Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Im sorry but i don t understand what can i do with pettester Documentation
- VCF South West - June 14 - 16, Davidson-Gundy Alumni Center at University of Texas at Dallas
- VCF West - Aug 2 - 3, Computer History Museum, Mountain View, CA
- VCF Midwest - Sept 7 - 8 2024, Schaumburg, IL
- VCF SoCal - Mid February 2025, Location TBD, Southern CA
- VCF East - April 2025, Infoage Museum, Wall NJ
-
Please review our updated Terms and Rules here
Cbm pet 8296-d black screen
- Thread starter Desperado
- Start date
daver2
10k Member
I wouldn't remove the upper bank of RAM if they are soldered in. If they are in sockets - yes. If soldered in no.
Good work on piggy-backing UE9 and UE10 and replacing them. So, it looks like they were actually faulty - and this was likely the cause of the PETTESTER page 0 and 1 RAM test faults that were identified.
You have now passed the VIDEO RAM test (test 1) and the page 0 and 1 RAM test (test 2) and are on to the ROM and keyboard test.
Read the PETTESTER manual to identify what this test is indicating please (the section headed "ROM/KBD TEST").
The ROM checksum indicates to me that you have BASIC 4 installed - and the checksums are correct for ROMs B, C, D and F. We can't checkout the EDIT ROM - because we have replaced it with the PETTESTER. Look for the value of the checksums changing whilst the countdown is counting down. From the video you posted I can't see any initially.
The keyboard displays (ff) are all wrong. They should be (00). This indicates to me that either UB12 is duff - or the physical keyboard is faulty. Physically disconnect the keyboard cable from the PET mainboard and see if the values go to (00). If they do, the keyboard needs fixing. If they don't, UB12 requires replacing.
From this screen you can press each key in turn on the keyboard - and you should see one (and only one) bit of the row of 00's change to a '1' indicating that the keypress was detected. The bit should return to a '0' again when you release the key.
I suspect that what you are observing on the VDU screen is the problem that you were having before - and this fault is temperature dependent. When you first turn the PET on, all the chips are cold - and the display works. As the chips heat up - one of them gets 'upset' and starts to misbehave. As I have stated before, it is either a chip you have replaced (in which case replace each of the chips that you have replaced - that should be in IC sockets - IN TURN and see if the problem goes away). If the problem DOESN'T go away, put the original chip back in the socket and continue for all the other devices you have replaced.
If the problem still doesn't go away, get yourself a can of freezer spray and spray each individual IC on the VDU section of the circuitry (schematics 7, 8, 9 and 10) and see if the problem goes away when you spray a specific chip. You have to do it one IC at a time and make sure you just spray one and not multiple chips. Build a little paper or thin cardboard 'shield' to prevent spraying multiple chips. You will need to do this as soon as the problem starts - so it could take a while. Beware that the problem could cause the PETTESTER to FAIL - in which case you may not observe any effect on the screen when you spray the faulty chip - and thus miss it.
Just out of interest, did you let the PETTESTER run for a while after the video fault started?
When the countdown gets to 0, it should start testing the dynamic RAM, so it would be interesting if it did actually start that test.
If course, you can check the CPU SYNC pin to see if the processor is still executing instructions, and pin 20 of UD7 (the PETTESTER ROM) to see if it executing instructions out of that specific ROM. This will demonstrate that the CPU-side of the PET is still running - but it is the VDU-side that is corrupt.
Incidentally, did you check your display against figure 7 of my PETTESTER manual to make sure what you are seeing corresponds to my documentation?
Dave
Good work on piggy-backing UE9 and UE10 and replacing them. So, it looks like they were actually faulty - and this was likely the cause of the PETTESTER page 0 and 1 RAM test faults that were identified.
You have now passed the VIDEO RAM test (test 1) and the page 0 and 1 RAM test (test 2) and are on to the ROM and keyboard test.
Read the PETTESTER manual to identify what this test is indicating please (the section headed "ROM/KBD TEST").
The ROM checksum indicates to me that you have BASIC 4 installed - and the checksums are correct for ROMs B, C, D and F. We can't checkout the EDIT ROM - because we have replaced it with the PETTESTER. Look for the value of the checksums changing whilst the countdown is counting down. From the video you posted I can't see any initially.
The keyboard displays (ff) are all wrong. They should be (00). This indicates to me that either UB12 is duff - or the physical keyboard is faulty. Physically disconnect the keyboard cable from the PET mainboard and see if the values go to (00). If they do, the keyboard needs fixing. If they don't, UB12 requires replacing.
From this screen you can press each key in turn on the keyboard - and you should see one (and only one) bit of the row of 00's change to a '1' indicating that the keypress was detected. The bit should return to a '0' again when you release the key.
I suspect that what you are observing on the VDU screen is the problem that you were having before - and this fault is temperature dependent. When you first turn the PET on, all the chips are cold - and the display works. As the chips heat up - one of them gets 'upset' and starts to misbehave. As I have stated before, it is either a chip you have replaced (in which case replace each of the chips that you have replaced - that should be in IC sockets - IN TURN and see if the problem goes away). If the problem DOESN'T go away, put the original chip back in the socket and continue for all the other devices you have replaced.
