Taito Operation Thunderbolt: color ram error

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

    • Taito Operation Thunderbolt: color ram error

      Recently purchased an Operation Thunderbolt cabinet cheap that was sold as, "It was working, but the monitor died".

      Figured I would remove the CRT, put a modern display in it and move on.
      Quickly realized I needed a converter board to get a display, so ordered a GBS-8220 and after a bit of messing around I got a picture.

      Unfortunately the picture displays an error indicating a failure in the color ram (see attached image).

      Looking at the board, it has 8x 2k static RAM chips (TMM-2018-35). I've ordered 8 replacements as they are cheap, but they are soldered to the PCB and I'm not looking forward to removing all 8.
      Is there any way to tell from the memory address in the photo which chip is likely to be the culprit?
      Or indeed if its likely to be the RAM at all?

      PS. I know the colours are horrible on the screen, but until I can get it to boot that's way down on my list of priorities :)

      Thanks.
      Images
      • color_ram_small.jpg

        653.98 kB, 1,512×2,016, viewed 270 times
    • The colours on the screen are messed up because your colour RAM is bad ;)

      The screen is telling you which address is wrong, the MAME driver can tell you which region it is

      It's either the RAM or the custom reading it, more likely the RAM. Especially if it's Sony or Toshiba.

      The MAME driver has actually got all the RAM listed. That makes it easy. The colour RAM is a pair of 6264 that are right next to an RGB mixer custom.

      I think Motorola calls their 8kx8 SRAM 6164, to confirm if it's compatible, Google the part code and find a datasheet with a pinout, compare it here:

      citylan.it/wiki/index.php/SRAM_8k_x_8

      This page also gives you a list with a lot of compatible part numbers so you can compare it there.
    • xodaraP wrote:

      The colours on the screen are messed up because your colour RAM is bad ;)

      The screen is telling you which address is wrong, the MAME driver can tell you which region it is

      It's either the RAM or the custom reading it, more likely the RAM. Especially if it's Sony or Toshiba.

      The MAME driver has actually got all the RAM listed. That makes it easy. The colour RAM is a pair of 6264 that are right next to an RGB mixer custom.

      I think Motorola calls their 8kx8 SRAM 6164, to confirm if it's compatible, Google the part code and find a datasheet with a pinout, compare it here:

      citylan.it/wiki/index.php/SRAM_8k_x_8

      This page also gives you a list with a lot of compatible part numbers so you can compare it there.

      Ok so 0x100000:

      Source Code

      1. AM_RANGE(0x100000, 0x100007) AM_DEVREADWRITE("tc0110pcr", tc0110pcr_device, word_r, step1_rbswap_word_w) /* palette */
      If i'm reading it correctly, MAME seems to imply its the palette RAM.

      Looks like the 6164 has the same pin out as the 6264, so I think i'll go that path.
    • doozer brought this down today and I removed both RAM chips and replaced them with sockets he brought down with him - the board had a hefty ground plane, but we got there. I've put a pair of sockets on the board and installed a pair of Fujitsu MB8464 chips that I'm hoping are fast enough.

      The board was super clean and in otherwise really great condition. Unfortunately the board doesn't use a JAMMA pinout and while the power connectors are in the correct places, the video pins are much further down the edge connector (pins 21/22 as opposed to 12/13)

      I tested the RAM I removed from the board (Toshiba TMM2063) and this was the result:

      RAM IC 1: Pins 05, 07, 26 and 27 not registering
      RAM IC 2: Pin 05 not registering

      I ignored the errors on both to see if we could get any kind of testing done. IC1 failed miserably and just errored out, IC2 failed about 50% of the way through the RAM test:



      This is the same test performed on the MB8464 ICs I replaced them with to show what should happen on working RAM chips:



      doozer is on his way back home now (I'm about 45 mins south) - so hopefully when he connects it up it's all systems go, otherwise I will be up there next week and I have offered to try to fix any other issues that arise.
    • It’s fixed!
      Game boots and plays fine. Both guns are giving feedback.

      Next step is to either fix the crt or replace it with a modern display.

      Thanks for your help today xodaraP!
      Much appreciated.
      Images
      • 9AFA3D60-1BD2-45DF-8326-A28123A8BF3B.jpeg

        862.61 kB, 1,971×2,046, viewed 94 times
    • Thanks guys much appreciated, I didnt know if this was a bit old to reply to!

      First the easy part.... Is this all good to replace the 2 Colour ram ICs with?
      ebay.com/itm/FUJITSU-MB8464A-8…LOT-OF2-NEW-/253518517057

      Secondly I have attached a screenshot of the error code and one of the Custom TC0110PCR IC.

      In the IC pic the red marking on the left, there appears to be a leg missing at pin 78, the "pad" is also showing continuity to pin 79, so Im assuming maybe someones had a crack already and stuffed it, that or it isnt needed to be connected to anything (doubtful of course) also the error message is quite garbled until I put pressure with a moistened figertip on the area marked to the right of the image.

      I'm starting to think maybe this custom is what is stuffed, im happy to reflow and try but am I wasting my time here do you think, I assume you cant purchase these anymore, would need to salvage of another board.

      Thanks again for getting back to me, I've gone from completly dead to this now, so progress is progress and I am enjoying it.

      Cheers

      Rivv
      Images
      • thunderbolterror2.jpeg

        338.08 kB, 1,920×1,440, viewed 31 times
      • TC0110PCR.jpeg

        451.96 kB, 1,920×1,440, viewed 17 times
    • Finally found some schematics, looks like 78/79 arent is use so maybe some hope after all.
      Checked continuity on that right side and most are fine, just cant find a spot to test 3 of the pins, so Im assuming it may be alright for now if I am reading this right.
      Images
      • opthunderschem.PNG

        88.57 kB, 470×583, viewed 29 times