• Please review our updated Terms and Rules here

Finally, I have it. Commodore PET 2001 with blue frame but its not working

SkyCaptain

Experienced Member
Joined
Apr 18, 2020
Messages
85
Location
Basel-Stadt, Switzerland
Hello good people of the Vintage Computer Forums

I finally have the crown jewel for my personal collection, a PET 2001 with blue frame, complete with chiclet keyboard and tape drive. And only minimal yellowing on the keys.

Its sadly not working... yet, but with all your help i managed to get two PETs working again, and this one seems less far gone than the others, but i dont know.

When i turn on the machine, i get a black screen. When I switch it OFF and ON again, i quickly see the screen filled with random characters but then it goes blank. The symbols are nice and crisp although not fully centred. So far so "good".

Before further diagnostics, i installed the Tynemouth Softwares PET ROM/RAM replacement board, and ran the PET Tester on it. It showed all RAM as "G"ood and the other screen with all the character set looked also ok. Although it was fuzzy, i checked with my other PETs and there, the character screen was also fuzzy. I cant entirely remember but it just looks like that.

I went on to switch it to BASIC 4 and 32k RAM and turned the machine ON again. And again I was presented with a black screen.

also for clarification, because i kinda dread working on and around the CRT screen... this is one of those PETs that has a video controller/board, right? I think i need a whole new set of schematics for this machine... does anyone have a layout map/schematic of the board? i only found one of the video board at zimmers. (I have that for the newer 2001 boards... i kinda need a layout otherwise i cant transfer anything i read on the schematics over to the actual board..)

anyways, pictures will follow when cleaned up!

Thank you all for any insight!
 
Well, if it is a 2001 it could be 1 of 3 video boards. It could also have a slightly later board in it as well (if a repair was carried out).

There could also be a host of different mainboards depending upon what type of ROM and RAM is fitted as well.

You probably need to look at what ROM and RAM is fitted and try and match the mainboard and video board numbers up with the schematics as appropriate.

I suspect Tynemouth do the same with their test board as I do with my PETTESTER EDIT ROM and don't enable interrupts. If this is the case, it could be one of the two PIAs or the VIA that is causing you problems when you try to boot up BASIC. Check for activity on the CPU /IRQ pin (pin 4). It should either be HIGH or PULSING - never permanently LOW.

There is also a "BLANK TV" signal that 'sneaks out' of PIA #1 (6520 CA2 pin 39). Check what the state of this signal is.

It doesn't sound a difficult fix - it would seem that most of the stuff is working if the Tynemouth test is giving it a good bill of health...

Dave
 
A 2001-8 with a chicklet keyboard does NOT have a CRTC video controller.
Since you are seeing random characters for a second and then the screen clears, it seems video section is fine and the computer is booting far enough to clear the screen.
Since you are seeing a normal display with the Tynemouth diags, you can be 100% sure your video section is fine.

It's probably crashing shortly after clearing the screen which would point to bad RAM or bad ROM.
Schematics can be found here http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/2001/index.html
Since you have a blue screen bezel, it should be one of those schematics, unless someone has swapped out the PCB with a later model.

Could you post some photos of the PET including a shot of the main PCB?
 
Last edited:
That’s a thought, does the Tynemouth board give you a checksum for each ROM?

It appears to have tested the RAM.

Is it worth using my PETTESTER ROM (in place of the EDIT ROM - assuming this is in a socket)?

This will test the video RAM, ROM checksums (apart for the EDIT ROM of course) and it will do a pretty exhaustive MARCH-C memory test in the RAM.

Clearing the screen is early on in the start-up process. This means it is getting into the EDIT ROM from a handful of instructions in the kernal ROM.

Normally, any crash from here onwards would be the DRAM, as the CPU tries to do a return from subroutine from the EDIT ROM back to the kernal ROM. However, if the RAM is good, this shouldn’t be a problem.

Interestingly, this is the second PET I am involved with that appears to do a similar thing. The other machine is an 8032. It (sort of) works with the test ROMs, the DRAM tests OK, the ROM checksums are OK - but the screen is black when running with BASIC 4 ROMs. At this time, the CRTC of the 8032 was driving the H and V monitor signals. The video signal was ‘black’.

Dave
 
Hi Guys! Thanks for your replies

Hutch, thank you for the link. I have used Zimmers before. What i cant find for this board is a layout view. Basically, a drawing of the whole board with the socket labels.
I think the RAM is ok, at least according to the PET tester i have on the ROM/RAM replacement board. And since it doesnt run with both ROM and RAM replaced, im not sure the black screen problem is caused by either of those. That is not to say that one of the original ROMs couldnt in fact broken as well.

daver2, So I have the one with 6550 RAM and 6540 ROM. I do have a PETTESTER ROM Is there something like this: http://www.zimmers.net/anonftp/pub/cbm/schematics/computers/pet/2001N/320351-6.gif for my board? that way all the stuff is labelled so i wont mess up.
I will check the IRQ on PIN 4 of the 6502 with my logic probe.
I will also check the PIN 39 of the 6520 in CA2 (there, such an overview would be helpful again, cause im not entirely sure how to identify CA2 on the board. Im sure one could figure it out, but theres some safety and confidence behind using a proper plan)

Thank you again for your replies!

Link to the first round of pictures!
 
