Skip to main content

I have an MTH GP-38-2 (20-2491-1) That has rapidly flickering lights on the numberboards. They still switch on and off fine with the remote. The flickering occurs at startup, stationary or moving. There is no parts diagram available on the website, but most others look to be 6v bulbs. I get 12 to 16v on the contact pad in the rear. I think I have the right ones. I don't have another PS2 diesel to compare voltages. Both front and rear do the same thing. I remember seeing a diagram on the forum showing wire colors and plugs for PS2 engines, but I can't find it with a search. The electrical schematic would be beyond my scope. Thanks for any ideas.

          John

Last edited by John H
Original Post

Replies sorted oldest to newest

To be certain of where the issue is, the board or the engine, you would need to test the board set on a MTH test fixture.  If the boards are fine then the issue is in the wiring of the engine. 

Check the wires that are connected via yellow wire nut.  I have seen this one other time and after removing the wire nut, re stripping the wire and then re connecting them together the issue went away.

This particular repair , I tested the boards and the flickering did not happen when the boards were out of the engine.  So I was certain the problem was somewhere with the wires in the piece.  So check each connection to ensure they are tight and making good contact.  I am a MTH Tech so I have the diagnostic fixture.  Email me If I can help further, but you will find there is a wealth of knowledge here on the forum.

You shouldn't be getting 12-16V.  The bulbs are 6V and are powered from the 8 pin connector.  I would make sure it is seated fully, and all the individual pins are down. Check the wire for nicks and routing through the engine.

 

I would try a factory reset.  I would also make sure the springs on the connector are clean and stretched to make good contact with the pad on the chassis.

 

If all that is still fine, but bulbs still flicker you can try changing bulbs, but it may be an electronic issues on the PS-2 5V board.   G

Last edited by GGG
Originally Posted by John H:

I was reading voltage across the pads. From pad to frame I get 4 to six on one, but it's negative. The front bulbs don't go through the pads, so I'm guessing it is a deeper problem.

          John

This is a big no no.  MTH Doesn't use AC ground as a reference like Lionel TMCC does.  You can blow the board it you do it on the input side.  One pad is Positive High DC Voltage, the other is returned to a DC ground via the FETs (transistor).

 

Check the 8 pin and the wiring.  Do the factory reset.  If still a problem than it probably is a board problem.   G

Is the flickering new behavior or perhaps you just picked up this engine used?

 

Does the flickering behavior change if you vary track voltage - I realize you're running command mode but if you happen to be able to vary the voltage between, say, 12-18 or so...

 

It can be tricky measuring the voltage at the 2-pin bulb connectors.  Even though the 6V bulbs are driven with a 6V DC signal, the signal is pulsed and a DC meter will typically read something less than 6V (say 3 or 4V).  Also, the measurement is only valid when loaded by a bulb due to the way the circuit operates; that is, if the circuit does not see a load, it is not "obliged" to put 6V across the 2 lamp wires.  For example, you might read 16V between the two wires going to the lamps. Seems like black-magic but that's the way it is. 

 

IF, however, your front and rear lamps are on the same circuit then you can measure the voltage on the rear pads (with the rear lamps disconnected) since the front lamps are connected.  The front lamps would provide a valid load to the circuit and hence the circuit would attempt to put 6V (albeit pulsed) across the 2 wires.

Normally, the number boards are separate on a diesel.  Front are paired on MARS 1 and Rear on MARS 2 as an example.  Or Ditch 1 or 2 if the model doesn't have Ditch lights.

 

Yes you could measure at the pin but again it is easy to short with a probe at that point.  Plus the Pulsed DC won't tell you much.

 

Lets step back, is this flashing like ditch light oscillation?  If you go to menu, control. ditch lights, you have options to turn on and off, auto, flashing.  Do these buttons effect the number boards?

 

Maybe someone used a different sound file and the output is effected.  This is only if the number boards are flashing.  If it is just a slight flickering, something else is the issue.  This is a 3V engine, so this behavior is abnormal.  G

I know the model doesn't have Ditch lights, but the ditch light outputs can be used for NB.  The Soundfile controls how the light outputs work.

