• Please review our updated Terms and Rules here

Commodore PET 2001-8 stuck on garbled screen

Unfortunately not. I would like to going forward but not found a device yet
OK... well the BASICs clear the screen as one of their first jobs... so you could select a BASIC and just keep pressing the reset button whilst you look for life at pin 9 of G2

If that doesn't show something then maybe contention on the data bus? You have removed all the ROM, RAM, PIAs and VIA?
 
OK... well the BASICs clear the screen as one of their first jobs... so you could select a BASIC and just keep pressing the reset button whilst you look for life at pin 9 of G2

If that doesn't show something then maybe contention on the data bus? You have removed all the ROM, RAM, PIAs and VIA?
Yes everything removed. Apart from character set video ram x2 and pet rom ram board. Do you mean press the reset on my Tynemouth board.
 
You don't have any of my devices. :)

Anyway...

You could also usefully scope all the data and address lines of the 6502 and check the traces are nicely square-ish and 5V peak-ish.

View attachment 1249767
I'll mention pins with square waves as others are just a line.
Video ram sockets have now been replaced.
Will do character rom when socket arrives.
Probe earth had become detached on all of these pics apart from pin 7 pic further down post #49
Pin 16 pic 1
Pin 14 pic 2
Pin 13 pic 3
Pin 12 pic 4
Pin 11 pic 5
Pin 10 pic 6
Pin 9 pic 7
Pin 23 pic 8
Pin 24 pic 9
Pin 25 pic 10
 

Attachments

  • 16705225920587439470519580498263.jpg
    16705225920587439470519580498263.jpg
    988.3 KB · Views: 0
  • 16705225605056860229427297293214.jpg
    16705225605056860229427297293214.jpg
    1.1 MB · Views: 0
  • 16705225216487257884793562453212.jpg
    16705225216487257884793562453212.jpg
    1.1 MB · Views: 0
  • 1670522477627172209160538703046.jpg
    1670522477627172209160538703046.jpg
    769.5 KB · Views: 0
  • 16705224424415905567712633031731.jpg
    16705224424415905567712633031731.jpg
    1.1 MB · Views: 0
  • 16705224287781979887499728578516.jpg
    16705224287781979887499728578516.jpg
    972.3 KB · Views: 0
  • 16705223948567026040684972818823.jpg
    16705223948567026040684972818823.jpg
    953 KB · Views: 0
  • 16705223145324349997653323129355.jpg
    16705223145324349997653323129355.jpg
    1 MB · Views: 0
  • 16705221845575629431587837157712.jpg
    16705221845575629431587837157712.jpg
    988.1 KB · Views: 0
  • 16705220922868355980686150098805.jpg
    16705220922868355980686150098805.jpg
    845.4 KB · Views: 0
Last edited:
Pin 26 to 33 in order
 

Attachments

  • 16705229587723793736720658862860.jpg
    16705229587723793736720658862860.jpg
    1.1 MB · Views: 0
  • 1670522933405372846996717553786.jpg
    1670522933405372846996717553786.jpg
    1.1 MB · Views: 0
  • 16705229151304199053481033930024.jpg
    16705229151304199053481033930024.jpg
    1 MB · Views: 0
  • 16705228929277414035215794716126.jpg
    16705228929277414035215794716126.jpg
    1.1 MB · Views: 0
  • 16705228814951575258225408482421.jpg
    16705228814951575258225408482421.jpg
    1.1 MB · Views: 0
  • 16705228709482926286853828983380.jpg
    16705228709482926286853828983380.jpg
    1.1 MB · Views: 0
  • 16705228610774087508741975236766.jpg
    16705228610774087508741975236766.jpg
    1.1 MB · Views: 0
  • 16705228452793028075978895189864.jpg
    16705228452793028075978895189864.jpg
    1 MB · Views: 0
Replaced CPU character set and ram sockets now.
I am quite pleased with my removal process. I have definitely improved.
Some improvement to performance but I am suspicious of all the ram that I have. Still not getting a full character set. Even with the two new substitutes in adapters are not working correctly so I have ordered two more just to be sure as I have no ic tester. If they don't produce the correct result then there must be a fault elsewhere.
 

Attachments

  • DSC_1141.JPG
    DSC_1141.JPG
    4.2 MB · Views: 11
  • DSC_1140.JPG
    DSC_1140.JPG
    3.9 MB · Views: 10
OK... well the BASICs clear the screen as one of their first jobs... so you could select a BASIC and just keep pressing the reset button whilst you look for life at pin 9 of G2

If that doesn't show something then maybe contention on the data bus? You have removed all the ROM, RAM, PIAs and VIA?
Ok so I have just been looking at G2 pin 9.
Pressing the reset button on the pet rom ram device for a split second I get abit of a activity. If I keep pressing and keep my finger on the reset button I sometimes get the following image. Only when I keep the reset button pressed down.
Pic 2 is where I am at now display wise. This is using a pair of ram that display the following on there own when placed in C4 one at a time. Together I get pic 2.
 

Attachments

  • DSC_1154.JPG
    DSC_1154.JPG
    1.1 MB · Views: 12
  • 16707708881666075291708368758564.jpg
    16707708881666075291708368758564.jpg
    2.6 MB · Views: 19
  • 16707709661772858311909175731922.jpg
    16707709661772858311909175731922.jpg
    2.8 MB · Views: 19
  • 167077099082670696891259658144.jpg
    167077099082670696891259658144.jpg
    2.6 MB · Views: 16
As an aside and totally unrelated to your fault, that image on the screen with all the left pointing arrows shows up nicely the H scan linearity defects which are intrinsic to a design without a magnetic linearity coil, with the character stretched on the left hand side and a little compressed on the right hand side compared to the center, making the biggest difference in the length of the arrow, comparing the extreme left and right sides. I my VDU I improved this by adding an additional diode.
 
