SoylentNews is people
SoylentNews
http://www.righto.com/2025/04/commodore-pet-repair.html
In 1977, Commodore released the PET computer, a quirky home computer that combined the processor, a tiny keyboard, a cassette drive for storage, and a trapezoidal screen in a metal unit. The Commodore PET, the Apple II, and Radio Shack's TRS-80 started the home computer market with ready-to-run computers, systems that were called in retrospect the 1977 Trinity. I did much of my early programming on the PET, so when someone offered me a non-working PET a few years ago, I took it for nostalgic reasons.
You'd think that a home computer would be easy to repair, but it turned out to be a challenge. The chips in early PETs are notorious for failures and, sure enough, we found multiple bad chips. Moreover, these RAM and ROM chips were special designs that are mostly unobtainable now. In this post, I'll summarize how we repaired the system, in case it helps anyone else.
(Score: 4, Informative) by VLM on Monday April 14, @05:03PM (3 children)
1) Reads like an Adrians Digital Basement video watches
2) Brave to smack the power as the first step to test the power supply; you can assume any RIFA brand caps in the RF filtering network have gone bad, plenty of electrolytics get questionable. He 'won' this time but I'd suggest testing the power supply separately first on any old equipment, computer or ham radio or anything old really.
3) People like to rip on shotgun troubleshooting but a fair first step is just look at every pin on the o-scope looking for weirdness and following up. Its fast and you can get pretty far with that. He saw there was a bus conflict and some chips were fighting making multiple voltage levels on the bus but didn't directly follow up on it and just tested all the chips; probably could have figured it out by reasoning about the o-scope waveforms.
4) He didn't explain why he knew his eprom programmer was to blame as opposed to the eraser; if I remember my days of playing with 27xxx eproms, they erased to bit high, all 0xFF so when he saw a 0x7A instead of a 0x72 he instantly knew it was a programmer failure not an eraser failure. Look a bit 3, it's still 1 and was supposed to be 0. He didn't necessarily need to use a different programmer, forced reprogramming probably would have set (cleared) that bit eventually in another attempt or two.
5) Note that he/they hit the logic analyzer; its an important step before hitting the analyzer to put the scope to each pin (see step 3) because the logic analyzer's voltage limits do not necessarily match the voltage limits of the mystery family chips onboard. They probably checked each pin with an o-scope before slapping the analyzer on. The fault could have just as well been electrical something on that pin being pulled up. Of course then step 112238 would not have had a 0 for bit 3 but whatever, you get what I mean in general. Kind of like OSI networking model the higher level stuff shouldn't be troubleshooted until the lower level stuff actually works.
That said troubleshooting stories are always cool and its hard to argue with success!
(Score: 0) by Anonymous Coward on Tuesday April 15, @03:01AM (2 children)
Informative and insightful. But a few points / augmentations / disagreements:
1) (huh? means something to someone, so that's fine)
2) Absolutely agree. I'd even dummy-load test the PS before connecting it to those fragile old chips, check for anything getting hot, check waveforms, give it some time to run, etc. I didn't know about RIFA caps, so thank you for that.
3) Well, the problem chips were all sharing a bus, so you pretty much have to isolate them to figure out which one is (or ones are) causing the problem (any one or several can load down a signal).
4) I guess you missed it- he said the first pass programming the EPROM was on a programmer that didn't provide enough programming voltage for the particular chip they used.
("foreshadowing" meant that there would be problems later due to the insufficient programming voltage.)
5) Absolutely agree. Makes me wonder: do any logic analyzers auto-sense the input voltage ranges? Or allow you to adjust the thresholds?
Absolutely agree, really really good article. I learned a lot, a lot. Not sure I would have figured it all out. Certainly would have taken me longer, and my lack of knowledge of those older machines might have demotivated me.
(Score: 2) by KritonK on Tuesday April 15, @05:45AM
Adrian's Digital Basement [youtube.com] is a youtube channel where said Adrian restores old computers and peripherals. He's even restored a PET or two. The rest of the comments are comments that Adrian or frequent viewers of his channel would have made.
If you are interested in retro computing, this channel is really worth watching.
(Score: 2) by ChrisMaple on Tuesday April 15, @02:32PM
There's a good chance that reprogramming (without erasing) at the below-spec voltage would shift enough charge to correct the bad bits. The risk is that they'd still be marginal; that they'd be erratic and the first to fail again as time passed.