Skip to main content

Hello Everyone. Working on Proto 2 5 volt boards. Looking from those that have repair expierence. I have replaced mutiple Capicitors already that charge the battery or BCR. What im looking for now what are the other issues people are seeing? What else is failing? I have a few other boards that all the capicitors test fine on but they still are dead. Someone mentioned bad diodes. They seem to test good as well. I have attached pics of a board i have seperated and numbered. Keep in mind this is an older 5v board and does not have the extra capicitor. I have also numbered top board and bottom board. If you can, please refer to top board or bottom board with the number assigned in the discussion. I am determined to repair them and get them working again. Thanks for all your help!

Attachments

Images (4)
  • 20210207_084046: Top board 1
  • 20210207_085002: Underside top board
  • 20210207_084332: Top of bottom board
  • 20210207_084413: Underaide of bottom board
Original Post

Replies sorted oldest to newest

Good luck.  There can processor failures, internal PCB trace failures, other microchip failures.  The heat used to separate the boards could damage traces and many of the SMD parts.

These boards are just beyond worth anything anymore.  Now if you enjoy beating your head against the wall, I applaud your effort, but working on PS-2 3V or PS-3 will be more worthy of your time.    G

@Paul Lytle posted:

...

I am determined to repair them and get them working again.

Do you have and know how to use an oscilloscope?

From what I can see, you did a nice job non-destructively separating the boards which is often the deal-breaker for guys attempting to mess with the PS2 5V board set.  But do you have a method to non-destructively remove the ~50 to ~100 pin surface-mounted digital IC chips of the complexity as found on the PCB?  When my eyesight was better, I used Chip-Quik for this purpose.

I see this as a matter of principle.  If looking at this from a business perspective - say from a tech who repairs train electronics for remuneration - then this is a no-win situation!  But if it is a personal mission in life to be the go-to-guy for what might be THOUSANDS of dead PS2 5V boards sitting in boxes (or the landfill), then hurray for you and let the games begin!  As I see it, the satisfaction of repairing something that has been written off as a lost cause is priceless.

There is some guy or guys on OGR that reverse engineered how sounds are stored in MTH PS2/3 engines.  This allows him/them to replace the stock sounds with whatever one wants.  I don't think any money changed hands and I've got to believe this was more than a weekend project!

One thing's for certain.  OGR is the right place to undertake the journey. 

@stan2004 posted:

Do you have and know how to use an oscilloscope?

From what I can see, you did a nice job non-destructively separating the boards which is often the deal-breaker for guys attempting to mess with the PS2 5V board set.  But do you have a method to non-destructively remove the ~50 to ~100 pin surface-mounted digital IC chips of the complexity as found on the PCB?  When my eyesight was better, I used Chip-Quik for this purpose.

Getting the boards apart is a major accomplishment, that allows you to get to many of the components that are otherwise out of reach.  For removing smaller chips, I've turned to my hot air tool, but the larger ones I'd have to invest in the nozzles specific to the chip, and I just don't have the call to do it enough to spent that money.

@GGG posted:

I think you will be held criminally liable when the head bashing starts trying to salvage 5V you provided.  G

You're probably right George, I'm calling my lawyer now to prepare my defense!

The problem I think is deeper.  There are parts that are obsolete on that board, let alone those with firmware loaded.  There are no design drawing out there. They are multi trace layer boards.  I have my process down to a science for 3V boards.  But still can't fix them all especially when the 5V and 3V buss is fused internal in the PCB.  Or the main processor looses a function from a short that is critical.  3V went obsolete for the same issues of specific chips no longer available.  Just compare it to a 3V and realize even if you fix one chip, another is about to fail on a 5V.  So unless your going to remove and replace all through hole and major SMD devices like diodes and FETs.  I think we had some one over a year ago going down this same road.  Never heard it was successful.

