Had a new one happen today I could use some help with. Some specs. I'm working in HO scale, running an NCE Powerhouse Pro 5 amp wireless system. Code 100 Atlas flextrack & Peco switches with feeders or soldered joints everywhere. I've done the quarter test literally everywhere to make sure of sufficient feeders (none spaced greater than 6' apart and then only with soldered rail joiners). Last night when I shut the trains down was having no issues. I had been running a BLI Niagara with a TCS Wowsound decoder the last couple of evenings in reverse while I worked on a car kit. The locomotive wasn't smooth in reverse and I hadn't really done any break-in running in reverse so figured it was the perfect time. When I finished for the night, I ran the locomotive forward to get it to my yard about 12 feet. No issues. There have been zero changes to the layout in months, so it's not like something changed since the last time I ran (again, last night).
Today I turned on the layout and decided I would back the Niagara onto the turntable to send it to the roundhouse. It backed up about 18", paused, the NCE cycled power(like there was a short) and the Niagara (TCS decoder), a Paragon 2 Mikado, Paragon 3 Mikado, and Proto 0-8-0 w/ Tsunami 1 all took off in reverse. These were not the only locomotives sitting on powered tracks. A Bachmann light mountain with a Tsunami 1 did not move, nor did a Bachman GP7 w/ Bachmann non-sound decoder, Proto F3's w/ ESU decoders (loksound & lokpilot), Proto PA w/ unknown decoder brand (potentially ESU), Paragon 2 Pacific, and Paragon 3 Mikado. It was only the four locomotives that reacted and each time it happened, it was all four locomotives that reacted like this. When they would freak out and start running in reverse, the emergency stop button on the throttle wouldn't work to shut it down and I had to hit the power each time. Upon powering back up it would take a variable amount of time of everything just sitting there before the same sequence would take place and the four locomotives would take off. I even tried running each of the "problem" locos immediately after powering up, but nothing prevented the eventual freak out.
I have only a single throttle so they can't be cued up on another throttle, and none of the four locomotives have ever been in a consist, so that can't be the issue either. I also only have the recall stack set to two addresses, so they aren't responding to a throttle setting hidden in the recall stack. Through trial and error, I eventually pulled the Niagara off the tracks and the issue stopped. After fiddling with it (basically poking, proding, and scratching my head), I placed the Niagara on the tracks and sent it running forward before it had a chance to misbehave (I had a hunch that reversing the engine or just letting it sit was somehow contributing to the issue). It seems to have worked. After several laps around the layout, I sent it to the yard, backed it onto the turntable, and put it in the roundhouse. No problems. I would love to make sure to never have this problem again, but I'm not sure where to start.
Has anyone experienced anything like this, and what did you do about it? Any ideas what to look at on that Niagara?
Any ideas would be appreciated. Thanks guys.
Mike
Mike,
I don't have an specific answer about your short issue. However, I do have a question: Do you have dual-mode disabled on all your decoders?
Tom
https://tstage9.wixsite.com/nyc-modeling
Time...It marches on...without ever turning around to see if anyone is even keeping in step.
Not on all of them. I know I had it disabled on the Tsunami 1 in the 4-8-2. I seem to remember doing that on the Paragon 2 locos, but can't confirm that. I don't recall disabling it with the ESU decoders or the TCS. I'm almost positive I did not disable it with the Paragon 3 locos or the Bachmann non-sound loco. Unfortunately I don't have JMRI yet so I can't just look them up. Guess what's on my holiday break to do list now?!
It may be your DCC system's equivalent of 'a lot' of decoders getting current inrush upon startup, and maybe that is making the system or the decoders misbehave.
It would/might be instructive to remove two or all-but-one of the offending locomotives and try again. If it does the same thing, there's something about the decoders because now you do not have 'a lot' of decoders demanding inrush.
This is obviously a SWAG, but maybe Randy can see something in what I propose. Of course, this is AFTER you have ensured all decoders are set to the CV29 value appropriate for your addresses that does not permit dual-mode ops. for larger addresses, extended ones, such as 566, you'd want a value of 34.
selector It may be your DCC system's equivalent of 'a lot' of decoders getting current inrush upon startup, and maybe that is making the system or the decoders misbehave. It would/might be instructive to remove two or all-but-one of the offending locomotives and try again. If it does the same thing, there's something about the decoders because now you do not have 'a lot' of decoders demanding inrush. This is obviously a SWAG, but maybe Randy can see something in what I propose. Of course, this is AFTER you have ensured all decoders are set to the CV29 value appropriate for your addresses that does not permit dual-mode ops. for larger addresses, extended ones, such as 566, you'd want a value of 34.
For what it's worth, this number of locomotives is pretty normal and had been in their places for the last month or so. Sometimes more, sometimes less. In my troubleshooting, I cut power to two staging tracks, cutting off 4 decoders in the process (3 sound, 1 non-sound) and the issue persisted until I pulled that Niagara. Not sure if it's an issue as trains have otherwise been running well, but could dirty rails/wheels cause this? Kind of like a how a high resistance connection in house wiring can cause a fire? I haven't cleaned in a while, but then again, I don't normally have to. Maybe a combination of that and the number of them on powered rails?
If it is the Niagara...maybe it's channeling its Tesla???
That was almost a too far stretch pull there Tom - mainly because thse days, when peopel hear "Tesla" they think cars.
Crandell has it right. The most common reason for locos to just take off is that they don't have DC mode turned off via CV29, and when the system recovers from a short, the DCC signal is unstable long enough for the decoder to think it is running on DC with one side of the signal greater than the other, a DC offset of sorts. Usually, disabling automatic DC conversion by adjusting CV29 prevents this.
It's possible that the temporarily corrupted signal causes runaways through other means as well, but the automatic DC setting helps a lot. Also a good reason to have circuit breakers and dividing the layout up so that only the part with the short cuts out so at least locos all over the layout wouldn't take off.
--Randy
Modeling the Reading Railroad in the 1950's
Visit my web site at www.readingeastpenn.com for construction updates, DCC Info, and more.
Thanks for all your help guys. Any idea what would make an otherwise fine locomotive short, specifically when run in reverse? Could a motor issue cause this?
I hear you about breakers, dividing the layout up, etc. Luckily for me, all of my locomotives are either at the main yard, or in the staging yard. In this case, all the engines that took off were all in the yard or roundhouse so the breadth of the issue was obvious.
Thanks again guys. CV29, here I come.
rrinker That was almost a too far stretch pull there Tom - mainly because thse days, when peopel hear "Tesla" they think cars.
Yea, but given the audience, Randy, I figured the more "mature" members would get the reference - albeit a stretch.
Water Level Route Any idea what would make an otherwise fine locomotive short, specifically when run in reverse? Could a motor issue cause this?
I don't think an engine just running in reverse would by itself ever cause a short. I would go back to where it happened and try to recreate it. It could be when running in reverse something happened - like say the tender wheels going through a turnout caused a short. Lotta wheels under the Niagara tender, could be the wheels made contact with a guardrail or something that wouldn't happen when going forward.
wjstixLotta wheels under the Niagara tender, could be the wheels made contact with a guardrail or something that wouldn't happen when going forward.
Edit: OK, I can reply, I just can't include a quote from a previous message. Oh well, wasn't completely on topic anyway.
Interesting, I can reply to any thread but this one, but it doesn;t show as locked.