If you get a blank screen using the Tynemouth RAM and ROM, then don't worry about the PET internal RAM or ROM at this time.
One quick thing to try would be to carefully pull the PIA's and VIA from their sockets and see if you get the boot up screen. If you do, one of those chips is bad.
 
dave_m, i will remove them all. I still have a working 6522 but the others I would need to order, if one is broken.

A quick question regarding the 6540 ROMs in this machine. I assume those are not compatible with the 2532 EPROMs that i have, and at this point i should probably start thinking about building 2732 adapters...?
Thanks again,
SkyCaptain
 
Dave_m's idea of removing the two PIA's and the VIA is possibly the best way to go to start with.

Ideally, you need to find out whether it is the ROM or RAM that is failing. Try just mapping the Tynemouth RAM on its own (and using the PET's ROMs) and then the other way round to see whether it only fails in one configuration or both.

The CA2 I referred to was a pin name not a board reference.

I have never seen a layout for these particular PET boards. However, it is not that difficult to make your own (for the ICs at least). The board is on a matrix consisting of the letters A, B, ... and numbers 1, 2, ... The IC reference is at the intersection of the letters and numbers. Just to confuse the issue, the larger (e.g. 40 pin chips) can occupy two 'letter' spaces.

Each device has a type (e.g. 74LS00) so it should be fairly straightforward (with a bit of effort) to make your own. It should make probing much easier (as you say) to have things identified from the start. Some of the discreet (resistors and capacitors) will be a pain - but don't bother about those to start with.

Which PETTESTER ROM do you have? One of the 'old' ones - or my later one (PETTESTE2KV04)?

Dave
 
Thanks for the photos. It's odd that someone scrubbed off the Commodore markings on the PCB. Since the badge is removed too, that makes me think this was one that was custom branded by a third party.
The blue screen bezel, OEM tape drive and 6550 RAM make it likely an early to mid 1978 model.
The number 3852 hand written on the board is the board serial number. There's another digit on the end I can't see. I'm curious what the full number is and serial number on the case.

When you are testing the RAM with the Tynemouth board, do you have the dip switches set to also replace the RAM? Are you sure you're testing the on-board RAM or is it testing the Tynemouth board replacement RAM?
 
To simplify matters you could remove all the 6540s and all the non-video 6550s... and the VIA and both PIAs! (The keyboard won't work but it should start).

You can also test your 6550s by swapping them in/out of Video RAM. Nice looking garbage that doesn't change when the CPU is missing is probably a good sign they are working.

Looks like it is BASIC 3.0? 025, 024,... etc... would be nice to have a photo of all the ROMs.

Curious that someone has scrabbled out the writing... should read "COPYRIGHT 1977 COMMODORE INTERNATIONAL"... looks like a legit 320008 board to me.

Looking at the chip dates... it is late 1978 (>week 38)... so the ROMs are probably factory?
 
Last edited:
Hi Sky,
I believe you mentioned that you tried the Tynemouth Basic 4 with 32K option and still got a blank screen.
If that is true, then something other than your PET memory is causing a problem. When Daver and others help you with that problem, then you can try using internal RAM and ROM. If more than one ROM or RAM is bad, I would advise you to use the Tynemouth board for all RAM and ROM rather than using a lot of adapters.
-dave_m
 
Welcome to VCFED Alan.

I don’t know, it is SkyCaptain’s machine... But it does seem to fit the bill!

Dave
 
Are we sure that this version of the PET can boot without the VIA? I think the easiest way to test the VIA is by substitution.

Alan
 
Are we sure that this version of the PET can boot without the VIA? I think the easiest way to test the VIA is by substitution.

Alan
It will boot to the opening screen, but of course there will be no cursor and nothing will run. It is just a quick test to see if screen comes up. If a screen comes up, one of the PIAs or VIA is bad.
 
It will boot to the opening screen, but of course there will be no cursor and nothing will run. It is just a quick test to see if screen comes up. If a screen comes up, one of the PIAs or VIA is bad.

Yes I understand the thinking and a working 320008 PET will certainly boot without the PIAs. I just wasn't sure about the VIA. It'll be interesting to see if SkyCaptain has made any progress.

Alan
 
Generally speaking, a PCB layout isn't that useful. In the schematic, all the parts are labeled with the grid location, row and column.
Example UJ1 is a chip at column 1 row J, UG6 is at location G6, etc...
 
Generally speaking, a PCB layout isn't that useful. In the schematic, all the parts are labeled with the grid location, row and column.
Example UJ1 is a chip at column 1 row J, UG6 is at location G6, etc...

But it’s helpful if you have it nonetheless - especially if you are trying to identify the discrete components or links (unless they are clearly marked on the PCB silk screen of course).

Dave
 
Alright, im back with some updates. Or, one update. I have completed the first suggested item, removing the PIAs and VIA.

With just the 6502 running, i get a black screen again. But when I replace ROM and RAM with the replacement board, it now boots up to BASIC.

ajgriff, thank you very much for that image! printed and filed ;)

the PET TEST ROM that I have is PETTEST4. I wanted to install it but i want to check back first if im at the right location. In the Manual for the PET TESTER it states that the EDIT ROM sits at "$Exxx". On the layout of the board, H3 is labelled with "E000"... is that the same location?

Thank You all for your inputs!
SkyCaptain
 
Yes, H3 is the EDIT ROM. E000 refers to the base address of the EPROM. The EDIT ROM responds to addresses $E000 to $E7FF - hence the reason I identified it as $Exxx.

Dave
 
Back
Top