• Please review our updated Terms and Rules here

DOA PETs. (2001-8, 4032 and 8032)

Eudimorphodon

Veteran Member
Joined
May 9, 2011
Messages
7,075
Location
Upper Triassic
Hi, all. It's my first post on the forum... I'll try not to be completely useless here.

This weekend I netted three former educational PETs from a warehouse dump near Berkeley, CA. (I'm fairly confident several members of this board were also there.) I selected three different units, primarily for their good looks. Unfortunately it appears I'm excellent at picking lemons. *sigh*

The first unit is a pretty good looking 2001 with the "calculator keypad". (Internally it seems like it's sort of an oddball as despite having the calculator keyboard it has an "N" motherboard with dynamic RAM, complete with punched-out holes where Commodore murdered the second bank of RAM socket pads.) The monitor lights up but there's moving garbage on the screen, IE, the screen is about half-filled with the "normal" character hash but there are stray dots running around within the characters. I picked this one up for a friend so I haven't fooled with it any further.

Second unit is a non-CRTC 9" 4032. It cleaned up very nicely but it's also DOA, this time with non-moving garbage on the screen. This is the one I had the most hope for and done the most with, but as of yet I've had no luck changing its behavior at all. So far I've tried some minor spot-cleaning of the circuit board (not a complete wash, but I tried to get some of the oxidation off the edge connectors and wiped some "gooey" looking dust blobs off of places where I thought it's possible they might be shorting something), and I've also pulled and reseated all socketed chips. It seems from reading the forums that magically works a fair amount of the time. No love here.

Last unit is an 8032, I don't know enough about them to say which motherboard. When powered on this unit makes the "Chripchirpchirp" noise but the screen never lights up. I've tried blind-typing "print chr$(7)" in an attempt to get another chirp with no luck, and another experiment (see #3 below) has me pretty convinced it's not getting to the BASIC prompt.

Anyway, I'd *really* like to get either the 4032 or 8032 working, preferably the 4032. Unfortunately my practical electronics experience is weak and the only tool I have on hand is an auto-ranging DVM, so... here follows the stupid newbie questions.

1: My first inclination is to concentrate on the 4032. Is there a good link out there describing where good test points are for doing basic tests like verifying voltages are on the board?

2: I've been kicking myself for not getting two identical units in order to have greater leeway for swapping parts. That said, given what I had at hand would there be any suggested recipe for swapping socketed chips between units in hopes that something that's bad one place is good another? Is any one chip more likely than any other to go bad? I'm guessing that the 6502, 6520, and 6522s should be freely swappable from one to the other, but ROMs are a bigger problem. Would it be worth trying the 2001's ROMs in the 4032, or should I try the other chips first? (And I don't think I want to swap anything without at least verifying voltages.)

3: Are monitors compatible between 9" and 12" PETs? After hearing the 8032 chirp so cheerfully but have the screen remain dark I stupidly risked putting the unit close enough to one of the 9" units for me to plug the video cable from the 9" into the 8032's motherboard. (Of course both units were powered on to supply the high-voltage lines) When I did that all I saw on the 9" screen was a dim pattern of free-scanning lines. Then I tried reversing it and plugging the 12" screen into the 9" PET's motherboard to see if I'd at least see garbage on it. But it stayed pitch black. When I restored everything back to normal nothing seemed worse for the wear. (Easy enough condition to be in when you're already broken, I guess.)

My conclusion based on those experiments was that the 8032 motherboard isn't outputting a video signal *and* its monitor is dead, but of course those conclusions could be wrong if the monitors are not compatible. Anyone swapped screens between a 9" and 12" before?

4: Does the fact that the 8032 go "chirpchirpchirp" at least suggest the CPU is good?

Anyway, it's a bit frustrating to have struck out so badly. I ended up with three sexy disk drive units along with the PETs, I'd love to be banging my head on trying to get one of *them* to work. Any pointers for getting further on fixing one of these things would be awesome.

Alternatively if someone would be interested in swapping me a working PET 4032 for a sick 8032 and an untested 2040 or 8050 (your choice!) I'd totally be willing to talk trades. Thanks.
 