 

If someone load a different sound file, with ditchlight features the numberboards would flash if horn blown, or could be set to flash all the time via the menu.

 

No more then 3 bulbs on a circuit.  So you might have issues trying to add 4 NB lights to a interior light that already has 2 bulbs.   Better to go with a 6VDC Regulator and make them fixed on when power applied to track.  Can you post a small video of the flashing/flicker?  G

Last edited by GGG
Originally Posted by Dale H:

How about putting a small NP capacitor across the light output for the board?

It is pulsed DC so he could use a regular polarized electrolytic.  Without seeing the nature of the flicker it's hard to determine if a practical-sized capacitor could store enough energy to ride out and smooth the pulsating light.  But assuming it were possible to "spread the wealth" using a capacitor, I'd think the effective voltage the lamps would see would be less than 6V and possibly be too dim.

 

In any event I would not go the capacitor route because of the history of the problem.  It was working and now it went south for some unknown reason. The flickering is just a symptom and the circuit might completely give up tomorrow.

Last edited by stan2004
Originally Posted by GGG:
Stan, You took it out of context.  There not driven from the same contact pad.  If configured as Ditch light the 2 circuits Ditch 1 and 2 work together.  Single soft key turns both on and off.  Actually same when configured as NB.  It is two separate wire and pin outs though.

 

Can you post a diagram of the buffer circuit?  G

G, understood on the 2 separate wire and pin outs. But we know that the two ditch circuits CAN operate independently (such as for alternating ditch lights).  That means there must be two separate circuits (FETs) on the board itself controlled by 2 separate signals from the processor chip.  I just find it curious that both circuits "failed" in the same way.  I was actually thinking that somehow the "firebox flicker" effect got activated but of course that's for a steamer and the NB's were working just fine.  I suppose it wouldn't hurt to reload the soundset.

 

Let's see where John wants to take this.  In his own words an electrical schematic would be beyond his scope so constructing a circuit might be off the table.  Another alternative might be to retrofit LEDs into the NB fixtures; I don't know how tight the quarters are for a GP38 NB fixture but that would electrically require just a few resistors since the added load of, say, 4 white LEDs would be less than 1 incandescent bulb; hence, it should not require a buffer to tap off the Interior circuit.

I like the led idea. I have dabbled with them with good success thanks to Dale H and Gunrunnjohn's excellent help. I will try to post a video over the weekend. Single digit weather hasn't made dairy farming much fun this week. I'm blown away by all the support from everyone and really appreciate it. The front lights are a little tight on a bracket close to the motor, but there is room to work in the rear. There is also an unused two pin connector near the front lights. If this is for the nonexistent beacon could I try hooking it to the front lights? Or does the beacon connect to the unused set of contact pads? Thanks again for all the interest.

          John

Unfortunately I don't have a GP38 with number boards to know how much space you have but the "problem" with some white LEDs is their light is focused too narrowly...rather than the omni-directional incandescent bulbs.  So maybe someone can assist in this regard.  Hmm, perhaps if MTH made a PS3 diesel (with LEDs) that has number boards using the same shell tooling that you could get those LEDs and just add a resistor or two to make them compatible with the PS2 incandescent circuits.  That would be your lucky day...

 

The beacon output on MTH diesels blinks once per second. I don't know if the pads you're referring to are beacon pads but a blinking NB is probably no better than a flickering one!

Originally Posted by stan2004:
Originally Posted by GGG:
Stan, You took it out of context.  There not driven from the same contact pad.  If configured as Ditch light the 2 circuits Ditch 1 and 2 work together.  Single soft key turns both on and off.  Actually same when configured as NB.  It is two separate wire and pin outs though.

 

Can you post a diagram of the buffer circuit?  G

G, understood on the 2 separate wire and pin outs. But we know that the two ditch circuits CAN operate independently (such as for alternating ditch lights).  That means there must be two separate circuits (FETs) on the board itself controlled by 2 separate signals from the processor chip.  I just find it curious that both circuits "failed" in the same way.  I was actually thinking that somehow the "firebox flicker" effect got activated but of course that's for a steamer and the NB's were working just fine.  I suppose it wouldn't hurt to reload the soundset.

