Image Map Image Map
Results 1 to 9 of 9

Thread: Old mask ROM issues - questions

  1. #1

    Default Old mask ROM issues - questions

    So I made a ROM dump board to dump/document some of the early 80's NEC PC-6001 cartridges that another forum member here and all dumped fine except for one. It has been giving us fits because everytime you dump it, it has different data. I've tried changing the dump method to wait longer, provide a pullup on the data ilnes, etc., but it doesn't seem to help. Internally these cartridges have a pair of 8K ROM's that I am assuming are mask ROM's. The computer supplies a mess of address lines and then the cartridge drives the data lines. My latest attack plan was to make it dump each byte 1024 times, so a 16K cartridge gets dumped to a 16M file I can post process on the PC. Most locations have the same 1K value and it looks like the lower 8K is pretty solid. The upper 8K IC however seems to have a lot of troubled memory locations where the 1024 dump contains varying characters. My technique so far has been to count up the values and the one that has the highest count is the "good" one, but I wondered if anyone here with intrinsic knowledge of mask ROM's and their failure conditions might have any better advice/algorithm to try.

  2. #2

    Default

    I know with EPROMs the VCC being off would cause data errors when reading. Not sure it applies to mask ROMs though.

  3. #3

    Default

    Good thought - I'm supplying the proper 5V to it.

  4. #4

    Default

    File attached showing errors and how they look.y.txt

  5. #5
    Join Date
    Sep 2006
    Location
    Silicon Valley
    Posts
    1,517

    Default

    Sometimes, you can get a better read by increasing Vcc slightly.

  6. #6

    Default

    The technique of varying the supply rail is called margining, as an example good EPROM testers usually go +/- 10%.

    Also check your address and control lines are good TTL levels >2.7V and maybe even try CMOS levels >3.5V.

  7. #7

    Default

    It is not my cartridge so I don't want to risk going higher unless the owner says to try it. I was going to try 3.3V to see its effect though...

  8. #8

    Default

    Problem solved! It was not getting enough power. I was supplying the VCC to the cartridge from an AVR port pin which worked great for 7 other cartridges being dumped, but wasn't enough for this one. Makes me wonder if the capacitor inside it is beginning to short/leak and causing it to need more current. Once I moved the power to the dump board VCC, I got 4 clean matching dumps of the cartridge in a row.

  9. #9
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    28,212
    Blog Entries
    20

    Default

    Heavy loads require more than a logic voltage for power. Use a power MOSFET to switch the +5 and drive the gate with the AVR.

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
  •