UK Vintage Radio Repair and Restoration Discussion Forum

UK Vintage Radio Repair and Restoration Discussion Forum (https://www.vintage-radio.net/forum/index.php)
-   Vintage Computers (https://www.vintage-radio.net/forum/forumdisplay.php?f=16)
-   -   Non-working Commodore PET 3016 (https://www.vintage-radio.net/forum/showthread.php?t=174829)

ScottishColin 29th Apr 2021 11:24 am

Re: Non-working Commodore PET 3016
 
Here's a .PNG of the labels in case anyone ever wants them.

https://drive.google.com/file/d/1OpO...ew?usp=sharing


Colin.

ScottishColin 29th Apr 2021 11:29 am

Re: Non-working Commodore PET 3016
 
Thank you. As a bit of a first-timer in all of this (and with plenty of time on my hands), I've tried to keep as much as I can.

I did count and have replaced 17 chips and added 18 sockets though.

The large front label is showing signs of wear and tear, paint splashes from somewhere and a bit of rust dribbling down it, but my plan is to mask it up and leave that on there if I can.

As my wife said yesterday, this has been a good lock-down project for me.

Colin.


Quote:

Originally Posted by Slothie (Post 1369155)
I appreciate the care you are taking to preserve the history of this machine by retaining as many of its original features. Too many retro restorations I have seen erase the marks of use over the years and return something that looks brand new but is not quite "right". It's a difficult balance, how much do you repair and how much do you leave? But I think that repainting the case in this machine is a good move given that the rust is so severe. If the existing front label can be preserved then that will be good too, as I imagine getting an authentic replacement that doesn't look odd will be hard, and presuming it is still OK then any minor scuffs are just "patina" :)


ajgriff 29th Apr 2021 11:46 am

Re: Non-working Commodore PET 3016
 
Securely refixing labels that have been removed by peeling can sometimes (not always) be tricky especially at the corners as the the original glue can lose its stickiness. If this proves to be a problem its worth using double sided carpet tape suitably trimmed with a sharp craft knife and metal rule.

I'm very much looking forward to seeing photos of the finished machine.

Alan

SiriusHardware 29th Apr 2021 5:33 pm

Re: Non-working Commodore PET 3016
 
Quote:

I think that repainting the case in this machine is a good move given that the rust is so severe.
I have to agree, lack of action / poor storage by the previous owner has basically narrowed the choices to 'live and let rust' or 'tackle the rust', with all that that implies. I have to say that what with all the electronic problems the machine turned out to have and now the poor physical condition of the unit, Colin could hardly have chosen a tougher first retro-computer project. It's very admirable that he just keeps battling so relentlessly onwards.

Slothie 29th Apr 2021 5:57 pm

Re: Non-working Commodore PET 3016
 
I'm astonished with the number of ICs that have needed replacing. I'm inclined to believe somehow the +5v got overvoltaged and took them out. It's not hard to do, only takes a stray screw or bit of wire in the power supply area....

ScottishColin 29th Apr 2021 8:57 pm

Re: Non-working Commodore PET 3016
 
Wait. You mean all projects aren't like this?

Colin.

Quote:

Originally Posted by SiriusHardware (Post 1369288)
Quote:

I think that repainting the case in this machine is a good move given that the rust is so severe.
I have to agree, lack of action / poor storage by the previous owner has basically narrowed the choices to 'live and let rust' or 'tackle the rust', with all that that implies. I have to say that what with all the electronic problems the machine turned out to have and now the poor physical condition of the unit, Colin could hardly have chosen a tougher first retro-computer project. It's very admirable that he just keeps battling so relentlessly onwards.


SiriusHardware 29th Apr 2021 9:24 pm

Re: Non-working Commodore PET 3016
 
Well... No.

Quote:

I'm inclined to believe somehow the +5v got overvoltaged and took them out.
...And yet it somehow missed all of the LSI (big chips), and every single one of the notoriously fragile RAM ICs survived. It's a mystery on par with the Marie Celeste, The Princes In The Tower, and the Even Number Of Socks In / Odd Number Of Socks Out puzzle.

