My Sega G80 PCBs Repair Thread

mjenison

Well-known member
Joined
Aug 1, 2002
Messages
1,574
Reaction score
422
Location
Streamwood, Illinois
Hi all,

I've had boxes of G80 PCBs for almost 20 years now that I *still* haven't gotten to, and by golly, this is going to be the year I do it! To motivate myself, I'm starting this thread to document the process, and hopefully pick up a few thread followers to help advise me along the way.

The boards:
~30 CPU boards
~15 USB sound boards
~30 Speech boards
~30 XY pairs
~30 EPROM boards
~10 Space Fury sound boards

All non-working.

Let's get started!
 
Ok, first we'll start with the CPU boards.

Looking over the boards, it looks like I'll be short on the decoder PROM, as a few of the boards are missing this from the socket. Anyone out there able to provide a source for these? All I have is a GQ-4X programmer; not sure if it can burn these or not.

Thanks to Shupac (David), I am armed with the Sega XY Diagnostics ROM to help debug these boards. The great thing about this is that the EPROM is security chip agnostic, meaning I won't have to burn one of each game CPU EPROM, put in a matching (and hopefully working) game EPROM board to match the security chip just to test the basics of the CPU board.

CPU1: Dead
CPU2: Dead
CPU3: Dead
CPU4: Dead
CPU5: Bad Z80. Replaced. Dead with good socketed devices.
CPU6: Board is labeled "good". Huh, wonder why this is in the pile? Putting in the diagnostics ROM and passes. Next step is to swap this into a working set to verify it can communicate with other boards successfully. For now, I swap in the Sega Multigame (because it doesn't require me to put together an EPROM board), and the multigame boots fine.

Next up: build a dedicated test bench so I don't have to swap all this stuff into my Eliminator 4-player any more.
 
Last edited:
What type of device is the decoder PROM? If 82s123 or 82s129/131 we have rewritable 512M EEPROM replacement options. (perhaps)
 
What I have found with the sound boards is they often have the same problem of not turning off the last sound played. The last tone just continues until a new sound starts. I also found the cause of this, plugging in the sound board connector one pin off.

Now I just need to find the cure. I only caused this problem on my personal board but have found it on many of the broken sets I have tested meaning this is a common mistake people make. I would love to have a documented fix.
 
Its an 82S123.

Then, we can experiment, perhaps, with the 512M EEPROM option. Though it might not fit correctly in the cage with the other boards as it sits relatively high off the PCB plane.

Personally, I like these as they are re-writable and I don't have stock anyway of the older PROMS. The 82sxxx are also rare and expensive. I'm pretty paranoid about programming them too as they are write-once devices. Good way to throw away $15 with a bad write. :eek: YMMV ...

See this thread ->

http://forums.arcade-museum.com/showthread.php?t=353715&highlight=82s123

"The 82S123-to-27512 is a universal replacement solution to replace a 82S123 bi-polar PROMS with newer easy to find fast EEPROMs. This kit is a very small adapter board that can take a cheap and easy to find W27C512 EEPROM and replace 82S123 bi-polar PROMS. "

http://www.arcade-cabinets.com/board_hacks/82S123-to-27512/
 
Ok, got a standalone power supply that I can hook up to a cage finished.

CPU7: Dead
CPU8: Dead
CPU9: Static (loud roar from the sound board after a few seconds). The last board that did this had this removed when I replaced the Z80, but not this one.
CPU10: Static
CPU11: Some activity with the Sega Diag ROM. A few flashing vectors here and there, will some being drawn nearly off the edge of the screen. I think David said for his Diag rom to work at least some RAMS had to be good, so maybe there's an issue with accessing the RAMs or maybe bad sockets.

Once I get all the CPU statuses determined, I'll start working on the first one in earnest. I mean, there's only 40+ devices on this board that can cause catastrophic failure...
 
Last edited:
I've had boxes of G80 PCBs for almost 20 years now that I *still* haven't gotten to, and by golly, this is going to be the year I do it! To motivate myself, I'm starting this thread to document the process, and hopefully pick up a few thread followers to help advise me along the way.
The boards:
~30 CPU boards
~15 USB sound boards
~30 Speech boards
~30 XY pairs
~30 EPROM boards
~10 Space Fury sound boards

Mark generously let me stop over last night and test a few G80 raster PCB sets. I don't know anything about Sega G80's and while Mark is focusing on vectors games, he taught me a lot in the 90 minutes I was there.

