Non-working Commodore PET 3016 - Page 43 - UK Vintage Radio Repair and Restoration Discussion Forum

ScottishColin · 10th Mar 2021, 1:32 pm

Progress with a capital P.

I replaced UH6 with a new chip and I now have something on the screen. I could have a little weep.

The screen is full of the semi-colon character, but I'm still happy.

Where do we go from here?

Colin.

ajgriff · 10th Mar 2021, 2:11 pm

Fantastic! That's real progress as the screen display will help you diagnose the remaining problem(s). I think (remembering I'm no expert) that displaying a single character repeatedly could indicate an issue with the CPU or the clock signal. Are you using the original or the replacement CPU? Is it properly seated?

Alan

ScottishColin · 10th Mar 2021, 2:26 pm

It does the same with both CPUs, and also with no CPU inserted - I have soldered in a new 40 pin socket during this thread though.

At least it means the monitor works - I was not looking forward to fiddling with the voltages in there.

Colin.

Quote:

Originally Posted by ajgriff

Fantastic! That's real progress as the screen display will help you diagnose the remaining problem(s). I think (remembering I'm no expert) that displaying a single character repeatedly could indicate an issue with the CPU or the clock signal. Are you using the original or the replacement CPU? Is it properly seated?

Alan

Slothie · 10th Mar 2021, 2:30 pm

Quote:

Originally Posted by ajgriff

Fantastic! That's real progress as the screen display will help you diagnose the remaining problem(s). I think (remembering I'm no expert) that displaying a single character repeatedly could indicate an issue with the CPU or the clock signal. Are you using the original or the replacement CPU? Is it properly seated?

Alan

Interesting... the screen code for ';' is 3B or 00111011 so its not just one stuck bit.
I thought we'd established the CPU is running, so I'd plump for dodgy ROM or a stuck buffer in the display logic.

ajgriff · 10th Mar 2021, 2:32 pm

The fact that the screen doesn't change with no CPU is a good starting point and much better than having to delve inside the monitor. Worth just checking that 5V is still reaching the CPU but I'm sure others will be along shortly to advise on further investigations.

Alan

ajgriff · 10th Mar 2021, 2:36 pm

Quote:

Originally Posted by Slothie

I thought we'd established the CPU is running, so I'd plump for dodgy ROM or a stuck buffer in the display logic.

Has the video ROM been checked previously along with the other ROMs?

Alan

ScottishColin · 10th Mar 2021, 2:49 pm

Great minds think alike. About to do it.

Colin.

Quote:

Originally Posted by ajgriff

Quote:

Originally Posted by Slothie

I thought we'd established the CPU is running, so I'd plump for dodgy ROM or a stuck buffer in the display logic.

Has the video ROM been checked previously along with the other ROMs?

Alan

julie_m · 10th Mar 2021, 4:04 pm

Quote:

Originally Posted by ScottishColin

I replaced UH6 with a new chip and I now have something on the screen. I could have a little weep.

The screen is full of the semi-colon character, but I'm still happy.

This is excellent news! It's certainly close to working now.

SiriusHardware · 10th Mar 2021, 4:37 pm

Yes, a great leap forward, and what a relief that the monitor appears to have survived despite having been exposed to all that damp.

I did think Colin had already read the content of the character ROM but it's not a bad idea to read it again. (I didn't get that one to check with the others).

I think it is reasonable to assume that the CPU is OK because we were able to execute NOPs, as shown by the fact that it ran smoothly through the address range when we did.

Mark1960 · 10th Mar 2021, 4:42 pm

Quote:

Originally Posted by Slothie

Quote:

Originally Posted by ajgriff

Fantastic! That's real progress as the screen display will help you diagnose the remaining problem(s). I think (remembering I'm no expert) that displaying a single character repeatedly could indicate an issue with the CPU or the clock signal. Are you using the original or the replacement CPU? Is it properly seated?

