It sounded like binding but when you mentioned it was now running OK forward I figured we were barking up the wrong tree.

Glad you whipped the problem. Also nice to see you report what the problem was. We all learn by these things.

I am glad that you are making some progress.

Something else I got to thinking about. What address are you running the start and stop loco on? I am NO DCC guru but have read about some addresses causing issues with the DCC 'wizardry'. Of course my old brain doesn't remember what these numbers are/were but maybe try a decoder reset and running it on #3.

I recall you were using Digitrax decoders and I believe to do a reset you program CV 8 to #8.

I would think if it is binding that is causing the loco to stop you would see increased current draw. You say it doesn't so I don't think that is the issue.

I have heard about and done the screw loosening thing with some success. Usually I will investigate what the cause is and file away some material to make it run well with the screws just snug; not honked down like the factory does.

I am stumped as to why the loco just randomly stops. You have swapped the decoder and have the same issues which makes me believe that the issue is in the loco which you have pretty much proven is OK.

I don't remember if your loco uses an 8 pin plug for the decoder install. I am wondering if it does, can you install the dummy plug and run the loco on DC to see how well it runs? Can you detect any slowing or other issues? I am wondering if it could be something funky in the motor.

The other thing that I would try is doing Internet searched for similar problems. This has helped me on more than one occasion.

Please, keep us posted as to what you are doing. These issues and their resolutions help the rest of us keep them rolling.

While watching the ammeter on your command station, what does it read when the loco does its random stops?

Back to what Jeff said about the decoder going into thermal shutdown, I would run the loco and pay close attention to what current it is drawing. If it is near the one amp rating of the decoder before it randomly stops then it would appear there is something causing current draw to increase.

It would then be binding maybe.  Very weird issue for sure.

The only other thing I have heard of causing stuff like this is when another throttle has the address of the loco in it and the two throttles fight over control.

If you took the caps out of the circuit then I don't imagine they are causing the issue.

Back to the drawing board.

Did you check stall current at any point in this operation?

Did the motor in the #2 loco that got the first decoder and ran fine have the suppression caps?

I am guessing not and would think the thing to do is cut them off of the 4-8-4 and see what happens.

I re-read this whole post and the only thing I can come up with at this point is to wonder if maybe you have a funky decoder.

I put a Digitrax sound decoder in a loco once that did the same thing as you described. The engine would run fine then slow down and stop.  Then it would take off again. I considered it being a thermal shut down but the locos stall current was tested prior to the decoder install and after the install, track current was monitored with a meter. Current draw remained well below the decoder's rating.

You mentioned replacing the motor. Did the replacement by chance have noise suppression capacitors on it? I have read that the caps can affect the decoder's back EMF feature and they recommend you remove the caps from the motor.

Obviously none of this would affect the pulling power.

Crosswire, it's hard for us to really help you because we are not onsite to see the goings on.

It sounds like you have a real stinker of a problem. I wonder why it isn't pulling better?

Something that I just thought of and maybe you already checked this; but I have run into poor pulling caused by the tender, lifting the rear of the loco up.

Is there any chance the drawbar or wiring is binding any at all? This can lift the rear on the loco up and kill traction.

I forgot to ask if you got the loco to run OK in forward?

A couple of thoughts;

sometimes a headlight, which only requires a minimal amount of current to operate will work fine while the motor which requires maybe a half an amp will not run.

This reflects a questionable connection between the wheels and track.

When there is  a wiring fault someplace a voltmeter may still show voltage present but not 'enough' current available to do any work.

I know you have re-wired recently so we'll assume that you whipped any wiring issues but maybe there is still some oxidation on the rails/wheels.

Hope this helps.

I am not a G scale 'guy' but recently repaired 3 Bachmann large scale locos.

The 2-4-2T required a new motor and Bachmann did not offer the part.

I found a perfect replacement on the Internet. Mabuchi #RS-385PH-2085

It fit perfectly. I hope this helps someone else looking for a motor

On the early K-4 they used a plastic 'hubcap' for the driver spokes. Over time these swell out and begin to hit the siderods. This causes a real waddle or even binding at times.

The black spoke thing can be cut out and the remaining metal spoked wheel can be painted black.

Tycodepot.com

A good place to find diagrams that will help you understand construction and assembly work is'

https://hoseeker.net/

That site will have what you need to understand.

I think Rich is right about trying a drop of oil on each motor bearing.

As soon as 'squeal' is mentioned I think motor lubrication.