Schematic for QSI DCRU

Pete, there are 3 5087 transistors.  2 of those 3 are for the relays(one each). Same as fixing a bottom board PS-1 that is stuck in one direction.

If that is not it, I have found the MC may need to be replaced.  More involved trouble shooting to figure which one.  G

Pete, Can you post a photo of your board(s)?  I may have one I'll send you.

I've got boards out of a Williams/Samhongsa 2-8-2 and 4-6-2 and it seems like a couple of other boards (maybe one from a Weaver 4-6-0).  I've replaced them with PS2 guts so they're just sitting in a box.  If you don't need them maybe you George?  I'll never use them.

G, I figured the transistors might be what is driving the relays but all three read OK using a multimeter in diode mode. Things are pretty well packed in there and I am hoping I won't have to pull components without being fairly sure they have failed.

Bob, I can take a pic but won't be able to post it until tomorrow. I appreciate the offer.

BTW I have a few Williams E units I could use but would have fabricate some circuitry to maintain the directional LED lighting. Not a big deal but something I am trying to avoid.

Pete

I imagine you don't have the signal coming from the logic chip than.  I have seen that failure.  G

G, I'm in the process of tracing the nand gate. Like you I figure it is either a transistor or that gate. Both halves of the flip flop toggle with each power down and the relays both work if I power each one directly from the -6V regulator.

Bob, Here is a pic of the board.

Pete

Pete, I'll take a look as soon as I eat some lunch, I just got home from (trying) to play golf.  Good thing it's free

Pete,

Here's what I have, seems like no 2 are alike:

The first board has a 2-prong plug just right/bottom of the NEC components and the upper left pin slots have 1/2 the "slots" (just above and left of the 220uf cap) as the other board:

The 2nd board has a 2-prong jumper plug left of where the connector on the above board is located:

If you want them both, send me an email with your address.

G where can i get a Replacement DCRU board for a QSI MTH Locomotive

Pete,  Instead of jumpers, you seem to have wires added to the motor control ports.  Did you add them for testing or is there a switch used controlling direction?

If switch I would make sure it works.

If you pull out those wires and test position 1 (far left as I look at you board against PCB grnd (position 5 you should get 5V that cycles as the board goes through F, N, R, N F.

Do the same for position 3.  If you don't have 5V or you don't get it to cycle one of the 2 micro controllers is bad.  If they do cycle properly and you can get the transistor to operate when you jumper at the leg, I would say a trace has gone bad between the transistor and the MC.  G 

Putting a picture to G's words, the 2 NAND gate outputs (pins 3 and 4 of the 4093 chip) go to the header.  They jumper over one slot to "B1" and "B2" as etched on the board.  B1 and B2 go to the Base of the 2N5087 transistors labeled 1 and 2.  The transistor outputs drive the relays labeled 1 and 2.  Your board has 4-wires in place of the header jumpers - apparently to manually set the relays?

Stan, George, the 4 wires you see on the bottom connector go to a DPDT switch marked direction. The two wires in the upper connector go to second switch (SPST) which acts as a lockout when open. It is in the closed position mimicing the jumper. Same goes for 4 wire switch. They are shorting the same pair of pins that the jumpers do. I will check the pins you mention to see if the 5v toggles. I appreciate the info on where the signal that drives the transistor comes from. Didn't get to work on today, hopefully will have it resolved tomorrow.

Pete

Pete, Do they short them together?  If so, not sure why it is installed.  Usually a switch is installed when the logic circuit fails or a pure manual direction control is desired.  In that case no need for the lockout switch.  Typically you just bring in the logic and ground and use the switch to alternate the 2 on the B1 and B2.

I would remove wires and add jumper and test the board that way.  I think your switch maybe preventing power to one of the Base inputs.  G

I have seen that before.  Replace the MC, the voltage is to low to activate the 5087 transistor.  G

Found the problem! It was a broken trace located under the big electrolytic and a diode near the center of the board next to the relays. 

Thanks guys. I now know a lot more about these boards than I did before.

Pete

Good catch Pete!  I doubt my "peepers" would have caught it.

Was it burnt or just broken?  A buddy of mine came over yesterday with some PC RAM and a better monitor than the one I had (he was cleaning house and was going to throw them away).  Pushing the memory sticks onto the motherboard (with it inside the case) isn't the best way of doing it, I can just hear those traces cracking off the MB.

I hope you can still use those parts, you should get them tomorrow.

Bob, I had to lift the cap and diode to be able to follow the traces. It was broken next to one of the plated through holes used to transfer the circuit from one side of the board to the other. By comparing the path from the working side to the non working side I knew there should be a connection there. A temporary jumper proved that was the case. 

Now that I have a better idea how these boards work I see they offer some advantages over my Williams E units in that they have circuitry built in for directional lighting so I will find a place for them someday.  

Pete

Check the zener diode near that trace.  That is usually what can cause the burned trace, and loss of logic voltage.  G

I repair traces with #30 wire-wrap wire.  It's pointless to fool around with tapes or other nonsense repairing traces.  For heavy traces, I just use heavier wire.

I suspect George is on to something, the trace doesn't just die for no reason, something happened.

John,

I was thinking more of a stress fracture when pushing on the card than an electronics issue, but as thin as those traces are any sudden "flash" could burn thru it.

Well, he said it was a trace and a diode.  I wonder if the diode croaked and took the trace with it?