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 13th Jan 2021, 12:18 pm   #101
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Couple of quick questions before I start.

1) I'm interested in why it's lower risk for the PCB to de-solder as you suggest rather than leaving the IC in? Not quibbling, just trying to learn.

2) Is there anyone/company in these difficult times that I can procure the socket and chip from? I'd rather buy from a smaller company than an Amazon for example, and in these tricky post-Brexit times, it looks easier with regard to shipping if I buy from a British company.

Thanks very much again for your continued help and patience.

Colin.
ScottishColin is offline  
Old 13th Jan 2021, 12:57 pm   #102
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

Leaving the IC in presents a very low risk, but won't really get you any further. Maybe you're asking why not just desolder pin 7 and see if you get an output from it then? The reason is that, as the short you had between pins 7 and 8 has apparently disappeared there is no reason now for the IC not to be able to generate that 1MHz output - unless that output pin is faulty.

If we conclude that the chip now needs to be replaced to rule it either in or out, then of the two possible methods - remove it in one piece or cut it to pieces and remove the pins individually - cutting it to pieces is the better method given the tools which you have.

16-pin IC sockets you can get from just about anywhere but if you want to go UK / British you could try ESR components in Cullercoats or Cricklewood electronics.

The IC, which appears to be a 74191 and not a later 74LS191 (check that by reading the number off the chip?) will have to be obtained from where ever you can find it available. If you are lucky someone here may have one in their box of bits.
SiriusHardware is offline  
Old 13th Jan 2021, 1:17 pm   #103
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

Here you go:-

https://www.cricklewoodelectronics.com/74191.html

However, I would make an enquiry with them to see if their ICs really are 'original' 74191 and not 74(something)191. If the one actually fitted in UG5 position is already a 74LS type then by all means just find and fit one of those.

If Cricklewood's ICs really are 74191 at that price I suggest you buy at least two, one to keep as a spare, as there is no telling how much longer 'original' TTL ICs will continue to be available for.

Before you do any of this, check one more time to see whether you do or do not have 1MHz out on UG5 pin 7. (With the power on, obviously). The there / not there nature of that fault worries me a little bit.
SiriusHardware is offline  
Old 13th Jan 2021, 1:23 pm   #104
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Non-working Commodore PET 3016

Langrex do both IC types at £1.95 a pair including postage.

Alan

Edit: You need to use their eBay shop to get the free postage.

Last edited by ajgriff; 13th Jan 2021 at 1:31 pm. Reason: Additional Info
ajgriff is online now  
Old 13th Jan 2021, 2:47 pm   #105
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Apologies - my question wasn't clear.

I meant why is it best that I snip all the legs and de-solder them, rather than not snip the legs and de-solder the IC intact?

Thanks.
ScottishColin is offline  
Old 13th Jan 2021, 2:50 pm   #106
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Here is the UG5 chip marking.

I was doing some continuity testing last night and intermittently got a connection between UG5 Pin 8 and Pins 1 and 2 of J10. Looking at the schematics, this doesn't look right to me. Might this mean I still have a short somewhere else?

Or should I continue down the replacing UG5 route?

Thanks.
Attached Thumbnails
Click image for larger version

Name:	IMG_20210113_134813.jpg
Views:	76
Size:	57.2 KB
ID:	224522  
ScottishColin is offline  
Old 13th Jan 2021, 3:34 pm   #107
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

That's an 'original' 74191 logic IC so ideally, replace it with an 'original' 74191 and not a 74LS191.

Specifically, what was the resistance you were seeing? As an experiment, pick any two random IC pins on the PCB and measure the resistance between them both ways.

In many cases you will see a resistance, probably in the K to M range, but what is happening there is that the meter current is flowing into one IC pin and then up through the IC to the 5V supply rail and down through the IC to the 0V rail, along both of those and then into the power pins of the second IC and through the IC to the other pin and back to the meter.

The question is not whether there is a connection between pins, but whether that connection has the resistance you would expect.
SiriusHardware is offline  
Old 13th Jan 2021, 4:19 pm   #108
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 ScottishColin View Post
Apologies - my question wasn't clear.

I meant why is it best that I snip all the legs and de-solder them, rather than not snip the legs and de-solder the IC intact?

Thanks.
If you snip the body you are then left with 16 individual 'one-pin' items to desolder. That's very easy to do, first time, every time. Melt the solder on a pin with one hand, remove the pin with tweezers in the other hand.

If you try to take the IC out intact you have to get rid of virtually every last bit of solder which might be holding the IC on because when you try to remove a complete IC you are trying to remove 16 pins all at the same time.

No matter how hard you try with your solder sucker there will be a little bit of residual solder left between the bottom of the wide parts of the IC legs and the top side pads which they rest on. If you don't crack that solder first (by pushing the IC from side to side) then when you try to take the IC off the PCB you may very well tear some of the top side pads off with it.
SiriusHardware is offline  
Old 13th Jan 2021, 4:23 pm   #109
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Thanks - I understand the soldering process now.

I'll see what the resistance is and report back before I go snipping legs.

