about?) clout the cab steps. So the rears of the steps were filed back slightly at 45˚ for extra clearance. Likewise I found I needed to carefully file away the rears of the brake shoes, again at 45˚, to get free running. For some reason this was more of a problem with the flanges of the rear drivers.
I realised the replacement phosphor-bronze wire pick-ups are a bit visible but they can be blackened with a Sharpie permanent marker to reduce the metallic glare. I found that pickup from the centre wheels is erratic or negligible because they are ever so slightly raised (by 0.1mm?); I think this is an old trick from the 1950s to try and stop the loco rocking on this axle (or is it?). I do wish Hornby had sprung this centre axle (why did they bother with jointed rods? Did they plan for a sprung axle and change their minds?).
I now have an EM Terrier capable of slow speed comparable to the Hornby M7. Unfortunately it is not as good a steady crawler as my Bachmann pannier.