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 26th Jan 2021, 12:55 am   #261
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by SiriusHardware View Post
It's only fair to warn you that Lead/Tin solder is deemed a hazardous substance due to the lead content - try not to rub it into your skin and definitely don't go mistaking it for liquorice.

However it is much, much better solder than any currently available substitute so the benefit outweighs the risk for occasional use IMO and for boards like this which were only designed with the working temperature for lead-tin solder in mind, it is essential.
I do s lot of 3D printing and have an SLA printer so I have plenty of gloves. I'll remember to put them on when I solder/de-solder.

Colin.
ScottishColin is offline  
Old 26th Jan 2021, 5:29 am   #262
TonyDuell
Dekatron
 
Join Date: Jun 2015
Location: Biggin Hill, London, UK.
Posts: 5,190
Default Re: Non-working Commodore PET 3016

Most of us here have been using lead/tin solder for years without wearing gloves (I can't do precision work in gloves!) with no major ill effects. So yes, don't chew solder, wash your hands before handling food but other than that don't panic.

I would also worry that the gloves could be a hazard if you touch the soldering iron with them and they melt onto your skin.

And I will never use anything but turned-pin sockets for my projects and repairs. The extra cost (and it's not too bad if you buy the full tubes from RS or somewhere like that) is minor compared to my time and stress tracing a 'silly' fault. I remember repairing a Whitechapel workstation where the processor, MMU, floating point chip, etc were all in turned-pin sockets but for some unknown reason the EPROMs were in 'normal' sockets. You guessed it, bad connections at the EPROM pins caused a lot of faults.
TonyDuell is offline  
Old 26th Jan 2021, 8:56 am   #263
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: Non-working Commodore PET 3016

Hi Tony, I think you meant to say decades rather than years

Going off topic, just for a change. Back in the day we used to avoid RS for sockets or connectors, preference was to use Farnell where we could specify the manufacturer, Augat or Cambion etc. RS usually labelled with their own brand but not to the same standard. I wore out many turned pin sockets in the days of uv eproms after a few months of software upgrades every week.

It was nice living in Birmingham, could get almost anything from the RS trade counter. One of the few nice things about Birmingham.

Now I’m just experimenting with old ceramic parts I avoid the turned pin sockets, they grip too tight for the first 2 or 3 goes, so putting the same old chip in 2 or 3 different new sockets starts to risk breaking pins.
Mark1960 is online now  
Old 26th Jan 2021, 10:03 am   #264
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

I've been using lead-tin solder since I was 11 and while I can't point to any definite side effects lead, once absorbed by the body, is not easily got rid of so it's best not to absorb it in the first place. I've never worn gloves either, but who can say whether my addled wits these days are just due to my being older, or at least partly due to accumulated lead?

Anyway - Like Tony I have been using lead-tin solder for decades and I am still here, so I think it's probably OK for Colin to use it every now and again.
SiriusHardware is offline  
Old 26th Jan 2021, 10:41 am   #265
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Non-working Commodore PET 3016

There have been many studies of lead absorption by the skin and unless the lead is finely powdered or formed into a compound there is little risk of exposure. (e.g. https://pubmed.ncbi.nlm.nih.gov/3238426/ )
Unless your soldering iron is set to 1740 degrees C there is no lead in the flux smoke given off by the soldering process (its still wise to avoid breathing it in as it is all forms of smoke).
Slothie is offline  
Old 26th Jan 2021, 12:48 pm   #266
dave cox
Nonode
 
dave cox's Avatar
 
Join Date: Jan 2009
Location: Bristol, UK.
Posts: 2,059
Default Re: Non-working Commodore PET 3016

The side contact sockets are rubbish, for sure, but can be mitigated, somewhat, by a few drops of contact cleaner/lube when putting the chips back in. This can slow the effect of the atmosphere getting into the gap and corroding the contacts. I don't recall seeing any ill effects from doing this ...

Even on a really good quality PCB, cutting out a 40 pin DIL socket is a pain I don't usually want to face! For best reliability the only solution is to solder the chip in!

dc

EDIT
Removing a socket WITHOUT a vacuum de-soldering station - nooooo !

Last edited by dave cox; 26th Jan 2021 at 12:53 pm.
dave cox is offline  
Old 26th Jan 2021, 3:00 pm   #267
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by dave cox View Post
For best reliability the only solution is to solder the chip in!
Unfortunately we expect to have to do quite a bit of diagnostic jiggery pokery on this machine so we need the flexibility of being able to remove and refit ICs. Sockets, but good ones, is the best compromise.