Hi, all. It's my first post on the forum... I'll try not to be completely useless here.
Welcome! Looks like this PET forum is going to pick up a bit as you folks get home with your treasures from that event ;-) Hope we can help.
... Unfortunately it appears I'm excellent at picking lemons. *sigh*
I wouldn't feel too bad; AFAIK these were returns from several local school boards, so it's likely that most of them have problems anyway.
The first unit is a pretty good looking 2001 with the "calculator keypad". ... The monitor lights up but there's moving garbage on the screen, IE, the screen is about half-filled with the "normal" character hash but there are stray dots running around within the characters.
Try the usual first thing, reseating and cleaning chips & connectors, or did you do that already in this unit?
...here follows the stupid newbie questions.
No such thing.

1: My first inclination is to concentrate on the 4032. Is there a good link out there describing where good test points are for doing basic tests like verifying voltages are on the board?
Not really AFAIK; just get the schematics and probe the regulators and some of the chips' +5 and -5V pins.

2: I've been kicking myself for not getting two identical units in order to have greater leeway for swapping parts.
Aren't the 2001-N and the 4032 essentially the same board? Any pictures?
That said, given what I had at hand would there be any suggested recipe for swapping socketed chips between units in hopes that something that's bad one place is good another? Is any one chip more likely than any other to go bad?
Usually ROMs, RAMs, I/O, glue TTL, 6502CPU in that order
Would it be worth trying the 2001's ROMs in the 4032, or should I try the other chips first? (And I don't think I want to swap anything without at least verifying voltages.)
Assuming they are the same boards, the ROMs should be compatible as a set but not individually since they're probably different versions of BASIC.
3: Are monitors compatible between 9" and 12" PETs? ... My conclusion based on those experiments was that the 8032 motherboard isn't outputting a video signal *and* its monitor is dead, but of course those conclusions could be wrong if the monitors are not compatible. Anyone swapped screens between a 9" and 12" before?
I can try it, but AFAIK they are not quite compatible; it actually sounds like the 8032 is almost working but the display is dead.
4: Does the fact that the 8032 go "chirpchirpchirp" at least suggest the CPU is good?
Yes. Assuming you didn't mistype "?chr$(7)" with a blank screen, it sounds like it locks up somewhere after the initialization and before the BASIC prompt; if they're socketed you might try removing the 6520s and 6522 one at a time and see if the beep works then.
Anyway, it's a bit frustrating to have struck out so badly.
Like I said, I doubt that many of the others would have been working either, although you could have been lucky like our other 'client' with the broken fuse holder ;-)

As a start I'd confirm the board numbers and the ROM numbers for reference, and download the relevant schematics here:
http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/index.html

Good luck!
 
Last edited:
Welcome to the forums :) On the bright side you're right there should be a number of folks who may be in driving distance who might be able to help or swap parts. Most any numbered chips will be compatible between the units (6502s can be freely swapped between different systems). You've probably found your way to a few of the recommended sites via google-fu but here's one start on troubleshooting Pet systems and some Pet schematics depending on how technical you're wanting to dig.

Swapping the socketed chips is just a simple check as over time the pins can have some corrosion and end up not getting the voltage or connections they need to work. It's always a good first test as you read. You can also search the forums as well for others with similar issues.

Unfortunately I'm not aware of compatibility between monitors, I thought they needed different roms from reading about programs which weren't compatible between the newer and older systems but I don't have a bunch of experience with these and never owned a bunch to fix or swap parts. On the bright side there a lots of members here with a great variety of experience and some Pet experts :) Glad you found us though and hope you enjoy your time anyway. It's a great grab for free even non-working!

- John
 
Hi Eudimorphodon!

Sounds like we got our PETs at the same place :) I also ended up with a 2001, a 4032, and an 8032. Of those, only my 4032 is working perfectly. The other two have various problems I won't go into here, but I'm working on fixing them.

As you've already suggested, the first thing to do is of course to check voltage levels. For example, my 2001 had a bad fuse socket, and no mains current was making it to the transformer. Luckily that was an easy fix, but it too was DOA until I figured that out. Start by testing power right at the big power supply transformer itself. According to MikeS, here are the expected voltages between the outputs on the transformer (quoting from him):

Between 4 - 6 should be ~16VAC
Between 4 - 5 should be ~8VAC
Between 5 - 6 should be ~8VAC
Between 7 - 8 should be ~16VAC

If your power is good, my second suggestion is going to sound very, very silly, but... it happened to me, so maybe it's happening to you! This is ridiculous and I'm a little embarrassed even to admit this, but when I turned on my 4032, I got nothing on the screen. Then I had a flash of inspiration and reached around behind the monitor to fiddle with the brightness knob. Yup, it had been turned all the way down. It was working fine, I just couldn't see it until I turned the brightness up :) Maybe, just maybe, that's what's up with yours?

