• Please review our updated Terms and Rules here

Commoodore Pet Cbm 3016

Desperado

Veteran Member
Joined
Nov 25, 2017
Messages
6,827
Hi guys!
I'm back very soon ;)
I hope this is the right time....
I told you in the other post of this pet ...This is 3016 Pet and when i turn on,on the screen there are only random characters and symbols ...
I suspect bad eprom....or video ram...
What do you think is the problem?
Thanks for any suggestion!
IMG-20190528-WA0003.jpg
 
This is the board....what iso must i download to verify if eprom are correct please?
Thanks
20190528_184142.jpg
 
This is the board....what iso must i download to verify if eprom are correct please?
Thanks
Welcome back!

In the other thread you asked if a monitor from an 8032 would work with your 2001 (no); does that mean that you have access to a working 8032 by any chance? That would be the easiest way, putting the 3016 ROMs into one of the spare 8032 sockets and calculating a checksum.

Failing that, do you have the means to burn an EPROM (2716 or 2532) and a blank EPROM? If so, there are several diagnostic ROMs that would check RAM and ROM for you.

Meanwhile, carefully move the ROMs and the other socketed chips back and forth in their sockets in case there's some oxidation on the pins.

FYI, when looking for schematics etc. this model is also known as a 2001N and, to add to the confusion, also as a 4016/32 (not to be confused with the 'FAT40' 4032 which is essentially a 40 column 8032).

Hopefully we'll get this one going with not quite so many posts ;-)
 
Last edited:
Welcome back!

In the other thread you asked if a monitor from an 8032 would work with your 2001 (no); does that mean that you have access to a working 8032 by any chance? That would be the easiest way, putting the 3016 ROMs into one of the spare 8032 sockets and calculating a checksum.

Failing that, do you have the means to burn an EPROM (2716 or 2532) and a blank EPROM? If so, there are several diagnostic ROMs that would check RAM and ROM for you.

Meanwhile, carefully move the ROMs and the other socketed chips back and forth in their sockets in case there's some oxidation on the pins.

FYI, when looking for schematics etc. this model is also known as a 2001N and, to add to the confusion, also as a 4016/32 (not to be confused with the 'FAT40' 4032 which is essentially a 40 column 8032).

Hopefully we'll get this one going with not quite so many posts ;-)

Hi Mike,
yes i ve an eprom programmer and some empy eprom...
Where i can find diagnostic please?
Thanks!
 
Good evening,
i burned pet tester and i putted in Ud9...ram seems to be god....
I verified all roms with eprom programmer and are ok.
Maybe must i change all white socket?
Thanks!
Ps:i just washed main board ;)20190529_181023.jpg,20190529_181026.jpg,20190529_204448.jpg
 
I'd suggest testing the continuity between the address line pins of UD9 and the remaining ROM sockets down the row. (They're mostly wired in parallel, only the select lines on pin 20 are different.) The way the PET board is laid out the address and data lines enter the row from opposite ends; since the PET tester ROM worked you can be sure the data line traces are okay for the whole row, but there's a possibility you have a broken address line somewhere between D9 and a downstream socket.

(A bad socket is probably a more likely possibility, but the PET board I repaired years ago was chock full of bad traces so I ran into exactly this problem.)
 
I'd suggest testing the continuity between the address line pins of UD9 and the remaining ROM sockets down the row. (They're mostly wired in parallel, only the select lines on pin 20 are different.) The way the PET board is laid out the address and data lines enter the row from opposite ends; since the PET tester ROM worked you can be sure the data line traces are okay for the whole row, but there's a possibility you have a broken address line somewhere between D9 and a downstream socket.

(A bad socket is probably a more likely possibility, but the PET board I repaired years ago was chock full of bad traces so I ran into exactly this problem.)


Thanks so much!
Tomorrow i'll check continuity on roms socket and i wanna change white socket.
I'll update soon:)
Byee
 
Good evening,
i burned pet tester and i putted in Ud9...ram seems to be god....

I hope you put the Pettester3 EPROM into the D8 EDIT socket. If the RAM memory test passes, it should go on to display the checksums of the ROMs. Please list that data and we can see if the ROMs are OK.

If OK, there is a fair chance the original D8 EDIT ROM was bad. Then burn another D8 EDIT chip using a 2716 with file 901474-01 (assuming a business keyboard) or 901447-24 (graphics) and give it a quick try. Otherwise the continuity testing will have to be done as suggested by the guys.

Also, if you have washed the board, make sure you dry it very well. Water is the enemy of printed circuit boards. For aerospace boards, if wet, they had to go a thermo-vacuum station for 24 hours to dry and outgas the water vapor.
-Dave
 
I hope you put the Pettester3 EPROM into the D8 EDIT socket. If the RAM memory test passes, it should go on to display the checksums of the ROMs. Please list that data and we can see if the ROMs are OK.

If OK, there is a fair chance the original D8 EDIT ROM was bad. Then burn another D8 EDIT chip using a 2716 with file 901474-01 (assuming a business keyboard) or 901447-24 (graphics) and give it a quick try. Otherwise the continuity testing will have to be done as suggested by the guys.

Also, if you have washed the board, make sure you dry it very well. Water is the enemy of printed circuit boards. For aerospace boards, if wet, they had to go a thermo-vacuum station for 24 hours to dry and outgas the water vapor.
-Dave

hI dAVE,
no i putted Pettester in Ud9 Kernal....doesn't correct?
 