Quote:
Originally Posted by dave cox View Post
Removing a socket WITHOUT a vacuum de-soldering station - nooooo !
Unfortunately true. Ajgriff's suggestion of cutting the socket spars at top, middle and bottom and then wobbling each separated side from side to side to gradually snap all the pins may be the least bad option, although then you don't have a lot of pin left to get hold of with the tweezers (would have to pick them out from the bottom instead).
SiriusHardware is offline  
Old 26th Jan 2021, 3:05 pm   #268
Refugee
Dekatron
 
Refugee's Avatar
 
Join Date: Apr 2012
Location: Worksop, Nottinghamshire, UK.
Posts: 5,549
Default Re: Non-working Commodore PET 3016

The lead problem is worst in battery factories where the plates need to be clear of all oil based products before assembly.
When working with solder there is a layer of flux over it so there is far less risk.
Pretty well all the solder related regulations are part of a "one kills all" policy that also covers lead acid batteries.
It is well worth avoiding breathing flux fumes as they are more likely to be harmful than the lead in our solder.
Refugee is offline  
Old 26th Jan 2021, 6:13 pm   #269
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Hi all. Socket removed at the cost of some partially lifted pads on top (see attached photos).

I have gone through continuity tests for the pads that don't look 100% and can confirm that I still have continuity to either their destination, or the first point on their travel to their destination, so I think I may have got away with it.

I have new sockets here, but I'm waiting for non-lead-free solder to turn up so I may be a few days before I can get the socket back it. I have again purchased turned sockets and normal sockets.

If there's anything else I can do in the meantime, please shout. Otherwise I guess we're waiting for me to solder the new socket in place.

Thanks all.

Colin.
Attached Files
File Type: zip 6502 socket removed 20210126.zip (2.91 MB, 75 views)
ScottishColin is offline  
Old 26th Jan 2021, 7:02 pm   #270
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

There's not much we can do until you have the chip back in (in a socket). Can you say which of the top side pads (which pin numbers) you think are lifted? You've given us perfect look-down views of both sides but because of that it is difficult to see which pads are separated.

Separated top side pads are potentially a problem, so you are going to have to try to avoid this kind of damage in future. Avoiding any sort of damage at all would be even better .
SiriusHardware is offline  
Old 26th Jan 2021, 7:41 pm   #271
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

12,24,25,27,32,35 & 38

12 tests fine to UC3/15
24 & 25 are connected underneath to UB3/8 and UB3/11 and test fine
27 tests fine to UE10/6&7
32 tests fine to UE 9/4&5
35 does not connect to anything that I can see on the schematics
38 tests fine to R50

I did try - promise....
ScottishColin is offline  
Old 26th Jan 2021, 10:22 pm   #272
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: Non-working Commodore PET 3016

Is there a divider chain on the video circuit that could be tested without the cpu fitted?

Any risk of leaving the address lines floating with the cpu not fitted?
Mark1960 is online now  
Old 27th Jan 2021, 11:43 am   #273
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Non-working Commodore PET 3016

The video/monitor circuitry seems to have problems. As I understand it even with a faulty or missing CPU a functional video system would display a screen full of at signs (code 0x00) as shown below.

I think the focus here is to get the CPU working before moving on to look at the display issues.

Alan
Attached Thumbnails
Click image for larger version

Name:	PET No CPU.jpg
Views:	59
Size:	180.5 KB
ID:	225401  
ajgriff is offline  
Old 27th Jan 2021, 12:36 pm   #274
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by ajgriff View Post
The video/monitor circuitry seems to have problems. As I understand it even with a faulty or missing CPU a functional video system would display a screen full of at signs (code 0x00) as shown below.

I think the focus here is to get the CPU working before moving on to look at the display issues.

Alan
Without the CPU running, the screen should show a jumble of random characters, whatever random values are in the video RAM. If it was all @ characters something is forcing the outputs of the video RAM to zero, or writing zeros to RAM. If you switch a PET off and on (so the monitor is already powered and warmed up) you briefly see a random jumble on the screen before the CPU clears it.
Slothie is offline  
Old 27th Jan 2021, 12:54 pm   #275
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by Slothie View Post
Without the CPU running, the screen should show a jumble of random characters, whatever random values are in the video RAM. If it was all @ characters something is forcing the outputs of the video RAM to zero, or writing zeros to RAM. If you switch a PET off and on (so the monitor is already powered and warmed up) you briefly see a random jumble on the screen before the CPU clears it.
I'm no expert as you know but from what I've read if the CPU isn't running the absence of clock phases managed by the CPU means that the random characters held in video RAM will not be displayed, thus the code 0x00 output.