I've got to rush to work, but I'll be around later this evening. Hopefully we'll figure something out!

-Twylo
 
I had a flash of inspiration and reached around behind the monitor to fiddle with the brightness knob. Yup, it had been turned all the way down. It was working fine, I just couldn't see it until I turned the brightness up :) Maybe, just maybe, that's what's up with yours?
Good point; we often overlook the simple and obvious.

Sounds like all you folks who grabbed PETs there should start a club for mutual support ;-)
 
Sounds like all you folks who grabbed PETs there should start a club for mutual support ;-)

You know, that's really not such a bad idea :) I'm sure almost all of us are local to the Bay Area. I'd be thrilled to help anyone with debugging and repairs. I get a kick out of this, it's fun.

I hope that folks will find these threads when they google for "Commodore PET". Maybe we can use the Vintage Computer Forums as a place to organize.
 
I don't know if board convention would favor separate replies for quotes from different people, so... many apologies if "lumping" is the wrong answer.

Hi Eudimorphodon!

Sounds like we got our PETs at the same place :) I also ended up with a 2001, a 4032, and an 8032. Of those, only my 4032 is working perfectly. The other two have various problems I won't go into here, but I'm working on fixing them.

Yeah, I saw your post and guessed you were at the same event. (I was the fat guy with the sunglasses.) I'm totally jealous of your working 4032. If you find yourself wanting to swap your almost-working 2001-N-32 for a spare-parts 4032... (Since you have BASIC 4.0 they're pretty much the same machine other than the label. Actually, I'd swap you both the 4032 and 8032 for a working 40 col. PET.) ;^)

(re the 2001) Try the usual first thing, reseating and cleaning chips & connectors, or did you do that already in this unit?

Haven't done much with the 2001 since I picked it up for a friend that wanted a calculator-pad PET or nothing. I'm not actually sure they're going to want it when they find out it has the non-original motherboard, but until I know I'd rather not dip into it for parts. One thing about it, though... I think the magic smoke may of escaped from part of the power supply section. It does get the garbage on the screen but it's "twitchy", and something near the front corner of the motherboard didn't smell good.

Not really AFAIK; just get the schematics and probe the regulators and some of the chips' +5 and -5V pins.

Yeah, I was just sort of wondering if anyone had a good drawing of the "Dynamic PET" board showing where the major subsections are. (Me=Lazy) I'll track down the regulators by number and probe them. Any particular place on the board better than another to use as a ground?

Aren't the 2001-N and the 4032 essentially the same board? Any pictures?

They're pretty much the same, but I could take some shots. (Don't have them in front of me now, however.) I'd swear that the monitor connector is in a slightly different place on the 2001 but I could be on crack.

Usually ROMs, RAMs, I/O, glue TTL, 6502CPU in that order

So the 6502 is probably the least likely candidate? Feh. That's the one chip I have that I'm fairly certain works. ;^)

Assuming they are the same boards, the ROMs should be compatible as a set but not individually since they're probably different versions of BASIC.

I'll see if I can borrow the 2001's set. Given the possible smoke escape though I worry about whether they're any good.

I can try it, but AFAIK they are not quite compatible; it actually sounds like the 8032 is almost working but the display is dead.
Yes. Assuming you didn't mistype "?chr$(7)" with a blank screen, it sounds like it locks up somewhere after the initialization and before the BASIC prompt; if they're socketed you might try removing the 6520s and 6522 one at a time and see if the beep works then.

Yeah. My vague understanding is that the CRTC PETs have to get far enough to initialize the CRT controller before they'll even display garbage. You don't happen to know if that's before or after the "Chirpchirpchirp", do you?

Good point; we often overlook the simple and obvious.

Sounds like all you folks who grabbed PETs there should start a club for mutual support ;-)

*snicker* Let's just hope we don't have to base the meeting agendas on the "7 Stages of Grief".

I do wish the problems with the 8032's monitor were as simple as a brightness knob but I'm pretty sure it's more than that. I did try spinning that knob on the back (which was *very* stiff, and I wasn't sure what it was for) with no dice. Also, my ears are still pretty good at high frequencies so I'm pretty sure I would of been able to hear a whine if it had come on. Next time I look at it I'll try to peek through the hole in the bottom and see if the tube heater is lighting up. Alas the monitors on both the 9" PETs are very pretty and bright and do a wonderful job displaying garbage. It would be interesting to know definitively if swapping one onto the 8032 would be a possibility. I'm sure it'd also be blasphemy, but... I want to play with these things, darn it!