In most cases you would be looking at one or two devices failed and maybe a bit of cosmetic damage to repair - I doubt that you will ever tackle another one as difficult as this.

Timbucus 29th Apr 2021 10:15 pm

Re: Non-working Commodore PET 3016
 
Quote:

Originally Posted by SiriusHardware (Post 1369386)
Well... No.

Quote:

I'm inclined to believe somehow the +5v got overvoltaged and took them out.
...And yet it somehow missed all of the LSI (big chips), and every single one of the notoriously fragile RAM ICs survived. It's a mystery on par with the Marie Celeste, The Princes In The Tower, and the Even Number Of Socks In / Odd Number Of Socks Out puzzle.

In most cases you would be looking at one or two devices failed and maybe a bit of cosmetic damage to repair - I doubt that you will ever tackle another one as difficult as this.

Now you just had to go and say that didn't you just as I start a major restoration project - even if Colin never goes through it again...

ScottishColin 3rd May 2021 5:12 pm

Re: Non-working Commodore PET 3016
 
1 Attachment(s)
OK - all back together and worked first time. I have yet to put the silver stickers back on but I will soon and I'll post some pictures.

Here's some before and after photos of the inside and outside. I can still see the postcode and the KCFE (Kendall College of Further Education) 'engravings' and I'm fine with that. I wasn't after a perfect restoration back to how it came out of the factory - I just wanted to show it a bit of love.

I'll post a couple more times on this thread with the summary and then I think we're done.

Colin.

Refugee 3rd May 2021 5:23 pm

Re: Non-working Commodore PET 3016
 
Don't say you are done.
It will most likely bight back with another fault if you do that.

ajgriff 3rd May 2021 5:31 pm

Re: Non-working Commodore PET 3016
 
A very pleasing outcome. The off-white looks about right too. Well done indeed!

Alan

ScottishColin 3rd May 2021 5:42 pm

Re: Non-working Commodore PET 3016
 
2 Attachment(s)
Thanks - it's so much brighter than the colour I got it, but I think it's pretty much what I remember.

fyi I used Halfords Gloss Black for the base and Halfords Peugeot Bianca White for the case.

I managed to keep the QC stams as well inside the case (not that anyone will ever see them) by judicious use of masking tape.

Colin.


Quote:

Originally Posted by ajgriff (Post 1370531)
A very pleasing outcome. The off-white looks about right too. Well done indeed!

Alan


SiriusHardware 3rd May 2021 5:43 pm

Re: Non-working Commodore PET 3016
 
Wow, superb improvement, you have really done the machine proud. It must have been the luckiest day of that machine's life when it found you, Colin.

ScottishColin 3rd May 2021 5:53 pm

Re: Non-working Commodore PET 3016
 
I have just noticed I have mounted the transformer 180 degrees the wrong way round....

Colin.

Timbucus 3rd May 2021 7:24 pm

Re: Non-working Commodore PET 3016
 
Amazing restoration job - I just have to say I have learnt a lot so thanks for sharing the journey with us.

Now though I will have to work harder on the project I am working on... I must start a thread as I had a breakthrough today!

Buzby123 3rd May 2021 10:37 pm

Re: Non-working Commodore PET 3016
 
Another big PET job ahead ?

https://www.facebook.com/groups/1233645553343772

Timbucus 3rd May 2021 10:41 pm

Re: Non-working Commodore PET 3016
 
I just noticed that and was going to tag these threads to him - Jim is in the Triton user group on there as well - he is Happy Little Diodes - I am still awaiting his next episode on that machine...

ScottishColin 4th May 2021 1:33 pm

Re: Non-working Commodore PET 3016
 
Here's my next problem then.....since I have cleaned up the CRT and its board and put it all back together, while the PET stills works fine, the monitor has acquired a little irritating 'bounce' when powered on.

Has anyone come across this and fixed it?

Thanks.

Colin.

