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 11th Jan 2021, 1:00 pm   #81
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

re #78 - I'm sorry; what does "clipping your meter from clock to ground" mean?

Thanks for your pateince.
ScottishColin is online now  
Old 11th Jan 2021, 1:30 pm   #82
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 suggestion is that you connect the meter between the clock line and ground using 'clips' so that you've then got your hands free. Depends on the availability of suitable connectors (ie clips) and corresponding places to clip them. Not too difficult with the ground connection using a crocodile clip or similar but may be a bit more tricky with the clock line.

You may have noticed that I'm wrestling with a PET 2001-8 at the moment so I share your pain!

Alan
ajgriff is offline  
Old 11th Jan 2021, 5:11 pm   #83
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

OK. I've gone through the whole thread to make sure I've missed nothing. Although I haven't made any advances, I have a couple of questions if that's OK.

Post 5 - I have not done this; we moved onto other things; is it worth me doing this?

Post 9 - all readings are consistent with the diagram

Post 18 - all readings are the same (within tolerance)

Post 38 - here's my questions. I'm confused.
I get readings on Pins 3 and 5 on J7 (c. 50Hz - varying) but only if I do not connect to earth. If I earth the meter, I get a zero reading. I cannot explain this - if anyone has any suggestions I'd be grateful.

The same is true using UH7 pin2 and UG10 pin 11 i.e. I get a reading when I connect the Red lead, but as soon as I ground the black lead, it goes to zero.

Post 43 - same readings

Post 46 - next question:
If I connect Black UG5 Pin8 to red UG5 pin 7, I get a zero reading
However, If I swap the leads (ie Red UG5 pin 8 to Black UG5 Pin 7), I get a reading of 1.298 M Ohms.
Does that make any sense?

Post 52
All readings for resistance from Black UG5 pin7 are between 0.001-0.003 Ohms

Apologies if this adds more confusion. I'm leaving desoldering UG5 Pin7 until last.....

Thanks

Colin.
ScottishColin is online now  
Old 11th Jan 2021, 5:31 pm   #84
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

Quote:
Post 38 - here's my questions. I'm confused.
I get readings on Pins 3 and 5 on J7 (c. 50Hz - varying) but only if I do not connect to earth. If I earth the meter, I get a zero reading. I cannot explain this - if anyone has any suggestions I'd be grateful.
The reading you are getting when you have the black lead grounded is the 'true' reading. (No frequency). What's happening is that the leads of the (ungrounded) meter are picking up the 50Hz frequency being radiated by your mains wiring. You can disregard this. You won't get any signals on vertical_drive and horizontal_drive until you manage to restore that 1Mhz clock, which is why we have been so focused on that.
SiriusHardware is online now  
Old 11th Jan 2021, 5:53 pm   #85
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

Quote:
Post 46 - next question:
If I connect Black UG5 Pin8 to red UG5 pin 7, I get a zero reading
However, If I swap the leads (ie Red UG5 pin 8 to Black UG5 Pin 7), I get a reading of 1.298 M Ohms. Does that make any sense?
No.

When you are measuring a piece of wire or a resistor, something like that, it should generally measure the same resistance both ways, a short-circuit especially should be a short-circuit both ways.

Perhaps we should just re-state the first rule about making resistance measurements, always do those with the power off, never with the power on. Attempting to measure resistance in the presence of voltage certainly can give some odd results.

The exception to the 'same resistance both ways' rule is when semiconductors (Transistors, diodes, ICs) are in the path between your probes as they can (and usually do) show a different resistance reading if you swap the probes around - however, a short-circuit between two pins should completely override any of that and should always look like a short-circuit with the probes either way around.
SiriusHardware is online now  
Old 11th Jan 2021, 6:33 pm   #86
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

PET powered off an unplugged.

I am getting the following:

UG5 Pin 7 Black Pin 8 Red - 1.35 m Ohm
UG5 Pin 7 Red Pin 8 Black - flickering between 0 - 20 - 60 m Ohms

I'm starting to doubt myself now, but I know I'm doing the readings right....

Are we at the stage where if you have the time, you may be able to take a look at the motherboard? or is there something else yet?