Alan

Interesting... the screen code for ';' is 3B or 00111011 so its not just one stuck bit.
I thought we'd established the CPU is running, so I'd plump for dodgy ROM or a stuck buffer in the display logic.

As I think Alan noted, if the screen is no different with CPU not fitted, then I would suspect the buffer UB3 or a fault in the video RAm UC4 and UC5.

I think if it was the display ROM then you probably wouldn’t see correctly formed characters.

Try checking the input and output pairs on UB3. Pin 3 compared to pin 2 etc.

ScottishColin · 10th Mar 2021, 5:09 pm

UB3 is a new IC - replaced earlier on in this saga.

Happy to still check it; are you saying to check all pins 2-9, 11-18 for waveforms?

Colin.

Quote:

Originally Posted by Mark1960

Quote:

Originally Posted by Slothie

Quote:

Originally Posted by ajgriff

Fantastic! That's real progress as the screen display will help you diagnose the remaining problem(s). I think (remembering I'm no expert) that displaying a single character repeatedly could indicate an issue with the CPU or the clock signal. Are you using the original or the replacement CPU? Is it properly seated?

Alan

Interesting... the screen code for ';' is 3B or 00111011 so its not just one stuck bit.
I thought we'd established the CPU is running, so I'd plump for dodgy ROM or a stuck buffer in the display logic.

As I think Alan noted, if the screen is no different with CPU not fitted, then I would suspect the buffer UB3 or a fault in the video RAm UC4 and UC5.

I think if it was the display ROM then you probably wouldn’t see correctly formed characters.

Try checking the input and output pairs on UB3. Pin 3 compared to pin 2 etc.

Mark1960 · 10th Mar 2021, 5:35 pm

Are you sure that was UB3 ? What was the post number that was referenced?

I think you need to compare the input pins with the output pins. One probe on pin 3, second on pin 2. Then pin 4 and pin 5 etc.

SiriusHardware · 10th Mar 2021, 5:51 pm

UB3 and UC3, which are the address buffers, have been replaced. I wonder if Mark is maybe thinking of the data bus buffers UE9 and UE10?

ScottishColin · 10th Mar 2021, 5:56 pm

post 435

I've actually replaced UB3 and UC3 at the same time.

Colin.

Quote:

Originally Posted by Mark1960

Are you sure that was UB3 ? What was the post number that was referenced?

I think you need to compare the input pins with the output pins. One probe on pin 3, second on pin 2. Then pin 4 and pin 5 etc.

Mark1960 · 10th Mar 2021, 6:00 pm

Sorry, looking at the wrong schematic. Video data latch, UF9, 74ls373, if I’m looking at the right schematic now.

SiriusHardware · 10th Mar 2021, 6:06 pm

We haven't worked in that area before, so UF9 has not been changed. It may be worth looking to see if there is an active 'video latch' signal going to UF9 pin 11. Depending on whether there is or is not, we'll proceed from there.

ScottishColin · 10th Mar 2021, 6:14 pm

OK - what am I looking for on pin 11?

Colin.

Quote:

Originally Posted by SiriusHardware

We haven't worked in that area before, so UF9 has not been changed. It may be worth looking to see if there is an active 'video latch' signal going to UF9 pin 11. Depending on whether there is or is not, we'll proceed from there.

SiriusHardware · 10th Mar 2021, 6:14 pm

A waveform, rather than the absence of one.

Also while in that area, can you have a look for activity on pin 10 of either UF7 or UF8.

ScottishColin · 10th Mar 2021, 6:25 pm

UF9/11 - consistent waveform at 1Mhz. It's a bit more rounded than square, but the timing, shape and frequency are consistent

Nothing on either UF7/10 or UF8/10.

Colin.

SiriusHardware · 10th Mar 2021, 6:29 pm

OK on the 'Video Latch' signal.

When you say you have nothing (presumably meaning no waveforms) on UF7(10) / UF8(10), what do you have instead? Steady high state or steady low state?

