Bringing a vintage IBM PCJr back to life

[Stonent]

A friend of mine offered up his family's old IBM PCJR (complete with second floppy, monitor, BASIC cartridge and Keyboard) for free on Facebook and I told him I'd take it.

So far, it doesn't seem to work.  I stripped it down to just the motherboard, power supply and monitor and here's what happens when powered on.

The monitor lights up the CRT slightly grey and a roughly 200HZ square wave audio tone comes out of the monitor and nothing more.

The computer is capable of driving a composite NTSC display so I tried that and got no video.  The audio tone from the audio out was still there.

Unfortunately this one has no socketed chips.

So it doesn't matter if I connect everything to it, or just run bare motherboard with power, the results are the same.

I've checked the external power brick and get the required 18V AC.  I have a second power brick and get about the same reading.

The DC power supply board seems to work fine as well.  I get all the required voltages on all the outputs and there doesn't seem to be any problems with load, burning parts, bad caps or anything like that.

Some random pokes around hitting the opposite corners of the chips usually gets me 5V readings.

Here's a high res picture of the board I found online:

http://bunniestudios.com/blog/images/ntw_nov_2012.jpg

The only thing I see that looks remotely wrong on the board is on my board, this little light blue part has a slight green tint on the bottom lead and the top lead looks darker.  The blue part itself doesn't seem damaged or burned (this is not the picture from my board).
Do you think that's a resistor or another diode.



One more thing. At one point I had left the board on for about 15 minutes continuously and suddenly every so often, the cassette relay on the board would click a few times randomly it seemed.

I'd really like to get this guy going again.  If you didn't already know, the PC Jr itself wasn't a commercial success only lasting a little over a year before IBM discontinued it. It had no ISA slots so you could not upgrade it using standard parts.  Tandy pretty much cloned the PC Jr when they created the Tandy 1000, but addressed the issue of no ISA slots and the proprietary peripheral connections on the back (all .1 inch header connectors).

Where would you proceed next? I have not found a service manual with schematics unfortunately.  There is a "technical reference" manual here : https://archive.org/details/IbmPcjrTechnicalReference (or here) http://www.retroarchive.org/dos/docs/ibmpcjrtechref.pdf but it doesn't seem to address the kinds of issues I'm seeing or provide any board level troubleshooting. However it does have some block diagrams.  Here is a general IBM PC repair manual http://ibm-pc.org/manuals/ibm/5150/Pcsim1-9.pdf

[JoeO]

Back in the day, I was able to fix many PCJr systems by scoping to the suspected bad TTL chip and then placing a duplicate chip over the suspect one.  Nine out of ten times I was able to locate the defective chip this way.

Reseat the socketed chips, then if the voltages are good, start looking around with a scope on the pins of the 8088.

[JoeO]

Today is your lucky day.  Section B of the PCJr TR has the schematic of the boards.  It is not included in the PCJr Techref on archive but it is in the TechRef I have.
I will scan them in on Sunday 6/28 and email them to you.

[Rasz]

this little light blue part has a slight green tint on the bottom lead and the top lead looks darker.  The blue part itself doesn't seem damaged or burned (this is not the picture from my board).
Do you think that's a resistor or another diode.

diode, might be corrosion from leaked electrolytic cap next to it?

[Shock]

Do you think that's a resistor or another diode.

 

[amyk]

Today is your lucky day.  Section B of the PCJr TR has the schematic of the boards.  It is not included in the PCJr Techref on archive but it is in the TechRef I have.
I will scan them in on Sunday 6/28 and email them to you.

You should post them on the forum or to one of the archive sites.

[poot36]

In something that old I would check all of the input filtering caps and check to see if any of the logic chips data or address lines are being held at 0 or 5 volts.  Also the component that I think you are talking about looks to be a zener diode so you can't test it with most multimeters (you can check if it is shorted or leaky at the lower voltage that your diode test function on your meter provides).

[JoeO]

Today is your lucky day.  Section B of the PCJr TR has the schematic of the boards.  It is not included in the PCJr Techref on archive but it is in the TechRef I have.
I will scan them in on Sunday 6/28 and email them to you.

You should post them on the forum or to one of the archive sites.

I do not know how to do that.  However if you do, I would gladly email them to you to post them to an archive or archives.

[Stonent]

Here you go.

https://drive.google.com/folderview?id=0B3nYGMOtVmrVfkoyU0VNVnBqd0doVkUwWWhTQ0hlUS1zZkxRaXRUMDlxZzIwX3FEMFVDcDg&usp=sharing

One PDF has OCR enabled, one does not.

[Stonent]

this little light blue part has a slight green tint on the bottom lead and the top lead looks darker.  The blue part itself doesn't seem damaged or burned (this is not the picture from my board).
Do you think that's a resistor or another diode.

diode, might be corrosion from leaked electrolytic cap next to it?

The cap there is in good condition as far as I can tell.  Nothing leaked or any sign of leakage.

So, if a diode, what kind?  407 on a resistor would mean 400M Ohm which seems a bit much.

[flynwill]

The diodes are in the cassette audio input circuit, CR3, CR4 Sheet 10 of the system board schematics, top of the page.  Extremely unlikely that they are causing your problems.  I can't find the part numbers in the available documents, but CR3 is the tiny glass one, probably 1N914, 1N4148 or equivalent.  The blue diode is CR4 and is indicated as a Zener, I would guess in the 3-4V range based on the circuit.  "407" is only part of the part number, it will be continued on the line above or below on the part.

I think the constant tone on the audio interface means that the initial memory or other diagnostic test failed. 

Good luck!

[Rasz]

interesting

[divelectservices]

I worked at a computer repair shop in the late 80's and we used to get many of these in from the area schools for service.
Most common faults were floppy drive issues and RAM memory issues.
I remember the original PCjr floppy drives were quite expensive so I figured out a way to modify a cheaper, generic drive to fit in the case.

Fire the unit up and feel all the chips...chances are you will find a RAM chip that is much warmer than the others (some use to get very hot!)

[Kjelt]

Nothing to contribute just wanted to share that I find it terrific to read these stories from the old days from real life technicians 
"Just put another 74 ttl on top of it", "feel which chip gets hot"  priceless tips.

[Stonent]

No actual update on my system, but I did learn some information from a video about repairing a Tandy 1000 EX.  The sound chip in the Tandy 1000 in an uninitialized state is not muted, so there will be sound coming out of it until the CPU initialized it and tells it to shut up.  In that particular video he determined the VLSI branded programmable clock generator had failed so the CPU was not doing anything.  He replaced it and it fired right up.  Now this exact chip is not used in the PC Jr. I seem to recall the PC Jr uses an Intel part for this.  I was certain that I was getting a CPU clock when I scoped it the last time I worked on it, but I will try it again the next time I clean up some a workspace to play with it again because the PCJr and Tandy shared sound chips (or were very similar)

[nightlynx]

I have what sounds to be the exact same issue (maybe I purchased the one in question?)  Did this ever get a resolution?

Thanks