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.

Reply
 
Thread Tools
Old 15th Jul 2023, 6:08 pm   #141
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,241
Default Re: Commodore PET 4032/8050/3022

Quote:
Originally Posted by ScottishColin View Post
I also get a reading of 19 ohms on my meter to pin 33 of UB13 and UB14 (6502). That seems odd.

Colin.
Is the resistance the same if you reverse the polarity of the meter leads?

Is there any slight change in the resistance if you measure at each of the ICs? Sometimes you can get a clue from seeing even just 0.1 ohm difference on the meter reading.
Mark1960 is offline   Reply With Quote
Old 15th Jul 2023, 7:48 pm   #142
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

Quote:
Originally Posted by ScottishColin
Do you think I should replace the 6502 socket?
That depends on just how confident you feel you are about desoldering and removing anything.

I would think the CPU socket would at least be easier to desolder and remove than a 40 pin IC so doing that first would ease you into the process, so to speak. I also think we are going to be doing a lot of swapping things in and out of the CPU socket (NOP Gen, Tynemouth Board, CPU etc) so it's for you to judge whether the original Commodore 'white' socket will be able to take that sort of punishment.

Socketing of the currently un-socketed large devices would at least allow you to test them in your other PET or anything else you have which you can try them in - with the exception of the CRTC controller. It might still be useful to be able to remove that so we can see what effect removing it has on any lines which don't look right.

I personally would fit conventional sockets, which tend to work better with ICs which have previously been soldered in as the pins have usually been clipped short. Plus, it is quite difficult to remove every last trace of solder from IC pins, which you really need to do so they will fit properly into a turned-pin socket.
SiriusHardware is offline   Reply With Quote
Old 16th Jul 2023, 8:44 am   #143
FrankB
Heptode
 
Join Date: Oct 2005
Location: Olympia, Washington, USA.
Posts: 661
Default Re: Commodore PET 4032/8050/3022

IIRC, Howard W. Sams had the service manual for that computer. (Possibly CC1? I last looked at the manual in the mid 80's).

That being said, I had all the full service OEM schematics and such as I was a CBM ASC. Not the easiest machines to work on either.

My fiancee moved my schematics into a 28' semi trailer for storage, and I have not been able to find them, as she was killed 12 years ago, and I just gave up on almost everything since then.
I must go into the storage building tomorrow though, and on the off chance it might still be in a file cabinet, I shall look.
FrankB is offline   Reply With Quote
Old 16th Jul 2023, 9:48 am   #144
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

Hi Frank - I'm very sorry to hear you had such a distressing experience. I can't imagine that is something you could ever fully recover from.

As you may have seen we think we have located an online copy of the standard service manual for this machine but if you think you can lay hands on anything over and above that which you might be able to share or if you, as someone who used to repair these, have any insight into the problems we are going to encounter on this machine then your input would be very much appreciated.
SiriusHardware is offline   Reply With Quote
Old 16th Jul 2023, 11:23 am   #145
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

Frank - any help you can give will be gratefully received - thanks very much.

Colin.

Quote:
Originally Posted by FrankB View Post
IIRC, Howard W. Sams had the service manual for that computer. (Possibly CC1? I last looked at the manual in the mid 80's).

That being said, I had all the full service OEM schematics and such as I was a CBM ASC. Not the easiest machines to work on either.

My fiancee moved my schematics into a 28' semi trailer for storage, and I have not been able to find them, as she was killed 12 years ago, and I just gave up on almost everything since then.
I must go into the storage building tomorrow though, and on the off chance it might still be in a file cabinet, I shall look.
ScottishColin is offline   Reply With Quote
Old 16th Jul 2023, 11:25 am   #146
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

Readings with black connector on r15

pin 40 <-> pin 33 - 27.5 Ω
Pin 40 <-> pin 32 - 15.1 Ω

Readings with red connector on R15

pin 40 <-> pin 33 - 27.3 Ω
Pin 40 <-> pin 32 - 14.9 Ω

Colin.

Quote:
Originally Posted by Mark1960 View Post
Quote:
Originally Posted by ScottishColin View Post
I also get a
reading of 19 ohms on my meter to pin 33 of UB13 and UB14 (6502). That seems odd.

Colin.
Is the resistance the same if you reverse the polarity of the meter leads?

Is there any slight change in the resistance if you measure at each of the ICs? Sometimes you can get a clue from seeing even just 0.1 ohm difference on the meter reading.
ScottishColin is offline   Reply With Quote
Old 16th Jul 2023, 8:22 pm   #147
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

OK progress I think. UB16 removed cleanly and a socket inserted in its place.

