My Timeharvest Neo Geo CMVS seems to be fried for a 2nd time

[bobrocks95]

My consolized MV-1C from Timeharvest has decided to die yet again. Seems like every time I sit down and decide to seriously play something on this, it kills itself.

I am getting garbled multicolored garbage on boot, with a cart inserted, and without. Metal Slug froze after a uniBIOS soft reboot, and when I turned it on again later it's been nothing recognizable since.
Here's a picture: https://i.imgur.com/9MjkuUP.png

So I have had the entire MV-1C board on this unit replaced once before, and now after light use it has broken again. Last time I thought maybe after the whole 10 minutes I spent trying out a multicart, that the 3.3V flash on it was enough to break something. The person I sent it to said the diagnostic BIOS they put in indicated the CPU was fried. Since then I've only used genuine original carts and that didn't make a difference.

So clearly something is horribly wrong with this CMVS. My first offhand guess would be the cheapo Chinese power supply it came with. It's super light, and supposedly 12V 3Amp center positive. I'm not really sure how to start checking a PSU to see what it's outputting, or how to verify in the circuit that it's doing its job correctly. Maybe I can probe on the board where 5V is actually expected and see if it's getting that?

1) Is anyone in the US able to flash me a diagnostic BIOS chip I can buy from them? Maybe JNX? Plenty of sellers in Europe but that's going to take a month or two considering the holidays and covid.
2) How can I test voltage from the power supply with a multimeter? How am I supposed to know if 12V is even what I'm supposed to be using, given that's just going off of what Timeharvest put on the back?
3) How did this piece of crap cost me $300...

I have a self-installed NeoBIOSMasta VMC that I've double-checked everything on, and I'm assuming I will be able to pop the diagnostic BIOS into its socket to use it?

[Josh128]

Pink garbage on boot very often indicates no BIOS, bad BIOS connection (when using the NeoBIOSMasta), or dead BIOS. I would 100% look in that direction first. Those things are NOTORIOUS for developing bad connections to the CPU from the upside down mounted CPU socket. So your EEPROM is socketed into the NBM PCB and not soldered? Even more chance its a bad connection.

[jd213]

Hmm, I haven't used mine much recently, but I do remember it being slightly picky with MVS carts, so I often had to re-insert them until it recognized them properly.

[bobrocks95]

Pink garbage on boot very often indicates no BIOS, bad BIOS connection (when using the NeoBIOSMasta), or dead BIOS. I would 100% look in that direction first. Those things are NOTORIOUS for developing bad connections to the CPU from the upside down mounted CPU socket. So your EEPROM is socketed into the NBM PCB and not soldered? Even more chance its a bad connection.

Thanks for the tip, I'll reseat the EPROM and the NBM. I hadn't heard complaints about it when I bought it but we'll see.

Would still like to know any test points I can check for voltage. I might replace the PSU either way, I dunno.

Hmm, I haven't used mine much recently, but I do remember it being slightly picky with MVS carts, so I often had to re-insert them until it recognized them properly.

I get the same garbage when booting without a cart inserted.

[bobrocks95]

Pink garbage on boot very often indicates no BIOS, bad BIOS connection (when using the NeoBIOSMasta), or dead BIOS. I would 100% look in that direction first. Those things are NOTORIOUS for developing bad connections to the CPU from the upside down mounted CPU socket. So your EEPROM is socketed into the NBM PCB and not soldered? Even more chance its a bad connection.

Reseated the NBM and the uniBIOS EPROM, checked continuity on the socketed EPROM with the socket pins on the bottom and it's all good, but still pink garbage. I guess I'll go ahead and order a diagnostic BIOS from JNX, it'll be useful enough to have even if I miraculously solve the problem before then.

In the meantime, anything else to try? I suppose I could hook the "original" BIOS chip back up (it is a Macronix chip with 4 legs cut off, definitely not original to the system) and remove the NBM, though again this is the second time this Timeharvest has self-destructed so I feel like the problem lies elsewhere.

EDIT: Cleaned the CPU pins at JNX's suggestion, still no go. I don't actually have a fiberglass pen though like he recommended, I just scrubbed a little with a q-tip and some crc contacts cleaner. It does definitely seem BIOS related based on your pic Josh but I'm not sure what to do. Maybe the EPROM with the UniBIOS on it is outright dead? Wouldn't figure there's a way to test that other than buying another one.

EDIT 2: Reconnecting the original BIOS chip gave different results each boot but still overall a garbled mess. I saw a solid blue screen, a solid urine yellow screen, a bunch of squares with vertical lines through them, some 1-pixel white dots that flew up the screen rapidly...

It's always possible I damaged the original BIOS chip or didn't reconnect it properly, though continuity between the nearby vias and the pins on the chip was good.

EDIT 3: Probing on the CPU VCC get's about 5.14V so it's a little more than expected but I don't know if that's a damaging amount. The JAMMA edge is getting +5.11V on its -5V pin and 5.11V on its 12V pin so I don't know if that's expected (not sure how the 12V line is used). 12V adapter that came with it shows 12.61V just probing the barrel connector while unplugged, and 11.8V in-circuit on the 12V points.