Mike's PCjr Page

Howdy all!

I'm excited to be joining the forum and I hope, in time, I will be able to add to it, but for now I am a man with questions in need of answers.

I have a PCjr that I acquired through the normal ways these days, eBay, and the listing was the standard "I have no clue" type auction, but the price was right and I fancy myself a tinkerer so I got it. As luck would have it, there was another auction with a system board only that was about to end with no bids, no reserve, and only $10 with free shipping. SO, long story short I now have a PCjr with a bad system board and a seemingly functional main board. I could simply throw out the old one and use the new one, but where's the fun in that? Besides these things are only getting rarer and rarer and it seems a shame to pitch a board that probably just has a bad IC or cap somewhere, so armed with my trusty multi-meter and scope I went probing. I'm using AkBKukU'S ATX to PCjr power adapter so I've discarded (aka put in the closet) the only power card and brick (which also didn't work). With that card in, and it powered up.... nothing happened, and I do mean nothing. No signs of life on the monitor. No beeps or signs of life from the computer beyond the floppy fan. So I started removing stuff. Floppy controller, 128Kb sidecar, and the 64k RAM expander all yielding the same nadda. Before I forget, I then plugged everything into the good board and it booted right up, no issues.

Next, one at a time, I swapped the socketed chips from the bad board with their counterparts from the good board. In total this included the CPU, the two ROM chips by the cartridge port, and the 8425E. When switched, all booted up normally. Hoping, but not thinking, that a solder join might be an issue, I went ahead and sockted the 8259A and the DMA 8225A. Shockingly that did nothing either, other than they are now socketed which makes me happy.

Now I am by no means an expert at electronics, but I do know enough to be dangerous, so I set about testing the voltage rails to make sure they seemed ok. Using the schematics in the technical manual I set about confirming that all the main chips get their +5VDC, which they do. I confirmed the votage regulator was working. Checking the resistance levels on the +5VDC, +12VDC, and -12VDC lines yielded a resistance level that led me to think that the caps on the board hadn't shorted out (but I guess that's still a possibility). Then using my rudimentary ability to read a system block diagram, I set about checking the chips in the startup routine. I've confirmed that the crystal generates the required 14.3Mhz signal. I've confirmed that the 8284 is correctly outputting the correct clock signals at the correct frequency. I've confirmed the correct clock signal is reaching the CPU on Pin 19, as well as the other chips that get that signal. I've confirmed that the 74LS74 (just below the CPU's pin 40), the 74LS34 (to the right of pin 40), and the 74LS08 (just above the joystick ports) are working in a similar manner to their counterparts on the working board. I mention these chips because based on my understanding of the system block diagram I believe these chips to be a part of the READY signal circuit.

I had read that some of the 5150's have an issue where if the RAM chip in Slot 0 is bad, the computer won't POST either. So, I'm in the process of removing the chips and socketing them and will try to swap out a known good chip with each chip on the board, one by one, to see if that's the issue. However, I honestly don't think the issue is with the RAM, I think there is something faulty in the processor initialization circuity but not knowing enough about the order in which the logic chips are initilized, I'm just guessing and not getting very far.

If anyone has any suggestions, things to try, needs some more details on something, anything please let me know. I'd like to try and save the main board if at all possible and put the good one in the closet for a spare in case something happens down the line that can't be fixed by off the shelf parts, like the CGA chip dies or something.

Thanks,
Travis

Glad to see another Jr member here, and not to worry, we ALL have issues now and then, as you said , its part of the fun..

If you take the MB back out at some time it would interesting to see some good pics of the MB, in sections close up...

Look for swollen caps, and also it it has been exposed to the elements, the possibility of a trace or two that have corroded and gone away, rare but possible
Also look at the 8284 chip , you mentioned you and the 14Mhz so it is probably fine,,,

Do you have a Jr Monitor plugged in or composite, And you get NO BEEPS, so, walk aroung the board checking chips for 5V on their powe pins..
it is possible the original Power board went bad and wiped out a bunch of chips as it failed, look for scorch marks and inspect the tops the chips
to see if there are any signs of them 'releasing the magic smoke'

This sounds like fun adventure..

chuck

Chuck,

Thank you so much for your reply! I still have to complete my 5V check on every chip, but so far every one I've checked has a proper ground and 5V line. However with that said, I've made some progress based on your input. Normally I'm not one to replace caps made from the 80's unless they're, as you said, bulging or leaking or in some other way showing signs of failing. However, I did notice that one capacitor was different than the rest. Based on that, I figured that someone replaced one of the caps at some point and, for some reason, didn't just replace them all. So I cursed the previous owner under my breath I replaced every electrolytic cap on the board.

Based on the assumption that someone had fiddled with the board, I starting looking for differences between the non-working and the working board (using that as my control board so to speak). Apon closer inspection, I noticed that whereas my working board has 8 Mitsubishi 5K4164ANP-15 DRAM chips, my non-working board had 4 Mitsubishi, 2 Motorola, and 2 Mostek RAM chips. Unlike Commodore, I've never known IBM to just go with whatever they had on the shelf so I find it unlikely that this was the original chip configuration, but I suppose it could be. However, I'm considering this more proof that someone tried to repair this board at some point and it either worked for a long time or didn't and they sold it on Ebay to me.

Regardless, since RAM is needed to help execute any startup initialization process, I decided to replace the hodgepodge of RAM chips with 8 working TI 4164-15s that I had. After doing all of that this evening, now when I try to power on the computer, instead of getting no beep and no sign of life, the monitor lights up as if it has a signal and I've progressed from the dreaded no beeps, to the dreaded 2 beeps! Which I take as good progress forward.

You asked about my video connection. I use the PCjr CGA & Serial breakout board from TexElect (https://texelec.com/product/ibm-pcjr-cg ... out-board/). They also have a board that converts the joystick and keyboard ports to a standard game port and either the RJ style PCjr keyboard connector, or allows for a PS/2 keyboard.

Finally, I searched for any signs that anything had released any magic smoke, and it all looks good. As you requested, I've taken some pictures of the top side of the board for your to look at. I'm not actually sure if I've properly shared them so take a look and let me know (https://photos.app.goo.gl/U3sJXTHRom3MYY7L9).

So with all this said, and suggestions on the two beeps?

Good job, getting there... I was surprised how clean the board is, very nice...
I am sure you are doing this, but make sure you have the plug in memory board when you boot up,
and I would test witha Basic Cart, just so that is has a boot path and not worry about a diskette..
this board is definetly worth getting going,,, it is in good shape, I was expecting a garage stored unit