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 10th Apr 2022, 8:38 pm   #21
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Perth PET

I think Colin will have been using the on-board PETTester utility. A copy of the manual is attached.

Alan
Attached Files
File Type: pdf PETTester ManualV04.pdf (3.33 MB, 68 views)
ajgriff is offline  
Old 10th Apr 2022, 9:00 pm   #22
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

Right, it does sound like the Daver2 test code and I would hope the RAM / ROM device does ship with the current version which the manual in #21 refers to. A bit remiss of the device provider to include the test code but not to provide a link to the detailed manual for the test code, especially when the author of the test code made a considerable effort to write a good manual for it.

The assumption is obviously 'Oh look, using this fixes the RAM fault so I'll just keep on using this'.

Colin, can you post a photo of the actual screen / display that the test is stopping on? The results may require a degree of interpretation. Based on what the manual says and my understanding of what you have said so far, the test is getting through the video RAM checks and then failing the Page 0 / Page 1 memory test.
SiriusHardware is online now  
Old 10th Apr 2022, 9:18 pm   #23
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

I will do so, but it doesn't stop. I get a screen of the character set as per the PDF then a screen full of the capital B. It then cycles between the two

Colin.
ScottishColin is offline  
Old 10th Apr 2022, 9:31 pm   #24
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

OK, could we see pictures of BOTH of those screens please. If that doesn't make it clear we may need a video, but one step at a time.
SiriusHardware is online now  
Old 10th Apr 2022, 9:59 pm   #25
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

See attached photos - character set, full screen letter B, full screen letter G (when RAM is swapped out to the Tynemouth RAM).

Colin.
Attached Thumbnails
Click image for larger version

Name:	PXL_20220410_204522509.jpg
Views:	81
Size:	98.0 KB
ID:	255116   Click image for larger version

Name:	PXL_20220410_204524696.MP.jpg
Views:	74
Size:	46.4 KB
ID:	255117   Click image for larger version

Name:	PXL_20220410_204547753.jpg
Views:	72
Size:	139.3 KB
ID:	255118  
ScottishColin is offline  
Old 11th Apr 2022, 12:16 am   #26
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

Hm. I'm not sure that the version of PETTEST in that gizmo is V4, because the behaviour of the page 0 / page 1 test is not as described in the manual.

Whether the test is good or bad the display should not be the same character all the way through the screen, rather the first 256 characters should be either 'G' or 'B', maybe a mixture, and the following 256 characters should be 'dots' if all the characters written and read back were the same, and something other than a dot where the character which was read back from the location was different from the character written to it. The third group of 256 characters should be either G or B and the fourth group of characters should be all dots if the third group were all 'G'.

So when you test with the gizmo's 'good' RAM I expect you would see at the second step 256 Gs, 256 dots, 256 Gs, 256 dots. That is if I am reading the manual correctly, and the version of Pettest is the same as the one we have the manual for.
SiriusHardware is online now  
Old 11th Apr 2022, 7:18 am   #27
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Perth PET

I’m sure Sirius is right and the Tynemouth board doesn’t have the latest version of PETTester installed. Also I think that one of Colin’s missing EPROMs does include V4 of Daver2’s test routine although my memory may be playing tricks. If it helps the V4 files can be found here:


Alan
ajgriff is offline  
Old 11th Apr 2022, 8:01 am   #28
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

Thanks for the reminder. I think we came to the conclusion that due to a quirk in the way the wiring for UD8 falls, combined with the way it is addressed, we could put it into a 'normal' EPROM like a 2716. I'll have to read back through the original thread. I know the Slothie test code (at UD9) definitely needed an adaptor.

I'll give Colin a day or so more to try to find the original ICs, in the meantime, Colin, you can make a start by measuring the supply voltages to the relevant pins of all of the RAM ICs, just in case it is something as simple as the loss of one of those supplies to the (or to some of the RAM). I think you know your way around the circuit now, but if in doubt, shout out.
SiriusHardware is online now  
Old 11th Apr 2022, 10:59 am   #29
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

I went back through the epic thread and found my original thinking-out-loud about the test EPROMs - Daver2's V4 code can go into a 2716 and be fitted directly as UD8, Slothie's (which has the specific very useful property of continually retrying the same location when it encounters a memory write-read fail) will have to go into a 2716 in an adaptor or possibly into one of the handful of Hitachi 2532s I acquired a while back, those should fit directly into UD9 position.

Unfortunately Easter is happening at a remarkably inconvenient time this year: I am being carted off to visit relatives I don't even know, so I may not be back into the usual rhythm of things until after the Easter break.
SiriusHardware is online now  
Old 11th Apr 2022, 1:43 pm   #30
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

All RAM chips (2114s) tested across pins 9/18 - all give 11.87v.

Are there other pins to test on the 2114s?

Colin.

Quote:
Originally Posted by SiriusHardware View Post
Thanks for the reminder. I think we came to the conclusion that due to a quirk in the way the wiring for UD8 falls, combined with the way it is addressed, we could put it into a 'normal' EPROM like a 2716. I'll have to read back through the original thread. I know the Slothie test code (at UD9) definitely needed an adaptor.