I am just saying pick some thing that the design was better to start and actually a desirable part.  It took me about a year and half to reverse engineer the 3V boards and fine how it worked, and it was a much simpler device. Then another year or so of finding common modes of failure.  We have no idea how the 5V fails.  We just know some are total dead, and other have direct shorts.  No real insight except the PV diodes, the rectifier, the swelling cap.  Audio amps, light fets and motor fets, no problem.  Everything else G

I finally got some time to dig into the boards. I started off with the boards that were causing an immediate short to the transformer when power was applied. There were 2 of them. Before separating the levels I removed the battery charge capacitor. I then removed the diode that covered where the pins were soldered into the board. After removed I was able to heat each pin and use my solder suction tool to remove all solder and loosen the boards. Note that the removal and connections on the board are very clean. Finally I was able to locate  failed diodes on the bottom side of the upper board. Before I go purchasing diodes. Any suggestion on the rating of the diodes and what would cause them to fail? 2 boards same exact issue.

Attachments

Images (5)
  • 20210225_113322
  • 20210225_113332
  • 20210225_115712
  • 20210225_115747 (1)
  • 20210225_120125
@Paul Lytle posted:

.... Finally I was able to locate  failed diodes on the bottom side of the upper board. Before I go purchasing diodes. Any suggestion on the rating of the diodes and what would cause them to fail? 2 boards same exact issue.

Can you mark in the photos exactly WHICH diodes failed (or did I miss it)?

And did they fail open or closed?  Note that if a power supply circuit, it's highly likely some of them are Schottky type.

Stan, I'm guessing the ones in the middle with the smutz on them are the failure items.   I notice in the other photos the discoloration of the bottom board where something blew.  The one diode in there seems to have different banding, so it's likely to be a different type.  Perhaps measuring some from boards that don't show failures will yield a guess as to diode type?  On a diode test on your meter, a Schottky should show a significantly different voltage drop.

Attachments

Images (1)
  • mceclip0

I have to smile here, I have a whole box of dead 5V boards, and I have no desire to work on them.  Paul, you can have them for the cost of shipping!

So did the box of dead 5V boards change hands?

If sitting on a pile of dead 5V boards with goal of fixing several/many,

1) I'd want to know if you can you power up JUST the smaller power supply board and reach some confidence level that it is working?  You'd think you could at minimum confirm a few different voltages.

2) Then, I'd think some kind of test fixture or test connector/harness would be handy that can mate the two boards with more separation than the final connectors.  In other words, something where you can probe/measure voltages on both sides of BOTH boards mated and powered.  This might entail some effort to make a clamp or whatever with spring-loaded "pogo" pins or the like.

OTOH, perhaps this means the glass is half-full.

I'd be more concerned if there was a box of dead logic boards...because those have got to be harder to repair given the combination of obsolete and/or custom digital chips that would be impossible to troubleshoot with "just" a meter and difficult to physically replace.

As per my earlier premise, MTH would be the other end of the "business" perspective.  I figure their "burdened" labor rate would be higher than you or GGG or the other usual suspects.  What would be interesting to see is the procedure/process by which those power supply boards ended up in the box.  I figure somewhere in there is a check to see if a Wincap (or whatever that cap was called) is installed...and if so replace.  For all we know, there might be another 1 minute test such as a particular diode that changing would solve some percentage of failures.  Etc..

So as a "business" proposition, there must have been a line-in-the-sand where MTH told their techs to spend no more than X minutes on a board then just toss it in the box.  We know the replacement parts cost of the diodes or capacitors is essentially negligible.  So it was all about time equals money.  From a "hobby" perspective time does not equal money.  That's what makes Paul's quest interesting.  I now step down from the soap box...

Addendum: I stumbled across this OGR thread where @tansqrx reports 20-30% success rate with PS2-5V board repair.  That's a very encouraging batting average!  He also posted a detailed troubleshooting checklist.

Last edited by stan2004

They are the PV diodes and rated at 1 amp and probably susceptible to age failure.  Make the DC for the various  other Power Supplies on the board.

Way back I saved a few with getting out the one bad diode, but a real pain and not always successful.  I also (once) separate boards and then reattached via jumper wires just to test, and it was not successful.

Under the rectifier you will find a rev code for the boards, you might want to work on the K, they were the last and best board.  G

Wanted to share with you something I learned. PS2 5volt board. Engine is programmed into DCS. Train starts and runs no problem but after some time running you cannot control it with your DCS or it moves to inactive. You then kill the power to the track and then power up again and engine is back in active again and starts and works for some time and then you have response loss again or engine is inactive in remote. I found that changinging the radio pickups seems to help. if you have an old failed board you can remove from old board and install in the board your having issues with. I was able to swap out while board is still in the engine by installing a wedge between both board which held the bottom board up so i could unsolder and remove the frequency filter and replace with others.