With the 6520 installed into the socket, I still get a low voltage on pin 40 of the 6502.

However, with the 6520 removed, pin 40 has a rock solid 5V now (5.076 to be specific).

I tried for fun the Tynemouth board in place of the 6502 but nothing on the screen and no chirp - where's next on the goosechase?

Colin.
ScottishColin is offline   Reply With Quote
Old 16th Jul 2023, 9:24 pm   #148
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

Well, that seems to have been a lucky hit. Remind me never to take you on in a game of Battleships.

I think it's back now to looking for activity on the CPU now that it should not be held in reset, so look again at the clock-in and two clock-out signals, also the states of or activity on the CPU RDY, SYNC and A0 lines.
SiriusHardware is offline   Reply With Quote
Old 16th Jul 2023, 11:38 pm   #149
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

Also, try putting that suspect 6520 into your breadboard (with the rows of legs on either side of the centre gap as usual) and measuring the resistance between its RESET pin and its D0 pin. (Pins 34, 33).

(There's no special reason to put it into the breadboard, it's just an easy way to keep it still so you don't have to chase it all over the bench with your meter probes).
SiriusHardware is offline   Reply With Quote
Old 17th Jul 2023, 8:18 am   #150
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

As well as the states of, or activity on, CPU Clock-in (pin 37), the two clock-out pins (pins 3, 39), RDY (pin 2), SYNC (Pin 7), and AB0 (pin 9), lines also check and report the states of or frequencies on:

UD3 pin 8
UD3 pin 5
UD3 pin 9
UD3 pin 2
UD3 pin 12

UE4 pin 10
UE4 pin 8

UE6 pin 8
UE6 pin 5
UE6 pin 9
UE6 pin 2
UE6 pin 12

UE7 pin 8
UE7 pin 5
UE7 pin 9
UE7 pin 2

UE5 pin 8
UE5 pin 3
UE5 pin 4
UE5 pin 5
UE5 pin 6
UE5 pin 11
UE5 pin 12

UD2 pin 6
UD2 pin 8
SiriusHardware is offline   Reply With Quote
Old 17th Jul 2023, 9:19 am   #151
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

The suspect 6520 gives a reading of 15.2 ohms.

A known working 6520 gives no reading (ie infinity).

I'll get on with the other readings today but work will intrude.

Colin.

Quote:
Originally Posted by SiriusHardware View Post
Also, try putting that suspect 6520 into your breadboard (with the rows of legs on either side of the centre gap as usual) and measuring the resistance between its RESET pin and its D0 pin. (Pins 34, 33).

(There's no special reason to put it into the breadboard, it's just an easy way to keep it still so you don't have to chase it all over the bench with your meter probes).
ScottishColin is offline   Reply With Quote
Old 17th Jul 2023, 9:26 am   #152
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

That's good, that was definitely our culprit as far as the load on the RESET line was concerned so good work there. Was that really the first one you chose to desolder? (With my luck it would have been the last).

Quote:
but work will intrude.
Same here, unfortunately, and I have been informed by SWMBO that I will be involved in a garage clearout this evening, so I might not be around much tonight.
SiriusHardware is offline   Reply With Quote
Old 17th Jul 2023, 10:31 am   #153
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

It was the top one on the motherboard so I started there. No other reason.

I was quite pleased how cleanly it came out - the 4032 motherboard is a lot easier to work on than the 3016. The 3016 had pads and I managed to lift a couple through over-use of the soldering iron.

This one has more solid through holes/soldering points (I'm sure there are much better technical terms for this) which meant that an iron and a solder sucker took me about 10 minutes to remove the 6520 and put the socket in (see attached photo).

Enjoy the clearout. I hope you find something interesting.

Colin.


Quote:
Originally Posted by SiriusHardware View Post
That's good, that was definitely our culprit as far as the load on the RESET line was concerned so good work there. Was that really the first one you chose to desolder? (With my luck it would have been the last).

Quote:
but work will intrude.
Same here, unfortunately, and I have been informed by SWMBO that I will be involved in a garage clearout this evening, so I might not be around much tonight.
Attached Files
File Type: zip PXL_20230716_185334016.zip (3.94 MB, 34 views)
ScottishColin is offline   Reply With Quote
Old 17th Jul 2023, 10:40 am   #154
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

Really neat job, well done. Unfortunately, your demonstrated prowess in this area does mean that we will casually ask you to do more of the same in future.