Maybe don't hold down the RESET button... I'm not sure how it is implemented on that board but it might not be a good idea to keep it pressed.

From the trace it seems that some decoding is going on.

I believe you have multiple PETs... perhaps you should test the Tynemouth Board in another PET to remove that as an unknown.

I would be tempted to burn a 6540 compatible ROM with daver2's diagnostic (The F800 variant) and put it in H7
and/or build a NOP Adapter and test the addressing. That would be my approach.

I'll leave it to others to make further suggestions.
 
Had some new ram come today for the video ram slots and this is the results with two different sets of these ram replacements. Seems to be worse in that I don't get any character. The replacements are MM2114N in adapters. Pic 3
 

Attachments

  • DSC_1166.JPG
    DSC_1166.JPG
    3.3 MB · Views: 5
  • DSC_1165.JPG
    DSC_1165.JPG
    2.4 MB · Views: 4
  • DSC_1164.JPG
    DSC_1164.JPG
    2.4 MB · Views: 5
Maybe don't hold down the RESET button... I'm not sure how it is implemented on that board but it might not be a good idea to keep it pressed.

From the trace it seems that some decoding is going on.

I believe you have multiple PETs... perhaps you should test the Tynemouth Board in another PET to remove that as an unknown.

I would be tempted to burn a 6540 compatible ROM with daver2's diagnostic (The F800 variant) and put it in H7
and/or build a NOP Adapter and test the addressing. That would be my approach.

I'll leave it to others to make further suggestions.
Tynemouth board works fine in my other pet.
 

Attachments

  • 16709474251244412580458673627877.jpg
    16709474251244412580458673627877.jpg
    800.9 KB · Views: 3
That's interesting. As far as your video RAM is concerned.... have you tested the 2114s? I've seen many non-working (possibly fake) 2114s over the last year or so... It might be worth randomly picking a pair of 6550s and using those. The fact we have seen rendered characters suggests the character ROM is working... Did you prove the old one wasn't?

I'm tempted to guess that the CPU is working with the ROM/RAM board... just not sure how we prove it...

To write to the video RAM requires the CPU to get through two sets of buffers...

1670962269274.png

The G5/G6 buffers and the B3/B4 buffers and at the same time make sure the H8/H9 buffers aren't interfering.

The next step is easy with a NOP generator as the address bus changes in a known way, however with a program running there should be some activity (but perhaps not across all the bits).

You could replace the ROM/RAM board and then look at E5, E4, E3 and E2 which are used for buffering the address bus.


1670962661505.png

Obviously all bits matter but BA11 might be a good place to start. Put your logic probe over E5 (and E4 and E3 and E2) and look for stuck bits.

Note: E5 is a set of inverters, E4 and E3 are non-inverting buffers, E2 is a set of AND gates. That's why BA11 goes through 2 gates... NOT NOT... obviously they had some spare gates at E5.
 

Attachments

  • 1670962641093.png
    1670962641093.png
    412.2 KB · Views: 2
Last edited:
That's interesting. As far as your video RAM is concerned.... have you tested the 2114s? I've seen many non-working (possibly fake) 2114s over the last year or so... It might be worth randomly picking a pair of 6550s and using those. The fact we have seen rendered characters suggests the character ROM is working... Did you prove the old one wasn't?

I'm tempted to guess that the CPU is working with the ROM/RAM board... just not sure how we prove it...

To write to the video RAM requires the CPU to get through two sets of buffers...

View attachment 1250076

The G5/G6 buffers and the B3/B4 buffers and at the same time make sure the H8/H9 buffers aren't interfering.

The next step is easy with a NOP generator as the address bus changes in a known way, however with a program running there should be some activity (but perhaps not across all the bits).

You could replace the ROM/RAM board and then look at E5, E4, E3 and E2 which are used for buffering the address bus.


View attachment 1250078

Obviously all bits matter but BA11 might be a good place to start. Put your logic probe over E5 (and E4 and E3 and E2) and look for stuck bits.

Note: E5 is a set of inverters, E4 and E3 are non-inverting buffers, E2 is a set of AND gates. That's why BA11 goes through 2 gates... NOT NOT... obviously they had some spare gates at E5.
So should I wait until the nop adapter arives or can I look at this now.
If bits are stuck does that mean just hi or lo readings without pulsing.
I have contacted the seller regarding the ram.
I think my BBC micro model B uses the same ram so I may try them in that.
 
Last edited:
So should I wait until the nop adapter arives or can I look at this now.
If bits are stuck does that mean just hi or lo readings without pulsing.
I have contacted the seller regarding the ram.
I think my BBC micro model B uses the same ram so I may try them in that.
You can look at the gates with your probe. I'm especially interested in E5.
 
OK... so now you have a NOP generator! Plug it in... make sure it the right way up and has the 6502 the right way up!

You should get regular flashing on the LED attached to A15

Now warm up your scope...

Wander your scope around the address lines that are exposed at H1 which is probably hopefully empty right now...

H1 pins 5, 6, 7, 8, 9, 10, 15, 14, 13, 11 and 18... we are interested in seeing nice regular frequencies

1671654227654.png

(If you are lazy... just use your logic probe)
 
This is what we are looking for....

1671655299186.png

e.g. (Not sure why I have two traces... but you get the idea)...

A0

A0.jpg

A10

A10.jpg
 
Last edited:
Back
Top