BLI locos stalling with sound on

[mmagliaro]

This is a DCC system.  I don't think for an instant, no matter what's going on in that decoder, that we have a situation where the sound increases the load "just enough" to lower the voltage to some critical threshold.  Not when it's 12v or so all the time.

A possible short... now Peteski may be on to something there.

STOP STOP STOP.
Run a pair of jumpers into the engine, clip them to the track, and run it through the turnout.  If the problem goes away,
then lost contact is causing it.  Why?  How?  Why does sound make it worse?  No idea.  But let's be sure of this first.

[victor miranda]

DCC or not.  resistance is real and the effects are the same in both systems.

in fact what you have just said is the reason the decoders do not handle brownouts.
it is NOT 12v at all times.
the track MIGHT have a nominal 12vAC on it.
now does ALL that get to the decoder through the pick-ups?

....

I've done a lot of tinkering trying to find the solution to this problem.

n-scale has a lack of reliable contact at the wheel and rail.

when you get into what most locos have for internal electrical contact,
you end up at any voltage below 12 volts at the decoder.

what voltage is on the rails.... not the problem.
If my locos are an example the ohm rating from one rail to frame is about one ohm.
my volt ohm meter could be misleading me...
that is on a good contact.  there is the other rail to add....
what ever the voltage drop inside the loco might be,
if you double the current draw, you will double the voltage drop...
most locos use about 100 miliamps and the sound decoder I met seems to be about that as well.

sound from a decoder is a problem because the sound stops when the decoder resets.
you will hear it.  you will know.

if the decoder resets with just a motor...
you might see the headlight flicker and most people think that is normal.
a couple of locos may bump and you would not notice.

I can tell you that a couple more ohms is all it takes to stall most locos at slow speeds.
the DCC system is not immune to this problem.

what is going on is a combo. pickup issues are the biggest part.
the decoder lacking any ability to avoid a reset and
the additional load of the sound with the motor making any pickup issue worse.

n-scale still wants for reliable pickups and you are looking at that as an issue here.

[victor miranda]

I don't know Victor, if you read my earlier posts I seem to mention power pickup problems few times too. 

The detailed questions I have been asking were phrased in a way which should let us know whether it was a pickup problem. The last set of (unanswered yet) questions I posed might shed some more light on this problem.

hi peteski,

I am aware you asked.
I am betting you will get an answer along the lines...

all the contacts are clean. the loco runs well everywhere else.

try thinking of the problem this way.
what would you decide is the fault if all the contacts are as solid and good as possible?

the wheels are clean.
the wipers are clean.
the wires to the decoder are screwed to the frame.

you have a 44 tonner with this problem...

victor
 

[nkalanaga]

I would think there's a flaw in the track.  Maybe not much of one, but I've seen almost invisible vertical or horizontal kinks cause problems for one, or a very few, locos, while others work fine.

I ended up debugging my first-run ME turnouts, with the warped frogs, using an ABBA set of Kato first-run F units, which Trainworx had turned the flanges on.  Any three of them, in any combination, ran fine.  ABA, ABB, BBA, even BAB, worked.

Put all four on the track, the second unit would derail every time, almost always the rear truck.  It took about a year of off and on work, but they finally went through everything, and would pull my passenger train, in either direction, with no problems.  And everything else ran smoother as well.

[peteski]

hi peteski,

I am aware you asked.
I am betting you will get an answer along the lines...

all the contacts are clean. the loco runs well everywhere else.

try thinking of the problem this way.
what would you decide is the fault if all the contacts are as solid and good as possible?

the wheels are clean.
the wipers are clean.
the wires to the decoder are screwed to the frame.

you have a 44 tonner with this problem...

victor

I don't have a 44-tonner with problems. Friend had one, but I modified and polished the contact surfaces and the problem is gone.

I'm now going to accuse you of what you always say about me: You didn't read my questions. 

The need to nudge the stalled loco  still seems like a power pickup issue.  Another good thing to observe would be to have the headlight on then check if the headlight stays on when the loco stalls, before it gets a manual nudge.

I'm still curious about how the loco behaves when running at high speed through the trouble spot with the sound turned on full blast.  Will is still stall or will it be able to run through? 

If the answers I will get are are like you think then they wouldn't be reading my questions either.

My experience with sound decoders is that they need a DC/DCC voltage of around 6V to keep the microprocessor alive.  If the track voltage is 12V that means it would have to drop by at least 50% to make the decoder stop working.  I suspect that the decoder (with sound enabled) and the motor are consuming less than 250mA.  So, at that current, to drop the 12V to 6V the resistance between the rails and decoder would have to be more than 24 ohms.

[bobthebear]

Hi Guys.
RESULT
Well...........after 4 hours yesterday we have a running loco.
We first checked the track power: voltage and amps. Then we checked current draw of the loco. All checked out fine.
The problem stems from the pickups. Unlike most n scale diesels today with pinpoint axles onto metal strips with conical indentations to transfer the power, the BLI PAs are different. They have stub axles which are held into a PLASTIC conical indent. Set into the plastic sideframe is a metal strip with HOLES which the axle passes through, and hopefully touches! We ran the loco until it stopped, this with the sound off (CV133), then, if a screwdriver was run along the railhead, as it touched a wheel, the loco set off. Any wheel! This means that all 6 wheels were not transferring current. With an ohm meter and the loco off the track all wheels were showing they were "connected". We bent the strips so that they touched the face of the wheels rather than hopefully making contact with the axles and success. Re-instated the sound and they work.
A rather poor design, and a shame for such a nice loco.

