Helping sort out Ed2's Key 2-8-0 odd performance issue

[Ed Kapuscinski]

My dad started having some issues with the beautiful Key PRR 2-8-0 that @cbroughton67 installed Loksound in a few years back recently and I want to brainstorm some potential fixes before I see him and the patient next weekend.

His club layout's DCC system was having an intermittant short (that did not seem to be caused by the engine because it was happening without it on the layout). After each short, the sound in the locomotive would start up and the headlights would come on but then the layout would short again and the cycle would repeat. After a few hard resets of the DCC system things settled down and ops got back to normal.

However, a short time after that the engine just stopped while running in reverse. The light and sound stayed on and playing.  He assumed that something happened to jam the gear train or stop the motor.   
 He changed the direction to forward and the engine responded, moving forward and the light and sound behaved appropriately.  He ran it forward a little distance, then tried reversing again; the light lit and the sound sounded appropriate.  The locomotive moved backward a short distance and stopped. The light remained on and sound was for the stopped engine.  He changed the direction to forward and the engine worked correctly. 

He ran the engine in the forward direction for a while and it operated properly.  It also worked fine in reverse for about 20' and then stopped.  He put it back in forward and it worked fine. 

When he got home he put the locomotive on the home layout (with a Digitrax Zephyr) for more testing. It ran fine when moving forward but, after reversing it it stopped after a distance.  He repeated the experiment several times with the same results.

He then tried it on DC. The locomotive "stuttered," momentarily started with light and sound, then stopped without light or sound, several times.  He assumed that the throttle was putting out pulsed power and "confusing" the decoder, so he dug out and attached a very old MRC Ampack which is straight DC through taperwound rheostat.   

The locomotive behaved in the same way; it stuttered in the forward direction for several tries,  but eventually started moving forward. It ran fine forward, but reverse was pretty much the same as on DCC; it stuttered on starting, then ran backwards for a distance, then stopped. 

We're not sure if the locomotive has a mechanical problem of a jam in the gears, or the decoder has a problem, or if it could possibly be the keep-alive doing something.  He's inclined to believe that the gears aren't the problem because of the distance it can travel before stopping.

So, what do you all think?

[EdKap2]

The Key H8 wasn’t the cause of the shorting; it was happening BEFORE the engine was put on the layout.  The system has had problems, but we thought they were corrrcted.  The consol was put on to replace the engine on the layout when the shorting started up.

[peteski]

The old MRC rheostat-based DC throttle is anything but "straight DC".  It is unfiltered fill-wave DC (basically "all pulses").  For reliably testing decoders on DC you need a better source of DC.  I normally suggest a 9V battery, but that might not be enough voltage for a sound decoder to actually run the loco. But I agree that running it on DC would be a good test.   If you could remove the decoder from the circuit (temporarily hardwire it as a DC loco), then you could easily test the mechanicals using the old DC throttle (or a 9V battery).

If the loco was sitting on the track while the shorts were occurring on that layout, the resulting voltage spokes could have "blew its brains".  Basically scrambled the decoder's programming.  Not  sure what keep alive is in it (a "real" keep-alive, or just few tantalum caps?). I would disconnect the keep-alive  and see if decoder's CVs can be read.  If it responds, I would then perform a decoder reset.

[Rasputen]

To me it sounds like there is little or no thrust control for the motor armature.  Thus the worm can push the armature too far when it is run in one direction (reverse in this case).  I would remove the shell and see how far the armature is moving when the direction is reversed.  Sometimes you can add thrust washers next to the worm, depending on how it is mounted.  Ideally the worm should never be pushing or pulling on the armature in either direction.
If it has a piece of tubing as a drive coupling, you may be able to simply shove the tubing on further, but this solution may not last forever.

[Ed Kapuscinski]

Ooh, good thought. I let him know to check that out.