I'm happy to pay of course.
ScottishColin is online now  
Old 11th Jan 2021, 7:19 pm   #87
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

If I understand you correctly the readings you state in your last post do not show a short circuit - by m ohms, do you mean Mega ohms? (It would be unusual for a meter like that to be able to read milliohms).

For an IC junction which does not have a short across it, it is quite feasible to get a reading of 1.34M in one direction and >20M in the other direction.

We've been proceeding so far on the understanding that you appeared to have a zero-ohm (or near zero ohm short) between pins 7 and 8 of UG5.

Your readings in #85, are those the readings, in Mega ohms, that you have been getting between those pins all along? It's a bit confusing when we read this, when you seem to be quoting the resistance in ohms elsewhere:-

Quote:
All readings for resistance from Black UG5 pin7 are between 0.001-0.003 Ohms
.
SiriusHardware is online now  
Old 11th Jan 2021, 7:28 pm   #88
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

The meter (I believe) auto adjusts (see photo) and shows no letter, a lower case k and an upper case M. I've tried to get my replies correct - I will go and double check the resistance readings that you have quoted.

Apologies if I have been confusing.
Attached Thumbnails
Click image for larger version

Name:	IMG_20210111_182605.jpg
Views:	70
Size:	62.3 KB
ID:	224395  
ScottishColin is online now  
Old 11th Jan 2021, 7:32 pm   #89
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

I have double checked and all the readings are as before - between 0.001-0.003 Ohms with no k or M on the meter.
ScottishColin is online now  
Old 11th Jan 2021, 8:02 pm   #90
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

And do you get that both ways between UG5 pins 7 and 8, or are you still getting an apparent high resistance between the pins when measured the other way around?
SiriusHardware is online now  
Old 11th Jan 2021, 9:24 pm   #91
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Photos attached, along with a link showing a short video - it's too big to upload I'm afraid.

The 1.360 M ohms reading is Black Pin 7, Red Pin 8.

The other photos show some of the varying readings Red 7 and Black Pin 8, as does the video.

https://drive.google.com/file/d/1R3X...ew?usp=sharing
Attached Thumbnails
Click image for larger version

Name:	IMG_20210111_201517.jpg
Views:	62
Size:	53.7 KB
ID:	224403   Click image for larger version

Name:	IMG_20210111_201747.jpg
Views:	62
Size:	50.9 KB
ID:	224404   Click image for larger version

Name:	IMG_20210111_201758.jpg
Views:	60
Size:	45.6 KB
ID:	224405  
ScottishColin is online now  
Old 11th Jan 2021, 9:51 pm   #92
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

On the basis of post #91 alone there is no short between pin 7 and pin 8 just now and the different Mohm readings in each direction are acceptable because you are measuring through a semiconductor device in two different directions (not just that one device, but the other 8-9 which are on the CLK1 line).

The reading of well under 1 ohm you reported in #89 - where did that come from, the same two pins? If the answer to that is yes we can only assume that short is intermittent.

Try doing as Mark1960 / ajgriff suggested. Take two strands of insulated wire maybe a foot long and bare 2mm at one end of each wire and about 2-3 inches at the other end of each piece of wire. 'Tin' the 2mm bared ends - (coat them with some solder).

Solder the bared and tinned 2mm end of one piece of wire to the pad of UG5 pin 8 on the underside of the PCB, and solder the 2mm end of the other piece of wire to the pad of UG5 pin 7 on the underside of the PCB.

Take the other ends of the wires, the 2-3 inch bared ends, and wrap one wire end tightly round one of your meter probes, and wrap the other wire end tightly around your other meter probe.

Now turn the meter on and it will constantly show you the resistance reading between those two IC pins without you having to hold the probes in place with your hands.

With your hands now free, you can go around carefully flexing, bending and pressing the PCB to see if that short comes and goes. It may help to put the meter into 'continuity' mode where, whenever there is a short between the probes, it makes a continuous beep sound, so you would be trying to find out whereabouts you have to press or flex the PCB to make the beep stop.
SiriusHardware is online now  
Old 11th Jan 2021, 10:59 pm   #93
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

The readings in post #89 are from UG5 Pin7 black to Red on all other pins that UG7 connects to (post #52 plus UH3 Pin 1).
ScottishColin is online now  
Old 11th Jan 2021, 11:37 pm   #94
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

OK. At the moment it appears you don't have a short between UG5 pin 7 and UG5 pin 8 and you do have continuity across all branches of the CLK1 line so let's backtrack a little to an earlier measurement.

