Decoder not responding

[RW&C]

I pulled out my Model Power Pacific this evening to run it on a test track. It has an NCE four-function decoder in it, which I've never had any trouble with.

I tried it first with DC because I didn't want to drag out the head unit (Prodigy Express). My power source was a half-dead 9v battery, which I'd been using to test a DC loco with no problems. The Pacific didn't budge. So I dragged out the DCC.

Edit: I did a little more poking at it, and got it to run. But only one way. If it faces left, it'll run in either direction, the light works, it's all good. Facing right, nothing. I tried switching the track wires, just switches the directions.

If I was confused before, I'm really scratching my head now. DCC is... phased AC, right? So why does it matter when I switch the wires? And why won't this thing work the other way?

When I set it up the other way, it won't even respond to DC. It's dead to the world facing right. Facing left, it works as well as it's ever done.

Any ideas? I'm incredibly confused here.

Edit: Now the headlight's on with it facing the other way. It's still not talking to the controller, but the light's on anyway.

 

[fast_ian]

...Edit: I did a little more poking at it, and got it to run. But only one way. If it faces left, it'll run in either direction, the light works, it's all good. Facing right, nothing. I tried switching the track wires, just switches the directions.

Firstly, the easy one; Yes, the DCC signal is (high frequency) AC, but the decoder "knows" which rail is which, so switching 'em will indeed reverse NDOT (normal direction of travel).

What changed when you "did some poking"?......

Do you have test leads? - I think I'd eliminate the track by connecting the DCC feed directly to the loco pickups - Probably won't make any difference but at least that's something eliminated.

Beyond that, I too am scratching my head! - Doesn't make any sense..... Particularly now the light is on in the opposite direction.

I also don't think you want to be "mixing" DC and DCC testing - "We" know (suspect?) many decoders don't like that very much and *maybe* that's messing with the decoder setup - Just a WAG btw.

I assume you've done a decoder reset? [On my Digitrax stuff set CV08=8] I think you also said it responded as expected to programming commands [Little "jerks" is response to commands] - On the program track or the main?

Very, very odd

Cheers,
Ian

 

[CSX Robert]

Firstly, the easy one; Yes, the DCC signal is (high frequency) AC, but the decoder "knows" which rail is which, so switching 'em will indeed reverse NDOT (normal direction of travel).
...

No, reversing the rails will not reverse the normal direction of travel.

...
Edit: I did a little more poking at it, and got it to run. But only one way. If it faces left, it'll run in either direction, the light works, it's all good. Facing right, nothing. I tried switching the track wires, just switches the directions.

If I was confused before, I'm really scratching my head now. DCC is... phased AC, right? So why does it matter when I switch the wires? And why won't this thing work the other way?
...

This sounds like one half of your comand station's booster circuit could be dead. If the circuit has failed in such a way that one rail output is at a constant potential and the other rail output is still switching, depending on how the circuit is designed you could get a pulsed DC output instead of the normal square wave AC output. With the pulsed DC signal, there would still be enough power to run the loco, although at a reduced speed. Most decoders only look at one rail to decode the DCC signal. If the engine is placed on the track so that the decoder is checking the rail that happens to be pulsing, then it can read the DCC signal and operate. Turn the engine around(or swap the wires) so that the decoder is now checking the rail that is not pulsing and it will no longer see the DCC signal.

When you were testing with a battery, you did not do any testing with the DCC command station still attached to the layout, did you? If so, that could have damaged the command station.

 

[RW&C]

Thanks to both of you for your help.

Ian- I checked for a good electrical connection. One of the tender trucks wasn't making a good connection, so I fixed that, no improvement. I haven't read all the paperwork yet but I haven't found a reset code for NCE.

CSX- I think you may have nailed it. I didn't have the base unit connected when I did the DC test, but I always noticed it ran a little warm. For this reason, I attached a small fan to the top. I took the base apart just now, and found that one of the four large MOSFETs on the board is discolored on the metal bit. More strikingly, the plastic above and below (even viewed from the other side!) is heat-damaged. The kicker? The heat damage is right below the fan. So either it happened before I hooked up the fan (and I got lucky since then plugging it in right) or it overheated while the fan was running.

Time to contact MRC. Maybe there's a warranty or something...

 

[fast_ian]

No, reversing the rails will not reverse the normal direction of travel.

My bad! I stand corrected.

I can't try it right now, but you're saying if I'm running forward, heading "East" and swap the rail connections when I re-apply power I'll continue Fwd & East?

Sorry for the bad-info, and even more sorry to hear of the burnt up Mosfet [Did you ever smell it? Or see the magic smoke?]

Cheers,
Ian

 

[RW&C]

Yeah, the loco hears "forward" and applies current to the motor in the orientation it's programmed to think is forward, regardless of how the rails are wired.

No problem. And yeah, I wasn't that thrilled to find it. No smell, no smoke. (I'm starting to think that it fried the first time I used it, which was a few years ago. I didn't know the smell of frying electronics from the dog then.) I emailed MRC, and they said that they'd fix it for $40, but to test it with another loco first. Looks like I'll be DCCing something else to confirm that it is in fact the base that's shot.