Now I'm interested, if for nothing else to figure out what boards in the 3 cages I have, are working or not.

Does anyone in the Chicago area have a fully working G80 game or test set up? If I could find that, we could swap boards in the cage one by one and I could at least figure out what is working and what isn't.

I was very surprised to hear (and witness) that a CPU board for a G80 raster and vector are 100% interchangeable.
 
2 questions:

1) where do i apply for being your G80 intern and helping you troubleshoot the boards (I can bring my own soldering station and G80 boards, but I have a lot to learn...)

2) Joe, after all of our special time testing gorf boards, why didn't you bring your G80 boards with you to test them?? I have a working Star Trek :)
 
2) Joe, after all of our special time testing gorf boards, why didn't you bring your G80 boards with you to test them?? I have a working Star Trek :)

I appreciate the offer, but I need a raster game, not a vector.

By KLOV definition, that's Astro Blaster, Space Odyssey, 005, Monster Bash. A couple other web sites mention Pig Newton and Sindbad Mystery.

I thought there were more than that, but I guess not. Time to check VAPS.
 
Last edited:
On with the show...

CPU12: Dead
CPU13: Dead
CPU14: Dead
CPU15: Static Roar
CPU16: Dead
CPU17: The Diag ROM reports a bad CPU RAM! Finally it pays off! Replacing one of the 2114 RAMs fixes it.
CPU18: Dead
CPU19: Static Roar
CPU20: Upon firing up, no activity. I assume its dead. Look at the screen, no activity. About to shut down the game when suddenly a flash of the LED, and the diagnostic rom comes up showing all good. Hmm, that's odd. Usually something delaying startup is capacitor related. I decided, just for giggles, to replace the capactors. There is one (sometimes two) electrolytics near the 555 circuit on the PCB. After replacing, upon retesting, the board starts right up. Could it be some of these boards just need a cap kit?
Stay tuned...
CPU21: Dead
CPU22: Tests good with Diag
CPU23: No activity, but I can hear vector chatter. Hmm...
CPU24: Dead
CPU25: Tests good with Diag

Getting near the bottom of the barrel, but running into CPU boards now that are missing some equipment (test switches, edge pins, socketed chips). I'll need to sacrifice a few CPU boards to complete some in better condition. In the immortal words of Spock: "The needs of the many outweight the needs of the few. Or the one."
 
Last edited:
Also in the words of Spock, "What a fascinating thread".

I had no idea there were raster G80 games; I always thought they were all vector.
 
It doesn't use a standard Z80 processor? Wow!
That page is very misleading. It does use a standard z80. The raster manuals mention it must be a z80A because a standard z80 isn't fast enough. However, I believe that may just apply to raster CPUs, as I've found many XYs with Z80s and they run just fine.

However, part of the bus goes through the security chip which does the mangling for the ld (addr), a instruction.
 
Last edited:
That page is very misleading. It does use a standard z80. The raster manuals mention it must be a z80A because a standard z80 isn't fast enough.

However, part of the bus goes through the security chip which does the mangling for the ld (addr), a instruction.

Ah, that makes sense. It wouldn't make a lot of financial sense for Sega to have paid for a custom version of the Z80, assuming Zilog would even have agreed to it.
 
And finally...

CPU26: Static roar
CPU27: No picture or led, but vector chatter and constantly triggers sound
CPU28: Diag reports bad RAM! Swap ram, now no activity?? Swap rams around, and diag reports good. Able to reproduce it several times, indicating that Sega's infamous crappy wipe sockets are the problem. How many problems will be attributed to these? Stay tuned...
CPU29: Dead
CPU30: Good w/ Diag ROM
CPU31: Halt light with Diag ROM
CPU32: Halt light with Diag ROM
CPU33: Dead
CPU34: Static roar

...and that's it.

Well, to be honest, I was hoping the Sega Diagnostics ROM would have been more helpful. There's not much it will be able to help with if the clock circuitry is bad, or the Z80 is dead, but maybe add an initial flashing signal to the test as a basic "clock and Z80 are executing" to quickly rule out those problems.

Next up I'll be completing my test bench, then start with checking clock signals, z80s, and so forth.
 
Back
Top Bottom