After the Plunge...

[ljudice]

Well the Digitrax DCS100, PR3, power supplies and cables came today - and after about 20 minutes I had full control of
the system through JMRI - no real issues at all...

I then pulled out my ONE DCC equipped loco and a FVM ES44AC and added a TCS 6 prong decoder to make two.

The first, a Kato SD70MAC had been picked up on Ebay, and I had no idea if the decoder worked or not.  So I turned to the GEVO.

Well,  in DecoderPro,  I was able to see it on the programming track and apparently read and write to it.  So off to the running track.

All I get is the unit going in ONE DIRECTION at very slow speed.  One or two times it seemed as if I could control the lights and also get it
to go forward REALLY slowly - even cranked up to 100%.

I know I am doing something stupid - it looks like the loco is acting as a DC unit and responding to what must be some constant voltage that is out there at all times to carry the signals to the decoders.

Also - what programming needs to be done to the TCS decoder?  I am presuming it knows nothing of what it is connected to.  All I set was it's number....

Thanks for your help,

- Lou

[peteski]

Did you check the thread 14 posts below this one?
https://www.therailwire.net/forum/index.php?topic=31798.0

[ljudice]

No but it sounds similar...

The Kato/Digitrax decoder unit runs perfectly, so it is definitely an issue with the FVM/TCS.  I installed it exactly as shown at the TCS
site, so next will check for shorting somewhere....

To revise the behavior - put on the track it goes backwards and won't stop (slowly).  Crank it up to full power on Forward and it sometimes creeps forward.


thanks,
lou

[C855B]

Was the Kato unit still on the rails when you tried this?

I experienced a strange problem recently where another decoder in the consist "crashed" into some sort of odd state that, while it ran fine, it caused another unit also in the consist to do exactly what you described. I had factory-reset and reprogrammed the "creeping" unit three or four times with no change, but as soon as I removed the other unit, it suddenly behaved itself. I then reset/reprogrammed the one I removed, and everything was fine after that. Both were TCS decoders.

[ljudice]

No, the Kato was off...

I removed the FVM shell again and the behavior changed to just sitting there with full control of FWD and REV lights and no movement.

There is no shorting between loco and the decoder, and it is definitely in the right way...

[C855B]

Have you done a full factory reset yet, to start fresh? I've had some issues recently with a number of different decoders - might be my own Digitrax/JMRI setup, for that matter - and resetting seems to cure most ills. This is especially true if I've poked at a bunch of CVs and found I twisted things up unintentionally.

[ljudice]

Have you done a full factory reset yet, to start fresh? I've had some issues recently with a number of different decoders - might be my own Digitrax/JMRI setup, for that matter - and resetting seems to cure most ills. This is especially true if I've poked at a bunch of CVs and found I twisted things up unintentionally.

Just tried it...  It actually wouldn't at first and was returning the wrong MFG code.  Then it suddenly seemed to reset.  But no change in behavior.
Turn on power - it starts moving and won't stop.  But the lights are under control.  Looks like a bad decoder to me?

BTW - dumb question....  If you set a long/active address (ie. to loco number),  do you also have to pick a unique short address or just leave it at 3? I assume it needs two unique addresses.

[peteski]

The long and short addresses are independent (you select one or the other in CV29).

Try this: unplug the decoder, flip it 180 degrees and plug it back in and see how it behaves. This will not damage the decoder (due to the way the connector is laid out).

[C855B]

[Ha! Pete beat me to it... here's what I had written, anyway:]

No - when you set a long address it also sets a CV that tells it to ignore the short address.

So it's still screwed up after reset and you tried running it as 003 (untouched)? Frankly, for grins I'd flip the decoder in the socket just on the off chance that it's upside down. I recall the instructions on my GP60 not being all that clear on the orientation, and I had to try it both ways on the first install after experiencing some sort of issue. Anyway, there's no harm in it being in wrong either before or after trying this.

[ljudice]

Yup, already tried that!   No go...

It's odd that the behavior is different with the shell off...  (no movement at all, but lights work)

Seems like a short or open somewhere...

[peteski]

Yup, already tried that!   No go...

It's odd that the behavior is different with the shell off...  (no movement at all, but lights work)

Seems like a short or open somewhere...

I would agree.  If you feel adventurous, you could take the chassis apart and see if you can examine the circuit board looking for shorts or opens.  The only FVM loco I have it the 4-4-0, so won't be much help with this one.

You can also unplug the decoder and do some continuity checks. There should be open circuit between pin 1 and 3, 1 and 4, 2 and 3, 2 and 4.  Pin 1 and 2 should have less than 100 ohms continuity between them - that is the motor. Pin 3 should be connected to the right rail pickup, and 4 to the left rail pickup. Pin 5 and 6 should not have any continuity to each other or to any other pins (those are front and rear headlight cathodes).

[C855B]

I'll bow out here and let Pete scratch his head over this one. Besides, my iced tea has worn off and it's time to toddle off to bed. Good luck.

[bdennis]

Try pulling the decoder out and run it in DC mode to make sure that the loco runs correctly.

[peteski]

I know I am doing something stupid - it looks like the loco is acting as a DC unit and responding to what must be some constant voltage that is out there at all times to carry the signals to the decoders.

- Lou

Lou, just to clarify - there is no DC voltage in a DCC signal. The positive and negative pulses average to pretty much 0V DC. The only way a model would see any pseudo-DC voltage would be when the DCC system was configured for running on address 00.  Most DCC systems no longer have that feature available.

[ljudice]

Ok, well this AM I put the decoder in a different FVM GEVO - and it works PERFECTLY....

The actual symptom on the other loco is:

      With DCC Decoder:
             Apply track power - unit moves slowly in reverse
             Set speed to ZERO - keeps going.
             Set to Forward - GOES FORWARD and can stop on command
             Set to Reverse - Goes backard and won't stop
             Lights fully controlled by DCC.

Any thoughts?