CPS3 : Errors 13/23/33 when flashing new games.

[penrhos]

I sold a CPS3 setup on eBay and it was thoroughly tested before I posted it out - I loaded all games from CD and played a couple of goes on each one (so about 6 hours on soak-test). Boxed it up in an anti-static bag and plenty of pink bubble-wrap.

Buyer received it, left positive feedback and it was fine for 48 hours then I got a "Item not as described" refund requested from eBay out of the blue...

On pressing him for more information than "It doesn't work" it appears he tried flashing "Red Earth" and it failed, then he said he got CD-Rom not ready (which I'd warned him about as drive is a bit slow to spin-up and to wait a few seconds for it to clear) but he power-cycled the PCB a couple of times instead. Now he's getting repeated error 13/23/33 and it won't write any games.

I had him return the PCB (even though I think he's damaged it) and of course eBay insist I pay return postage, even though he'd left positive feedback and admitted he'd been re-seating the simms.
It arrived last night and I'm pretty sure its the same simms as I sent it out with but not 100% sure he hasn't swapped any over.

The error messages 13/23/33 point to simm #3 being at fault so I took it out and put it in the spare slot in my CPS3 setup:-

• erased OK,
• write test OK,
• checksum OK,
• I can program it with any file from the CD calculate checksum, reprogram it and the checksums match,
• put it back into the original pcb and "Error 13".

So I decided to take all the simms out of the PCB and perform the same tests on them:

• 2 * 64Mb Fujitsu simms- OK,
• 3 * 128Mb Fujitsu simms - OK,
• 1 * 128Mb Intel/Sharp simm - erases OK, programs OK but write test fails.

But the intel/sharp simm was in slot #4 not #3?

Re-filled the PCB with known good simms and it now works fine. So put on soak-test for a few hours and re-programmed it with no issues at all.

I can't prove that he's swapped any simms or he's damaged them so I've had to issue a refund.

Anyone come across these intel/sharp 128Mb simms and are they any less reliable to the Fujitsu ones?

 

[Mitsurugi-w]

They are no less reliable, they are just less common.

Also, I can tell you from experience testing every repro simm that the error does not always point to the simm slot. Some times it can be very difficult to determine the bad simm unless you have a few spares to swap out one at a time.

 

[penrhos]

They are no less reliable, they are just less common.

Also, I can tell you from experience testing every repro simm that the error does not always point to the simm slot. Some times it can be very difficult to determine the bad simm unless you have a few spares to swap out one at a time.

So what's the best way to test/verify a simm?

I've been using the free slot and performing an "Erase", checksum, write with any full sized file from CD, checksum and repeating the process a few times to see if the checksum's match. then doing a few "write tests" to check for any NG or fails. Finally loading up SF3:3S and leaving it to run for a while before playing a game.

 

[Mitsurugi-w]

Your process would work well also for an individual simm.

 

[penrhos]

Your process would work well also for an individual simm.

OK - I'll assume testing these that way has identified the issue and it looks 99% certain its the Intel/Sharp simm.

I'll get the hot-air rework station out and re-flow the chips in case the simm has been "flexed" and caused a dry-joint on one of the legs (as the fault is intermittent).

I need to dig it out anyway as I have six 32Mb simms and a faulty 128Mb Fujitsu I might be able to salvage.