Thanks chaps, I appreciate the feedback re- cardan shafts I will revisit this approach especially because it’s technically the correct scale solution. I also believe that the minimum quoted radius of 4.3m is unlikely to be present at my local track so hopefully my calculated bogie angle is greater than I’m likely to encounter in practice.

Thanks Howard, I’ll draw it up and ‘ see how it sits’. Actually it wouldn’t be especially expensive to try a mock up, I’ll give it a go. Thanks for your words of encouragement:0)

Hi peter, I hear you, I see most 5” gauge locos use motors mounted in the bogies which nestle between the flanges of the locomotives wheels. My issue is ld like to stick to 3 1/2” gauge which leaves a bout 83.5mm between the wheel flanges. It would be very tight to get a small mabuchi 540 motor in there with step down gearing and final drive on to the axles, I also doubt durability of such brushed DC motors. Modern trends suggest using brushless motors together with dedicated electronic drives, i recon I could get one of the smaller motors in, I just worry about obtaining sufficient gearing to produce sufficient torque. And whilst there is no ‘strict requirement’ for my loco to be passenger hauling, the club does seem to encourage this capability.

One option I am considering is to mount the brushless motor above the bogie, so that it protrudes into the superstructure of the locomotive then use toothed belt drive, as you suggest to provide the drive and gearing to the axle below.

I like the idea of using drive shafts because that’s how the real early dmus were driven, it also allows me to use just one main motor located in the superstructure driving through a long lay shaft mounted beneath the chassis.

pros and cons in both solutions, simple belt drives from multiple motors and speed controllers  (costly) or one motor with more involved mechanical drive shafts. I’m having to weight this up.

i should add that I appreciate all of your comments, it’s really helpful having others critique my thoughts on design, especially as there are so few smaller diesel models being made.

Hi Duncan, Agreed multiple controllers not required for permanent magnet brushed motors, but I think sensored brushless motors are likely to give better control and low speed torque.

I understand the idea of using the outside frame,I was going to make it purely cosmetic, but I’m not sure about hiding belt drives, I was thinking of using 10mm toothed belts, that would be quite a gap to hide. I do like the idea though, thank you, especially as the in board belt and pullies take up significant space between the wheels.

Interesting idea Peter thanks, I have to confess I was thinking of a single stage epicyclic gear box on the end of the motor, but your idea opens up some options. I would like to power both axles but I might be able to squeeze a second pulley along side the first.

The search for a gearbox begins.

The dmu telescopic prop shaft could indeed be driven by an under-slung motor or even from a large motor mounted above, lots of room up there! Then belt drive to a longitudinal shaft below.

I’ll try both options on my drawn scheme.

Another idea is to use this Chinese made bevel gear box, this one has 8mm dia. shafts but I believe they make one with 12mm shafts which would be suitable for the DMU. The down side is the ratio is 1:1 so the gearing will have to take place elsewhere.

On 17 November 2025 at 10:47 noel shelley Said:

Several thoughts ! If the cardan shafts transmit the motor speed then load is small, if the shaft transmits gearbox speed then the load/torque will be much higher. The angle you mention will be no problem. I transmit 300Hp at 3000rpm, as the shafts in the drive pass through the gear boxes the UJs get bigger, the final drive UJs are about 6″  Worm and wheel reduction gives good reductions but on power to the motor stopping there may be little or no over run ie it stops dead ! This does depend on the worm and wheel design. Tooth belts or small chains will work. Good Luck. Noel.

Thanks Noel, your experience is very helpful, I would prefer to run the cardan shafts at axle speed, this equates to a max of 1500rpm, I worry about ‘whip’. Power in my ‘small’ loco will be limited to 1kW max. I calculations suggest  likely running load will be significantly less than this.

I like the simplicity of the DMU loco running gear, my plan is to use an OS 120 2 stroke engine to drive a PMDC motor as a generator, use the DC to feed a similar single drive motor through a set of switched series resistors to limit max current. The PMDC drive motor drives a lay shaft run longitudinally under the chassis rails via step down toothed belt and pullies, with cardan shafts at each end linking to the bogies. I’ll let you know if this ever gets off the ground. I am determined that my loco shall be engine driven despite the additional complexity.

I like it. 5″ gauge certainly offers more options due to the extra space.

I have considered building a class 20 at 5″ gauge or a class 35 ‘Hymek’ at a push but even these are are chunky vehicles at 5″. I am going to persist with my chosen smaller gauge even though it might be my undoing :0)