Sega CD FMV audio issue

Discussion in 'Repair, Restoration, Conservation and Preservation' started by Flynn, Sep 11, 2019.

  1. Flynn

    Flynn Newly Registered

    Joined:
    Jan 9, 2016
    Messages:
    2
    Likes Received:
    0
    I am restoring a Sega PAC-S10 unit for the Pioneer Laseractive system. The PAC-S10 unit was working before, but I wanted to recap it and replace the dead battery to restore it. I am experiencing a strange audio issue. Now, I say that it was "working" before, but I just tested a cart, Sega CD, and Mega LD before I started the restoration (all seemed to work). I did not look for this particular issue, so I do not know if the issue was already there or not.

    The issue: Sega CD audio when playing full motion video (FMV) is mute. Also, some sound effects on Sega CD games won't play. All Genesis cart audio is fine, all Mega LD audio is fine. CD audio tracks on Sega CD games play fine. Other sound effects on Sega CD games play fine. It seems like just a subset of the audio of the Sega CD games won't play (always the audio on FMV).

    This issue is not specific to the PAC-S10. Through searching, I have located at least 2 other gamers out there that experienced the exact same symptoms on their model 1 & model 2 Sega CD systems. Because my unit was recapped, I don't think it's a capacitor. I have traced all the caps from the pads to their first connection points successfully, so I don't think I damaged any traces while recapping.

    I'm pretty sure the problem lies in the SSUB board in the PAC-S10. That board contains the main Sega CD PCM audio chip 315-5476A. I know that the chip provides 8 more audio channels to the Sega Genesis system.

    All that said, does anyone have an idea where to look first to find the audio issue? Can anyone confirm what the PCM 315-5476A chip actually controls (audio-wise)? If it adds the 8 extra channels of audio, and the audio CD tracks from the Sega CD games play fine, then 2 of the 8 channels are working fine (which would make me think it's not the chip).

    Any input would be appreciated.

    -Flynn
     
    Last edited: Sep 12, 2019
  2. Flynn

    Flynn Newly Registered

    Joined:
    Jan 9, 2016
    Messages:
    2
    Likes Received:
    0
    Progress. I have been able to repeat this exact problem in a Sega CD model 1 system that I had around for parts. The issue that causes the audio CDs to play fine, audio CD tracks to play fine in Sega CD games, some Sega CD sound effects to play while other effects don't play, but audio in FMV is completely mute (all while the Sega CD game plays fine) is the Pseudo SRAM in the audio circuitry.

    Looking at block diagrams of the Sega CD and the PAC-S10 units, I see that "most" audio goes from the main Sega CD processor to the Sega 315-5476A PCM audio chip. That chip interfaces with two other main ICs. One is a digital to analog converter (DAC) and the other is to 2 PSRAM chips that hold 256kbits of SRAM. When you play audio CDs or audio tracks in a Sega CD game, the SRAM is not needed and so audio CDs and audio tracks in the games play fine (as does the actual CD game since non-audio data is not handled here). However, when you play MPEG files or other digital audio files (FMV and some audio samples), the main Sega 315-5476A audio chip needs to use the SRAM. I took a Sega CD motherboard from a parts unit and put it into my "mostly working" Sega CD unit. The parts unit mobo was fine and it played all Sega CD audio fine (no issues at all). Then, I cut the VCC connections to the 2 PSRAM chips to simulate a non-working set of chips. The result was the exact problem I am having as described in my first message, no FMV audio but most other audio and the actual game plays fine.

    After reconnecting the VCC connections to the chips, everything worked fine again.

    Now, for the PAC-S10 restoration, I need to find out if the proper voltage is getting to the PSRAM chips (again just recently recapped, so I'm thinking yes, but hoping no because that would be an easier fix). Or, if in a very rare case, the PSRAM chips have gone bad. Or, in an even more rare case, the Sega 315-5476A chip is only partially working. There is also the possibility that the traces from the PSRAM chips are compromised, but the board did not have any corrosion on it when I recapped it, so I'm leaning against that.

    Luckily, if it does happen to be the PSRAM chips, they are still readily available and cheap. But, if it's not the chips and it's the voltage, then that's an easier fix.

    -Flynn
     
    Last edited: Sep 14, 2019

Share This Page