Erratic running of Hornby loco with Bachmann decoder

[Topfeed]

Having replaced the motor of Hornby R3205 'Rood Ashton Hall', it ran-in perfectly in DC with no decoder but, when fitted with a Bachmann 36-566, it starts off OK but then speeds up without any further controller input. It runs faster in reverse than forward direction and is more marked at higher speed settings. I changed the decoder for another of the same type (most of my locos are fitted with them and I've never had a problem), without any improvement. Controller is Digitrax DCI51 Zephyr Xtra. Can anyone help, please? 

[RAF96]

Any chance you could try it with a different controller - friend, club or local model shop maybe. That would help point at where the problem could lie, although if you have other locos running with these decoders under control of the Zephyr then it points to the replacement motor.

[34theletterbetweenB&D]

Why was the motor replaced? Were there any operating problems with the previous motor?

 

Are there any capacitors in the motor circuit? I'd remove any such to see if that makes a difference.

[Topfeed]

Model was purchased as a non-runner, motor replaced with a brand new one, all wiring in loco and tender replaced and checked with a circuit tester. Capacitor was removed. I have done several similar refurbs with excellent results.

[john new]

Apologies if this is what you have already tried. The key to fault tracing a problem like yours is can the problem be replicated?

 

Applying the logic gained from testing software issues over many years at work suggests the next steps are to try either or both (a) that decoder swapped into another loco (b) a tested ok decoder from another loco put into the Hall. Your next step after that wil be dependant on what occurs.

 

In your case (a) above would suggest a chip fault if either the Hall runs properly with a different chip and/or the same issue occurs with that chip put on another loco and with option (b) something in the specific loco to be further traced (but eliminating the initial chip as a likely source of the problem) if the Hall is still problematic.

 

Edited June 7, 2019 by john new
Better explanation for what I meant

[Topfeed]

22 hours ago, john new said:

Apologies if this is what you have already tried. The key to fault tracing a problem like yours is can the problem be replicated?

 

Applying the logic gained from testing software issues over many years at work suggests the next steps are to try either or both (a) that decoder swapped into another loco (b) a tested ok decoder from another loco put into the Hall. Your next step after that wil be dependant on what occurs.

 

In your case (a) above would suggest a chip fault if either the Hall runs properly with a different chip and/or the same issue occurs with that chip put on another loco and with option (b) something in the specific loco to be further traced (but eliminating the initial chip as a likely source of the problem) if the Hall is still problematic.

 

Thanks John, your systematic approach is the right one. I guess I was just hoping someone else might have had the problem and found a solution for me to piggy-back on. I tried the decoder in another loco, and it worked perfectly, so that's not the problem. Then I tried running the loco without a decoder, in DC, and it runs forward with no trouble but, when I switch to reverse, it stalls and needs a huge power input to get going. So, it seems either the new motor is faulty or the motor drive/running gear is binding. Thanks for your input.   

[Butler Henderson]

The Bachmann ESU Lokpilots based /rebadged decoders often need tweaking to their Back EMF CVs to get acceptable running in steam locos - they are usually fine in diesels. Bachmann did produce a list of CV setting for the earlier 36-553 which covered a number of their own models (56xx, Hall, G2A, Ivatt 4 2-6-0, Jubliee, Std 4 2-6-4T  2-6-0 & 4-6-0, 9F)

 

Load control parameter "K" (CV54) usually defaults as 48 and Bachmann  recommended this be set to 12, with the exception of 15 (Std 4 4-6-0) and 25 (9F). The higher this is set the stronger the effect of Back EMF control

Load control parameter "l" (CV55) usually defaults as 32 and Bachmann  recommended this be set to 63, with the exception of 32 (9F), 56 (Jubliee) and 62 (G2A and Std 4 4-6-0). The higher the momentum of the motor (large flywheel or bigger motor diameter, the lower this value has to be set.

You may to be able to switch Back EMF off so narrowing down easily whether that is the problem.  ESU list the option as part of the Extended Configuration #1 CV49 and is switched off by reducing the existing CV value by 1 (and switched back on by adding 1 to the value)

[Topfeed]

On ‎07‎/‎06‎/‎2019 at 09:06, RAF96 said:

Any chance you could try it with a different controller - friend, club or local model shop maybe. That would help point at where the problem could lie, although if you have other locos running with these decoders under control of the Zephyr then it points to the replacement motor.

 

Having taken another look at the motor, it runs smoothly in one direction but needs a huge initial 'kick' from the controller to get it to start in the other, which explains why it appeared to speed up spontaneously as it got under way. Strange that it was ok when tested originally, but failed at some point after it was fitted. Thanks everyone for your input.  

[Topfeed]

Fitted another new motor today and the loco now runs very smoothly in both directions, so RAF96 was spot on. Thanks to all.