PS-2 locomotive only goes forward when horn button is held down?

Have you tried a different transformer to run the engine?

Time for a trip to the JWA RR.

What engine?  PS-2 what?  Odd that it takes the DC offset an allows that to pass to the motor fet to run the train?  I have never seen that and can't even figure out how that is possible.  G

maybe the controller is toast?

How’s the tether?

Thank you for your comments, in response:

PS-2 upgrade on a SS 4-8-4.  Previously working fine for quite a few total hours.  Has been run on medium to slow speed on a regular basis.  Beautiful brass beast and a great puller.

Controller works fine for other PS-2 upgrades and factory PS-2 & factory PS-3 locos steam and Diesel.

Tether is fully engaged and the cord appears intact.

Only ever used in conventional.

It sounds like it's time for the test-set to come out of the closet and check the board out.

Since controller seems ok it appears the engine thinks it is in DC conventional mode.  So a net positive voltage (horn) is FWD and a net negative voltage (bell) is REV. Since the controller appears to be ok driving other PS2 engines, there is a simple track voltage measuring circuit that might be giving faulty data to the processor chip that calculates voltage level and voltage type. I think it needs to be looked at by a tech.

I don't know Stan, still does not make sense to me.  DC conventional operation is odd ball and not well documented.  DC via a TIU is fine but the engine know it is in DCS mode.   This implies somehow a low level .5V offset or change, is used to drive the motor fet?  The processor either tells it to go or not, and with a small Voltage level like that it would not want to go fast at all.

I have never seen a board do this.  I have seen in conventional with AC the whistle or bell cause the engine to speed up some, but not the motion as described.

There is a chip that takes the processor direction and drives the motor fet, so maybe some sort of damage that caused it to fire the fet with a DC offset but the relay has to also be told to change position for the motor fet to drive in reverse. So I still think unlikely.

In my mind somehow it shorted so the track AC is on the motor leads so motor is stalled.  Then the DC offset come in and it moves with the voltage difference.  So maybe the main rectifier shorted for PV.  Are the lights on? Normal or bright? Out?

Are you sure this is PS-2 not PS-1?  Is this conventional or DCS operation?  Have you tried a conventional or DCS reset?  G

To be clear, I have not seen this either.  So I'm speculating. I'm suggesting the engine "thinks" it's in DC mode due to circuit damage.  I suppose it's worth a try to factory/feature reset the engine, but I suspect the engine will not respond to the Horn-Bell button sequences since pressing either button simply makes the engine move rather than triggering the respective sound.  But since it takes only a few seconds to try, why not.  So press the reset sequence and miracle of miracles if he all of a sudden get a double-horn-honk acknowledgement sound!

I think it is in DC mode because that's the only conventional mode where interruptions in track voltage do not change the F-N-R-N sequence. I think the engine needs to go to a tech.  My advice to said tech would be to look at the circuit that converts the AC track voltage to a low-voltage DC signal that is read by the processor chip using A/D conversion.  This is done with an op-amp.  It is via this circuit that the processor decides the conventional engine is in AC or DC mode when power is initially applied to the engine.  And then once in DC mode, it is this voltage that the processor chip averages to get the speed command.  Hence with regular AC on the track, the speed command is zero.  And when horn or bell is pressed, the average voltage is no longer zero so speed command is no longer zero with the direction taken from the polarity.  I believe there's something screwy going on in this circuit.

OP did not mention if he has DCS capability but it's possible that this engine would work fine in command mode!  The engine decides it is in DCS mode by looking for the watchdog signal via a different signal path than the track voltage measuring circuit.  Now that would be a new one!  Usually you hear about engines that work fine in conventional but not in DCS.

That is not how it works stan.  Input AC is directly converted to DC for PV and DC for the Power supply chips.  RAW Track signal does go to an opamp to work in conjunction with audio amp to detect offset for bell/horn.  Track AC also does go to the processor via signal.  None of this is taught or known by ASC, and in most causes not chips that could be repaired.  Regardless, how is it triggering the motor fet and relay with a small offset.  From my perspective depending on how fast it moves with the offset signal present the voltage is getting to motor directly.  We still do not know for sure which engine, nor how it is operated.  So this is all speculation. I would like to see the board to actually observe the actions. G

@GGG posted:

(1)... RAW Track signal does go to an opamp to work in conjunction with audio amp to detect offset for bell/horn.

(2)...Regardless, how is it triggering the motor fet and relay with a small offset.  From my perspective depending on how fast it moves with the offset signal present the voltage is getting to motor directly.

(3)...So this is all speculation. I would like to see the board to actually observe the actions. G

(1) OK. we'll just have to agree to disagree on this one.  I'm suggesting that any DC offsets in the track voltage are detected/measured by the processor chip.  The processor chip is the master-of-the-universe such that EVERYTHING goes thru the processor where all decisions are made about what sounds to play (including bell and horn), how to drive the motor, when to flip the relay between Fwd and Rev, etc.

