PET Alive! - Pet Hospital

[harry]

Harry,
This is not good. I do not think a working CPU will do this (stay in instruction fetch forever). You stated you already re-seated the chips. So do you have a spare 6502?
-Dave

Hi Dave, I have some other machines I could remove a 6502 from , I have taken out the whole board now , and I am checking all over the underside, there is a small mod underneath, a small wire connects pin 3 of uc3 to pin 11 on ud 4, and pin 16 of uc3 has its trace removed to isolate the pin, other than this mod everything underneath looks good. this board has nearly
all ic's soldered on, except for the cpu, and rom UD7 .and the character rom. should I now replace the cpu and check readings again?

 

[harry]

Hi Dave, I have some other machines I could remove a 6502 from , I have taken out the whole board now , and I am checking all over the underside, there is a small mod underneath, a small wire connects pin 3 of uc3 to pin 11 on ud 4, and pin 16 of uc3 has its trace removed to isolate the pin, other than this mod everything underneath looks good. this board has nearly
all ic's soldered on, except for the cpu, and rom UD7 .and the character rom. should I now replace the cpu and check readings again?

I should also say, I noticed the trace to pin 3 on UC3 on the top of the board next to the ic has been cut, don't know if this helps? or why it has been done? ....Harry

 

[dave_m]

I should also say, I noticed the trace to pin 3 on UC3 on the top of the board next to the ic has been cut, don't know if this helps? or why it has been done? ....Harry

Harry,
I assume some repair was performed in the past due to a shorted or open trace. Don't worry about it except to note if it is a "neat" job. Hopefully there are no big blobs of solder that may short pins, etc?

Too bad so many ROMs and I/O devices are soldered in. This may make it a little tougher to troubleshoot. The 6502 is still in doubt and may be working, but since you have a replacement available, it should be OK to swap and see what changes.

You are doing good. We can get you replacement ROMs but it will be a job to remove the old ones if necessary (24 pins). We can guide you through the process. Using something called "Solder Wick" will make the job possible without overheating the board. If you are not a 'solder guy', it may be possible for you to find an electronics engineering student at your local university who can do it for you.
-Dave

 

[harry]

Harry,
I assume some repair was performed in the past due to a shorted or open trace. Don't worry about it except to note if it is a "neat" job. Hopefully there are no big blobs of solder that may short pins, etc?

Too bad so many ROMs and I/O devices are soldered in. This may make it a little tougher to troubleshoot. The 6502 is still in doubt and may be working, but since you have a replacement available, it should be OK to swap and see what changes.

You are doing good. We can get you replacement ROMs but it will be a job to remove the old ones if necessary (24 pins). We can guide you through the process. Using something called "Solder Wick" will make the job possible without overheating the board. If you are not a 'solder guy', it may be possible for you to find an electronics engineering student at your local university who can do it for you.
-Dave

Dave, and Steve, thanks for all your help so far, tomorrow I'll bring back another cpu, I may even be able to loan a working 8032 board, which would show if the crt is still working ok, quess for me its time to call it a day! many thanks great stuff! ....Harry

 

[channelmaniac]

I have a PET to bring back from the dead... however I'm in need of the 6550 RAM chips. The board is missing a couple and I could use a couple spares.

Also, does anyone have a short memory test program I can type in and run on it?

 

[cosam]

I have a PET to bring back from the dead... however I'm in need of the 6550 RAM chips. The board is missing a couple and I could use a couple spares.

Have you tried that site, www.arcadecomponents.com I think it was... oh, wait ;-)

Also, does anyone have a short memory test program I can type in and run on it?

Just so happens we just had one pass through in another thread: http://www.commodore.ca/manuals/PET_RAM_test.htm. Not that short I guess, but pretty thorough by the looks of it.

 

[carlsson]

Regarding the memory test, I realize it is tailormade for an 8K PET since it reports the chip numbers I2 - J8. If your PET is a 16K or 32K model, the program technically will work but the outputs may not make sense.

Besides, I wonder if X=164 should not be X=168 in that listing. I converted the values to binary patterns, and 168 better mirrors 87. On the other hand, 87 could be 91 too...

 

[harry]

Dave, and Steve, thanks for all your help so far, tomorrow I'll bring back another cpu, I may even be able to loan a working 8032 board, which would show if the crt is still working ok, quess for me its time to call it a day! many thanks great stuff! ....Harry

Hi back on the case! some good news , I just fitted a working MB ,.... and the screen comes on , keyboard works good too , at least it proves my MB is faulty.and the crt board appears ok. this working MB is slightly different, all 40pin ic.s are plug in, as are all the roms,and the video connector is in a different position? ( maybe I should make him an offer). I could carefully remove the 6502, and replace the one in my board, or is there any checks I should do first? I brought the heart rate monitor as well, ( sorry " scope "I should say) this hospital case is definitly having its affect, I'm starting to talk to my patient, ( my computer I mean ). .......Harry

 

[cosam]

this working MB is slightly different, all 40pin ic.s are plug in, as are all the roms,and the video connector is in a different position?

Empty ROM sockets in your non-working board needn't be a worry. The new board most likely has some expansion ROMs fitted which aren't necessary for normal operation. But there were 40-pin chips missing? Which ones?

