Need help with two CMVS MV1C boards

[JLL]

Hello and happy new year

I need help with two consolized MVS MV1C boards (which I'll call A and B). They come with one supergun adapter (made in china) for the inputs and the MVS-CBOX-A in the back for component and RGBS output.
Both boards have been confirmed partially working (correct video output) but:

• board A has distorted / saturated / missing audio
• board B has no audio at all

While investigating board A, I saw that the spot AC1 had a 470uf capacitor soldered on but the specs tell me this needs 47uf. Being a spot fairly close to the sound chip, I thought this was likely the culprit so proceeded to desolder it and replace it.
However, during this small surgery, I apparently lost the video (or more than that ?) - now the screen is entirely black (connected to an OSSC over component, the OSSC doesn't detect anything at all whereas before it would detect a signal immediately)
I was being very careful and nothing out of the ordinary happened, there is no obvious damage to the board etc

I tested the supergun/cbox-a parts with board B, video still works so something definitely happened to board A while I was replacing the capacitor.

I tested the video output plugs on board A with a multimeter, I see some voltage and activity but those values seem vastly different from what I get on board B.

I can also sense the chips heating up on the board, which tells me there's proper power flowing through but beyond that, I'm a bit lost.

I also tried to see if just the video was lost by plugging in a cart but no sound ever came out of the speakers (before that issue, the neo-geo boot up chime would be audible)

How can I go about debugging this ? I have basic soldering/desoldering tools and a multimeter but unfortunately no oscilloscope.
Any help appreciated.

Thank you.

 

[Neo Alec]

A diagnostic bios could be a help for both. What's the bios situation on these? Is there an adapter for a drop-in chip?

 

[BIG BEAR]

Put the 470uf capacitor back where it was and check your volume pots.
BB

 

[JLL]

Both boards still have their original bios chip soldered on, although board A seems to have had some modifications done to it (I see two legs lifted up). When board A video was working, I could see an unibios 3.3 menu at startup so that was probably the modification done to it ?
I could technically desolder the bios and put an IC socket but I'm also reading that MV1C has a different pinout and doing that without a neobiosmasta would be a pain in the ass.

I will try with the 470uf back on again and see if there's any sound coming back.

 

[maki]

board A:
clean the slot like it was your genitals before you put them on ebay for sale
AC1 should be a 4.7uF cap according to console5

Neo Geo MVS - TechWiki

wiki.console5.com

on mvs-scans it shows as small cap, no way this was a 470uF IMO

MV1A - MVS

www.mvs-scans.com

don't have one here at the moment to check myself

board B:
check if it had a 5V mod done, this will disable on board audio and requires a stereo mod PCB (passive, or better yet the active one)
also, you can check if the problem is on the digital part or the analog part quite easy with a logic probe

 

[JLL]

Both boards still have their original bios chip soldered on, although board A seems to have had some modifications done to it (I see two legs lifted up). When board A video was working, I could see an unibios 3.3 menu at startup so that was probably the modification done to it ?
I could technically desolder the bios and put an IC socket but I'm also reading that MV1C has a different pinout and doing that without a neobiosmasta would be a pain in the ass.

I will try with the 470uf back on again and see if there's any sound coming back.

Actually board A has a custom chip with unibios 3.3 on it (just checked, I didn't realize the chip was different)
Board B still has the original bios.

Put the 470uf capacitor back where it was and check your volume pots.
BB

Just tested with the 470uf back on and tried to fiddle with VR1 while booting up the board with a cart on. Couldn't hear any sound at all either.
It should also be noted that for the sound, it is tapping directly on the backside of the board for stereo output with a left/right and ground cable, connected to the cbox part at the back that came with the CMVS.

board A:
clean the slot like it was your genitals before you put them on ebay for sale
AC1 should be a 4.7uF cap according to console5

Neo Geo MVS - TechWiki

wiki.console5.com

on mvs-scans it shows as small cap, no way this was a 470uF IMO

MV1A - MVS

www.mvs-scans.com

don't have one here at the moment to check myself

board B:
check if it had a 5V mod done, this will disable on board audio and requires a stereo mod PCB (passive, or better yet the active one)
also, you can check if the problem is on the digital part or the analog part quite easy with a logic probe

Unfortunately those boards are not mine, I'm just trying to repair them for a friend so cannot sell them.
As per your link, AC1 should be 47uf for the MV1C (both boards are MV1C)
And I agree it is a small cap on all pictures I could find online. The one that was soldered on was a big one, hence the potential issue.

As for board B, there's no 5V mod that I can see (board B is largely untouched compared to A, except for some sketchy bits around the back up area but all the chips still look like the original ones). I will look at getting a cheap logic probe to add to my tools and will report back.

For the time being, I will try to focus on restoring board A to a working state (with video at least). Worst case, I was given the green light to salvage parts from one and make the other a complete, working board, if necessary.