[kewatin]

Greetings &thanks
 bobthebear thanks for pursuing the problem along with all the other brilliant minds in resolving this problem.bending the brass pickup strips solved the problem on mine also.at 75yrsof age and 2 lens implants i had to go a little different route, i slipped an exacto knife blade &gently twisted the brass strips inward.went to BLI &checked with their exploded view but with my old eyes and not wanting to break anything,took the easy way out for me. i just hope it lasts but the loco now runs like it is intended.it seems victor was right that it seems in n scale at least to usually be a pick-up problem.being a new member here on this forum has for sure paid  results and gentlemen ,i can't thank you all enough for your time and effort.
   regards&later  KEWATIN

[rrjim1]

Hi Guys.
RESULT
Well...........after 4 hours yesterday we have a running loco.
We first checked the track power: voltage and amps. Then we checked current draw of the loco. All checked out fine.
The problem stems from the pickups. Unlike most n scale diesels today with pinpoint axles onto metal strips with conical indentations to transfer the power, the BLI PAs are different. They have stub axles which are held into a PLASTIC conical indent. Set into the plastic sideframe is a metal strip with HOLES which the axle passes through, and hopefully touches! We ran the loco until it stopped, this with the sound off (CV133), then, if a screwdriver was run along the railhead, as it touched a wheel, the loco set off. Any wheel! This means that all 6 wheels were not transferring current. With an ohm meter and the loco off the track all wheels were showing they were "connected". We bent the strips so that they touched the face of the wheels rather than hopefully making contact with the axles and success. Re-instated the sound and they work.
A rather poor design, and a shame for such a nice loco.

The latest run of BLI PAs, at least the 2 that I received, have a new truck design with pointed axles and contacts.

[bobthebear]

The latest run of BLI PAs, at least the 2 that I received, have a new truck design with pointed axles and contacts.

Ah, maybe I'm not the only one who has had problems then!

[Santa Fe Guy]

Well done bobthebear. Sometimes we look to find faults in obscure places and it is usually something simple (well sort of). We should remember to do root cause analysis before we head down the wrong path.
Great result.
Rod.

[up1950s]

All six axles on one side failed electric ............... damn . I wonder if the trucks have enough lateral slop , and if that would have helped seal the continuity deal instead of . Max has pointed out years ago that even with the cone design that the axle to frame pb strip must float freely . Did yours float ?

[peteski]

All six axles on one side failed electric ............... damn . I wonder if the trucks have enough lateral slop

I highly suspect not.  So, if there is any twist in the track (along its center-line axis) something like this can happen.  I still don't get it why the sound being on or off made a difference. The only thing I can think of is that the large electrolytic cap installed in sound decoders might have had just enough capacity to let the loco coast through the problem area with the sound off. The extra 80mA or so being while the sound was on might have discharged it quick enough to stall completely.

It is nice to see a solution and at least partial explanation of the problem.

[bobthebear]

. I wonder if the trucks have enough lateral slop , and if that would have helped seal the continuity deal instead of . Max has pointed out years ago that even with the cone design that the axle to frame pb strip must float freely . Did yours float ?

The strips are held firmly in the plastic truck sides, so don't float at all. Another thought Chinapig made yesterday: to remove a small bit of the plastic at the top of the cone, so the axle HAS to sit on the metal strip. If the problem returns, that's what we will do next. But for now, I'm happy.
Thanks for all your feedback Guys!
Cheers, Bob.

[RBrodzinsky]

How did you bend the strips to contact the wheel faces?  I have an E8A which doesn't run well, at all (used to be great). It goes about an inch, shorts / resets, runs a bit, shorts/resets.  I sent back to BLI, twice, with "no issues found", yet even on straight track. Long out of warranty, now. Its B unit runs fine.   After reading this thread, I went and checked, and I see very intermittent connectivity between the wheels and the pickup wires from the trucks. I also found the wire from the front right wheels had a bare spot in the insulation, but that doesn't seem to be related (was not shorting to frame or anything).

[peteski]

How did you bend the strips to contact the wheel faces?  I have an E8A which doesn't run well, at all (used to be great). It goes about an inch, shorts / resets, runs a bit, shorts/resets.  I sent back to BLI, twice, with "no issues found", yet even on straight track. Long out of warranty, now. Its B unit runs fine.   After reading this thread, I went and checked, and I see very intermittent connectivity between the wheels and the pickup wires from the trucks. I also found the wire from the front right wheels had a bare spot in the insulation, but that doesn't seem to be related (was not shorting to frame or anything).

Are you sure that those are shorts?  Maybe it stalls because the circuit opens (not shorts)?  Intermittent connectivity would be more of an open rather than a short problem.