CBM PET 2001 Restoration

A little update on my giant struggle.

I've built x1541 's universal RAM/ROM expansion that Toast_r was so kind and sent me the board and GAL for.

To make sure that the exp.board itself is working I've put it into my VIC-20 and it seemed to be fine at first.

As the PET didn't react positively to the expansion board, I took the advice of Toast_r to build a NOP generator.

I've found two track related issues with the NOP generator and fixed them. Now all A,BA,BD,BBD,SEL,CS lines are good.

The signals on some lines are a bit noisy, but within the tolerance of TTL standards.

When I place back the CPU, ROMs, 4x RAMs, and the VIA, the machine is still showing the garbage screen.

I did the " detlef mod" on the BA11 lines to accomodate 2732 type ROMs in all sockets - except char_gen - but those lines are fine.

As I am able to tell, the contents of the ROMs and their order should be all good:

LOW ACTIVE ( BA11 )

H1: 901447-1 $C000

H2: 901447-3 $D000

H3: 901447-5 $E000

H4: 901447-6 $F000

HIGH ACTIVE ( BA11 Inverted )

H5: 901447-2 $C800

H6: 901447-4 $D800

H7: 901447-7 $F800

I've tested the RAMs in my DIY tester, and I think 2k should be enough, but not so sure anymore.

Now it would be good if the expansion board was working, but it clearly isn't - still something wrong with the GAL programming...

( Can't make it work in the VIC again. The RAM expansion was working once, but the ROM wasn't. I've reprogrammed the GAL, now nothing

seems to work. )

WOOOOOOOOOOOOOOOOOHOOOOOOOOOOOOOOOOOOOOO!!!

Thank you Toast_r, detlef and x1541 !!!

oh my god, i was really at the edge of crying for a moment.

The problem was with two address tracks, that the NOP generator method recovered. It didn't boot up with 2k of RAM, so I used the RAM expansion and enabled 4k and that did the trick.

Thank you all, again!

Now I am looking for two 6520 chips, both are fried...

A bit more explanatory images on what went wrong in this PET ( just a couple of things really )

I've received today two PIAs for the PET ( MC6821P types instead of the MOS6520 ).

I've got now 2x 6520 chips that are may or may not be working, and 2x MC6821P chips.

I am using the universal RAM expansion to rule out any RAM issues ( as the computer was reporting only 4095 bytes free ).

Without any of the PIAs installed, the full 32k RAM is showing up.

If I place in [G8] PIA the garbage screen comes back. Populating [B8] PIA seems to be indifferent.

Cycling through [G8] all 4 PIA options that I have, one of them presents a black screen, the other three just garbage.

I would appreciate any inputs on this matter,

thanks

I've bought them from a pinball service, and they claim they test everything before shipping...

Zitat von detlef

And both chips show a garbage screen when you place them at G8 ?

My first idea is that they are blocking the address or data bus so that the CPU can not write to the screen memory.

I would check with the NOP generator if the chip selects are still ok.

And with the RAM/ROM board if there are any weak signals on the data bus.

Both tests with G8 in place.

I can look at the schematic later if there could be another cause.

Alles anzeigen

All four chips produce garbage screen when placed to G8.

I'll check with the NOP and the expansion card. Thanks for the tip!

Final BCD decoder and BUS driver replaced. Case closed.

Yesterday the PET went home, because "it belongs to a Museum!"

I am going through the row. PC-100 check, PET check. Coming up next, Trash-80 to Treasure.

oh no...