If the problem still doesn't go away, get yourself a can of freezer spray and spray each individual IC on the VDU section of the circuitry (schematics 7, 8, 9 and 10) and see if the problem goes away when you spray a specific chip. You have to do it one IC at a time and make sure you just spray one and not multiple chips. Build a little paper or thin cardboard 'shield' to prevent spraying multiple chips. You will need to do this as soon as the problem starts - so it could take a while. Beware that the problem could cause the PETTESTER to FAIL - in which case you may not observe any effect on the screen when you spray the faulty chip - and thus miss it.
Just out of interest, did you let the PETTESTER run for a while after the video fault started?
When the countdown gets to 0, it should start testing the dynamic RAM, so it would be interesting if it did actually start that test.
If course, you can check the CPU SYNC pin to see if the processor is still executing instructions, and pin 20 of UD7 (the PETTESTER ROM) to see if it executing instructions out of that specific ROM. This will demonstrate that the CPU-side of the PET is still running - but it is the VDU-side that is corrupt.
Incidentally, did you check your display against figure 7 of my PETTESTER manual to make sure what you are seeing corresponds to my documentation?
Dave
Last edited:
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Hi Dave, keyboard was sconnected from board !
daver2
10k Member
>>> Hi Dave, keyboard was sconnected from board !
So that means that UB12 is probably dead. Replace it and see if you get a row of 00's displayed instead of ff's.
>>> Unfortunately, at moment, i don't have any ice spray
Order some.
These are expensive https://uk.rs-online.com/web/c/faci...-cleaners-protective-coatings/freezer-sprays/
Cheaper alternatives are available - but make sure it is designed for electronics use as opposed to anything else.
>>> Can i use ice covered by plastic maybe???
Unlikely, it is not cold enough.
Dave
So that means that UB12 is probably dead. Replace it and see if you get a row of 00's displayed instead of ff's.
>>> Unfortunately, at moment, i don't have any ice spray
Order some.
These are expensive https://uk.rs-online.com/web/c/faci...-cleaners-protective-coatings/freezer-sprays/
Cheaper alternatives are available - but make sure it is designed for electronics use as opposed to anything else.
>>> Can i use ice covered by plastic maybe???
Unlikely, it is not cold enough.
Dave
daver2
10k Member
Well done...
I would suspect a bad UB12.
It is interesting that the display now behaves itself though...
UB12 also has one line (pin 18 VERT DRIVE) dedicated to the VDU display.
The DRAM test has seen 32k - which is correct to the amount of RAM installed.
Let the DRAM test run - again, read the PETTESTER document to see what 'good' looks like when it is running...
Some of these individual tests take a while to run - the tests are fairly exhaustive - so just be patient whilst they run.
If you get to this test, I suggest you leave it run for an extended time in the corner of the room whilst you drink some vino and do something else! The longer it runs the DRAM test successfully - the more convinced we will both be that the PET is now pretty stable.
After that, put the EDIT ROM back in place of the PETTESTER, cross your fingers and turn on! Of course, the keyboard will not work (no UB12) and you probably won't get a flashing cursor either, but the screen should clear, you should get the little 'ditty tune' played and BASIC should announce itself with the correct amount of RAM. Get a replacement UB12 and then we can test the keyboard out properly...
Dave
I would suspect a bad UB12.
It is interesting that the display now behaves itself though...
UB12 also has one line (pin 18 VERT DRIVE) dedicated to the VDU display.
The DRAM test has seen 32k - which is correct to the amount of RAM installed.
Let the DRAM test run - again, read the PETTESTER document to see what 'good' looks like when it is running...
Some of these individual tests take a while to run - the tests are fairly exhaustive - so just be patient whilst they run.
If you get to this test, I suggest you leave it run for an extended time in the corner of the room whilst you drink some vino and do something else! The longer it runs the DRAM test successfully - the more convinced we will both be that the PET is now pretty stable.
After that, put the EDIT ROM back in place of the PETTESTER, cross your fingers and turn on! Of course, the keyboard will not work (no UB12) and you probably won't get a flashing cursor either, but the screen should clear, you should get the little 'ditty tune' played and BASIC should announce itself with the correct amount of RAM. Get a replacement UB12 and then we can test the keyboard out properly...
Dave
Last edited:
daver2
10k Member
Is your EDIT ROM faulty?
Can you download the correct EDIT ROM file from Bo Zimmers's website and use your EPROM programmer to check the EDIT ROM out.
Dave
Can you download the correct EDIT ROM file from Bo Zimmers's website and use your EPROM programmer to check the EDIT ROM out.
Dave
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Is 901474-04 correct edit rom?
daver2
10k Member
Yep (assuming your mains electricity supply is 50 Hz).
Dave
Dave
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
I verifyied now and edit rom is ok....
daver2
10k Member
Can you put the PETTESTER back in place of the EDIT ROM please.
Power up the PET and video the PETTESTER screen please all the way from the RESET to the first pass of the memory test.