 

Let's see where John wants to take this.  In his own words an electrical schematic would be beyond his scope so constructing a circuit might be off the table.  Another alternative might be to retrofit LEDs into the NB fixtures; I don't know how tight the quarters are for a GP38 NB fixture but that would electrically require just a few resistors since the added load of, say, 4 white LEDs would be less than 1 incandescent bulb; hence, it should not require a buffer to tap off the Interior circuit.

Stan,  Not completely accurate.  There are 2 pins and 2 FETs, but the processor fires them together on PS-2.  No independent control.  PS-3 provides the additional control.  So if he had a processor issue, or a buffer issue, the pair can flicker together.  Hence, not 2 independent failures.  G

Last edited by GGG
Originally Posted by GGG:

Stan,  Not completely accurate.  There are 2 pins and 2 FETs, but the processor fires them together on PS-2.  No independent control.  PS-3 provides the additional control.  So if he had a processor issue, or a buffer issue, the pair can flicker together.  Hence, not 2 independent failures.  G

So for diesels with alternating ditch lights (i.e., fired separately), where are the left and right ditch lights hooked up to on PS2?

Originally Posted by gunrunnerjohn:

They're on two separate pins on the 8 pin connector, pins 3 & 5.

 

That's what I thought.  So I think they are 2 independent circuits since the processor can make them alternately flash.  But to the matter at hand, if the NB lights are hooked to these 2 pins, then I find it curious that these 2 independent circuits went south in the same way...going from steady-on light to a flickering-light.  Still would like to see a video of it as I wonder if it will look like the firebox flicker which would be a wild turn of events (diesel engine magically transforms into a steamer!).

Originally Posted by gunrunnerjohn:

I wasn't aware that any of the outputs were not individually controlled.  It would be easy to verify, connect a board to the test-set and load the sound file in question.  You can see outputs for every light pin, and see what they're doing in response to operator inputs.

 

Ok, Find a way to turn off Ditch 1 and not ditch 2.  Do the same with forward markers and rear markers.  As I stated they are 2 separate pins and FET, but I guess we are arguing about the definition of independent.  G

 

Well, it's hard to say for certain but that looks suspiciously like Firebox flicker on a steam engine!  Here's a video I just took of the firebox flicker bulb on a steam engine - I wish I had a NB panel to project it through but I don't.  The bulb attaches to the same pin (a Ditch light output) as your Number Board light.  If it walks like a duck, talks like a duck, etc.

 

Since you've already tried Factory Reset, I'm thinking your next move ought to be reloading the soundset.

 

 

Attachments

Videos (1)
ogr flicker on ditch
Originally Posted by John H:

Having only been in the DCS business for a short time, I haven't gotten into  the DCS Loader program yet. But I did manage to get a video. I will try the soundset download, but it may take me awhile.I am in a rural area so most of my help has to come from Barry's book and you guys.

          John

Ok, you definitely have a flicker, but it is more than just NB, so are your interior lights.

All your lights are run off a Positive voltage coming from a common purple wire.  There may be a wire nut were they are tied into.  Since most lights are flickering I would investigate the PV purple wire.

 

Check the 12 pin connector is fully seated.  You may want push the purple wire connector down with a small tipped screw driver.

 

Does this do it if smoke is off?  The heater element could contribute since it runs off the purple also.  I would also check for any pinched or crushed sections in the purple wire.  May be 2 coming out of the connector, one heading forward and one to the rear.

 

It also could be the light buffer (chip) that controls all the light FETs. 

 

First think to check though is an intermittent with the purple wire.  G

Last edited by GGG
Originally Posted by GGG:
Originally Posted by gunrunnerjohn:

I wasn't aware that any of the outputs were not individually controlled.  It would be easy to verify, connect a board to the test-set and load the sound file in question.  You can see outputs for every light pin, and see what they're doing in response to operator inputs.

