As for it just taking off in reverse, is the analog mode still engaged? If so the decoder needs to be reprogrammed to recognize DCC only.
Dr. Frankendiesel aka Scott Running BearSpace Mouse for president!15 year veteran fire fighterCollector of Apple //e'sRunning Bear EnterprisesHistory Channel Club life member.beatus homo qui invenit sapientiam
Put the loco on the programming track and enter a value of 34 into CV29, and see if that corrects the problem.
Why it will not run when you turn it around doesn't sound like a decoder problem, but something else with your layout wiring.
I have no idea has to why when you turn the loco around you have problems???
I would do a reset of the decoder first. C8 to 8 then remove the power to the loco. This will set the decoder back to factory defaults.
NOW with out the loco on your track, push the PROG/ESC button 4 times= display should read use program track- hit enter and the display will read 1=STD , 2=CV, and 3=REG, NOW PUT YOU LOCOO ON THE TRACK. push 2 for CV,display will read back the mfg ID # ??? hit enter and it will read back the decoder version #?? hit enter and it will read back set up ADDR, hit 1 for yes, read back will ask address enter 1 for yes and you can enter any two digit address the display will then ask if you want to activate this address, hit 1 for yes FOR A 2 DIGIT ADDRESS or just hit enter one more time and it will ask to set up a long address, enter any 4 digit address and enter, the display will then ask if you want to activate this address, enter1 for yes. Now you should have control of your locomotive via what ever number you programmed into the address.
Re-reading your original message, let me see if I'm understanding it:
1. You say the locomotive will run in one direction (forward?) but not in the other direction?
2. When you physically turn the locomotive around it won't run at all?
If that's what you're saying, it sounds like you don't have the Power Cab connected to the track correctly through the small panel that NCE provides. Double-check your connections to the panel from the Power Cab, and from the panel to the track.
Are you using the FLAT interface cable and not the coiled cord? Do you have the Power Cab plugged into the LEFT socket on the interface panel?
Make sure you have the track wires connected correctly on the rear of the panel.
If none of this solves your problem, you may need to email or phone NCE and see what they think may be wrong.
If the OP can run the loco one way I don't see how he has a track power problem.
OP,do you have a second loco to test with?
When you turn the loco around do you still have the power LED showing on? (The red LED on the panel)
Do you have this problem on a single piece of test track ( just a 2 foot section would work)
this will make sure you don't have some problem with your layout.
Since DCC has no means to control direction using track polarity, something is really wrong. There might be a wiring error, or the motor hasn't been disconnected from track power.
Maybe a trace hasn't been cut that should be?
What type of auto reverser do you have for your reverse loop? Joe
It sounds to me like one side of the H-bridge that drives the track output of the Power Cab might not be working.
Sorry. Hit post before I was finished typing.
RunawayIf one side of the h bridge wasn't working wouldn't one of the track rails be dead?..
Not necessarily, it could be stuck at one potential or the other. Normally, on most DCC systems, the output to each rail switches between positive voltage and ground. When one rail is at positive, the other is at ground and vice-versa. If one rail is stuck at one potential, the decoder will still get power because of the other rail switching, but it will only be half of what it should be(it will actually get full power, but only half of the time). Most decoders detect the DCC signal by monitoring only one of the rail inputs to the decoder. If the rail that the decoder is monitoring is stuck at one potential, then the decoder won't see the DCC signal. If this is the problem, if you check the voltage with a voltmeter, it should read about half of what is expected.
I didn't think of this earlier, but try measuring it with the meter set to DC. If one rail is not working, then you won't have AC.
This does make sense. If one side of the output wasn't working, there wouldn't be a proper DCC signal on the track. The decoder might work, but the negative or positive part of the signal is missing, making it impossible to drive the motor in one direction.
With just the positive signal, the decoder would probably power up. It might even recognize the DCC signal and respond accordingly. Some decoders put the raw DCC signal right on the motor, switching the polarity accordingly. (That is why some buzz no matter what you do.) It is possible that the Tsunami uses a positive and negative rail going to the motor controller.