modelsof1900's Get off yer duff Challenge (Part III)

[modelsof1900]

Sorry, not written for a long time and there is not a progress with my N&W class M.
I see all the progress on your models and I'm not sure if I can finish this loco.
Really, after some heavy mistakes with decoders I was sure that the problem was solved. Because not so good running of loco I replaced the motor and now the gear with large tooths sounds louder then before and the loco does not run many better - until today where I new decode problem stopped all activities. For a long time I did not found a non-fixed short in the system until I removed all additional power feeders on the loco and removed all the not used wiring again.
However I see that the running performance of the model is not that what this decoder should do. There are speed changes in acceleration and more yet while deceleration and that also with two different decoders.
What could be the reason for this untypical performance? Now I'm working since four weeks with this model and I'm not one step closer to finishing as before.
Maybe that I start tomorrow with a simple basic wiring - two wires from rail to decoder and two more from the decoder to motor, without lights, without sound. And maybe that I will start with a 100 percent new wiring hoping that I can eliminate possiblely undetected shorts or creeping current between lights, sound and motor powering. This is the first time that I did not found a reason for a problem in spite of tests over tests and attempts with different parts in this digitalizing project.
In moment I'm not very lucky with this model. But here the last picture of last mechanical modification at this model, a new motor with a divided power line from motor to gear. This works well, smoothly and gently.