Skip to main content

Well, that's a bummer, but you are made progress narrowing it down.  Since you've examined the smoke units six ways from Sunday, take a closer look at the SFC board.  See if there's ANYTHING that might be shorting to the chassis or other wires.  Obviously, there's a reason for the issue.

I think one of my text tests would be to rig up a wire to simulate a smoke unit with an 8 ohm resistor and a 4.7k resistor to simulate the thermistor.  Use a real smoke motor since they check back-EMF of the motor.

Here's what I'm suggesting.  If this rig still causes the thermal runaway, then it's not the smoke units, it's something else that is being triggered by the load of the smoke unit.  This is a simulated smoke unit, it's how my RCMC test set tests the smoke outputs of the RCMC.

Attachments

Images (1)
  • mceclip0

Do not be surprised if the SCF board is defective. Even new. I say this is I had a problem with a new Mohawk. I went through 3 control boards where there was a smoke unit issue. I sent them back to Lionel to be tested and ask if they would pull a brand new one from parts and test it. Sure enough a new one was bad. It seems like the whole batch of replacement boards were programmed wrong in China. I'm sure Lionel by now has read this thread, ask if they would pull another one from stock and bench test it there.

A question, do all the smoke features work properly?  The SFC3 board controls the whistle steam and the blowdown smoke.  You just need to follow the trail.  You can swap a different Legacy RS-Lite board into the locomotive for test.  It'll sound funny, but that will tell you if it's the one with the problem.

The SFC3 brings the serial data in from the RCMC and forwards it to the sound board

Jeff,

I am guessing that when you say the micro processor on the SFC board, you are referring to the A52A.

Looking at the 4-wire input from the RCMC board, I can not imagine this is the cause.

Did you happen to test the SFC board with either one or both of the aux smoke units plugged in and the connection to the RCMC board not plugged in?  The test would be if the microprocessor on the SFC would get hot.  I realize the sounds, etc would not work, but it might further prove the "new" SFC board is bad.

One other thought I had was to plug the main smoke unit into the SFC board (if the cable are the same and they would reach).  We know the main smoke control units are OK and work with the RCMC board.  In any case, as much testing and work you did on the two aux smoke units, I can't imagine those are the issue. I'm starting to be in the camp that the SFC board is bad, or at least not programed correct, causing it to eventually fail.

Hi Dave,

That was the last test I was going to do...but did not (smoke unit without rcmc).  I also had considered plugging one of the main stack smoke units into the sfc but wasn't sure if they were too different.

Yes the a52a is the chip that gets hot. I can't believe a Fourth board would be bad...unless the 3 we've gotten from Lionel to replace the original have all been misprogrammed.  Plus that still wouldn't explain the original sfc failure...which was that same chip that blew.  Would a misprogram cause that to heat up like that or would something just not work?

      I installed the new rcmc board friday evening and tested again with the 2 aux smoke units disconnected.  As expected all other functions worked correctly.
      I plugged in only the whistle smoke unit to the sfc board and powered up....the chip on the sfc board began heating.  So I removed this smoke unit from the engine again and went over every wire with a fine tooth comb. Could not find anything amiss visually.  Went ahead and ohmed out each pin again to each other and compared the readings to what I got when doing the same on the main smoke units.
      Ah ha....finally found something different.  On the main units between the HTR pin and COM pin I had open line or a high reading of like 2 megaohms which is basically no connection.  However, on BOTH of the aux smoke units I was getting a reading of around 29 Ohms between these 2 pins!  I thought ok...maybe we're finally on to something here since I shouldn't be getting this reading between these 2 pins if I don't on the main ones right?
     I confirmed it multiple times....Definitely 29 Ohms between HTR and COM pins on both aux smoke units,  but not on the 2 main units.  All other pins seemed to give similar readings on all 4 smoke units as noted below:
     Between raw and htr = 8 Ohms
     Between fan and com = 11 Ohms
     Between sens and com = 7.5 k Ohms.             when connected to a board...when.                   disconnected 53 to 56 k Ohms.
     Between sens and htr = open line
     Between raw and any other pin = open.             line.
     So there definitely seemed to be an anomaly here...htr and com giving me 29 Ohms on the 2 aux units but not the stack units. I went ahead and pulled both of those smoke units apart again...but still didn't really find anything to explain this behavior.  I tried moving the thermistor a bit to make sure it wasn't maybe  touching the element when assembled and I cleaned the top of the smoke unit pcb with alcohol thinking maybe there was a film of smoke fluid causing a short somehow.  Reassembled them...and now I no longer read 29 Ohms between HTR and COM on either of them.  All four smoke units now read the same.
     But...I don't really know what I did to change or fix anything since I didn't see the thermistor touching the element unless it was when it was assembled... or the fluid was in fact causing a problem.
     Went ahead and tried them again on the sfc board.  The chip still heats up after about 2 minutes...and neither whistle or blowdown smoke work.  The whistle smoke appears to be heating up now but the fan did not spin when blowing the whistle.  Blowdown appeared to be totally dead still as I didn't feel any warmth nor did the fan activate when activating the feature from the cab2.  Plus we still are getting a hot chip on the sfc board after some time. It's not instant now like before but it's getting hot enough that i feel if I left the power on much longer it might pop.  Note that it doesn't warm at all with the smoke units disconnected no matter how long the engine is powered up.
  So I guess my questions now are this:
     What is the normal reading for the connection between HTR and COM?  I'm assuming open or high megohms...but since it still isn't working I want to confirm normal before we try a 5th sfc. Defintiely not 29 Ohms though correct?
     Why is the chip still heating up on the sfc if the smoke units are ok now?  Damaged board from prior tests?  Mis-programmed?
     Any thoughts on what was causing there to be a connection on those pins initially and what I may have done to correct it? Is it corrected or is it likely to suddenly act up again without warning?  Should we be replacing both of the aux smoke units?
     Do you think we cooked the sfc again from all the prior testing and now we might be good to go with another new sfc?  Or do you think this sfc is programmed wrong and thats why it's still not working?
     This is one strange engine...its not gonna succumb to us without a fight until the very end I guess.
Last edited by PRREnola

Thanks for the update.

