UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Computers

Notices

Vintage Computers Any vintage computer systems, calculators, video games etc., but with an emphasis on 1980s and earlier equipment.

Closed Thread
 
Thread Tools
Old 27th Jan 2021, 2:39 pm   #61
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Commodore PET 2001

Considering the number of unknowns at this point, I'd be considering pulling the cpu and connecting a microcontroller to the pins, and using that to read the roms, test the rams etc in situ. Precise timing wouldnt be necessary as long as you ensure phi2 is low while data/address is changing and high when reading/writing. You could then checksum the ROMS and compare those with the roms in the archive.You could also "single step" the hardware and check enable lines and buffers etc.
Slothie is offline  
Old 27th Jan 2021, 2:53 pm   #62
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Commodore PET 2001

Considering the number of unknowns at this point, I'd be thinking about taking out the CPU and plugging in a microcontroller to the vacant socket. You could then read the ROM, test the RAM and "single step" the hardware to see if the enable lines etc are being properly operated. Timings wouldnt need to be precise since this machine has static RAM as long as you remember to pull phi-2 low while address and data are changing and put phi-2 high while reading/writing data.
Before he packed up shop for COVID, Tynmouth software had a board to do this, but there's no reason an Arduino Mega or PIC couldn't be pressed into use for this purpose, I guess it depends on which you are more familiar with. Unfortunately an MK14 has only 16 i/o lines.....
Slothie is offline  
Old 27th Jan 2021, 5:22 pm   #63
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Commodore PET 2001

I haven't noticed aj mentioning an interest in microprocessors / micro toys in the past, so maybe now would be a good time to ask whether he does have more specialised micro test gear (such as a logic analyser, device reader / programmers etc) otherwise it might be best to focus on checks which can be done with a meter and scope.
SiriusHardware is offline  
Old 27th Jan 2021, 5:51 pm   #64
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

Quite right Sirius. The first sentence of the opening post says it all really:

Quote:
Originally Posted by ajgriff View Post
Vintage computing really isn’t my thing.

Alan
I hope to have made it clear throughout that I'm on a learning curve (red L plates) with the PET. You're also correct in assuming that I don't have EPROM readers, logic analysers, programmers etc. My tools for this are effectively meters, oscilloscopes, the time to do some research, a basic knowledge of electronics and not much else apart from a soldering iron of course.

I'll be back with an update later.

Alan
ajgriff is offline  
Old 27th Jan 2021, 7:03 pm   #65
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Commodore PET 2001

You may, as inferred earlier, get to the point where you will have rule the PROMs in or out, but if you don't have the means to do so yourself there are a number of us here who could read them for you if you didn't mind the risk of posting them off-site for a few days.

It's a shame Slothie is currently separated from his own PET which is similar to yours otherwise he would be able to test parts for you, although I think he'd have to fix his own first.

As I mentioned, if scope and meter are your weapons of choice then I think some old-school methodical continuity checking to rule out possible problems with socket contacts is in order. I've heard, many times, that the IC sockets used on these are shoot-on-sight. You might even consider just changing all of the sockets as a preliminary to finding whatever the real fault may be and just accept your great good luck if changing the sockets actually fixes it.

I'm certain you do have the tools and skill set needed to do that.
SiriusHardware is offline  
Old 27th Jan 2021, 7:45 pm   #66
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

Thank you for your encouragement and offer of help with PROMs if I get to that point. Out of interest I just did a quick count via the parts list and there are 38 IC sockets with a total of 972 pins!

Alan
ajgriff is offline  
Old 27th Jan 2021, 8:28 pm   #67
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Commodore PET 2001

Ouch, that's quite a few. The ones I have been involved with so far usually only had the 40-pin and PROM ICs in sockets, with the rest direct mounted.
SiriusHardware is offline  
Old 28th Jan 2021, 10:58 am   #68
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

Quote:
Originally Posted by ajgriff View Post
Out of interest I just did a quick count via the parts list and there are 38 IC sockets with a total of 972 pins!
Correction needed here as I misinterpreted the parts list. There are actually 30 socketed ICs with 780 pins. Still daunting enough.

Alan
ajgriff is offline  
Old 28th Jan 2021, 2:51 pm   #69
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Commodore PET 2001

Just stumbled across this looking for something else, its a troubleshooting guide for the 2001/8 - it seems to be the same board but with upgrade ROMS so the rom numbers are different, but everything else applies. I think you've already covered most of this but there might be some nuggets of useful information there:
http://www.dasarodesigns.com/project...dore-pet-2001/
Slothie is offline  
Old 28th Jan 2021, 3:07 pm   #70
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

Thanks again Slothie. Yes it's a very useful guide for someone like me who had never encountered a PET 2001-8 before. As it happens it was a good starting point as I may have mentioned in a much earlier post. Unfortunately the site owner seems to have drifted away from PETs and no longer makes or sells any of the diagnostic/upgrade kit he originally developed.

Alan

PS It is the same board (320008) as the one fitted to mine. Everything's identical.

Last edited by ajgriff; 28th Jan 2021 at 3:29 pm. Reason: PS
ajgriff is offline  
Old 28th Jan 2021, 3:32 pm   #71
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Commodore PET 2001

Ah I didn't recall seeing it in this thread but then I dont remember everything Yes, its a shame he has stopped making bits, so many have, and so many places are out of stock for obvious reasons.
Slothie is offline  
Old 28th Jan 2021, 3:47 pm   #72
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Commodore PET 2001

My suggestion of blanket replacing all sockets was based on the idea that there were probably only a handful of 40-pin and PROM sockets to replace, as on Colin's PCB in the other PET thread.

