Image Map Image Map
Page 2 of 5 FirstFirst 12345 LastLast
Results 11 to 20 of 42

Thread: New PET 2001, garbled screen.

  1. #11
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    1,525

    Default

    Excellent...

    Pin 40 of the 6502 CPU should be high (/RESET). We don't want the CPU to be permanently held in a reset state...

    Pin 2 of the 6502 CPU should be high (RDY).

    Pin 6 of the 6502 CPU should be high (/NMI). We don't expect any Non Maskable Interrupts.

    Pin 4 of the 6502 CPU should be either permanently high - or mainly high and pulsing low for brief periods (/IRQ). We don't expect this line to be permanently in interrupt.

    Pin 37 of the 6502 CPU should have a 1 MHz square wave clock on it (PHI0). Clock input.

    Pin 39 of the 6502 CPU should have a 1 MHz square wave on it (PHI2). Clock output.

    Pin 7 of the 6502 CPU should be oscillating (SYNC). I think it should be mainly low and pulsing high briefly for every instruction it fetches/executes. I may have the sense inverted though (I am writing this from memory). EDIT: Just checked the 6502 data sheet and this signal should deb mainly low pulsing high as I originally stated.

    See where this leads us first...

    Dave

  2. #12
    Join Date
    Nov 2012
    Location
    Richmond Hill, Ontario, Canada
    Posts
    884

    Default

    I checked the 320349 schematic because the 320351 schematic wasn't available. As such, I only checked J10 and J11 and they both have the correct voltage. Is this schematic roughly the same?

    Here are my readings:

    Quote Originally Posted by daver2 View Post
    Pin 40 of the 6502 CPU should be high (/RESET). We don't want the CPU to be permanently held in a reset state...
    STEADY HIGH

    Quote Originally Posted by daver2 View Post
    Pin 2 of the 6502 CPU should be high (RDY).
    HIGH

    Quote Originally Posted by daver2 View Post
    Pin 6 of the 6502 CPU should be high (/NMI). We don't expect any Non Maskable Interrupts.
    HIGH

    Quote Originally Posted by daver2 View Post
    Pin 4 of the 6502 CPU should be either permanently high - or mainly high and pulsing low for brief periods (/IRQ). We don't expect this line to be permanently in interrupt.
    STEADY HIGH

    Quote Originally Posted by daver2 View Post
    Pin 37 of the 6502 CPU should have a 1 MHz square wave clock on it (PHI0). Clock input.
    SQUARE WAVE

    Quote Originally Posted by daver2 View Post
    Pin 39 of the 6502 CPU should have a 1 MHz square wave on it (PHI2). Clock output.
    SQUARE WAVE

    Quote Originally Posted by daver2 View Post
    Pin 7 of the 6502 CPU should be oscillating (SYNC). I think it should be mainly low and pulsing high briefly for every instruction it fetches/executes. I may have the sense inverted though (I am writing this from memory). EDIT: Just checked the 6502 data sheet and this signal should deb mainly low pulsing high as I originally stated.
    STUCK LOW

    Thanks again for assisting.

  3. #13
    Join Date
    Dec 2005
    Location
    Toronto ON Canada
    Posts
    6,807

    Default

    The 2001 PETs essentially came in two flavours: the original blue bezel/chiclet keyboard/built-in tape drive 2001 which used static RAM chips and came in 4 versions depending on the RAM & ROM chips used, and the later 2001-N AKA 30xx and 40xx which used dynamic RAM chips.

    Looks like in this case the original 2001 static RAM board was replaced with the later 2001-N dynamic RAM version; some later SRAM 2001s came with the new (normal) keyboard sans tape drive, but AFAIK no PETs with built-in tape drive and chiclet keyboard ever originally came with that board.

    Assuming that's the original power transformer then the additional board probably supplies the additional +12V and -5V that the replacement 2001-N board requires for the DRAM chips (the 2001 only used 5V).

    The mysterious edge connector may well be a vestige of when there was a 2001 board with its expansion edge connector in there.

    Not sure where you got that 320351 number; that's the parts list AFAIK; 320349 should be the correct schematic for that board.

    m

  4. #14
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,343

    Default

    Quote Originally Posted by snuci View Post

    'Sync' STUCK LOW
    The CPU is not executing instructions. Maybe bad 6502 CPU, but maybe CPU is executing the dreaded KIL instruction (illegal instruction) that causes a hangup. This may be due to fetching instructions from bad ROM or somehow fetching from RAM.

    Another thought: Sometimes a bad 6520 PIA or 6522 VIA can cause the garbage screen. Carefully remove the three chips, and on power-on, see if you get the Commodore message before the screen freezes. That means that one of them may be bad. Good luck. The guys here will get this PET running again.

  5. #15
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    1,525

    Default

    +1 for Dave's suggestion to remove the PIA & VIA chips next (morning Dave).