For those that need a little more info -

The RAW and HTR are the leads to the the heating element.  They should not have any connection to anything else (these leads are isolated from all other components).  You should get ~8 Ohms across the heating element.

You will see resistance between the motor leads that are FAN+ and COM (~8.3 Ohms).

You will see resistance across the thermo sensor leads that are SEN and COM (~56K ohms).

Here are couple of things to try:

Remove the aux smoke units and test the motor directly with a DC power source.  The smoke motor should spin freely with just a couple of volts.  I will put up to 5 volts on the motor.  If you have a way to measure the DC amp pull, that would be great.

The other thing you could try is to de-solder one of the motor lead and then test the smoke unit in the engine.  If it continues to have a problem, you know it is either the heating element or the sensor.

@DaveGG posted:
The RAW and HTR are the leads to the the heating element.  They should not have any connection to anything else (these leads are isolated from all other components).  You should get ~8 Ohms across the heating element.

You will see resistance between the motor leads that are FAN+ and COM (~8.3 Ohms).

You will see resistance across the thermo sensor leads that are SEN and COM (~56K ohms).

Correction, the common (pin-3) is the positive side of the fan drive, and it also connects to one side of the thermistor.  Here's a schematic of the wiring of a single Legacy smoke unit with a thermistor.  A dual smoke unit simply repeats this same design, the two halves are electrically isolated.

Attachments

Images (1)
  • mceclip0

@DaveG I get 11.6 Ohms across com and fan...and across the motor itself on all units including the 2 main units.

I did briefly put a weak 9v batrery to the motor on the whistle smoke and it spun...but I only brushed the wire against the terminal since I figured 9v might be a bit high for it. Bit it did spin so I think the motor is ok.  I can retry it with 3v or 5v if it still doesn't work with the replacement sfc board.

Last edited by PRREnola

The way I test smoke fan motors is with around 2VDC from a bench supply.  If the motor won't spin on 2VDC, it's bad.  Also, if the motor is drawing 50ma or more with 5VDC, it's also bad.  The RCMC in the VL-BB actually measures the back-EMF from the smoke fan motor and verifies proper operation.

I will also hasten to add, a bad smoke fan motor would NOT have caused this issue.

John is right (he usually is) about the fan motor not causing the issue.

Did both of the aux smoke units initially show resistance reading between the HTR and the COM prior to your disassembly/cleaning?

I still cannot figure out why plugging in the smoke units would cause the SFC board to go to failure (get hot).  Unless it already has failed and plugging the smoke units in, just allows it to heat-up.

@PRREnola posted:

Yes both of them showed about 29 Ohms between htr and com.  I checked it so many times because I couldn't believe it that they'd both be reading the same but they were.  Still don't know what I did to "clear" it either.  That's the worrisome part.

Well, before I put yet another SFC in there, I'd completely disassemble those smoke units and make SURE that there is no way there is a connection.  I'd also inspect the PCB very carefully to make sure no contamination is causing continuity between the pins.  My top suspect would be under the 5-pin connector to the smoke unit, it's where those traces come very close together.  You can gently pry up the connector plastic body and clean up anything that has run under those pins.

Sorry for the delay in response...things have been quite hectic lately...been trying to cover 3 roles at work...and it's crazy.

John is indeed correct...it has finally decided to behave itself and is working beautifully!  I got everything back together Thursday evening including getting the shell back on and we are all good.  Everything us working correctly now...including whistle and blow down smoke.  Whatever caused that initial short circuit between the element and common is still a mystery...that took disassembling the smoke units at least 3 times...but it's stayed cleared for several laps around the track and across a few days time...so I'm hoping we've solved it once and for all.

@DaveGG I thought about it maybe being a faulty plug where a pin was touching...but I couldn't find anything conclusive.  Nor could I make it happen on demand.  About the only thing I can come up with was perhaps the thermistor was too close to the element and when it got hot the 1st time it flexed just enough for 1 leg to make contact with the element?  I did also clean both boards with alcohol...so there could have been something under the connectors that was causing it too...its just so strange that BOTH aux units had the exact same issue....  thats what I really can't wrap my head around.  To be quite honest I'm still expecting it to happen again...crossing my fingers it does not..but without a provable root cause it has left me with a bit of doubt.

Anyway...both my father Lee and I want to send a huge THANK YOU to @gunrunnerjohn and @DaveGG for their time and persistence in sticking with us and putting up with our constant questioning while we worked through this 4 month saga. Definitely one for the books.   

Ps. If anyone has any interest in parts from the pcb "graveyard" I've got 2 dead sfcs and a rcmc up for grabs.  If not I may harvest the rectifiers and regulators off the sfcs then I guess they get tossed...as I'm not aware of any way of getting replacement for that microprocessor that got cooked.

Attachments

Videos (2)
20230928_213910
20230928_214533
Last edited by PRREnola

Add Reply

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