SiriusHardware 4th May 2021 2:21 pm

Re: Non-working Commodore PET 3016
 
Look carefully: Is the picture height expanding and contracting slightly, or is the whole picture staying the same height but shifting upwards and downwards by just one or two video lines?

ScottishColin 4th May 2021 2:47 pm

Re: Non-working Commodore PET 3016
 
The whole picture is shifting slightly up and down.

Colin.


Quote:

Originally Posted by SiriusHardware (Post 1370922)
Look carefully: Is the picture height expanding and contracting slightly, or is the whole picture staying the same height but shifting upwards and downwards by just one or two video lines?


SiriusHardware 4th May 2021 3:00 pm

Re: Non-working Commodore PET 3016
 
And when it does so, does it stay shifted for a significant amount of time or is it only a very brief blip / jump / shift and then back to normal? How often is it doing it - several times a second - several times a minute?

ScottishColin 4th May 2021 3:14 pm

Re: Non-working Commodore PET 3016
 
Very brief shift up and down. It does it several times a minute.

Colin.



Quote:

Originally Posted by SiriusHardware (Post 1370946)
And when it does so, does it stay shifted for a significant amount of time or is it only a very brief blip / jump / shift and then back to normal? How often is it doing it - several times a second - several times a minute?


SiriusHardware 4th May 2021 4:00 pm

Re: Non-working Commodore PET 3016
 
One more question: Is the fault progressive? In other words, is it mild or not there at all when the machine is first switched on, and then does it begin and become more more noticeable as time goes on?

or-

Is it there more or less as soon as the machine is turned on, and does the frequency - the number of times it happens per minute - remain more or less constant no matter how long the machine is on for?

ScottishColin 4th May 2021 4:25 pm

Re: Non-working Commodore PET 3016
 
The latter. I've tried to get a video to show you what's going on. It's more obvious (and annoying) in real-life, but you can see it at its worst about 30s into the video.

https://drive.google.com/file/d/16oI...ew?usp=sharing

Colin.

Quote:

Originally Posted by SiriusHardware (Post 1370974)
One more question: Is the fault progressive? In other words, is it mild or not there at all when the machine is first switched on, and then does it begin and become more more noticeable as time goes on?

or-

Is it there more or less as soon as the machine is turned on, and does the frequency - the number of times it happens per minute - remain more or less constant no matter how long the machine is on for?


Timbucus 4th May 2021 4:29 pm

Re: Non-working Commodore PET 3016
 
That looks like vertical hold being lost for a moment - missed frame pulses or the like?

ScottishColin 4th May 2021 4:31 pm

Re: Non-working Commodore PET 3016
 
Perhaps I shouldn't have cleaned the already working CRT board, but it was filthy....

Colin.

Quote:

Originally Posted by Timbucus (Post 1370986)
That looks like vertical hold being lost for a moment - missed frame pulses or the like?


SiriusHardware 4th May 2021 4:50 pm

Re: Non-working Commodore PET 3016
 
Well, let's think. Colin never mentioned the monitor doing this prior to the stripdown (he probably would have noticed it before then) so I think what we probably have here is a case of physical disturbance or change during the makeover rather than a new electronic fault.

I think something physical has changed - could be the position of a pot / preset, or the legs of a component 'connected' together by liquid which has not dried out, etc.

To begin with can you check that the thick rubberised cable leading from the tube cap to the top of the line output transformer is more or less as it was before, with the rubber cable kept well away from anything else on the way between the tube cap and the LOPT (Line OutPut Transformer).

Also try just reseating the display connectors at both ends of the cable which carries the video signals from the main board to the monitor. (Do that with the power off).

We'll need to find the right circuit for your particular monitor to see if there is a vertical hold or vertical position pot which may have just gone a bit noisy or had its value changed slightly.

Timbucus 4th May 2021 4:59 pm

Re: Non-working Commodore PET 3016
 
Quote:

Originally Posted by ScottishColin (Post 1370989)
Perhaps I shouldn't have cleaned the already working CRT board, but it was filthy....

Colin.