10th Mar 2021, 1:32 pm	#841
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,816	Re: Non-working Commodore PET 3016 Progress with a capital P. I replaced UH6 with a new chip and I now have something on the screen. I could have a little weep. The screen is full of the semi-colon character, but I'm still happy. Where do we go from here? Colin.

10th Mar 2021, 2:11 pm	#842
ajgriff Nonode Join Date: May 2004 Location: Halifax, West Yorkshire, UK. Posts: 2,587	Re: Non-working Commodore PET 3016 Fantastic! That's real progress as the screen display will help you diagnose the remaining problem(s). I think (remembering I'm no expert) that displaying a single character repeatedly could indicate an issue with the CPU or the clock signal. Are you using the original or the replacement CPU? Is it properly seated? Alan

10th Mar 2021, 2:32 pm	#845
ajgriff Nonode Join Date: May 2004 Location: Halifax, West Yorkshire, UK. Posts: 2,587	Re: Non-working Commodore PET 3016 The fact that the screen doesn't change with no CPU is a good starting point and much better than having to delve inside the monitor. Worth just checking that 5V is still reaching the CPU but I'm sure others will be along shortly to advise on further investigations. Alan

10th Mar 2021, 4:37 pm	#849
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,574	Re: Non-working Commodore PET 3016 Yes, a great leap forward, and what a relief that the monitor appears to have survived despite having been exposed to all that damp. I did think Colin had already read the content of the character ROM but it's not a bad idea to read it again. (I didn't get that one to check with the others). I think it is reasonable to assume that the CPU is OK because we were able to execute NOPs, as shown by the fact that it ran smoothly through the address range when we did.

10th Mar 2021, 5:35 pm	#852
Mark1960 Octode Join Date: Mar 2020 Location: Kitchener, Ontario, Canada Posts: 1,297	Re: Non-working Commodore PET 3016 Are you sure that was UB3 ? What was the post number that was referenced? I think you need to compare the input pins with the output pins. One probe on pin 3, second on pin 2. Then pin 4 and pin 5 etc.

10th Mar 2021, 5:51 pm	#853
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,574	Re: Non-working Commodore PET 3016 UB3 and UC3, which are the address buffers, have been replaced. I wonder if Mark is maybe thinking of the data bus buffers UE9 and UE10?

10th Mar 2021, 6:00 pm	#855
Mark1960 Octode Join Date: Mar 2020 Location: Kitchener, Ontario, Canada Posts: 1,297	Re: Non-working Commodore PET 3016 Sorry, looking at the wrong schematic. Video data latch, UF9, 74ls373, if I’m looking at the right schematic now.

10th Mar 2021, 6:06 pm	#856
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,574	Re: Non-working Commodore PET 3016 We haven't worked in that area before, so UF9 has not been changed. It may be worth looking to see if there is an active 'video latch' signal going to UF9 pin 11. Depending on whether there is or is not, we'll proceed from there.

10th Mar 2021, 6:14 pm	#858
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,574	Re: Non-working Commodore PET 3016 A waveform, rather than the absence of one. Also while in that area, can you have a look for activity on pin 10 of either UF7 or UF8.

10th Mar 2021, 6:25 pm	#859
ScottishColin Octode Join Date: May 2012 Location: Perth, Scotland Posts: 1,816	Re: Non-working Commodore PET 3016 UF9/11 - consistent waveform at 1Mhz. It's a bit more rounded than square, but the timing, shape and frequency are consistent Nothing on either UF7/10 or UF8/10. Colin.

10th Mar 2021, 6:29 pm	#860
SiriusHardware Dekatron Join Date: Aug 2011 Location: Newcastle, Tyne and Wear, UK. Posts: 11,574	Re: Non-working Commodore PET 3016 OK on the 'Video Latch' signal. When you say you have nothing (presumably meaning no waveforms) on UF7(10) / UF8(10), what do you have instead? Steady high state or steady low state?