(Maybe if we did form a Bay Area PET club I could get a fellow member to try my 8032's board in their case with a working monitor.)

Anyway, thanks for the pointers. It looks like my next step is try to visually identify the outputs of the power regulators and see if I'm getting clean outputs in all the proper places. After that, well... hrm. It's tempting go willy-nilly swapping the socketed chips but the 2001 is in at least as sorry a state as the 4032 so it might be an exercise in futility. It would be awesome to have a known-good set of ROMs as a baseline.
 
Last edited:
Hi Eudimorphodon,

Don't despair! Just about everything that could go wrong with a PET is fixable, if you can figure out where the problem lies.

Where are you located? If you want to get together and trouble shoot, let me know. I can bring my tools. I don't have a lot, but I've got a 60MHz analog oscilloscope, DMM, USB logic analyzer, and an EPROM burner.

Note: I can't burn the correct 6540 sized EPROMs, but it's pretty easy to make an adapter so you can use easy-to-find 2716 EPROMs in place of the 6540 ROMs.
 
Hi Eudimorphodon,

Don't despair! Just about everything that could go wrong with a PET is fixable, if you can figure out where the problem lies.

But finding the problem's the hard part. ;^b

(It's a shame "garbage on screen" is such a generic issue.)

Where are you located? If you want to get together and trouble shoot, let me know. I can bring my tools. I don't have a lot, but I've got a 60MHz analog oscilloscope, DMM, USB logic analyzer, and an EPROM burner.

Note: I can't burn the correct 6540 sized EPROMs, but it's pretty easy to make an adapter so you can use easy-to-find 2716 EPROMs in place of the 6540 ROMs.

Thanks. I'm in the south bay near Cupertino. If you're not a huge distance away I could probably use all the help I can get.

The friend I got the 2001 for was pleased as punch with it so I guess it's off the table for a parts donor. (Turned out he wanted it solely for looks, not necessarily for having perfectly original innards.) He really knows his stuff when it comes to these things (he's actually built and sold an Apple II product through ReactiveMicro.com) so he'll undoubtedly be able to fix his and is talking about turning out some ROM adapter boards boards if we need them. But it'll probably be a while before he has time to get to any of it. It'd be nice to at least rule out/explore some other issues while waiting for that.

I did get the suggestion that cold solder joints under the flyback transformer could be why the 8032's monitor doesn't light up. I'll have to see about making a discharge tool so I don't execute myself trying to reflow them.
 
So just as a quicky update...

I checked voltages in the 4032 tonight. I managed to find the -5v regulator up next to the RAM banks but I'm embarrassed to say I couldn't find the +12 or +5 regulators. Maybe I'm just bleary after a long day. (I'm also having some trouble reading the labels on some of the IC/packages.) Anyway, measuring the legs on one of the RAM chips seemed to indicate that power is probably "reasonably okay". (I got about 11.85v on the 12v line, -5.1 on -v, about 5.05 on +5.) I also checked and found 5v on pin 8 of the 6502. So it looks like power isn't the culprit here.

I might hit Anchor Electronics tomorrow. They claim on their to have 6502s, 6520s, and 4116s in stock. I can try swapping the CPU and a PIA and see if I get lucky without risking making the 8032 worse, and I was also thinking I could try that "press-piggyback a RAM chip" trick.

I've read in a couple places that shorting pin 5 of the user port to ground should bring the PET up to a monitor instead of booting to BASIC. Has anyone tried it? (Is pin 0 of the user port a suitable ground?) It feels like a long shot to me but perhaps if it's just the BASIC ROMs that are bad...

Speaking of the user port, anyone know a good household solvent to clean edge connectors? I've tried the eraser trick but the oxidation on this PET scoffs at such things.
 
seriously, it is never lupus. Oops I mean its never the cpu. I have never ever ever seen a 6502 go out on Apples, Pets, etc etc. I've dealt with many 100's. Its never lupus.
 
A PET Meetup sounds like fun, as I am sure almost everyone of the PETs taken will probably need some work.
 
I agree with leaknoil. Although I haven't dealt with 100's of machines, out of those I have troubleshooted, none turned out to have a faulty 6502. If it did, probably the majority of the board was fried as well.
 
seriously, it is never lupus. Oops I mean its never the cpu. I have never ever ever seen a 6502 go out on Apples, Pets, etc etc. I've dealt with many 100's. Its never lupus.