Quote:

Originally Posted by Timbucus (Post 1370986)
That looks like vertical hold being lost for a moment - missed frame pulses or the like?


No you were correct - it is better to remove any dirt and then correct any issues as dirt can cause worse faults as it can trap damp etc...

ScottishColin 4th May 2021 5:08 pm

Re: Non-working Commodore PET 3016
 
1 Attachment(s)
You've put me on the right track; thank you. I've fiddled with the green circled pot on this photo and that seems to have stabilised things.

Colin.


Quote:

Originally Posted by SiriusHardware (Post 1370996)
Well, let's think. Colin never mentioned the monitor doing this prior to the stripdown (he probably would have noticed it before then) so I think what we probably have here is a case of physical disturbance or change during the makeover rather than a new electronic fault.

I think something physical has changed - could be the position of a pot / preset, or the legs of a component 'connected' together by liquid which has not dried out, etc.

To begin with can you check that the thick rubberised cable leading from the tube cap to the top of the line output transformer is more or less as it was before, with the rubber cable kept well away from anything else on the way between the tube cap and the LOPT (Line OutPut Transformer).

Also try just reseating the display connectors at both ends of the cable which carries the video signals from the main board to the monitor. (Do that with the power off).

We'll need to find the right circuit for your particular monitor to see if there is a vertical hold or vertical position pot which may have just gone a bit noisy or had its value changed slightly.


SiriusHardware 4th May 2021 5:17 pm

Re: Non-working Commodore PET 3016
 
Looking around other corners of the net there is mention of an earth bonding strap or wire between GND on the main computer section and GND on the monitor. Does your machine have that feature?

Edit: Just caught Colin's post re: Pot adjustment. Let's hope it really was that simple. Try to watch it for at least half an hour without blinking, otherwise you'll never really know. ;)

SiriusHardware 4th May 2021 5:23 pm

Re: Non-working Commodore PET 3016
 
If I'm looking at the right component layout diagram and then the right circuit diagram that would appear to be R621 (maybe that's marked on the PCB) - if so that's the vertical height control, so it is at least involved in the vertical aspect / section of the circuit.

ScottishColin 4th May 2021 5:27 pm

Re: Non-working Commodore PET 3016
 
1 Attachment(s)
Circled from above here - there's no component markings on either side of the monitor PCB.

Colin.

Quote:

Originally Posted by SiriusHardware (Post 1371022)
If I'm looking at the right component layout diagram and then the right circuit diagram that would appear to be R621 (maybe that's marked on the PCB) - if so that's the vertical height control, so it is at least involved in the vertical aspect / section of the circuit.


Slothie 4th May 2021 5:48 pm

Re: Non-working Commodore PET 3016
 
Its possible during cleaning you unintentionally moved it and it just needed readjusting to get vertical lock, or maybe its noisy. Lets hope its finished trying to get your attention!

SiriusHardware 4th May 2021 6:07 pm

Re: Non-working Commodore PET 3016
 
OK, I had the wrong circuit layout and diagram.

For reference, this layout is the one which matches the PCB and layout of Colin's monitor PCB.

http://www.zimmers.net/anonftp/pub/c...deo-layout.gif

Note the awkward error on that diagram - while the parts are shown, correctly, as they would look from above, the copper trace diagram is shown as it would look from below the board instead of, as it should be, looking through the PCB from above. The line output transformer is shown upper left, but the PCB pads it should be soldered into are shown upper right. The copper trace outline needs to be flipped over left / right.

Just to complicate matters further there are three distinct versions of the monitor circuit built on that PCB, namely

http://www.zimmers.net/anonftp/pub/c...01/video-1.gif

http://www.zimmers.net/anonftp/pub/c...01/video-2.gif

http://www.zimmers.net/anonftp/pub/c...01/video-3.gif

It would take a bit more investigation to work out exactly which one corresponds to Colin's but luckily his tweak to R20 (which is the 'vertical height' control in all three versions) seems to have sorted it out for the time being.

