CSX Robert
Rich
Alton Junction
A faulty decoder has definitely been ruled out as the problem -- so the only things left are the circuit board and motor.
My bet would be on the circuit board. Rip it out and hard wire in a JST harness so decoder swaps are easier if such should ever be required again.
When given a choice, I always opt for removing a factory light board and hard wiring a JST harness.
CSX Robertbut one thing I would like to point out is that the value in CV 19 certainly can be greater than 127. When a loco is placed in to a consist in reverse, 128 is added to the consist address and that value is programmed into CV 19, so if the loco was placed in the consist in reverse and the consist address was 120 the value stored in CV 19 would be 248.
This thought had occured to me, and I was going to mention it as a possibility. However, while it makes perfect sense to me, the high number CV 19 values that were reported (above 127) did not seem to correlate with the CV 19 values he reported from the problem unit.
maxman CSX Robert but one thing I would like to point out is that the value in CV 19 certainly can be greater than 127. When a loco is placed in to a consist in reverse, 128 is added to the consist address and that value is programmed into CV 19, so if the loco was placed in the consist in reverse and the consist address was 120 the value stored in CV 19 would be 248. This thought had occured to me, and I was going to mention it as a possibility. However, while it makes perfect sense to me, the high number CV 19 values that were reported (above 127) did not seem to correlate with the CV 19 values he reported from the problem unit.
CSX Robert but one thing I would like to point out is that the value in CV 19 certainly can be greater than 127. When a loco is placed in to a consist in reverse, 128 is added to the consist address and that value is programmed into CV 19, so if the loco was placed in the consist in reverse and the consist address was 120 the value stored in CV 19 would be 248.
Something erratic is causing a whole bunch of erroneous values to be entered into various CVs including CV 19.
richhotrainIn this instance, it really doesn't matter what value is in CV19. Something erratic is causing a whole bunch of erroneous values to be entered into various CVs including CV 19.
"you got that right, driver"
maxman richhotrain In this instance, it really doesn't matter what value is in CV19. Something erratic is causing a whole bunch of erroneous values to be entered into various CVs including CV 19. "you got that right, driver"
richhotrain In this instance, it really doesn't matter what value is in CV19. Something erratic is causing a whole bunch of erroneous values to be entered into various CVs including CV 19.
Speaking of driver, where is that night owl, mlehman?
mlehmanI don't see how the motor could be the problem, but enlighten me if you think so.
I mentioned above that a friend had problems getting one of his units to operate properly when I was trying to speed match it, and that this was traced to a motor problem. The "problem" was that the brushes had worn down to the point where the brush springs were contacting something they shouldn't, so he replaced the motor. He's had a couple locos like this, all with Kato motors.
But then again his were high mileage units.
richhotrainSpeaking of driver, where is that night owl, mlehman?
Not sure. But I think I heard chanting coming from the computer speakers so maybe there's a decoder demon excorcism ceremony going on in his basement.
maxman richhotrain Speaking of driver, where is that night owl, mlehman? Not sure. But I think I heard chanting coming from the computer speakers so maybe there's a decoder demon excorcism ceremony going on in his basement.
richhotrain Speaking of driver, where is that night owl, mlehman?
Sorry about the late start
I was gathering some eye of newt, bewitching powder, and owl's bane...
Rich,
Thanks for the reminder on that earlier thread. I remembered the baling wire pics, but had forgotten the context they appreared in - and about the things we thought might be motor related.
Funny what my extended experiment in sleep deprivation will do to the memory...
I'm starting to think again about the motor, although not ruling out the lightboard, given the post Rich remembered for me.
The recommentations to bypass the lightboard seem to be the only way to resolve whether it or the motor is at fault here. I wanted to avoid that because of the marker and numberboard lighting in the Stewart FT. Although not really prototypical, I'm a sucker for lighting effects, even if the should only rate a "nice try." Guess I'll figure something out. It'll be easy enough to swap ends on this consist by reprogramming, since being an early FT all visible markings on both FT A-B pairs are of it as 540. That way the 541 chassis can lead with it's working lights if I flub it up.
The thing I am sure of is the decoder isn't the issue. In 541's chassis, it's happy as a clam.
Thanks for all the help, guys, but let me know if you think of anything else as may not get this done until later today. Will report back the results of this mad science experiment.
Mike Lehman
Urbana, IL
I will just reiterate one thing. Before you rip out the lighting board and hard wire the 540, I would strongly suggest that you reprogram the two decoders and clear all consisting. While you are at it, why not change the loco long addresses to some other long addresses, just temporarily. I would really like to confirm that consisting is not the problem, and that loco address 540 is not somehow in conflict.
I suppose the lighting board could be the problem, but my attention is now on the motor, not the lighting board.
I may fiddle with consisting later. Now suspect it's not the issue, as the 541 chassis acts just like it should at all times. If it was consisting, that would seem to indicate both decoders would be similarly afflicted?
I had a little time while cooking breakfast and thought I didn't need to puzzle out converting the lighting, just needed motor control. I hacked the decoder into the chassis after removing the lightboard. All the same odd behavior is there, including the indecision about which direction that you can play with by turning the throttle up and down.
It will take off and run, then alternatively start sputtering and stalling. Gotta be the motor, I'd think.
The one thing I didn't get it to do was change CVs. Possibly the motor issue could be aggravated by interaction with the light board, even though it doesn't seem to be the culprit here? Later I will check consisting to see if there is any difference.
mlehman All the same odd behavior is there, including the indecision about which direction that you can play with by turning the throttle up and down.
You do have analog (DC) operation ability turned off, correct?
mlehman Rich, I may fiddle with consisting later. Now suspect it's not the issue, as the 541 chassis acts just like it should at all times. If it was consisting, that would seem to indicate both decoders would be similarly afflicted?