Welcome back!

The fact that you can get my PETTEST ROM to work indicates that the CPU, RAM and the EDIT ROM socket and address decoding works OK.

As Dave has said, if the RAM checks out OK, it should move on to a page that displays the ROM checksums and enables you to checkout the keyboard. Unfortunately, one of my outstanding actions for the PETTESTER is to fully document it. I will get around to it :)!

If you can get us a screenshot of this page (the last page that should be displayed as part of the PETTEST ROM) and which version of the ROMs you have, we should be able to work out whether they are as expected or not.

If everything looks good (as Dave has already stated) then it is likely that your EDIT ROM requires reburning.

The other possibility could be that you have an issue with the interrupt lines somewhere (my PETTEST doesn't check the interrupts).

Incidentally, for your other PET, did you see my suggestion about finding a TV repair shop in your area? They should be able to repair your monitor for you, assuming the logic board is working.

Dave
 
hI dAVE,
no i putted Pettester in Ud9 Kernal....doesn't correct?

Paninaro,
There would be no proper Reset Vector if you did that. Did you somehow download the original version of Pettest instead of the new one by daver2 as pointed to by MikeS. The original version does run at the kernal address ($F000).
 
Welcome back!

The fact that you can get my PETTEST ROM to work indicates that the CPU, RAM and the EDIT ROM socket and address decoding works OK.

As Dave has said, if the RAM checks out OK, it should move on to a page that displays the ROM checksums and enables you to checkout the keyboard. Unfortunately, one of my outstanding actions for the PETTESTER is to fully document it. I will get around to it :)!

If you can get us a screenshot of this page (the last page that should be displayed as part of the PETTEST ROM) and which version of the ROMs you have, we should be able to work out whether they are as expected or not.

If everything looks good (as Dave has already stated) then it is likely that your EDIT ROM requires reburning.

The other possibility could be that you have an issue with the interrupt lines somewhere (my PETTEST doesn't check the interrupts).

Incidentally, for your other PET, did you see my suggestion about finding a TV repair shop in your area? They should be able to repair your monitor for you, assuming the logic board is working.

Dave


Hi Dave, sorry but i can't see page where display the Rom on pettester....i putted pettes in Ud9 and i can see only full "g" page and page with characters...
If i insert pettest in ud8 i have page full of strange symbols...
 
Paninaro,
There would be no proper Reset Vector if you did that. Did you somehow download the original version of Pettest instead of the new one by daver2 as pointed to by MikeS. The original version does run at the kernal address ($F000).


Ahhh ok i understod now!
Where i can find this new version please?
Thanks!
 
However i verified all eproms with programmer and are ok....
I suspect that white sockets are bad...do you suggest replacement?
 
Ok, so back to the beginning...

Exactly which version of PETTEST have you downloaded and from where? Please post the link of what you have downloaded.

There are two (2) versions of PETTEST in existence. So it all depends on which one you downloaded as to where to put it (kernel or edit rom socket) and what features they test.

If you downloaded my version, it fits in place of the EDIT ROM. It will test the VDU memory, page 0 and page 1 RAM, and will then display the ROM checksums and keyboard matrix.

If you suspect the white sockets, then my tester will display a differing 16 bit checksum to what is expected. People have also found faulty ROMs and sockets by the checksum value changing (it is constantly calculated and displayed).

Dave
 
Ok, so back to the beginning...

Exactly which version of PETTEST have you downloaded and from where? Please post the link of what you have downloaded.

There are two (2) versions of PETTEST in existence. So it all depends on which one you downloaded as to where to put it (kernel or edit rom socket) and what features they test.

If you downloaded my version, it fits in place of the EDIT ROM. It will test the VDU memory, page 0 and page 1 RAM, and will then display the ROM checksums and keyboard matrix.

If you suspect the white sockets, then my tester will display a differing 16 bit checksum to what is expected. People have also found faulty ROMs and sockets by the checksum value changing (it is constantly calculated and displayed).

Dave


Hi Dave,
i downloaded from zimmer this:

http://www.zimmers.net/anonftp/pub/cbm/firmware/computers/pet/other/petester.bin

I changed rom white sockets but nothing change...now i wanna change white cpu socket...
 
Exactly which version of PETTEST have you downloaded and from where? Please post the link of what you have downloaded.

The screenshot looks like the original version I wrote (or possibly the minimally patched version that included a routine to initialize the CRTC in PETs so equipped) that runs in UD9.

I doubt the CPU socket is a problem since a flaw there *probably* would have prevented any version of the PET tester from running, although there certainly are edge cases the old version won't see. (As noted, it's useless for telling you if there's a problem with the other ROM sockets, and it also wouldn't tell you if there was an address line issue affecting RAM.)
 
The screenshot looks like the original version I wrote (or possibly the minimally patched version that included a routine to initialize the CRTC in PETs so equipped) that runs in UD9.

I doubt the CPU socket is a problem since a flaw there *probably* would have prevented any version of the PET tester from running, although there certainly are edge cases the old version won't see. (As noted, it's useless for telling you if there's a problem with the other ROM sockets, and it also wouldn't tell you if there was an address line issue affecting RAM.)


Yes it's probable but for greater security i change also cpu socket...
If ram and rom are all ok...what could be the problem? Maybe video ram or logic chip?
 
Back
Top