However the alternative is to check continuity from every socketed pin to its destination: Equally daunting, and if an intermittent connection happens to be good at the very moment you measure it, you could miss it altogether.
SiriusHardware is offline  
Old 28th Jan 2021, 3:57 pm   #73
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

Thanks Sirius. I confess that I'd rejected the idea of replacing all the sockets, coward that I am!

Alan
ajgriff is offline  
Old 28th Jan 2021, 4:17 pm   #74
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,485
Default Re: Commodore PET 2001

As all of your ICs appear to be socketed - normally a good thing - you might consider investing in a relatively low cost logic IC tester, ideally one which can test RAM ICs as well, and then you could at least get an idea of whether the bulk of the ICs are OK. The big LSI ICs are really only verifiable by substitution, and the PROMs would need to be read and verified by some means.

If you want to start with the PROMs (and you don't mind sending them offsite) drop me a PM and we can take it from there. If you'd prefer to exhaust other avenues first, that's fine of course.
SiriusHardware is offline  
Old 28th Jan 2021, 5:59 pm   #75
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

There are 30 odd soldered ICs in various flavours as well of course but I'm trying to pretend that they don't exist. I'm in the midst of preparing a rather lengthy post summarising the state of play as I see it. Hopefully it will help me and everyone else to understand where I'm going with this particular challenge.

Alan
ajgriff is offline  
Old 28th Jan 2021, 7:04 pm   #76
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

I always try to adopt a logical approach to fault finding but sticking to this philosophy has proved difficult with the PET. This is ironic in a sense as computers rely on ‘logic’ for their operation. In an attempt to bring some sensible methodology back into this saga I’ll try to summarise the current position. This is for the benefit of my own sanity as much as anything else so please bear with me.

Just a reminder that the fault condition was and is a screen full of random characters at switch on with a failure of the PET to move on from there.

To my mind these are the positive indicators:

1. The motherboard is in excellent unmolested physical condition and the pins of the ICs I’ve lifted haven’t exhibited signs of oxidation/corrosion.

2. The PSU is working well with the correct voltage on all lines.

3. The monitor and its internal circuitry are fine and the connections to the main board are now good.

4. The character generator (PROM) is functional.

5. The VIA processor is doing its job.

6. The two Video RAMS are working as evidenced by the completely random nature of the display characters generated.

7. The 555 delayed reset circuit is working as it should and substituting the timing capacitor has no effect.

8. The 6502 CPU and its associated decoder are fully active although the signal on A12 may indicate that the system is locked in a loop.

Moving on to some of the unknowns:

1. It’s entirely possible that one or more of the system PROMs is corrupt. Notwithstanding Sirius’s kind offer I don’t currently have the means to adequately investigate this possibility.

2. The status of the two PIA processors is unknown.

3. One or more of the sysem RAMs may be faulty.

4. I suppose it would only take one poor contact out of 720 socketed IC pins to cause a problem. Not a cheering thought.

Proposed next steps after all this verbiage:

1. In theory the PET should run without the PIA chips in place so I’ll try it with them removed just in case.

2. The two Video RAM chips are ok and the system only needs the four RAMs in locations L1, L2, J1 & J2 to boot. I’m going to try to confirm the identity of four working RAMs by substitution into the Video RAM positions. It should be possible to identify any faulty chips from the resultant non-random nature of the characters displayed. Once I’ve got four good chips I’ll run the machine with just the four essential system RAMs. If by some miracle the PET then boots to BASIC I can then replace unverified chips in pairs until all faulty chips reveal themselves.

Assuming I still end up with a random character display most fingers will then be pointing in the direction of the system PROMS although I won’t have completely discounted the possibility of poor pin contacts or failure of a minor IC/component.

I’ve given some considerable thought to available options if I reach a stalemate with the PET but will save that analysis for another day.

Thank you for your patience if you’ve stuck with me so far.

Alan
ajgriff is offline  
Old 28th Jan 2021, 7:55 pm   #77
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Commodore PET 2001

(2) is a sensible course of action. A zero page ram fault could cause looping given interrupt vectors are stored there as well as numerous pointers.
Slothie is offline  
Old 29th Jan 2021, 4:49 pm   #78
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

Lo and behold ...................
Attached Thumbnails
Click image for larger version

Name:	PET Running.JPG
Views:	100
Size:	169.6 KB
ID:	225542  
ajgriff is offline  
Old 29th Jan 2021, 6:02 pm   #79
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Commodore PET 2001

Oh nice was it one of the PIA's?
Timbucus is offline  
Old 29th Jan 2021, 7:18 pm   #80
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Commodore PET 2001

Although I took the photo without either PIA fitted they're now back in place and I've got a cursor plus a working keyboard.

Failure to boot to BASIC was caused by at least two faulty RAM chips and there may be more. I used the methodology outlined in 'Next Steps 2' of my earlier post. Initially I isolated one faulty chip and ensured that the four chips needed for the PET to boot were all good. However it still didn't get past the random character screen until I removed the other 12 system RAM ICs. I now have to identify the remaining faulty chip(s).

Alan
ajgriff is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 2:34 pm.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.