Problem with TCS CN-GP and LL GP 20

Ristooch Dec 31, 2010

  1. Ristooch

    Ristooch TrainBoard Member

    I just completed installing a CN-GP decoder into an n-scale Life Like GP 20 that I have had on the shelf for a number of years. To my recollection the loco ran fine in analog mode on my DCC (Digitrax DB150/UT-1/UT-2 Genesis) layout. All my existing locos run fine.

    I have done several hard wire and drop in decoder installations with very few problems, so I consider myself experienced. Here's the situation:

    1. Installed the TCS-CN-GP decoder in the GP 20 following instructions. Everything went well with the installation.
    2. As soon as I turned on power to my test/programming track, the GP20 went forward at full throttle with the forward light on. As far as I can tell there are no shorts because the locomotive did not “sing” when I turned on track power, nor did the DB 150 beep to indicate a short.
    3. I have assumed that, if there was a short somewhere, I would have heard something and/or “let the smoke out” of the decoder.
    4. I decided to try a factory reset. I did notice, that when in programming mode and after I enter the CV value, the locomotive would move forward a bit. The DB 150 beeped to indicate that the CV was accepted, just like normal.
    5. I tried changing the locomotive address a couple times and each time the locomotive moved a bit and the DB 150 beeped just like always, indicating that it accepted the CV change.
    6. Still, as soon as I turned on track power, the locomotive took off with its forward light on.

    What do you think might cause this? Have I overlooked something and caused a short? I would appreciate any help you can provide.

    BTW I have emailed this same information to TCS tech support but thought I would post to the forum and see if any of you have any helpful ideas. Thank you for your assistance and Happy New Year.
  2. Scott Stutzman

    Scott Stutzman TrainBoard Member

    The decoder is faulty. I had a couple do the same thing a while back.
    TCS did promptly replaced them and I installed the new ones. No problems since them.
  3. Ristooch

    Ristooch TrainBoard Member

    Dear Scott:
    Thanks for the quick reply! I guess I was half-hoping I had overlooked something that I could easily fix. But it's reassuring to now the problem is not of my causing. Thanks again and best wishes in the New Year.
  4. jagged ben

    jagged ben TrainBoard Member

    Incidentally, that's normal for these decoders.
  5. Ristooch

    Ristooch TrainBoard Member

    Thank you for your response, Ben. That's good information. Since this was my first CN-GP installation I was unsure as to whether or not this was normal. All my other locos "lurch" very slightly so that's what I expected.
  6. bob303

    bob303 TrainBoard Member

    All -

    I'm not sure if it is appropriate to add to such an old thread. (Newbie here - perhaps this will be moved to the right spot) My recent experience (Oct 2022) with TCS CN-GP in LL 1998 GP-20 has been very good so far. The first couple I had the boards (PCB) a bit loose, but I eventually found that if I would add solder on both sides of the "engagement tab" on the decoder I could sneak up on a very tight fit by filing the solder off and testing, filing, testing. The snug fit was a bit tricky and I'll have to redo the first couple now that I have a few that are tight, but not too tight. I elected to add solder to the PCB/decoder tabs rather than bend the frame with pliers or spread it with a screwdriver. Both squeezing and spreading were the TCS install suggestions but I felt too clumsy to get the right amount of force. The solder, file and test seems to work without risking breaking the frame tabs that hold the PCB/decoder board.

    When the board was loose - (and I didn't realize it as it seemed initially good to me) - the operation was off and on. So for me, I think the source of my problem was getting the right amount of a snug fit.

    Warning - this is all newbie experience and may be off base - please feel free to correct if you see something amiss.

    This may be obvious to everyone else: I think another good tip (sorry I don't recall the source) is to solder the motor wires on without trimming them (so extra long), reassemble the frame, Kapton tape and then after testing and seeing how much wire length you need, detach from PCB, cut to length and resolder the motor wires to the front PCB. Challenge as these pads are extra small (smaller than the rear pads) and close together. I bought a 10x loupe in addition to my Optivisor to look very carefully at the pads after soldering and cleaning flux.

    I also use flux despite TCS saying not to use any flux. I clean after finishing with rubbing alcohol.

    I'm using the LL 1998 GP-20 mechanism / chassis as the basis for a number of ARR GP7u chop nose units so will be adding an Atlas GP7 sill, long hood, 2x height winterization hatch, new cab from another GP, 2 additional exhaust stacks, Atlas Alco trucks (really they used these), and chopped nose. Will see if I have the patience for drop grab iron installation. The GP7L may be the same mechanical approach or use the very similar Atlas GP7 mechanism/chassis but I'm focusing on the GP7u to start with. See John's Alaska Railroad website for prototype photos in the Roster Section if you are interested. I think the soon-to-arrive Atlas GP20 Gold would be a easier way to get DCC sound in the GP7 chop nose and may try that in the future. Maybe a Zimo 500 but that is in the distance.


Share This Page