 

Ok, Find a way to turn off Ditch 1 and not ditch 2.  Do the same with forward markers and rear markers.  As I stated they are 2 separate pins and FET, but I guess we are arguing about the definition of independent.  G

 

Well, ditch1 and ditch2 are really one function that requires two outputs, so I think that tracks.  However, I agree that the markers could and maybe should be independent, never thought about those.

Originally Posted by stan2004:

Well, it's hard to say for certain but that looks suspiciously like Firebox flicker on a steam engine!  Here's a video I just took of the firebox flicker bulb on a steam engine - I wish I had a NB panel to project it through but I don't.  The bulb attaches to the same pin (a Ditch light output) as your Number Board light.  If it walks like a duck, talks like a duck, etc.

 

Since you've already tried Factory Reset, I'm thinking your next move ought to be reloading the soundset.

 

 

Stan what engine is that?  The MTH engines have a firebox glow (steady)  Many times tied in with interior light (cab).  G or 1 gauge have an ashpan flicker.

I did not think the MTH O had flicker.   G

Originally Posted by gunrunnerjohn:
Originally Posted by GGG:
Originally Posted by gunrunnerjohn:

I wasn't aware that any of the outputs were not individually controlled.  It would be easy to verify, connect a board to the test-set and load the sound file in question.  You can see outputs for every light pin, and see what they're doing in response to operator inputs.

 

Ok, Find a way to turn off Ditch 1 and not ditch 2.  Do the same with forward markers and rear markers.  As I stated they are 2 separate pins and FET, but I guess we are arguing about the definition of independent.  G

 

Well, ditch1 and ditch2 are really one function that requires two outputs, so I think that tracks.  However, I agree that the markers could and maybe should be independent, never thought about those.

John, When I first got started I thought you could have oscillating lights front and rear, but you can't  Both Ditch 1 and 2 respond to a whistle and flash.  The PS-3 has separate control.  Either way there are common points that could cause both to react similarly.  the processor, or the buffer.  Since he actually has more then just number boards flickering, I would start with the PV Purple input.   G

Originally Posted by GGG:

 Stan what engine is that?  The MTH engines have a firebox glow (steady)  Many times tied in with interior light (cab).  G or 1 gauge have an ashpan flicker.

I did not think the MTH O had flicker.   G

Yup, One Gauge Hudson...PS2 board with bulb coming off the "ditch" pin on the 8-pin.  I've always called it firebox flicker but if the correct name is Ashpan flicker then I stand corrected.

Originally Posted by GGG:

Since he actually has more then just number boards flickering, I would start with the PV Purple input.   G

Let's see what John says.  I saw the cabin flicker but I assumed it was the NB lights leaking back into the cab.  Initially he said he could turn off the NB lights using the remote and presumably he would have commented if the remaining lights flickered.  Of course if turning on the NB lights is what causes all other lights to start flickering in tandem, then I'll have to press ctrl-alt-del on my thinking...

Last edited by stan2004
Originally Posted by gunrunnerjohn:

The only way I get firebox flicker on my MTH locomotives is to add it.

 

Since the "Ashpan flicker" comes off the ditch lines I assume you'd need a loco with the PS2 board in the engine since we have that minor nuisance of the tether.  Now there was some guy on OGR who was talking about building some kind of mux device to send more lighting info from the tender to the engine and that might do it.  Oh wait...that was YOU!

John, I know you've tried this, but for good measure try the Feature Reset, and then the Factory Reset again.  I think the engine has lost its mind and has activated the Ashpan flicker steam engine function.  The same pin on the PS2 board which drives the Number Board light on your diesel becomes the Ashpan flicker light on a steamer.

How is that possible without the steam sound file?  Certainly the processor can be acting up, so the feature and factory reset should be tried.

 

When you trace the wires from the number boards do they go to a single contact pad on the engine frame? Or do they go to a pad in the rear and a pad in the front?

 

You still want to check the purple wire connector and pin, then also make sure the springs are stretch on the connector so they engage the contact pad firmly.  If that doesn't resolve it, I would reload the sound file.  If that doesn't fix it, you should have a tech test the board out of the engine.  That halves the troubleshooting to either a board hardware or software issue, or an engine wiring/bulb issue.  G

Post
×
×
×
×
Link copied to your clipboard.
×
×