*snicker*

Ironically the one 8 bit computer I ever had spontaniously die on me, a TRS-80 Model 4P, did so because of the Z-80 going "poof". (I repaired that one with a Z-80 out of a Timex Sinclair ZX-81. Which I'm sure has its fans but which I considered a perfectly good Z-80 trapped inside in a cheap doorstop.)

I'm still going to see if I can get a replacement for each of the major chips, just to have them. Seems like the hardest part to find off the shelf is the 6522. My local HAM-advisor has suggested to me that if I get to the point of having to replace an entire bank of DRAM that it might be preferable to sub 4164s for the 4116s... (It can be done in a hamfisted manner by bending up a couple pins and jumpering one back into the adjacent socket pin. Apparently 4164s draw only about 1/3rd the power of a 4116 and only need the single 5v supply. He's already mulling plans to turn his 8k 2001 into a 32k machine with a 4108->4164 swap plus a bit of logic.) but I'll buy a few 4116s for troubleshooting anyway.

Unfortunately that still leaves me with un-replaceable ROMs so.. I've been thinking about it and I imagine the PET startup sequence is well documented enough that *in theory* perhaps what I could do is make a single modern-EPROM adapter and write a dinky little program to write into it that would just write "HELLO WORLD!" over and over to the video RAM in a tight loop when the machine was powered up. (I'm pretty but not *entirely* useless when it comes to assembly. Granted most of my experience was with Z-80 assembly, and that was a long time ago.) That would at least tell me if the machine is alive enough to make it worth building a complete set of ROM replacements. Of course, finding a fellow PET-grab-ee and being able to compare/borrow ROMS with a working unit would be a more direct approach.

If none of that works then I think I'm at the point of either giving up or trying to beg/borrow a scope and learning how to use it. Short of replacing every 74xxx chip on the board one-by-one it's unlikely I'll be able to make much progress on the glue logic with what I have available.
 
I agree with leaknoil. Although I haven't dealt with 100's of machines, out of those I have troubleshooted, none turned out to have a faulty 6502. If it did, probably the majority of the board was fried as well.

I was trying to be funny and it probably isn't even funny unless you've watched too much of the tv show House. 100's may be a bit of an overstatement. There have been many though and it never turned out to be the cpu. In fact, it is almost never any of the socketed IC for some blasted reason. Unless its just bad contacts of course.

I've only tested two of the pets I got there and so far one worked and one didn't. Another I know isnt going to work because it actually has no motherboard. Its looks really pretty though. For some reason I decided to grab it for the case. I was at the end of the list and all the pretty ones were gone fast. I only really wanted 8032's and pickings were slim. One I got looks like it spent time under water. I couldn't find anymore 8032 and so grabbed a couple 4032 with the large monitor. I tried to only grab ones the case was still screwed closed. Most were open. That didn't seem like a good sign.
 
I only really wanted 8032's and pickings were slim. One I got looks like it spent time under water. I couldn't find anymore 8032 and so grabbed a couple 4032 with the large monitor.

If you end up with multiple working (or even "mostly working") 4032s I'll happily trade you the 8032 I snagged for one. It's not super-pretty but it's not terrible. It's the later lower-profile model.
 
...I'm still going to see if I can get a replacement for each of the major chips, just to have them. Seems like the hardest part to find off the shelf is the 6522.
Looks like this place has 'em, even on sale! (And I believe he's on this forum as well). No 6520s listed but they're equivalent to a 6821 which he does list:
http://www.arcadecomponents.com/cpusupport.html
Unfortunately that still leaves me with un-replaceable ROMs so.. I've been thinking about it and I imagine the PET startup sequence is well documented enough that *in theory* perhaps what I could do is make a single modern-EPROM adapter and write a dinky little program to write into it that would just write "HELLO WORLD!" over and over to the video RAM in a tight loop when the machine was powered up.
A more or less standard 2716 will work in those ROM sockets without any adapter.
Of course, finding a fellow PET-grab-ee and being able to compare/borrow ROMS with a working unit would be a more direct approach.
Like I said, it doesn't have to be the same model, or even have socketed ROMs; you can put your ROMs in one of the spare sockets and read/dump/checksum them. An EPROM programmer or even another system that uses 2532-compatible ROMs would do as well, of course.
Those aren't really the classic bad-ROM symptoms though.

How about some pictures of the motherboards and the numbers on those ROMs, so we have a better idea of what we're talking about?
 
Back
Top