Attachments

Images (2)
  • 20210419_223310
  • 20210419_222515

I have three remaining PS2 5 volt board engines - two diesels and one subway (my beloved R36 Worlds fair).  All of my other PS2 5 volt trains have died and were upgraded to PS2 3 volts or PS3,  a number with flaking speakers and others for no apparent reason.  However, these three just keep on going, one is 20 years old and the other two are from 2002/2003.  I have no ability to change capacitors or anything else on the board, I just make sure to run them at under 16 volts (to not over tax them). Is there anything else you guys would do or should i just have my dealer upgrade them now before the inevitable failure? Or is that true - is it inevitable that these will fail before the ps2 3 volt boards?

@Paul Lytle posted:

John, I was able to get 3 repaired so far. I have about 8 I haven't looked at yet. I've had some failures on some as well. It is a great learning experience.

That's good, what is your average time to repair one of the boards?

I just make sure to run them at under 16 volts (to not over tax them). Is there anything else you guys would do or should i just have my dealer upgrade them now before the inevitable failure? Or is that true - is it inevitable that these will fail before the ps2 3 volt boards?

Nothing is inevitable, the 5V boards may outlive you.  While it's true that they are more prone to failure than 3V boards, there's no certainty that a specific 5V board will fail, some may last another 20 years.  I just run them until something goes wrong.  I do inspect them and replace any suspect caps that I can reach.

Nothing is inevitable, the 5V boards may outlive you.  While it's true that they are more prone to failure than 3V boards, there's no certainty that a specific 5V board will fail, some may last another 20 years.  I just run them until something goes wrong.  I do inspect them and replace any suspect caps that I can reach.

The ones I've seen fail (and it's usually a blown cap) were made in 2000-2001. The 2002-2003 5V boards seem to not suffer from that as much, although it's really random chance. I've seen more RK steam and diesel failures with PS2 5V than Premier steam as well (except the Premier FEFs, which fail frequently). I can't say why.

All my stuff is either PS2 3V, PS32 boards (things I've upgraded) or PS3. The oldest PS2 3V locomotives I have are about 17 years old and are running fine.

@Lou1985 posted:

The ones I've seen fail (and it's usually a blown cap) were made in 2000-2001. The 2002-2003 5V boards seem to not suffer from that as much, although it's really random chance.

This makes some sense, as computer motherboards from around that time are notorious for leaking capacitors and fail at a higher rate.  The story I heard is there were some less than reliable manufacturers that dumped loads of them in the Asian markets around that time.  I’ve had two fail (years ago now) and a couple of monitors from that time frame.  Meanwhile stuff before and after work fine.  

I also have a set of high end speakers made in 2000 or 2001 that have built in subwoofers.  The subwoofer quit working in one and the other started sounding like a motorboat.  Both had bulging power supply capacitors, replaced with some Nichicon ones and they’re as good as new.

Maybe that was the Y2K bug they were talking about?

Last edited by rplst8
@Paul Lytle posted:

... I found that changing the radio pickups seems to help. if you have an old failed board you can remove from old board and install in the board your having issues with. I was able to swap out while board is still in the engine by installing a wedge between both board which held the bottom board up so i could unsolder and remove the frequency filter and replace with others.

I am impressed that you discovered this "failure mode."

The two items you circled are crystals.  And if I understand what you're saying this could be a temperature phenomenon since the frequency of a crystal oscillator drifts with temperature in a somewhat defined/predictable manner.  We're talking in the parts-per-million category but it's a temperature dependency nonetheless.  That's why it might work for a while until the circuit board warms up and the frequency of the crystal drifts enough to cause a problem.  In fact, if you run across another board set that behaves as your describe, then take a can of compressed air, or can of component cooler, and give a blast right to the crystal.  If all of a sudden you restore DCS communications, then you know you're on to something.

Add Reply

Post
This forum is sponsored by MTH Electric Trains
OGR Publishing, Inc., 1310 Eastside Centre Ct, Suite 6, Mountain Home, AR 72653
330-757-3020

www.ogaugerr.com
×
×
×
×
Link copied to your clipboard.
×
×