Topic: NeoGeo 6 Slot Repair Log  (Read 1966 times)

Author Message

0 Members and 1 Guest are viewing this topic.

Offline Womble

  • Amiga 4000
  • ******
  • Posts: 1190
  • Kudos 33
  • Gender: Male
    • View Profile
    • Aussie Arcade
NeoGeo 6 Slot Repair Log
« on: November 26, 2009, 11:43:51 PM »
Had Scratcha's NeoGeo 6 Slot on the bench today, fricken massive thing but unwell, it booted up giving this RAM error..



These error messages can be read like this...

..the board tried to write XXYY to the two ram chips in the memory stack at the location stated, XX to the low addressed chip and YY to the high addressed chip (or perhaps the otherway round, it doesn't really matter). So from the above error you can infer that both chips are faulty, but only if you assume that its not a problem in the RAM controller logic, which is also possible. NeoGeo boards have an onboard battery that leaks corrosive shite when they age and the tracks on these boards are microscopic, they don't like the battery acid and they just corrode away to dust, track damage is a common problem.

The second problem with NeoGeo boards is finding which RAM chips the error message is talking about, half the problem there are so many board variants so chip location advice you find for a 2 slot board is not much use for a 4, or even a 6.

Anyway - with a bit of googling I found that the address the error message refers to determines which chip type is considered faulty.

Address 0000 means the board thinks the surface mount 62256 RAM chips are the problem, if that was the error for this board I would consider it scrap, or a boat anchor, its big enough.

Thankfully this error was at address 8000 which meant it was the Sony CKX5814 chip pair in the top right-hand corner of the lower board. It also helped that my logic probe showed the lower of the pair had its I/O lines D3 to D7 stuck low. The other chip looked healthy based on pin activity alone.




Desoldered this chip and tested on my eprom burner/chip tester and it showed up as faulty. Comparing the datasheets for this chip and the chips I had on my scrap boards showed I could use bog standard 6116 chips as replacements, so I dropped one in and fired the board back up to see what the error message would give.



So now the board is writing 00 to one chip and getting 00 back, but still writing 00 but getting 04 back from the second chip in the pair.

Replaced the second chip (which also tested as faulty) and the board booted to the crosshatch screen, which is what a NeoGeo board should do without a game cart inserted, or in this case without the cart interface board connected.



Connected it all back up, plugged it into my cab and after a bit of faff seating and reseating the carts I got it to work perfectly.



Anyone know a good method for cleaning the contacts in the cart slots? The egde of the ROM cart is easy enough, but the contacts on the slot may be a pain.

Oh and one of the two carts gives no sound, I assume it has a very dead EPROM inside, but thats for another day.

The final thing to do was to remove the battery, its was starting to leak but had only reached the leg of a resistor at the moment, the battery is not really needed as all it does is keep the housekeeping RAM alive - ie how many coins have I taken, game stats etc etc.

I removed the battery from the board, another 6 months to a year and it would have made a real mess, it had already eaten through the upper layer of the board down to the copper ground plane in places.



Scraped away all the crud and neutralised the alkaline muck then gave it a clean up with PCB cleaner. The board works fine without its battery, Gamedude reckons the battery is only for book-keeping functions (ie how much money has been taken, length of games, etc etc).

Gave the slots a go over with the paper trick and it seems to have worked, much more stable, originally the slightest touch on the cart would put lines through the gfx and or crash the board - crap connections clearly.

Shes fixed!
Sic Transit Gloria Atari