For example, even in conventional mode, when it is working right if you hold the horn button for an extended interval the processor will play an extended tail sound of the horn which is kind of neat.  This suggests it is the processor that is measuring the time that DC offset is present and deciding what horn sound to play.  Likewise, the processor must be looking at the track voltage (and any DC offset) to decode the Horn-Bell combinations to fire the coupler, activate PFA/PSA, and so on.

(2) Again, we can agree to disagree.  I do not see how a track voltage with negative offset (i.e., Bell button pressed) can get thru the bridge rectifier and maintain its negative polarity.  That is, a bridge rectifer converts either polarity to "positive" voltage in this case.  The ONLY way the engine can go backwards in a PS2 engine is if the processor flips the relay which in effect flips the positive PV voltage to negative essentially right at the motor.

(3) Now this we can agree on.  This is all speculation!  I would like to see the engine in the hands of a capable tech (like you of course!).  Again, if the OP does go this route and the tech chooses to openly discuss on this thread, I will dig up a PS2 board and fire up the scope to elaborate my speculations.

Otherwise, I believe there's an even-odds chance the engine will operate in DCS command mode since measurement of the track voltage is not used.  Though there may be an issue with the processor properly detecting loss of track voltage for the purpose of shutting down the engine and performing related shutdown chores (such as saving any changed parameters like the engine address, and finally electrically disconnecting the battery).

Stan, I agree with #1 and #2.  I did say that the processor gets track input and DC offset input.  There also is a Transformer signal from the PS T1.  There are no Bell and whistle buttons for DC offset on a DC power supply.  There are no operating instructions for conventional DC operation of a PS-2 engine other than the #1 gauge.

I am implying the processor would not see a slight change in DC voltage as a signal to run the motor from neutral.

If you look at the G scale instructions.  2 methods for operating in DC conventional for motion.  Raise throttle and you get motion in that direction.  Clearly the level of DC voltage measure will determine what speed the processor signals to the motor drive circuit.  Want to reverse direction, either turn throttle down but not off until engine stops.  Move direction button to change polarity and increase voltage.  So I believe at certain DC voltage the board is on producing sounds and such, but not motion because voltage level not high enough for the processor to determine you want it to go forward.  Some of the G guy can confirm this.  Believe PS-3 HO also works this way.  You have to raise voltage high to get it moving, then lower voltage for the speed you want.  Otherwise it sits in neutral, not motor stall.

Or with voltage high, you just move direction button and switch polarity.  The engine will naturally slow down and stop first, then switch direction and move in reverse back at set speed of track voltage.

So the behavior of the processor action and motor as described with low DC voltage offset doesn't seem compatible to me for the processor to think that is a motor drive signal with DC power input.  G

This professionally installed PS-2 up graded 2 rail SS 4-8-4 has only been used conventionally with a MTH Z-1000.

Never used with DC track power.

The throttle, horn and bell functions were all normal prior to the lead truck derailment and shorting.  Estimated total hours of use well under 20.

GGG, I have PS2 F series AA units on my G scale track with straight DC power.

I turned on the transformer and raised the handle too much. The units moved right away.

I lowered the voltage and they just sat and idled. As soon as the voltage is raised just right amount, they go forward.

Is that what you needed?

I think the main difference is the engine is in forward and just doesn't move. When you throw the direction switch, it is in reverse, yet doesn't move until voltage is raised. It doesn't appear to be stalled, just that it's in one direction or the other and so are the lights.

I don't hear any motor hum, and there seems to be a threshold where the power starts going to the motors.

It does seem very touchy about the amount of voltage to decide to start moving. My throttle might not be capable enough. A slight touch, and the engines move very slowly. I usually run in command mode, so this is different for me.

@Tom Tee posted:

This professionally installed PS-2 up graded 2 rail SS 4-8-4 has only been used conventionally with a MTH Z-1000.

Never used with DC track power.

The throttle, horn and bell functions were all normal prior to the lead truck derailment and shorting.  Estimated total hours of use well under 20.

So is the plan to send the engine in for repair?  I suppose you can wait a bit to see if anyone else comes forward with a proverbial "Get out of Jail" card to restore your engine's functionality...but I suspect you may be waiting a long time.

My diagnosis is a failure in the voltmeter circuit in the engine.  The PS2 electronics has what amounts to a built-in voltmeter that can measure both AC and DC voltages.  It is implemented by a combination of some basic circuitry to scale down the "high" AC or DC track voltages to lower voltages that can be processed by a low-voltage (3 Volt) processor chip using an A/D (analog-to-digital) converter.  Then there's a bunch of software to convert the voltage measurements to engine commands for sound and motion.  I'm hoping that the fault is in the basic scale down circuit because these are inexpensive parts that a skilled tech should be able to diagnose/replace.  If the fault is in the processor-side of the voltmeter circuit then it's game-over since the processor chip is effectively irreplaceable.

As for the discussion about conventional DC operation, I see it this way:

Again, everything goes through the processor chip.  The motor receives no voltage unless the processor says so!  The relay (that flips motor voltage polarity) only changes if the processor says so!