However, if you start videoing from a power-up condition, we will miss the first part of the test. You can temporarily short pins 1 and 2 of the 8-pin LM555 timer IC (UD16) together to generate a 'soft' reset.
Please hold your video camera as steady as you can (ideally on a tripod or supported by something to keep your hand steady) and ensure that autofocus and all of the clever features are switched OFF. I want to see what the display looks like, and I can't do that if the camera is constantly autofocusing.
You may have to enlist some help for this to make your life easier? The other option is to solder a normally-open pushbutton between pins 1 and 2 of UD16 and then you can easily soft reset the machine rather than power cycling it.
Dave
Power up the PET and video the PETTESTER screen please all the way from the RESET to the first pass of the memory test.
However, if you start videoing from a power-up condition, we will miss the first part of the test. You can temporarily short pins 1 and 2 of the 8-pin LM555 timer IC (UD16) together to generate a 'soft' reset.
Please hold your video camera as steady as you can (ideally on a tripod or supported by something to keep your hand steady) and ensure that autofocus and all of the clever features are switched OFF. I want to see what the display looks like, and I can't do that if the camera is constantly autofocusing.
You may have to enlist some help for this to make your life easier? The other option is to solder a normally-open pushbutton between pins 1 and 2 of UD16 and then you can easily soft reset the machine rather than power cycling it.
Dave
daver2
10k Member
Sorry, but you haven't done what I have asked you to do.
I would like you to FULLY record from the soft reset through the various test screens, through the countdown screen (counting down to 0) and then through the first FULL memory test (until it says "pass 000001.".
The fact that there is not a full screen of "g" characters is actually irrelevant PROVIDING it matches the screen display within the PETTESTER documentation. I keep saying to check the documentation...
Yes, the video recording is of a sufficiently good quality to see what I need to see.
You can step and pause the video you take to see exactly what the PETTESTER screens are showing. I really must increase the delay between screens some time. Perhaps for the next release!?
Dave
I would like you to FULLY record from the soft reset through the various test screens, through the countdown screen (counting down to 0) and then through the first FULL memory test (until it says "pass 000001.".
The fact that there is not a full screen of "g" characters is actually irrelevant PROVIDING it matches the screen display within the PETTESTER documentation. I keep saying to check the documentation...
Yes, the video recording is of a sufficiently good quality to see what I need to see.
You can step and pause the video you take to see exactly what the PETTESTER screens are showing. I really must increase the delay between screens some time. Perhaps for the next release!?
Dave
daver2
10k Member
That, my friend, is spot on and there is nothing on that video that I am concerned with 
! Thank you.
The only thing wrong is the keyboard display codes - but that is expected at this time because UB12 is missing.
You have two (2) 6520's on the PET mainboard. The IEEE488 interface uses one at UB16. I would be inclined to remove UB16 and put that into UB12. This will allow you to test the keyboard with the PETTESTER at this stage. Wait until you get the "countdown" message. You should observe a line of 00's for the keyboard display with no keys pressed and you should see a single 1, 2, 4 or 8 display on one of the digits when you press a key. It doesn't matter which keys you press, just run across the keyboard keys from top left of the keyboard to bottom right and observe that one digit changes for each keypress.
If that is OK - then you have a working UB12 - but a missing UB16 that you need to replace.
If that doesn't work correctly, then that indicates another faulty 6520!
Dave
! Thank you.The only thing wrong is the keyboard display codes - but that is expected at this time because UB12 is missing.
You have two (2) 6520's on the PET mainboard. The IEEE488 interface uses one at UB16. I would be inclined to remove UB16 and put that into UB12. This will allow you to test the keyboard with the PETTESTER at this stage. Wait until you get the "countdown" message. You should observe a line of 00's for the keyboard display with no keys pressed and you should see a single 1, 2, 4 or 8 display on one of the digits when you press a key. It doesn't matter which keys you press, just run across the keyboard keys from top left of the keyboard to bottom right and observe that one digit changes for each keypress.
If that is OK - then you have a working UB12 - but a missing UB16 that you need to replace.
If that doesn't work correctly, then that indicates another faulty 6520!
Dave
Desperado
Veteran Member
- Joined
- Nov 25, 2017
- Messages
- 6,827
Ok Dave, thanks again!That, my friend, is spot on and there is nothing on that video that I am concerned with
The only thing wrong is the keyboard display codes - but that is expected at this time because UB12 is missing.
You have two (2) 6520's on the PET mainboard. The IEEE488 interface uses one at UB16. I would be inclined to remove UB16 and put that into UB12. This will allow you to test the keyboard with the PETTESTER at this stage. Wait until you get the "countdown" message. You should observe a line of 00's for the keyboard display with no keys pressed and you should see a single 1, 2, 4 or 8 display on one of the digits when you press a key. It doesn't matter which keys you press, just run across the keyboard keys from top left of the keyboard to bottom right and observe that one digit changes for each keypress.
If that is OK - then you have a working UB12 - but a missing UB16 that you need to replace.
If that doesn't work correctly, then that indicates another faulty 6520!
Dave
Maybe can i have non compatible edit rom for this pet???