ScottishColin 4th May 2021 7:24 pm

Re: Non-working Commodore PET 3016
 
2 Attachment(s)
Stickers back on, using the suggestion from ajgriff of double-sided tape (which I happen to have in 6mm, 9mm and 12mm wide rolls).

I think that's the last thing now. I'll post that summary later on tonight to round this all off.

Colin.

ScottishColin 4th May 2021 8:57 pm

Re: Non-working Commodore PET 3016
 
OK - here's the overall summary of this entire mammoth thread which started on the 1st January this year and has 1835 posts (including this one), and has 35,450 views.

I had a broken 3016 and had it since 2012. It had power and I could see the monitor's orange colour in the back of the CRT tube, but that was it. The first person to reply was SiriusHardware, and I know lots of other people helped out, but he stayed with me all the way. I could not have done this without his help, patience, and logical approach. It's even more remarkable when he said he's never even seen a PET.

I didn't want to get it back to how it left the factory - it had a few dents and bruises and I wanted to keep them where I could. Its source had been scratched into the case - Kendall College of Further Education and I've managed to keep them through the rub-down/re-spray. I also managed to keep the QC stamps on the inside of the case that no-one will ever see, but it seemed important to me to keep.

I've learnt so much - I've put a blog post up here which lists all that I have learnt, and has lots more photographs.

https://colinjhaynes.wordpress.com/

I've replaced 17 chips and added 18 sockets (the difference is a chips that checked out OK but their sockets didn't, including the 6502).

The chips replaced were:

UA7
UA8
UB2
UB3
UC3
UC4
UC9
UE7
UE8
UE10
UF7
UF8
UF9
UG5
UH1
UH3
UH6

And Sirius kindly burnt me a new UD7 PROM as that one was faulty.

I spent pretty much exactly £300 - £40 on point, £150 on sockets and chips, £80 on meters and on a USB oscilloscope and the remainder on sundries (an Arduino, motherboard stand-offs, leaded solder, user-port interfaces etc).

We dealt with a faulty IEEE port and a faulty datasette #2 port. I got advice on how to remove and reattach the stickers, how to discharge a CRT (which seemed important to know really).

It seemed that others got interested in this thread and I had more than one private message encouraging me along.

I'd really like to thank every single person who chipped in with help, advice, testing code, kind queries on my duff scope readings and general tolerance of my sometimes naive questions. It's been a brilliant lockdown project and there's one PET rescued to add to the collection now.

Thanks everyone - I think we're done now until it breaks again. Anyone who ever manages to travel to Perth must give me a shout so I can buy you a pint.

Colin.

Mark1960 4th May 2021 9:55 pm

Re: Non-working Commodore PET 3016
 
Congratulations on getting it both working and cleaned up. I’m guessing a large portion of the spend was on shipping due to placing orders on one chip at a time.

ScottishColin 4th May 2021 10:14 pm

Re: Non-working Commodore PET 3016
 
There was a lot of P&P in there for sure, but having thought it through, there really was no other option other than to only order when it was clear that one (or a pair) had failed really.

Colin.


Quote:

Originally Posted by Mark1960 (Post 1371148)
Congratulations on getting it both working and cleaned up. I’m guessing a large portion of the spend was on shipping due to placing orders on one chip at a time.


SiriusHardware 4th May 2021 10:27 pm

Re: Non-working Commodore PET 3016
 
Quote:

And Sirius kindly burnt me a new UD7 PROM as that one was faulty.
Actually, you sorted that yourself as you got the replacement from 1980s_john. ;)

It has been a great pleasure to observe not only the computer but you yourself making steady progress as the project progressed - you probably didn't notice, but in the beginning we were being quite specific about what to look at and how to do it. By about half way through you were correcting our (...my...) circuit diagram reading blunders and near the end we were asking you to do things and you just... did them.

It's an old machine, so if it hits any more bumps in the road you know where to find us, but quite honestly, I'm not sure you'll need us any more.

All the best, and enjoy your PET.


All times are GMT +1. The time now is 5:45 am.

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