Thanks @Rasputen

[EdKap2]

Peteski,
I used a "recent" (compared to the Ampack) MRC Railpower 1370 to perform the DC test.  The engine did an on/off/on/off a number of times before doing anything, so i assumed that the pack was pulsing the voltage, so at that point i swapped in the old pack.  The Railpower may be filtered DC, I don't know the technical specs for it, but i bought it a number of years ago to replace the Ampack for better operation.

The "keep-alive" is a TCS KA-4c. 
Ed3 got the guts for the whole upgrade, for me, and EVERYTHING is "top of the line."

Rasputin,
I will take a look at the drive train.

[Lemosteam]

As @peteski mentions, I would flip it into a cradle and use a 9 volt battery to inspect the rods and linkage, reversing the polarity of the battery to check both directions.

I am wondering if the gage of one of the drivers is out allowing a pin or rod to make contact.

On thing that was not specifically mentioned is the distance in reverse the loco would travel on the track before it stopped. Like how many driver revolutions did it take or is it before one full revolution happens?

[mmagliaro]

There are many good suggestions in here, but I suggest you back up a whole step..

Since it reliably stops in reverse on the track, that is your repeatable failure (which is a lot better than a nasty
intermittent thing that doesn't always happen!)  So the first thing I would do would be to run it on the track in reverse, and
when it stops, see if the drivers are locked. 

Can you gently nudge them with a toothpick, to see if they
wiggle a little bit in each direction, and if all the rods can also wiggle?
If a wheel or rod is stuck, this is a straight-up mechanical jam, and for now, we can pause diagnosing the decoder, motor, or track pickups.

If this confirms that it is physically jammed, can you take a close, sharp photo of each side?  That would really help.

(and no, the MRC 1370 isn't anywhere near smooth DC)

[EdKap2]

John and Max,
Thanks for the ideas; i will check
The engine will travel well more than 1 revolution of the drivers; it will go several feet on the layout before stopping.
Thanks for the info on the MRC pack.  It seems all the DCC equipped engines do not run smoothly.

[u18b]

While I have had my share of electrical issues...

I'm thinking this may be mechanical.

As Max said.... check the drivers when it stops.

I would suggest also taking a photo of it stopped (both sides).   
And doing this several times.

If it ALWAYS stops with the siderods in the same position- that is not a co-incidence and indicates a mechanical bind.

[Ed Kapuscinski]

Ok boys, had a chance to check it out tonight.

[ Guests cannot view attachments ]

It ran a few laps around the layout backwards with no problem but eventually it manifested.

I checked it out and it does appear to be mechanical. I dropped the gear cover and it cleared itself.

My theory is that there's a burr or something that just eventually causes an issue. I applied a drop of oil for insurance.

[EdKap2]

That was the first time the gearcase had been opened by us.  I've oiled the rods and axles several times, but not the gears.
After Ed oiled the locomotive, i brought it home and applied a small smear of LaBelle gear grease with teflon.  I'm letting it work in to the gearbox by slow run on the layout.  We'll see what happens.

[peteski]

That was the first time the gearcase had been opened by us.  I've oiled the rods and axles several times, but not the gears.
After Ed oiled the locomotive, i brought it home and applied a small smear of LaBelle gear grease with teflon.  I'm letting it work in to the gearbox by slow run on the layout.  We'll see what happens.

The problem as described is unlikely caused by insufficient lubrication, but one can be hopeful.
If the problem is indeed resolved, then I would still think that it was not the lubrication, but moving of some pieces of the mechanism during disassembly that fixed the problem.

[lock4244]

Ok boys, had a chance to check it out tonight.

(Attachment Link)

It ran a few laps around the layout backwards with no problem but eventually it manifested.

I checked it out and it does appear to be mechanical. I dropped the gear cover and it cleared itself.

My theory is that there's a burr or something that just eventually causes an issue. I applied a drop of oil for insurance.

Have you tried lettering it for Conrail? That fixes everything!

[EdKap2]

Pete, you're probably right.  I was expecting to find some shavings or dirt in the gearbox, but it was clean.  Maybe jiggling the gear alignment relieved the issue.