I'll give Colin a day or so more to try to find the original ICs, in the meantime, Colin, you can make a start by measuring the supply voltages to the relevant pins of all of the RAM ICs, just in case it is something as simple as the loss of one of those supplies to the (or to some of the RAM). I think you know your way around the circuit now, but if in doubt, shout out.
ScottishColin is offline  
Old 11th Apr 2022, 2:32 pm   #31
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

Um, weren't the main RAMs 4108s in that machine? I certainly remember discussions about whether it would be OK to replace them with pin-compatible 4116s.
SiriusHardware is online now  
Old 11th Apr 2022, 3:00 pm   #32
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

I was just checking to see if you were paying attention....

Of course you're right - they are 4108s.

Colin.


Quote:
Originally Posted by SiriusHardware View Post
Um, weren't the main RAMs 4108s in that machine? I certainly remember discussions about whether it would be OK to replace them with pin-compatible 4116s.
ScottishColin is offline  
Old 11th Apr 2022, 3:04 pm   #33
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

Power connections to the main RAMs (if they are 4108s) is
Pin 16 = GND / 0V
Pin 8 = +12V
pin 9 = +5V
pin 1 = -5V

So you need to measure the voltages on pins 1, 9, 8 with the black lead always on pin 16. Repeat for each 4108 chip.
SiriusHardware is online now  
Old 11th Apr 2022, 3:34 pm   #34
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

I got that a bit wrong didn't I...

All check out OK after a bit of scraping.

Pin 8 - all +11.97
Pin 9 - all +4.97
Pin 1 - all -5.037

Colin.
ScottishColin is offline  
Old 11th Apr 2022, 10:33 pm   #35
SiriusHardware
Dekatron
 
Join Date: Aug 2011
Location: Newcastle, Tyne and Wear, UK.
Posts: 11,482
Default Re: Perth PET

OK, I think the best way forward is for me to make replacement diagnostic EPROMs because that's the easiest way to get steady state read / write waveforms to look at under fault conditions. The way the Gizmo's built in test routine cycles between video test / RAM test, that's not ideal.

As I mentioned I am heavily involved with family over the next two weeks so this may be delayed. In the meantime maybe you can get the machine working by enabling the 'Gizmo' RAM, try out the SD2PET - if you actually did get one - and give us your impressions about that?
SiriusHardware is online now  
Old 11th Apr 2022, 11:03 pm   #36
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

Enjoy the time with the family....

And thanks very much for volunteering to sort the test EPROMs. If you need to buy anything please let me know. I promise not to lose them this time.

I'll start a new thread for the SD2PET.

Colin.


Quote:
Originally Posted by SiriusHardware View Post
OK, I think the best way forward is for me to make replacement diagnostic EPROMs because that's the easiest way to get steady state read / write waveforms to look at under fault conditions. The way the Gizmo's built in test routine cycles between video test / RAM test, that's not ideal.

As I mentioned I am heavily involved with family over the next two weeks so this may be delayed. In the meantime maybe you can get the machine working by enabling the 'Gizmo' RAM, try out the SD2PET - if you actually did get one - and give us your impressions about that?
ScottishColin is offline  
Old 14th Apr 2022, 11:34 am   #37
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Perth PET

I'm aware that Sirius has been investigating PET testers via the VCFED forum and will leave it to him to update us when he gets the opportunity.

In the the meantime I wonder if Colin could confirm that the PET works ok with the Tynemouth board's RAM activated but with the Tynemouth ROMs de-activated? In other words are the PET's ROMs still operating correctly? Just trying to eliminate some variables.

Alan
ajgriff is offline  
Old 14th Apr 2022, 12:26 pm   #38
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

I saw the thread but decided to leave well alone otherwise we'll end up with suggestions both here and there.

In answer to your question, yes - Tynemouth RAM/onboard ROM combination works.

I am getting a new and odd cursor being left on the screen (see attachment), but I think I'll try (with help) to fix the current issues one at a time.....

Colin.


Quote:
Originally Posted by ajgriff View Post
I'm aware that Sirius has been investigating PET testers via the VCFED forum and will leave it to him to update us when he gets the opportunity.

In the the meantime I wonder if Colin could confirm that the PET works ok with the Tynemouth board's RAM activated but with the Tynemouth ROMs de-activated? In other words are the PET's ROMs still operating correctly? Just trying to eliminate some variables.

Alan
Attached Thumbnails
Click image for larger version

Name:	PXL_20220414_111912750.jpg
Views:	58
Size:	86.0 KB
ID:	255329  
ScottishColin is offline  
Old 14th Apr 2022, 6:05 pm   #39
ajgriff
Nonode
 
ajgriff's Avatar
 
Join Date: May 2004
Location: Halifax, West Yorkshire, UK.
Posts: 2,583
Default Re: Perth PET

Would I be right to assume that the cursor issue disappears when using the Tynemouth board ROM?

Alan
ajgriff is offline  
Old 15th Apr 2022, 10:41 am   #40
ScottishColin
Octode
 
Join Date: May 2012
Location: Perth, Scotland
Posts: 1,762
Default Re: Perth PET

No. It happens with all versions of the ROMs on the Tynemouth board as well as the on-board ROMs. I've tried a different 6503 and that doesn't fix it either.

Colin.

Quote:
Originally Posted by ajgriff View Post
Would I be right to assume that the cursor issue disappears when using the Tynemouth board ROM?

Alan
ScottishColin is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:59 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.