But, while I also don't think consisting is the problem, it could be. By reprogramming the two decoders on the programming track, zeroing out CV19, and changing the long adresses to different numbers, you could eliminate consisting issues and conflicts with long address numbers. I would still do it.
But, in the end, I have a feeling that it is the motor in 540 causing the problem.
Robert,
Yeah, I appreciate the confirmation CV19 can be greater than 127.
Rich, et al,
I reset and consisted the locos under different #s (542/543). The 540 chassis (542) behavior was pretty much the same. The 540 chassis slows. It also hangs up going from forward to reverse and vice versa, which a nudge to the flywheel overcomes. And the slowing issue is worse in the forward direction, something I was just starting to realize previously and may not have mentioned. This seems to fit with 540's long stint as lead unit.
The only difference was that the CVs don't change -- at least after test running to this point. I suspect that's because this happens by some interaction with the factory lightboard, which is now out of the chain between the decoder and motor.
Mike, I was hoping that you would run the two locos as separate units, not in a consist. But, that said, it is interesting that the CVs remain unchanged which, to me, would be consistent with the change in the 540's long address. I continue to suspect the number 540 as a possible problem within the stack, although I concede that pulling the light board could also be the reason that the CVs remain unchanged.
With the light board gone and the CVs remaining unchanged, you are still experiencing motor problems. So, you may have to rebuild or scrap the motor if that is, indeed, the source of the problem. With the motor still acting up, and the light board gone, a balky motor could have influenced the light board circuitry to send false signals to the decoder, but that is just a SWAG on my part.
Do all of the CV values remain accurate, not only CV 19 and CV29 but also CV2, CV5 and CV6?
maxman mlehman All the same odd behavior is there, including the indecision about which direction that you can play with by turning the throttle up and down. You do have analog (DC) operation ability turned off, correct?
Just catching up to cover all the bases...
Yes, analog ops always disabled as a matter of habit.
OK, misunderstood about doing them separately and not consisted. Got the whole mess boxed up with 540, awaiting potential parts from ebay, I suppose. I'm fairly certain it's not a command station issue, though. The fact that 541 was never affected and that the new 542/543 pair I made with both locos also points only toward the 540 chassis, I am certain it's a motor issue. Otherwise, the command station would need to know which of the "bad CVs" to feed to make sure only the 540 chassis was affected, no matter what it's long address was.
AFAIK, no other CVs changed here at the end. I suspect interaction between the shorts in the motor and the bridge rectifier on the light board was the likely cause of most of the CVs changing. Weirdly enough, as drastically strange as CVs with a mind of their own are, I suspect that's merely a sideshow to the real problem, which is the motor.
Mike, I pretty much agree with you. Once you get that motor repaired or replaced, hopefully you can add back that light board.
Speaking of repair, is that really possible with the Buhler? Or most other small motors these days? Maybe there's a service that does them?
Other than that, trolling on ebay for a good deal looks like it will eventually yield results for ~$50, although buy it now is pushing $100 -- and I can afford to be patient.
BTW, should I also consider Kato-motored examples? They seem fairly common. I've had good luck with my only Kato locos, my modified NW2M narrowgauge roadswitchers. Main concern is being different enough to they won't run well together.
Mike, do you have the specs on the Buhler (Buehler) motor?
I see them on eBay for a lot less than $50.
Yeah, the one that fits the FT???
I could probably figure it out, but maybe someone here knows? Having fkywheels assembled to it would be a plus.
The $50 solution I referred to was lifting the shell up and driving a new chassis underneath. I figured the motor alone would eat up most of $50 and I'd have another B dummy and spare parts to boot by buying another FT A-B set.
Can you get to the motor brushes to see if they're the problem?
Edit: And Bowser has the Buehler motors listed for $37.03. See item 29 here: http://www.bowser-trains.com/Prices/Bowser%20Stewart%20Prices.htm
Edit 2: Plus $9.95 shipping
I too had a problem witha NCE decoder, i called NCE and they had me reset the max speed. but they said that they had ran into this before, and if the problem returned the cause was most likely a motor with a lot of electrical noise. they said that some manufacturers have bad motors from time to time.
John Sawaska
maxmanCan you get to the motor brushes to see if they're the problem?
Hmmm, I dunno...but I'm not the sort to let taking something apart stand in his way of further investigation. I'll see if I can work up the appetite after being shut-out by the little devil in our last bout.
maxmanEdit: And Bowser has the Buehler motors listed for $37.03. See item 29 here: http://www.bowser-trains.com/Prices/Bowser%20Stewart%20Prices.htm Edit 2: Plus $9.95 shippingSad
That's both good to know and about what I expected. I'm liking my idea of patiently finding a bargain and buying a motor with free extra parts better now. Thanks for the data point.
John,
Yeah, and thanks for that data point. It will save me some time on hold, because I was considering seeing what NCE says. Now that I know, no need to bother. I take their word on it, just wondered how much they ran into it. Enough, I guess.
Anyway, thanks to everyone who's participated so far. It ain't over yet, but no real deadlines and minimal budget. Some weird things can happen with DCC. Don't panic or jump on the first solution, although it may even prove to be the best in some way. I may not make it past brush-check. But I'm sure I can drive a new chassis, motor included, under the present shell.
I saw several Kato and Athearn replacement motors offered on e-Bay -- both with and without flywheels, if any of these would fit.
cacole,
Thanks for the tip. I didn't see any FT-specific ones, but will keep looking. For the money and the small difference between buying a new FT A-B set at a reasonable price and the motor alone, I'm guessing I'll probably go with the new chassis because of the extra parts value, which always makde a good addition to the parts/scrap box. But you never know, might find a good deal on a motor.