• Please review our updated Terms and Rules here

Looking for help with a Commodore Pet 2001

R

rglenn

Experienced Member
Joined
Feb 16, 2014
Messages
104
Location
PA
I am trying to restore a PET 2001. I started with a blank screen so I removed the socketed chips and re-inserted them. That produced a screen full of random characters which led me to pull the RAM chips (except 4 of them) and put them back one at a time until I found the troubled one. Well, I did find a bad RAM chip, removed it and it boots up now. The issue I have now is that it displays the boot screen verbiage with prompt 2x on the screen.

pet.jpg

any guidance would be appreciated.
 
rglenn said:
I am trying to restore a PET 2001. I started with a blank screen so I removed the socketed chips and re-inserted them. That produced a screen full of random characters which led me to pull the RAM chips (except 4 of them) and put them back one at a time until I found the troubled one. Well, I did find a bad RAM chip, removed it and it boots up now. The issue I have now is that it displays the boot screen verbiage with prompt 2x on the screen.

View attachment 49269

any guidance would be appreciated.
Click to expand...

This might be either a fault in the video RAM address multiplexer (very likely) or a couple of slow video RAMs (unlikely because they should be both slow, but I've seen it happen) if you have the 2001 revision with 6550 RAMs.
Do you have an oscilloscope or at least a logic probe?
In that case I can give you more precise suggestions based on signals that you should probe around the video RAM addressing logic.
If you search on youtube, you can also find my videos about repairing PETs (and 2001s in particular) to have more hints on how to proceed if you have some test instruments.
If you don't have them already, schematics can be found here:
ftp://www.zimmers.net/pub/cbm/schematics/computers/pet/index.html
First of all, try this command and report what's the output:
? PEEK(32768+512)

HTH
Frank IZ8DWF
 
Last edited:
+1 for an address line logic fault to the video RAM somewhere.

It could be a faulty RAM - but (as stated) you would have to have two faults - therefore unlikely.

What is more likely is that the first 'half' of the video RAM is being read twice by the video circuitry.

Dave
 
daver2 said:
+1 for an address line logic fault to the video RAM somewhere.

It could be a faulty RAM - but (as stated) you would have to have two faults - therefore unlikely.

What is more likely is that the first 'half' of the video RAM is being read twice by the video circuitry.

Dave
Click to expand...

Yes, the question is only: is it the counter 74177 or the multiplexer? Or one gate of the end of screen clear&reload signal? He needs either an oscilloscope or a logic probe... or start replacing quite a few chips.

Frank
 
I agree, D7 ( 74177 ) or D4 ( 74LS157 ). I'm tending towards the D4 or a bad connection at pin 14 of D4. My reasoning is that it isn't messing with the vertical sync. This needs a working counter D7. It is just repeating the first part of the address space. It looks to be linearly going through the address space until it gets the the higher address bits. This points to as fault on A9, the highest address. The address space looks to be written correctly, just not reading out the correct address.
All this points to D4, the address mux, or a bad connection of A9 to D4 ( I'm assuming the first schematic is what I'm referring to ).
Dwight
 
Dwight Elvey said:
I agree, D7 ( 74177 ) or D4 ( 74LS157 ). I'm tending towards the D4 or a bad connection at pin 14 of D4. My reasoning is that it isn't messing with the vertical sync. This needs a working counter D7. It is just repeating the first part of the address space. It looks to be linearly going through the address space until it gets the the higher address bits. This points to as fault on A9, the highest address. The address space looks to be written correctly, just not reading out the correct address.
All this points to D4, the address mux, or a bad connection of A9 to D4 ( I'm assuming the first schematic is what I'm referring to ).
Dwight
Click to expand...

Very good point about vertical sync!
Yes, it can be UD7 or UD4, but we don't yet know what PCB version he has :)

Frank
 
The result of the PEEK command above was 21
I do have the 2001 revision with 6550 RAMs
The board has PET MAIN LOGIC ASSEMBLY 320008 printed on it.

I am now getting some additional characters on the screen.
 
rglenn said:
The result of the PEEK command above was 21
I do have the 2001 revision with 6550 RAMs
The board has PET MAIN LOGIC ASSEMBLY 320008 printed on it.

I am now getting some additional characters on the screen.
Click to expand...
Whoops, aren't you getting a repeat of what you type in?
 
rglenn said:
The result of the PEEK command above was 21
I do have the 2001 revision with 6550 RAMs
The board has PET MAIN LOGIC ASSEMBLY 320008 printed on it.

I am now getting some additional characters on the screen.
Click to expand...

Do you have a way to check that UD4 Pin 12 is maybe stuck to 0V or almost and instead UD4 Pin 13 is toggling? That should be checked during power on reset sequence or when running a simple basic program like:

10 FOR A=32768 TO 33768
20 B=PEEK(A):NEXT
30 GOTO 10

Anyway, symptoms indicates UD4 as bad, if BA9 signal was bad, it would not even boot to basic correctly probably.

Frank IZ8DWF

EDIT: could even be one of UC3 or UC4 that have a shorted A9 input, so since they're socketed, I would try to swap both of them with other 6550s and see what happens.
 
It could still be a shorted input lead on one of the RAMs A9, pin 11. It is not as likely.
It is also possible that SA9 is stuck High as well. The symptoms just indicate that SA9 isn't changing and not which value it is stuck at.
Dwight
 
Dwight Elvey said:
It could still be a shorted input lead on one of the RAMs A9, pin 11. It is not as likely.
It is also possible that SA9 is stuck High as well. The symptoms just indicate that SA9 isn't changing and not which value it is stuck at.
Dwight
Click to expand...

Correct, SA9 might be stuck even to +5V
 
Reviving this very old thread. I didn't have confidence in my soldering skills to attempt this until now. I replaced D4 as suggested and it did in fact correct the screen doubling however, now I have random characters and a doubling when I type (though not the same character and 1 line away). As always, any suggestions would be appreciated.
 

Attachments

  • 20230818_123910.jpg
    20230818_123910.jpg
    2.2 MB · Views: 17
  • 20230818_124600.jpg
    20230818_124600.jpg
    2.2 MB · Views: 17
Note - the random characters were appearing before I replaced D4
 
Have you tried running my PETTESTER firmware?

This makes diagnosis of a PET somewhat easier...

You can't (unfortunately) just burn an EPROM for the 2001 without an adapter. If you have one of the ROMulators, some of these are equipped with my PETTESTER and will work in a 2001.

Dave
 
Which version of the 2008 have you actually got? EDIT: Ignore my question - 320008!

The EDIT ROM is at address $E000 to $E7FF. EDIT: H3.

Dave
 
Turns out I don't have the adapter needed. I'll see if I can buy one somewhere. So far I have found info and pics of them but not any actually offered for sale.
 
You must log in or register to reply here.
Back
Top