 

[JLL]

Here are some pictures of board A as it came to me (and it was working)

Back-side, around the center, a trace seems to have been cut (there's still continuity in the other traces)


Back-side, back-up area


Back-side, two resistors were added on top of the original ones ?


Back-side, PC25 and 26 are missing


Back-side, a bridge that makes the fuse irrelevant ?


The custom bios chip with unibios 3.3

 

[JLL]

I did some comparison tests between board A and B and I noticed that the NEO-GRZ chip on board A heats up very quickly, to the point I can not keep my finger on it without burning myself (within just a few dozens of seconds after powering the board on). All other chips seem fine (NEO-YSA2 heats up a bit, 68k doesn't seem to heat up at all, NEO-DCR-T not warm at all either)
The two RAM chips on the backside of NEO-GRZ (RAM5 and RAM6) also seem to heat up very quickly but RAM1 and RAM3 don't

The NEO-GRZ on board B (which is working fine at the moment except for no audio) is barely warm to the touch after several minutes.

 

[maki]

Back-side, a bridge that makes the fuse irrelevant ?

Yes, it does.
Is that the board with the hot GRC?
If so, it sounds like its toast/fubar.

 

[JLL]

Pics above are from board A which also has the hot chip yeah.
Is there any way to make sure ? Honestly, nothing eventful happened that could explain the board/chip dying like that. I'm hoping it's just a benign symptom that could be easily fixed.
Some power issue somewhere maybe ?

 

[maki]

Pics above are from board A which also has the hot chip yeah.
Is there any way to make sure ? Honestly, nothing eventful happened that could explain the board/chip dying like that. I'm hoping it's just a benign symptom that could be easily fixed.
Some power issue somewhere maybe ?

That board has a jumped fuse, quite likely IMO that the chips was burned a long time ago, its not supposed to get hot, just warm.

 

[JLL]

So I put the board A aside for the time being and got a logic probe (Elenco LP-560) to test board B.

If I understand this correctly, according to the datasheet for the BU9480F I need to test pin 6 to see if there's any incoming signal from the YM2610 (which is integrated into NEO-YSA2)

And apparently there is no input signal (Hi: off, Lo: on, pulse: off, sound: constant low, which tells me logic 0 / no pulse activity)
On the NEO-YSA2, is the signal coming out of pin 59 (labeled SDATA) ? (logic probe gives me the same result when testing pin59)
If so, does this mean I have a bad NEO-YSA2 chip ?

EDIT: just tested all inputs on P1 and P2, everything works. I find it hard to believe that a defective NEO-YSA2 would only take the sound out
EDIT2: cleaned the slot and cart connectors with IPA, still nothing

 

[BIG BEAR]

I was hoping it wouldn't come to that but yes,it is possible.
I've had some where the sound area was fine and some inputs were bad and vice versa.
I just end up replacing the YSA2 using a board I assign as parts
BB

EDIT: just tested all inputs on P1 and P2, everything works. I find it hard to believe that a defective NEO-YSA2 would only take the sound out

 

[JLL]

Damn ok, is there anything else to try ?
I really don't want to desolder two of those chips, without any guarantee the transplant would work

 

[BIG BEAR]

I've never tried this but I was thinking if you could power up both boards at the same time and run a wire from board A's pin or a known working mv1c to board B?
In theory, it should work.
BB

Damn ok, is there anything else to try ?
I really don't want to desolder two of those chips, without any guarantee the transplant would work

 

[JLL]

I think I've found the culprit: for some reason, I have continuity between SDMRD and SDROM. On board A (which had sound, albeit broken) there's no continuity.

Can someone confirm that SDMRD and SDROM are not supposed to be connected ?

 

[Xian Xi]

Was the video working via component to the OSSC prior to the cap removal? I thought no component encoders on Neo's worked on the OSSC inputs.

 

[JLL]

Component video was working with the OSSC yes (board B still outputs component video just fine to the OSSC as well)

I haven't had time to progress on this yet, should have more time this weekend and will report back what I found.

 

[JLL]

It was absolutely invisible to the naked eye (and thanks a million to my Pixel 7 phone with a 8x digital zoom) but I could see there was actually a bridge between the two pins of SDMRD and SDROM on the NEO-YSA2



This avoided me a very long and painful desoldering session.
I just cleaned the pins, used some flux and melted that tiny bit which reflowed nicely.


And it works ! When I booted the cart, the neo geo chime was absolutely perfect
I have no idea why that little bridge was there in the first place. It looked pretty old (not something that was done recently) so I'm thinking the board was likely defective from the start and no one bothered trying to fix it until now.

As for board A... (I've spent too much time on this already and my friend was eager to have his MVS back so I cannot work on board A anymore)

Thanks to everyone for your time and help

 

[Neo Alec]

Nice work. I use my phone as a magnifying glass for soldering too. Looks to me like someone accidentally dropped hot solder on it.

 

[maki]

I made a similar bridge on the G0 that caused only one slot to get recognised.
Not uncommon these self made bridges

Don't trust your eyes, trust your continuity tester