Atlas GP35 and TCS CN decoder issue

[Tom4884]

I installed a TCS CN decoder in an Atlas GP35, and it checks out fine on the programming track,  can read and write CVs ok.  When I put it on the layout,  it runs just fine in reverse, but when I switch it to forward it immediately starts tripping the breaker.  I have run several other locomotives in both directions without issue so I am convinced it's something in the installation,  just don't have any clue what to look for in this case. I have installed many of these decoders before and never seen anything like this.  Any help or suggestions greatly appreciated!

Thanks,  Tom

[bigdawgks]

Having installed one recently where I was having issues, my recommendation is to check whether it runs the motor with everything wired up but outside of the frame. I was running into an issue where there was just too much friction when I installed everything into the frame (possibly thanks to the wire going down to the bottom motor brush) and after a little bit of filing and refitting it was OK.

[jagged ben]

How well did you isolate the motor leads from the frame?  Running in one direction torques the motor housing one way in its saddle.  It's possible the motor housing is rotating just enough in the forward direction to make contact with something it shouldn't.

[Lackawannae8]

How well did you isolate the motor leads from the frame?  Running in one direction torques the motor housing one way in its saddle.  It's possible the motor housing is rotating just enough in the forward direction to make contact with something it shouldn't.

id SAY BINGO!!! Those motors have a tendency to torque in one direction depeding on the fit tolerence.
 I wrap kapton tape around the entire motor brush caps then put it back inside the bird cage.
i would check that out before you smoke a deco.

[MK]

How well did you isolate the motor leads from the frame?  Running in one direction torques the motor housing one way in its saddle.  It's possible the motor housing is rotating just enough in the forward direction to make contact with something it shouldn't.

I would say DOUBLE BINGO.  Having installed a few of these, check the seating of the motor, especially those locos that uses a two piece motor saddle instead of the one piece variety.  If not just right they bind, sometimes one direction only.  That would explain the high current draw and eventual tripping of the breaker (a good thing).

(I don't know which motor saddle type the GP35 uses.)

[peteski]

I think that if one of the motor's contacts contacted the frame, that would blow out the H-Bridge transistors on the decoder, not trip the DCC booster's breaker. Well, unless the breaker's trip current is set low enough to trip without damaging the decoder's transistors.

[Tom4884]

Well I am not sure what I fixed, but it seems to be working fine now.   I took it completely apart,  checked all the solder joints, made sure nothing had poked thru the Kapton tape around the brush caps, then reassembled the locomotive and Presto!  It`s working fine.  I ran it for almost an hour forward and reverse with no problems.  Thanks for all the suggestions!

Tom

[bigdawgks]

Yep that pretty much summarizes my experience too. I think there's just the possibility of the motor binding when you reassemble everything. There's really not a lot of frame clearance in many hood diesels. And it seems it doesn't take much motor resistance for these decoders to trip.

[peteski]

Yep that pretty much summarizes my experience too. I think there's just the possibility of the motor binding when you reassemble everything. There's really not a lot of frame clearance in many hood diesels.

True, because they were never designed with isolating motor in mind. The motor power feeding strips actually contacted the chassis halves to route power to the motor. Back in the day Aztec frames milled for DCC had a slot milled in the side of the chassis for routing the motor power wire (along with a channel milled on top of the frame).

[bigdawgks]

Still waiting for the day when they invent zero thickness wires  Or some way to more easily route electrical connections that doesn't create a spaghetti of wires. 3D printed circuits perhaps?

[peteski]

Still waiting for the day when they invent zero thickness wires  Or some way to more easily route electrical connections that doesn't create a spaghetti of wires. 3D printed circuits perhaps?

Flexible circuit PC boards are excellent wire substitute. They are used extensively in consumer electronics (cell phones, etc.) and I have seen some limited use of those in N scale models.

[Sumner]

Flexible circuit PC boards are excellent wire substitute. They are used extensively in consumer electronics (cell phones, etc.) and I have seen some limited use of those in N scale models.
[/img]

True, I use ....



...the ones above quite a bit.



Haven't been able to find them in US/Canada but for a while you could get them out of England but now it looks like you have to be in the EU go get them easily.  It is a shame as they really come in handy for decoder installs and ....



... other uses like above.   I have some more info on my site ( HERE ).

Sumner

[GGNInNScale]

I have had two Atlas diesels (older models) that bound-up after cutting and pasting to change to DCC.  I found that the Kapton tape that I put in on one end caused the flywheel to just touch the frame opening.  Not much visible misalignment, but in one direction, the flywheel jammed against the frame; in the other it was fine.  I marked the flywheels with a felt tip pen, then ran the engine on clips away from the rails so I could see what was happening- the flywheel brass showed through the black ink.  The fix was to put Kapton tape at both ends of the motor so it was centered/aligned in the plastic carrier and filed the openings in the frame just a bit.  Problem went away.