    If that doesn't kick the beast into life, then either swap the 6502 CPU next, or manufacture a NOP generator from a couple of 40 pin IC sockets and a few passive components. This little device will (effectively) cause the ROM and RAM to be ignored and forces the 6502 to execute no operation instructions. If this now gives you pulses on the SYNC line, then the CPU is good and we need to look into the ROM and RAM next. If the CPU doesn't generate SYNC pulses, it is probably dead and needs replacing. Buying a known good 6502 CPU will just bypass this check if you don't fancy making anything!

    Dave
    Last edited by daver2; October 13th, 2017 at 12:10 AM.

  6. #16
    Join Date
    Nov 2012
    Location
    Richmond Hill, Ontario, Canada
    Posts
    884

    Default

    Quote Originally Posted by MikeS View Post
    Assuming that's the original power transformer then the additional board probably supplies the additional +12V and -5V that the replacement 2001-N board requires for the DRAM chips (the 2001 only used 5V).
    That makes total sense. Out of curiosity, I will see if I can measure the voltage on the S100 slot (I should have an S100 breakout board somewhere) to see if the S100 voltages are there.

    Quote Originally Posted by MikeS View Post
    Not sure where you got that 320351 number; that's the parts list AFAIK; 320349 should be the correct schematic for that board.
    The board has markings that say "Assy No 320351". You can see it in this picture here. I don't see any other markings on the board (but I didn't check the bottom).

    Thanks for chiming in Mike.

  7. #17
    Join Date
    Nov 2012
    Location
    Richmond Hill, Ontario, Canada
    Posts
    884

    Default

    Quote Originally Posted by daver2 View Post
    +1 for Dave's suggestion to remove the PIA & VIA chips next (morning Dave).

    If that doesn't kick the beast into life, then either swap the 6502 CPU next, or manufacture a NOP generator from a couple of 40 pin IC sockets and a few passive components. This little device will (effectively) cause the ROM and RAM to be ignored and forces the 6502 to execute no operation instructions. If this now gives you pulses on the SYNC line, then the CPU is good and we need to look into the ROM and RAM next. If the CPU doesn't generate SYNC pulses, it is probably dead and needs replacing. Buying a known good 6502 CPU will just bypass this check if you don't fancy making anything!
    Following this annotated board, I will see if I can swap C5, C6 and C7. I'll see if I have a working PET with socketted chips that I can swap with along with the 6502. I can also ensure those chips are working. I'll let you know how I make out.

    I appreciate the help guys.

  8. #18

    Default

    Quote Originally Posted by snuci View Post
    Following this annotated board, I will see if I can swap C5, C6 and C7. I'll see if I have a working PET with socketted chips that I can swap with along with the 6502. I can also ensure those chips are working. I'll let you know how I make out.

    I appreciate the help guys.
    Just for the record, I've repaired a few (three of them) 2001N boards all showing garbled screen initially and it was never any of the CPU/PIA/VIA.
    After checking the power supplies and the vital signals for the CPU (reset, clock, irq, nmi), I always try to dump all ROMs and check they're identical to the dumps found here:

    http://www.zimmers.net/anonftp/pub/c...computers/pet/

    I've found a few bad ROMs in two out of three machines.
    If that doesn't solve the problem, then it's time to start probing around with logic probes and/or oscilloscope.
    Using a pettester.bin image in place of UD9 ROM also might help.

    HTH
    Frank IZ8DWF

  9. #19

    Default

    Mine had a bad PIA and VIA, bad video RAM, a couple bad 7400 series chips, and a bad gate somewhere. I haven't found the bad gate, that would require desoldering half the board. It's pulling down an unused signal which causes the system to halt. I shorted the line to +5V hoping that someday the bad gate would burn up, but so far it hasn't.

    Mind you, I suspect the PIA and VIA failures were due to socket insertion errors by the previous owner. But really, I can't rule out that happening for any PET. These have always been the kind of machine that just anyone can poke around in.

  10. #20

    Default

    Quote Originally Posted by KC9UDX View Post
    Mine had a bad PIA and VIA, bad video RAM, a couple bad 7400 series chips, and a bad gate somewhere. I haven't found the bad gate, that would require desoldering half the board. It's pulling down an unused signal which causes the system to halt. I shorted the line to +5V hoping that someday the bad gate would burn up, but so far it hasn't.

    Mind you, I suspect the PIA and VIA failures were due to socket insertion errors by the previous owner. But really, I can't rule out that happening for any PET. These have always been the kind of machine that just anyone can poke around in.
    The last 2001N that I repaired had the following problems:
    bad reset circuit
    2 x 2114
    2 x LS157
    1 x 74177
    1 x LS373
    1 x LS00 <--- after that, it started to work, it caused most of the other faults but the 2114 that were just bad MOS technology products.
    1 x 6522 (bad port bits, the machine started but IEEE 488 wasn't fully working).

    To find a shorted line, I have an HP-547A

    HTH
    Frank IZ8DWF

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •