PowerCab with PSX and PSX-AR

[nickelplate759]

I have the same 1.8A wart that Steveruger45 has.  I'm pretty sure it's the PowerCab itself that is shutting down on a short, as it goes into a continues trip-reboot-tripp-reboot-... loop, which is what the PowerCab doc says will happen if a short isn't cleard.

I think there are two separate issues here -
1. when there's a short, the PowerCab shuts down before the PSX trips
2. Certain engines pause when entering a PSX power district from a reversed PSX-AR district.

It's the second problem I really need to solve the most, but both are important.  By the way, neither the PSX nor the PowerCab show a short indication when the pause happens.


Just to clear things up - engines with KeepAlive sail through with no problems.  Engines with good all-wheel pickup (modern Atlas & Kato for example) also work fine.
Engines with less reliable pickup - like my MRC/Model Power Mike, or an OLD Atlas/Kato RSD12 (pre-bronze pickup strips)  show the problem.

I'll check software version on the PSX today.  The docs that came with it say "Software Rev K 4/15".

I'll also try tinkering with CV65 (delay) and see if that helps.

By the way, DCC Specialties web site is awol - are they still in business?
[edit] Web site is alive, but their security certificate is out of whack apparently.  In a Google search they show up by IP address - https://66.39.46.164/ [/edit]

[peteski]

George, the Power Cabs bootup messages show its current firmware revisions.  You have to be quick as they flash on the screen fairly fast.

[Maletrain]

OK, since you have the highest current rated wall wart, getting one with a higher rating will not help.  From what you describe, it seems that it is the wall wart that is switching off and back on when there is a short.

One thing that might work is to replace the PSX with one of those current-limiting light bulb things (or equivalent) that NCE sells for the PowerCab track bus.  That would at least prevent the PowerCab from tripping, but I don't know if it would still allow the PSX-AR to function.  Cheap to try and a really cheap solution if it works.

But, there are some other funny things going on that might really be the cause of the whole problem.

For one thing, it seems odd that the PSX will trip when the loco leaves the AR section, but not when it enters the AR section.  You did say that you have 2 PSXs and 2 ARs, so my first question is whether the two ends of the AR section connect to the same PSX, or to the 2 different PSXs.  If the same PSX, then maybe try swapping PSXs and see if the problem persists.  If so, try swapping the ARs and see if the problem persists.

One other thing I should ask is how you wired the PSX-ARs to the PowerCab.  Are they hooked up in parallel to the PSXs or in series with the PSXs. (They should be in parallel, and might produce this problem if in series.)

[nickelplate759]

OK, since you have the highest current rated wall wart, getting one with a higher rating will not help.  From what you describe, it seems that it is the wall wart that is switching off and back on when there is a short.

One thing that might work is to replace the PSX with one of those current-limiting light bulb things (or equivalent) that NCE sells for the PowerCab track bus.  That would at least prevent the PowerCab from tripping, but I don't know if it would still allow the PSX-AR to function.  Cheap to try and a really cheap solution if it works.

But, there are some other funny things going on that might really be the cause of the whole problem.

For one thing, it seems odd that the PSX will trip when the loco leaves the AR section, but not when it enters the AR section.  You did say that you have 2 PSXs and 2 ARs, so my first question is whether the two ends of the AR section connect to the same PSX, or to the 2 different PSXs.  If the same PSX, then maybe try swapping PSXs and see if the problem persists.  If so, try swapping the ARs and see if the problem persists.

One other thing I should ask is how you wired the PSX-ARs to the PowerCab.  Are they hooked up in parallel to the PSXs or in series with the PSXs. (They should be in parallel, and might produce this problem if in series.)

The Two PSX-AR sections attach only to the mainline PSX unit - not to each other, and not to the second PSX (it's on a branch, and attaches only to the first PSX).
Everything is wired in parallel.

Other data:

Both PSX boards have a date code of 12/15
On boot up, the mainline PSX identifies its software rev as Long-Short-Long : rev. K.  This agrees with the manual that came with it.
I haven't checked the branch PSX, but since it's part of the same PSX-2 board it ought to be the same.

The PSX-AR have software rev N.`
The PowerCab firmware is rev 1.65B

[Maletrain]

I could understand if the PSX was tripping with the train on one side of the AR, but not the other, because the PSX senses the short on only one side of its circuit, and the short could be through the other side plus one side of the AR (assuming a tiny rail gap offset).  But, the current for the short all goes through the PowerCab and its switching power supply in the wall wart.

But, you say a simple short from rail-to-opposite-rail still does not trip the PSX before the PowerCab or its wall wart.  That really suggests some sort of timing issue with the PSX.  Have you tried that "quarter test" on both PSXs with the same result?

[nickelplate759]

So, another experiment. If I force the PSX-AR to trip using tweezers (no quarter handy), there is no problem, ever.  The engines (even the ones with unreliable contact) keep right on moving as they should.

If I cause a dead short, the PSX trips (set to 1.27 A) as it should, and restarts in about a second or so.

I'm thinking the issue is really related to relatively poor contact on certain engines, but I'm still perplexed on why it's happening at the block boundary between the PSX-AR and the PSX.   Could poor contact make tripping unreliable?   Would staggering the insulating gaps make a difference?

One other observation - when the pause happens, it seems to be when a wheelset is right at the block boundary.  On my unmodified MRC 2-8-2 (which has a factory MRC sound decoder), it's always when the lead tender axle touches the block boundary.   

Perplexed in Portland

[Maletrain]

When you say "forced the AR to trip using tweezers" I assume you mean that you had the AR set to not match the mainline phase on the exit end of the reversing section and put the tweezers across the insulating gap in one rail.  If that is the case, did you do that for both rails and get the same result, or is that for just one of the two gaps?

[nickelplate759]

When you say "forced the AR to trip using tweezers" I assume you mean that you had the AR set to not match the mainline phase on the exit end of the reversing section and put the tweezers across the insulating gap in one rail.  If that is the case, did you do that for both rails and get the same result, or is that for just one of the two gaps?

You assume correctly.   Also, I tried putting the tweezers diagonally across the gap (from L rail to R rail) to force it to trip.   Even tried it with the engine running in the reversing section to see if it hesitated - it did not.

Tried forcing the trip on both rails, same result.

I also tried setting CV55 =1 on the PSX to delay the breaker trip.  No change.

I'm coming to the conclusion that the real culprit here is just poor contact from the locos that have the issue, but I still can't figure out why it's only in the direction from the (reversed) PSX-AR into the PSX.  Once in a while they'll get through with no issue, for example.

[Maletrain]

What is the track like near the PSX to AR gaps? Is it just plain track, or are there turnouts or diamonds or other things that might cause a short while the loco is on the gaps?

[nickelplate759]

Turnouts.

Frogs are live with polarity matching the way the turnout is thrown controlled by  tortoises

[Maletrain]

I assume that there are turnouts on the reversing section end that is having a problem, but not on the end that is not having a problem?

You say that the hesitation happens when the lead tender axle is on the gap.  Where are the 2-8-2's drives, and also the pony wheels at that point?  I am wondering if there is a transient short there that coincides with the short that reverses the AR, and it is not enough time/current to trip the PSX, but when combined with the AR transient short, is enough to trip the wall wart.  It seems odd that the whole locomotive gets across the gap before there is a short that reverses the AR.  It is as if the locomotive has no pick-up at all.

Maybe it would reveal something if you slowly pushed the locomotive across the gap by hand, with its speed set to zero by the decoder.  Also, checking the locomotive for continuity between its various wheels and the tender wheels that are supposed to be the same phase would show that they are not electrically connected, and thus explain why the AR does not reverse on the first set of drivers. Or, even the pony truck should reverse the AR when it drops into the rail gap and makes the connection (unless you filled the gaps with plastic to keep that from happening).

[Maletrain]

For others who are following this thread and have various versions of PowerCabs and wall warts, I am providing one more bit of information I received from the NCE group:

There are no differences in the trip current settings in the firmware with different versions of the PowerCab.  Very early PowerCabs didn’t have the software overload shutdown, but once the feature was added it’s the same for all later versions of the firmware.

[peteski]

Shouldn't any metal wheel (regardless whether it picks up power or not) crossing the gap in the track (even just in one track) between the main and reversing block trigger the reverser? Heck, even placing a screwdriver across a gap in one rail should trigger the reverser. It should trigger as soon as it sees mismatched DCC phase across even one of the rails.  At least that is how I understand it to work.  It is not about the load across the track - it is about the phase (polarity) mismatch in each block

You should be able to trigger the reverse by simply running any freight car with FVM wheels on it across the gap.

[nickelplate759]

Pete,

The gaps are filled with plastic rail joiners, so there is actually a short dead spot  - maybe .030".

I wonder if shaving down the plastic so the wheel can bridge the gap would help.

[Maletrain]

Shouldn't any metal wheel (regardless whether it picks up power or not) crossing the gap in the track (even just in one track) between the main and reversing block trigger the reverser? Heck, even placing a screwdriver across a gap in one rail should trigger the reverser. It should trigger as soon as it sees mismatched DCC phase across even one of the rails.  At least that is how I understand it to work.  It is not about the load across the track - it is about the phase (polarity) mismatch in each block

You should be able to trigger the reverse by simply running any freight car with FVM wheels on it across the gap.

I agree, unless the rail gaps were filled with an electrical insulating material like a piece of styrene or thick glue and then filed to match the rail profile.  Some people do that to make sure the gap doesn't close and cause a short.  Others do it because they don't like the sight of the gaps.  That insulating material can provide a way for a metal wheel to only touch one side of the gap or the other, because it prevents the wheel from dropping into the gap to bridge it for sure.  That is why I added the last comment in my post to Nickelplate759 about whether he had filled his gaps.