In DC operation, the MTH PS2/3 engines have a speed-vs-voltage relationship roughly as shown above.  There is a small voltage window where the engine will indeed just sit there with idle sounds.  IIRC this small window is about 2-3V.  Go above this window and the engine starts moving.  Go below this window and there is not enough voltage to run the electronics so the engine starts playing the shutdown sound and turning off.  There is symmetrical behavior with negative DC voltages except the processor flips the direction relay.

I believe the voltmeter circuit is broken.  On powerup, the engine decides if it is to be a conventional AC engine or a conventional DC engine - how it decides to be a DCC or DCS command-control engine is not relevant for this discussion.  The broken voltmeter is somehow interpreting the AC track voltage to be a DC track voltage and going into DC mode.  Then, when looking for the DC voltage to set the speed, it is somehow "exaggerating" the modest DC offsets from Horn and Bell buttons on the controller thinking the DC voltage is above the threshold to move the engine.  All speculation.

@stan2004 posted:

I'm hoping that the fault is in the basic scale down circuit because these are inexpensive parts that a skilled tech should be able to diagnose/replace.  If the fault is in the processor-side of the voltmeter circuit then it's game-over since the processor chip is effectively irreplaceable.

Remember Stan, even the "skilled tech" working without any service information may have difficulty tracking down/fixing something like this.  There is also the point of diminishing returns on how long you're going to spend trying to track it down before simply replacing the board makes more sense.

Stan, It is not 2-3V of DC for motion.  The processor won't start up at that level and even if it was running, that is not a high enough voltage to come out of neutral.   Joe confirmed how the boards work in DC as I mentioned.  Please let us know which chip is this simple converter?  I just do not see how the Processor if it thinks it is in DC but see 0 VDC, or let's say it think 12VAC is 12VDC.  If so it should move.  Otherwise a small 3-5V DC offset is not high enough to determine that is the command to go.  The engine would sit at idle with 8 to 9V until it got a high voltage spike.  Frankly, it is more likely the main rectifier got damaged, and the motor leads have AC on them and the DC offset causes motion.  But as I stated before, track power and Signal transformer power do go to the main processor.  So if it took a hit, no can fixxy.  I just don't think the DC offset is of the level in DC voltage that the processor would think it should command motion.

If OP wants to send me the board I can test and diagnose.  But I think a diode took a hit and the offset is getting on the motor leads circuit to drive the motor. G

@GGG posted:

Stan, It is not 2-3V of DC for motion.  The processor won't start up at that level and even if it was running, that is not a high enough voltage to come out of neutral.   Joe confirmed how the boards work in DC as I mentioned.

I said the zone of 0 sMPH speed command in DC is 2-3V.  This is the voltage range that Joe said is small and finicky.  As shown in my simplistic curve above, this 2-3V zone starts at some voltage - I'm guessing around 8-9V DC or whatever is enough to be able to create 5V digital power for the electronics. 

So for any track voltage below 8-9V DC, the engine will shutdown.  Then for voltages above that there is a 2-3V zone where speed command is 0 sMPH.  Then around 10V the engine starts to move.

Again, I believe the voltmeter in the particular engine is faulty.  If the OP sends the board to you, then I'll volunteer to power up a PS2 board set and track down the voltmeter circuit.  This will involve looking up the locations of the A/D pins on the processor chip and working backwards to find the scale-down circuit that converts the track AC to a DC voltage.  Remember I'm not an ASC tech so I don't have a PS2 test fixture which I assume would help with component level diagnosis.

But if the OP doesn't want to send it to you or someone else who will commit to troubleshooting on this thread, then it's moot and chalk it up to an unsolved mystery.

@gunrunnerjohn posted:

Remember Stan, even the "skilled tech" working without any service information may have difficulty tracking down/fixing something like this.  There is also the point of diminishing returns on how long you're going to spend trying to track it down before simply replacing the board makes more sense.

Absolutely.  There is a fuzzy gray area between the hobby side and the business side. Chatting on OGR about curious electrical issues with trains or speculating about why an engine only goes forward when pressing the Horn button is to me the hobby side.  No charge.  Or perhaps just 2-cents worth.

Sending a board in to a service center is the business side (a.k.a. Show Me the Money!).  Even the 2-way shipping alone is a whack on the wallet. 

Stan, you are picking a complex failure that we don't even know if the board is coded to work that way.  Level versus rise, doesn't know power type but does or doesn't know voltage etc....  A never seen failure?  Certainly by me. Plus the weakest component the audio amp survives the AC Spike but the processor doesn't?  The Z-1000 brick is a slow breaker.

How it moves when offset applied also matters.  If it moves immediately and without speed control that would be more of an indicator that the offest is getting on the motor lead directly.  Otherwise, processor would signal a speed control type motion with accel/deaccel.  G

I have no issues with your questions and incredulity.  As stated it is just my opinion based on what might be happening based on the stated behaviors...and without the benefit of troubleshooting training procedures provided to ASC techs, 

If the OP decides on repair, I genuinely hope it's something simple and quick/inexpensive to repair like a bad diode or whatever. Again, if whoever does the repair exhausts the "usual suspects" and chooses to post additional symptoms/behaviors, I will now stand back and stand by - no charge!