SOLVED - Help me revive a Sega Mega Play - part 2: 315-5330 custom chip needed (present in some Master System consoles)

    This site uses cookies. By continuing to browse this site, you are agreeing to our Cookie Policy.

    • SOLVED - Help me revive a Sega Mega Play - part 2: 315-5330 custom chip needed (present in some Master System consoles)

      Ok so I've replaced the suspicious I/O chip on my dead Mega Play board thanks to @rewrite who provided the chip:
      arcade-projects.com/forums/ind…present-in-some-megadri/&

      Unfortunately this changed nothing, still a weird fixed voltage on one of the data pins of the Megadrive area Z80.

      However continuing probing revealed the /WR signal of the video RAM for the system VDP (BIOS) was floating. It's coming directly from the nearby 315-5330 custom chip, which it seems you can find in some Master System consoles, but again must be rare since none of my consoles has that chip.

      And again I'm not even sure the chip is faulty but I see no reason why the /WR signal of a RAM wouldn't be either high, low or pulsing.

      [EDIT]
      Of course VRAM has been tested good out of circuit, so not it dragging /WR signal where it shouldn't be, but for the peace of mind I installed a known working one to rule that out of the equation. Still it works fine on an other board.

      [EDIT 2]
      System Z80 never acknowledge the first VDP interrupt which would be consistent with a VRAM access problem I think.
      Looking for:
      - Champion Wrestler (Taito)



      "I'd 1cc games but I have a real life."
      01010011 01000101 01000011 01010010 01000101 01010100 00100000 01001101 01000101 01001110 01010101 00100000 01000111 01010101 01011001

      The post was edited 2 times, last by Apocalypse ().

    • yavuzg wrote:

      315-5330 is also found on Sega Mega-tech PCB. I don't mean to sacrifice a megatech to save a megaplay but just saying if you have a megatech you may try to use it just to know if the failed part is indeed 5330.
      Yeah I know that but as you said killing a Mega Tech isn't an option and I don't even have one to swap the chip otherwise I would have compared the behaviour of the /WE signal coming out of it before all.
      Could you please check the /WR signal (pin 41) of the 315-5330 on your Mega Tech? To me it should always be driven to a known state (never floating or high impedance).
      An other option could also be to wire 2 RAM chips ("classic" Master System way) to get rid of the custom chip, but that's a really messy and time consuming job.

      I've also tried to recreate the /WR signal from the two /WR0 and /WR1 signals the VDP generates by using a AND gate but no luck. Either it's more complicated than a AND (but I don't see why) or the chip has other internal failures (possible).
      Looking for:
      - Champion Wrestler (Taito)



      "I'd 1cc games but I have a real life."
      01010011 01000101 01000011 01010010 01000101 01010100 00100000 01001101 01000101 01001110 01010101 00100000 01000111 01010101 01011001

      The post was edited 1 time, last by Apocalypse ().

    • Apocalypse wrote:

      yavuzg wrote:

      315-5330 is also found on Sega Mega-tech PCB. I don't mean to sacrifice a megatech to save a megaplay but just saying if you have a megatech you may try to use it just to know if the failed part is indeed 5330.
      Yeah I know that but as you said killing a Mega Tech isn't an option and I don't even have one to swap the chip otherwise I would have compared the behaviour of the /WE signal coming out of it before all.Could you please check the /WR signal (pin 41) of the 315-5330 on your Mega Tech? To me it should always be driven to a known state (never floating or high impedance).
      An other option could also be to wire 2 RAM chips ("classic" Master System way) to get rid of the custom chip, but that's a really messy and time consuming job.

      I've also tried to recreate the /WR signal from the two /WR0 and /WR1 signals the VDP generates by using a AND gate but no luck. Either it's more complicated than a AND (but I don't see why) or the chip has other internal failures (possible).
      I'll check the signals tonight and report back ;)
    • The /WE signal seems HIGH most of the time. However, there are also periodic spikes (see below) which does not seem to be related with the game selection or game currently playing. Infact this 5330 chip is in the "selection menu handling" part of the PCB. Since the menu is always a static list of games, the signal does not seem to change.

      I can check whatever else you might be interested in...

      Also, I do happen to have a NTSC master system 1 and checked the inside. No 5330 inside. It seems this IC may exists in later models. My PCB revison is 837-6067 (year 1986) and according to SMS power site this IC seems to be found in models produced in 1988...

      Good luck ;)





    • Thanks a lot @yavuzg, mine is busted for sure...
      I'm probably the only one in the entire world having that chip failed. Bad luck.
      Looking for:
      - Champion Wrestler (Taito)



      "I'd 1cc games but I have a real life."
      01010011 01000101 01000011 01010010 01000101 01010100 00100000 01001101 01000101 01001110 01010101 00100000 01000111 01010101 01011001
    • I may know someone with a parts SMS. I will ask if they still have it, and if they can check for that IC.

      The abstract from this page makes me think that, with a working reference model, you could recreate this IC: smspower.org/forums/10098-3155246And3155330Pinouts

      Since it is used to help reduce the part count, maybe you could just restore the original dual PSRAM configuration that it eliminates as well.
    • Hatsune Mike wrote:

      Since it is used to help reduce the part count, maybe you could just restore the original dual PSRAM configuration that it eliminates as well.
      Yep I recalled that possibility earlier in the thread but it's not going to look nice. Of course you could make a circuit that plugs directly in the socket of the faulty 315-5330 but the 0.07" pitch means I can't use a piece of breadboard (even if I found 0.07" breadboards then how to fit the 0.1" DRAMs). And I'm not designing a PCB for the unique time in history of mankind someone is going to need it.
      Looking for:
      - Champion Wrestler (Taito)



      "I'd 1cc games but I have a real life."
      01010011 01000101 01000011 01010010 01000101 01010100 00100000 01001101 01000101 01001110 01010101 00100000 01000111 01010101 01011001
    • 8bitforlife wrote:

      I have a Sega master system. Its American though does it need to be a Japanese version?
      I have no idea. The widespread rumour is that chip can only be found on NTSC consoles (that would include US or JP) but I keep seing pics of French models (RGB only) with it too:
      dragoncity17.wordpress.com/201…ster-system-1-switchless/
      Looking for:
      - Champion Wrestler (Taito)



      "I'd 1cc games but I have a real life."
      01010011 01000101 01000011 01010010 01000101 01010100 00100000 01001101 01000101 01001110 01010101 00100000 01000111 01010101 01011001
    • Short version: SOLVED

      Not so long version:
      - I trashed that MP board
      - I finally ended up reproducing the 315-5330 custom chip for a friend (with a handful of TTL chips)
      Looking for:
      - Champion Wrestler (Taito)



      "I'd 1cc games but I have a real life."
      01010011 01000101 01000011 01010010 01000101 01010100 00100000 01001101 01000101 01001110 01010101 00100000 01000111 01010101 01011001