With power on, meter in frequency mode, do you now see 1Mhz on UG5 pin 7?
SiriusHardware is online now  
Old 12th Jan 2021, 1:12 pm   #95
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

Zero reading.

I'll get on with soldering some wires on and gently playing with the motherboard as you have suggested.
ScottishColin is online now  
Old 12th Jan 2021, 2:25 pm   #96
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

The suggestion for making resistance measurements hands free was based on the fact that you appeared to have a short between UG5 (7) and UG5 (8) at least some of the time. I'm not sure that is the case any more.

If:
-There is no short between UG5 (7) and UG5 (8)
and
-There is no 1MHz signal on UG5 (7)

The most likely cause of that is the 'UD' output on UG5 (pin 7) faulty, because we know the counter chain works as far as the UC (2MHz) output.

The other possibility is that UG5 is being prevented from counting up past 7 (0111) so the UD output on pin 7 never changes state. Does anyone have a handle on what's happening on the left hand side of UG5? - it looks as though it can be reloaded with 0, though I don't know under exactly what circumstances.
SiriusHardware is online now  
Old 12th Jan 2021, 8:26 pm   #97
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

OK Colin, could you please try these steps.

With power off, check to see that there is no short between UG5 (7) and UG5 (8).

With power on,

Check the voltage on UG5 pin 11.

Look to see if there is a frequency present on UG5 pin 11. If there is a steady frequency there, could you also please state the duty cycle (%) figure?
SiriusHardware is online now  
Old 12th Jan 2021, 10:51 pm   #98
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Non-working Commodore PET 3016

No continuity signal between UG5 (7) and UG5 (8).

Voltage on UG5 (11) 5.072V

No frequency present on UG5 (11) - I take it there should be one?
ScottishColin is online now  
Old 12th Jan 2021, 11:48 pm   #99
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

I wasn't sure. The fact that it seems to be at steady logic 1 (+5V) means that UG5 is not being continually reset back to a count of zero which I thought it possibly might be.

The way that chip (UG5) is wired up it should be working as a simple 4-bit upcounter. The _INIT line input going to the LOAD pin (11) can zero the counter by going low (0v) momentarily but as that is not happening, you should have a continuous binary count on the four outputs QA(8Mhz), QB(4MHz), QC(2Mhz), and QD (1Mhz, pin 7).

As you don't have the 1Mhz on pin 7, suspicion now falls back onto UG5 itself, so my question for you now is how would you feel about trying to change that IC?

Here's what that would involve.

1) Finding and ordering a replacement 74191 IC, and a socket for it.

2) If you don't already have a pair, order the smallest, sharpest, most needlepointed pair of electrical sidecutters you can find. The tips of the jaws have to be narrow enough to fit into the gaps on either side of an IC pin.

The steps below are to be done ONLY when you have a replacement IC and socket already sitting in front of you.

3) Snip the legs of the original 74191 IC high up, right next to the body of the IC, (never low down, next to the PCB pads and tracks). Once you've snipped all the legs you can remove the IC body leaving just the 16 legs sticking up out of the PCB. Melt the solder on each leg in turn and as the solder melts, use a pair of tweezers to withdraw the leg from the PCB. When all of the legs have been individually removed in this way use your solder sucker to clear the holes.

4) Fit the IC socket ensuring that you put the 'notch' at the right end, solder it in and fit the new IC.

Sacrificing the IC (as above) is the best way to remove it without damaging the PCB. You can replace the IC but you can't easily replace or repair the PCB, so if you must damage something it's better for it to be the IC.

If you had a pump assisted 'proper' desoldering iron I would suggest trying to remove the IC in one piece but I don't think that is the best approach for you at the stage you are at, and with the tools you have.
SiriusHardware is online now  
Old 13th Jan 2021, 10:00 am   #100
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Non-working Commodore PET 3016

Just looking around for suitable cutters to snip chip pins with I stumbled across something called 'cuticle nippers'. They look very similar to conventional sidecutters but are usually somewhat smaller.
SiriusHardware is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 4:23 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.