I could carefully remove the 6502, and replace the one in my board, or is there any checks I should do first?

Couldn't hurt to transplant the CPU, but I'd check the above first.

 

[MikeS]

Yeah, like I suggested another (working) PET is sometimes the most useful 'tool' ;-) ; you're half-way home already.

Also very useful is a NOP adapter that endlessly steps through all addresses; just takes a couple of 40pin sockets and a little soldering, and can reveal various addressing-related problems.

 

[MikeS]

Empty ROM sockets in your non-working board needn't be a worry. The new board most likely has some expansion ROMs fitted which aren't necessary for normal operation. But there were 40-pin chips missing? Which ones?

I think what he meant is that the other board is the previous 80-col only version which is laid out differently and has sockets for the ROMs and 40-pin ICs, whereas everything's soldered on the universal board except the CPU, CG and editor ROM (and the two empty optional ROM sockets)

 

[harry]

I think what he meant is that the other board is the previous 80-col only version which is laid out differently and has sockets for the ROMs and 40-pin ICs, whereas everything's soldered on the universal board except the CPU, CG and editor ROM (and the two empty optional ROM sockets)

Hi Mike, you have put it much clearer than I did, .. thats exactly how things are. thanks ..... I going to swap the cpu and see what happens... .Harry

 

[harry]

Hi Mike, you have put it much clearer than I did, .. thats exactly how things are. thanks ..... I going to swap the cpu and see what happens... .Harry

ok , I put the 6502 from my non working board, into the working board, and powered up, screen comes on, everything working? my cpu not the problem?
any ideas?

 

[cosam]

Hi Mike, you have put it much clearer than I did, .. thats exactly how things are.

OK - you had me wondering! ;-)

ok , I put the 6502 from my non working board, into the working board, and powered up, screen comes on, everything working? my cpu not the problem?
any ideas?

The plot thickens... I'm thinking along the lines of memory or maybe the buffers between that and the CPU. Seeing as the data bus is essentially stuck, you should be able to test the buffers (UB9 and UB10) quite easily, at least in one direction, by comparing the inputs to the outputs. It's at least something to try until someone has a flash of inspiration. ;-)

 

[dave_m]

It's a little unlucky that the CPU is good as it is so easy to swap. It's starting to look like a bad system ROM or the CPU can't read the ROM because of struck buffers as cosam says. Still a lot of troubleshooting left to do. Try to check to see what ROM 'Chip Enable' is activiated (low state). They are pin 20 of UD06 thru UD10. That might be a clue to which ROM may be bad.

 

[MikeS]

... the address lines 9- 14 seem to have contention with hi,lo, and pulse, pins 15 &16 are High, pin 17 , hi/lo and pulse pins 18, 19 20 all low

I'm a little confused; when you say "have contention"do you just mean they're toggling?

Also, pins 18, 19 & 20 (which you say are all low) are in fact address lines A9 -11 which "have contention"?

Any chance you could look at the 6502 pins again, with the 'scope this time, and also the address and data lines of any of the ROMs, in particular looking for signals that are always high or low or don't go as high or low as the rest?

 

[harry]

OK - you had me wondering! ;-)


The plot thickens... I'm thinking along the lines of memory or maybe the buffers between that and the CPU. Seeing as the data bus is essentially stuck, you should be able to test the buffers (UB9 and UB10) quite easily, at least in one direction, by comparing the inputs to the outputs. It's at least something to try until someone has a flash of inspiration. ;-)

Hi Steve, the logic probe gives the same results on UB9 as UB10 ie.. p10 low, p20 hi, all ther pins hi/lo, + pulse,

 

[harry]

It's a little unlucky that the CPU is good as it is so easy to swap. It's starting to look like a bad system ROM or the CPU can't read the ROM because of struck buffers as cosam says. Still a lot of troubleshooting left to do. Try to check to see what ROM 'Chip Enable' is activiated (low state). They are pin 20 of UD06 thru UD10. That might be a clue to which ROM may be bad.

Hi there Dave the rom c/s lines are all hi, except for UD 7 pin 20 is very busy? probe shows hi/lo +pulse??

 

[harry]

I'm a little confused; when you say "have contention"do you just mean they're toggling?

Also, pins 18, 19 & 20 (which you say are all low) are in fact address lines A9 -11 which "have contention"?

Any chance you could look at the 6502 pins again, with the 'scope this time, and also the address and data lines of any of the ROMs, in particular looking for signals that are always high or low or don't go as high or low as the rest?

Mike, I could get the scope out and see in more detail ,but I have just done some checking for Dave . The c/s on UD7 concerns me why there is a pulse and hi/lo lit at the same time?
another ic which looks in trouble is the 6545 crt controller , I'm just going too fast to cope right now, not so easy for anovice like me!

 

[MikeS]

Mike, I could get the scope out and see in more detail ,but I have just done some checking for Dave . The c/s on UD7 concerns me why there is a pulse and hi/lo lit at the same time?
another ic which looks in trouble is the 6545 crt controller , I'm just going too fast to cope right now, not so easy for anovice like me!

Relax...

It kinda looks like your PET's going around in circles, i.e. a loop in the E space; any chance you could replace that UD7 ROM with the one from the other board (that's the one that's socketed, right?)

You just might be lucky after all.