I'm pretty sure my 'Programming the 6502' book is buried a few archaeological layers down at the back of the garage. I don't know if we will be travelling that far back in time today.
SiriusHardware is offline   Reply With Quote
Old 17th Jul 2023, 8:01 pm   #155
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

PDFs attached showing voltages and frequencies for post 150. Some good, some less so and some of the voltages look a little low to me?

Colin.
Attached Files
File Type: zip 4032 - 20230717.zip (1.93 MB, 32 views)
ScottishColin is offline   Reply With Quote
Old 17th Jul 2023, 9:06 pm   #156
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

Hmm, odd results.

In your post #79 you had Clock In, and the 2 x clock outs all at 1MHz which is what they should be, but in your more recent captures in post #155 they appear to be a lower frequency. I notice that on the capture on those results the sample rate is stated as 1MHz which is exactly the same frequency as the frequency you are trying to measure, so that usually won't work well. (Generally the sample rate has to be significantly higher than the highest frequency you are trying to sample).

Could you try scoping those three signals (Clock in, Clock out x 2) again and this time alter the horizontal sweep speed setting so there are at least 10 cycles of each waveform on the screen before reading the frequency value?

You need not post each capture for the time being, just report whether the signal is static-low, static-high or a waveform. If the waveform is roughly 50/50 don't post the capture, just state the frequency.

On the plus side, continuous activity on AB0 means that the CPU is at least trying to run through the addresses, read instructions and execute them. It probably isn't managing to do so, otherwise you would at least get the start up 'chirp'.

We just need to try to reconcile the frequency measurements first.
SiriusHardware is offline   Reply With Quote
Old 17th Jul 2023, 9:29 pm   #157
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

OK - waveforms on pins 37, 3 and 39.

All three rock solid 1 Mhz with c. 10 waveforms on the scope.

The waveforms on 3 and 39 are not beautiful (a sloping rising reading), but they're consistent.

Colin.
ScottishColin is offline   Reply With Quote
Old 18th Jul 2023, 8:29 am   #158
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

OK, sorry to have to ask but could you now re-check the other signals (or states) in post #150 and again, make sure you have a good number of waveform cycles on the screen, not just one or two, when making each measurement - alter the horizontal sweep speed to achieve that whenever necessary. As before, no need to post images just now unless what you see seems out of the ordinary.

P.S. - I did find my Sybex 'Programming the 6502' (Rodnay Zaks) book some 20+ years after I last saw it, so the spelunking session down in the garage wasn't a complete bust.
SiriusHardware is offline   Reply With Quote
Old 18th Jul 2023, 9:12 am   #159
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,380
Default Re: Commodore PET 4032/8050/3022

Quote:
The waveforms on 3 and 39 are not beautiful (a sloping rising reading)
Are you using your scope probes in X1 mode or X10 mode? (There is usually a X1 / X10 switch on the probe itself to select this).

If you are using X10 (which is the preferred setting to use) then you may have to adjust the frequency compensation trimmer (which is usually on the body of the probe or built into the BNC plug at the scope end of the probe lead) to get the 'squarest' possible trace with the probe touched to a known good clean squarewave.

Scopes usually have a 'CAL' output terminal which outputs a squarewave signal to be used for exactly that purpose.
SiriusHardware is offline   Reply With Quote
Old 18th Jul 2023, 9:24 am   #160
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,740
Default Re: Commodore PET 4032/8050/3022

x10. I have recalibrated using the test signal to a clean square wave - it's just noticeable on 3 and 39 that they're there, but not square.

I'll re-do the readings shortly.

Colin.


Quote:
Originally Posted by SiriusHardware View Post
Quote:
The waveforms on 3 and 39 are not beautiful (a sloping rising reading)
Are you using your scope probes in X1 mode or X10 mode? (There is usually a X1 / X10 switch on the probe itself to select this).

If you are using X10 (which is the preferred setting to use) then you may have to adjust the frequency compensation trimmer (which is usually on the body of the probe or built into the BNC plug at the scope end of the probe lead) to get the 'squarest' possible trace with the probe touched to a known good clean squarewave.

Scopes usually have a 'CAL' output terminal which outputs a squarewave signal to be used for exactly that purpose.
ScottishColin is offline   Reply With Quote
Reply

Thread Tools



All times are GMT +1. The time now is 11:18 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.