I am reasonably certain I am doing something wrong, and lack the experience to self diagnose. Any help/suggestions would be greatly appreciated. I have a very simple Inglenook sidings tabletop layout I am putting together. The track (Kato Unitrack) is laid out. The switches are power routing, but since its a one engine show, I am not seeing that being an issue, either way though, once I get it working, I will be placing feeders on each siding. Currently I have one set of feeders connected to the NCE Power Cab Panel. I am using an Atlas U23B DCC Loco. I believe the factory decoder is a Lenz. When I power the system up, the Power Cab lights up and finds the loco at address 3. The light on the loco does come on. When I apply acceleration, the loco starts to hum, but will not move until it seems like a lot of acceleration has been applied. (at 128 steps, it doesn't budge until 40-50. 28 steps requires a similar threshold). Load doesn't seem to impact this 0 cars, 1 car, 2 cars all get about the same results. It will reverse direction properly 1 or twice, and will run a few feet, after which it will shut down. No lights, no humming, no movement. Unplugging the power to the Cab and rebotting it will usually allow me to repeat the process, but cleary something is wrong. I did run the loco around my LHS before I purchased it, and it ran well there. It definately was not as loud, and had none of the shut off issues. Their test track was also Kato, though they had no switches. There DDC test system was Digitrax. Not having any experience with this, I am assuming the problem lies with my setup. The NCE oOwer Cab is basically straight out of the box, and I am using the bundled power supply. I haven't used it before other than to test a Loco a few monthes ago, and didn't have any problems then, though that was very brief. Thanks
I also have the NCE PowerCab and am also modeling in N-scale. I have not used the PowerCab much yet, but I'll suggest a couple of things. 1. Is there any chance you inadvertently set the momentum to a high number? This will cause the loco to start very slowly, no matter how much throttle is applied. Try setting momentum to zero. 2. As for running a few feet then stopping, have you considered the possibility of dirty track or dirty wheels on the loco? If either are dirty, a DCC loco may stop abruptly as you describe. Sometimes, tapping the layout near the stopped loco or giving it a little push will be enough to restore contact between the loco and the track, and the loco will start moving again. If it does re-start, and the momentum is set to a high number, the re-start will be very slow. 3. As a test, try connecting the PowerCab to a section of track on your workbench and testing the loco there, completely separate from your layout. This will eliminate any possibility of an electrical problem on the layout causing the loco to run poorly. My first DCC loco ran pretty awful until I cleaned the track and cleaned the loco's wheels. - Jeff
Most likely it's the loco, particularly with the Atlas decoder in it, not the PC. Do what Jeff has suggested above first. As for the decoder in your loco, I have a pair of RS11's that had factory decoders (now TCS) in them and it took a lot of programming to get them to run as I wanted. It wasn't that they didn't run well in themselves, they were not set up like the others I had with NCE and TCS decoders. These factory decoders don't support the CV6 (midpoint) setting, but do CV2 (minimum) and CV5 (maximum). I can't recall if the acceleration (CV3) and deceleration (CV4) are user defined. I suggest you have a look at the CV settings by placing the PowerCab in programme mode with the loco on the track by itself. Press; prog/esc > 4 > enter (programme mode) > 2 (CV) then in turn 2, 3, 4, 5 and 6 pressing enter after each CV number then after each value is shown. This doesn't change the existing values. Note the values and report back. The buzz you hear is the PWM (pulsed) output from the decoder to the motor which was around 2000hz (pulses per second) so is in the audible range. Later decoders have the PWM frequency much higher and outside the audible range and generally adjustable.
Update: Thanks for the quick responses. I don't believe its the track. To call this a layout that this point was pretty optimistic. It is basically new assembled unitrack on a piece of plywood so that I could test the operation (particularly the car spacing and magnetic uncouplers) before doing anything to be permament. I wanted to be sure, though, so I tested another loco - a Bachmann DD40AX. This ran very well, no stalls, throttle seemed to work, lights functioned, etc. Based on this, I am assuming the issue is in fact with the loco itself since everything else is functional. After running the Bachmann, I put the Atlas back on the layout (they both use the same defualt address since I haven't changed them, and now the loco is completely unresponsive. I have unplugged the power supply a few times to boot the DCC system with no luck. I did try to view the CVs as Gats suggested, but I don't see entires for any of them. I am guessing the loco and NCE are not communicating at all at this point. Could I have done something by removing the Atlas and running the Bachmann that would cause this? Thanks
I don't know the Atlas U23B loco, but I have several other Atlas DCC fitted locos and one problem point can be with the spring contacts from the motor brush fittings which contact the underside of the decoder board. These can give unreliable contact which may explain your problems. Be careful if you dismantle the frames to check this that either of the spring contacts don't touch the frames when you re-assemble or your chip can self destruct - BTDT A small bit of insulating tape in the inside of the chassis frames will prevent this. Some people recommend replacing these contact strips with wires soldered between the pads on the decoder boards and the motor brushes, but I haven't found that necessary so far. Jim.
I agree. This proves the track and PowerCab are working fine. By entering the programming mode and getting no response back from the decoder could be one of three things; the wheels are making poor contact with the rails, the motor is making poor contact with the decoder (most times the front light will be on), or the decoder is inoperable (fried). I would venture the motor tabs are not making contact with the decoder as Jim has mentioned above, which stops any programming occurring as the decoder needs to see a load (the motor). That would explain not being able to see any information. Did you get a 'CAN NOT READ CV' message at any time? This would caused by any of the above reasons. Not likely.
Also make sure DC is turned off on the decoder? Try to do a decoder reset. Also very important is to make sure the track is clean. Here is an excellent site for NCE http://www.members.optusnet.com.au/nswmn2/DCC.htm
Good News! Looks like it is working now. Not 100% sure what the issue was between the loco and the cab, but between cleaning the wheels, checking all of the electrical connections, and letting it sit for a day or two, they are now communicating. Setting the momentum to 0 has helped significantly. I am not sure how that got changed, but with it at 0, the loco is now responive with no stalls and the noise level has significantly improved. Now I just have to figure out why my uncouplers work when I am pushing the cars with my fingers, but not with the loco. That's a little of topic for a DCC forum, so I'll post it over on the N-Scale. Thanks for the help!