I owe you a lot of beer.....
ScottishColin is offline  
Old 13th Jan 2021, 5:13 pm   #110
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

Pin 8 of UG5 is 0V and of course the 0V rail goes to every other device and almost every connector on the PCB, so it would not surprise me if you found either a dead short or a resistance reading from there to quite a few other places on the PCB.

I must stress that if this was a rare, £30, 40 pin soldered IC we were talking about we would be considering almost every other option before chopping such a device to pieces without being certain whether it was faulty or not.

In the case of UG5 though, we are talking about a cheap and apparently still easily available IC, so it's no major loss if you cut the original UG5 up and it turns out not to have been the cause of the fault after all. Preservation of the PCB, rather than the IC, is the priority here.
SiriusHardware is offline  
Old 14th Jan 2021, 4:38 pm   #111
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Not checked the resistance yet - fyi I ordered yesterday from Cricklewood Electronics at 15:30 and they've turned up in this morning's post. Can't beat that service.
ScottishColin is offline  
Old 14th Jan 2021, 4:45 pm   #112
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

No need to do any resistance checks really, you'll just find that UG5 pin 8 is directly connected to a lot of other points on the PCB, for example the 0V power connections of all the other ICs.

Sounds like very good service from Cricklewood and by the carriers (Royal Mail?) - good for them. As a matter of interest, what is the exact number on the new devices (including the two letters before the number?).
SiriusHardware is offline  
Old 14th Jan 2021, 5:24 pm   #113
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Dn74191n
ScottishColin is offline  
Old 14th Jan 2021, 5:48 pm   #114
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

Dn? Not DM?
SiriusHardware is offline  
Old 14th Jan 2021, 5:58 pm   #115
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Apologies. Eyes getting old - DM.
ScottishColin is offline  
Old 14th Jan 2021, 7:16 pm   #116
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,484
Default Re: Non-working Commodore PET 3016

That's great, not only the right type of IC but made by the same IC manufacturer as the original IC, (National Semiconductor) so you'll be keeping it as original as possible.

Next step is to snip out the original IC - remember to do that high up beside the IC body, not low down beside the PCB, and then remove the IC body. Working on the top side, heat one of the pins and its solder pad, when the solder melts lift out the pin with tweezers, repeat until all pins are removed. Then, use your solder sucker to clear the remaining solder from all 16 holes. It's been said before, but to emphasise, solder suckers work best when there is a decent amount of solder to remove.

Before attempting to clear a hole put a shallow hump of fresh lead-tin solder on the solder pad - top side or lower side, whichever is easiest to get at. If the hole only half-clears the first time you try, put another hump of fresh solder back on the pad and try again and eventually you'll get lucky and the hole will clear properly. If this sounds a bit hitty-missy, it is. A manual solder sucker is not the ideal thing to have to be using but you can only work with what you actually have.
SiriusHardware is offline  
Old 14th Jan 2021, 10:13 pm   #117
Mark1960
Octode
 
Join Date: Mar 2020
Location: Kitchener, Ontario, Canada
Posts: 1,265
Default Re: Non-working Commodore PET 3016

Just want to add a warning to be very careful when using a manual solder sucker, they tend to kick when the spring is released and if the soldering iron is touching the pcb and the solder sucker tip kicks the soldering iron, the combination of the heat and shock can damage the pad. Sorry for repeating the warning, just want to make sure you avoid damage to the board.

I would also agree with sirius using the soldering iron from the component side of the board when removing the IC pins with tweezers. This is less likely that you’ll try and pull a pin where the solder is not fully melted on the top side, which could pull the pad.

Then when sucking the solder from the holes work from the solder side of the board. This might just reduce the cumulative effect of heat on the adhesive holding the pads to the board, but also if you do damage a pad its easier to repair a bottom side pad than a top side pad.

Last edited by Mark1960; 14th Jan 2021 at 10:16 pm. Reason: Noted sirius already said to work from top side to remove pins.
Mark1960 is offline  
Old 14th Jan 2021, 11:21 pm   #118
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 Mark1960 View Post
when sucking the solder from the holes work from the solder side of the board. This might just reduce the cumulative effect of heat on the adhesive holding the pads to the board...
Good point - For practical purposes you have to apply heat to the top pads when lifting the pins out, so it might be better to switch to the underside pads when it comes to clearing the holes.

As Mark said, the chief danger with boards this old is that if you heat a pad for too long you will burn it right off the board (the adhesive which attaches the pad to the PCB will give up and the pad will just float free off the PCB).
SiriusHardware is offline  
Old 15th Jan 2021, 2:05 am   #119
Timbucus
Octode
 
Join Date: Mar 2019
Location: Barry, Vale of Glamorgan, Wales, UK.
Posts: 1,362
Default Re: Non-working Commodore PET 3016

It does not hurt to leave the board for an hour for the adhesive to cool and solidify again before proceeding...
Timbucus is offline  
Old 15th Jan 2021, 2:50 pm   #120
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

OK - done it but I am a little concerned. Please can you take a look and give me some advice. This is the top of the board - have I done bad things to pins 8 and 9?
Attached Thumbnails
Click image for larger version

Name:	IMG_20210115_134645_1.jpg
Views:	90
Size:	102.1 KB
ID:	224662  
ScottishColin is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 11:29 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.