All a bit academic really as this PET hasn't produced a display of any kind so far.

Alan
ajgriff is offline  
Old 27th Jan 2021, 2:25 pm   #276
Slothie
Octode
 
Join Date: Apr 2018
Location: Newbury, Berkshire, UK.
Posts: 1,287
Default Re: Non-working Commodore PET 3016

I looked at the schematics and its very different to the 2001 I am familiar with because of the use of dynamic memory and all the timing surrounding that. the phi2 does go to the video circuit but on cursory examination that mostly seems to be to do with access to the video RAM. But you may be right that it needs it for some reason, or if phi2 is stuck high or low it might blank the screen. I did think these machines had a CRTC chip but it seems that was later models (4000 series)
Slothie is offline  
Old 27th Jan 2021, 2:56 pm   #277
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Non-working Commodore PET 3016

Quote:
Originally Posted by Slothie View Post
I looked at the schematics and its very different to the 2001 .....
Certainly very different in many respects although there do seem to be some similarities in the video circuitry of the 3016. I'm straying into areas here that I don't fully comprehend so probably best if I remain quiet until something more mundane crops up.

Alan
ajgriff is offline  
Old 27th Jan 2021, 3:56 pm   #278
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Solder's here so I'll get started later on and test continuity when I'm finished and report back.

With regard to the damaged pads (see posts 269 & 271), should I put some solder on those pads (they're all on the top side) before I solder the rest? Or will the solder drain through if I'm soldering from the bottom anyway?

Thanks.
ScottishColin is offline  
Old 27th Jan 2021, 5:13 pm   #279
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

The solder won't run through from the bottom to the top if you have separated the top pad from the conductive barrel inside the hole. In that case there are a couple of dodges, one being to solder the IC socket pin both on the bottom side of the PCB and the top side. You can't really do that with a conventional socket because you can't see the pins on the top side, they are hidden by the body.

Even with a turned-pin socket where you can see a bit of the IC pins on the top side you will need a pencil thin soldering iron tip to solder the top pads to the socket pins.

There is another dodge, which is to repair the damage by passing a very fine strand of bare wire through the cleared hole, bending it over sideways as it emerges from the hole on the top side and bottom side and soldering it to the pads, taking care not to fill the hole with solder because you need to be able to fit the socket afterwards. The IC socket pin has to be able to pass through the hole alongside the fine wire.
SiriusHardware is offline  
Old 27th Jan 2021, 8:19 pm   #280
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

Rough sketches showing a cross section through a single VIA to illustrate repair methods. First sketch, how things are normally. The orange areas represent copper PCB pads and tracks. The gold coloured element is the component pin, grey is solder, brown is PCB material.

Second sketch, what happens when a top side pad gets separated from the VIA. There is no longer a continuous metal connection between the copper sleeve going through the barrel so solder applied to the lower pad and pin won't flow up through the hole and up onto the top of the pad and around the pin.

You therefore have to reinstate the connection between the copper barrel and the top side pad somehow.

Third sketch uses a thin strand of bare wire passed through the hole (shown green here) to repair the missing connection between the top and bottom pads. The strand has to be thin enough to let the socket pin pass through the hole as well. I would suggest baring around 3cm of the end of a piece of 7-strand insulated wire and peeling away just one strand and using a bit of that for the repair.

Forth sketch reinstates the connection between the top and bottom pad by using the socket pin itself as the bridge between the top and bottom pads but to do this you have to solder the pin to both the bottom and top side pads. You can't do that if you use a conventional socket because you won't be able to get at the pins on the top side, so really a turned pin socket is the one to use if attempting this method, and even then you need a fine tipped iron which can get to the pin and top side pad without burning / melting the plastic frame of the socket.
Attached Thumbnails
Click image for larger version

Name:	Via_Repair_1.jpg
Views:	66
Size:	22.3 KB
ID:	225423   Click image for larger version

Name:	Via_Repair_2.jpg
Views:	70
Size:	32.9 KB
ID:	225424   Click image for larger version

Name:	Via_Repair_3.jpg
Views:	66
Size:	30.7 KB
ID:	225425   Click image for larger version

Name:	Via_Repair_4.jpg
Views:	69
Size:	25.9 KB
ID:	225426  
SiriusHardware is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 2:55 am.


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.