Crazy problems on 2 MV1-1 boards

Hi guys, this is my first post here in this forum.

I´ve read a lot over the internet during the past weeks trying to solve my problems, and decided to ask for more help. I´ve already posted a thread on NLG forum but got no answers.

I live in Portugal, and I´ve found a pair of arcade cabinets at good price with several damage to the monitors and game PCBs due to rodent and bird dejections sitting on the boards for the past 20 years… I´ve fixed both monitors, and sold one cabinet.
Now I have a working jamma cabinet, and two SNK MV1-1 Boards both with serious damage due to battery leak and animal cohabitation.

They are the MV1-1 version with a LSPC-A0, PRO-CO, NEO-ZMC2, NEO FO, PRO-BO, NEO-GO, NEO-EO, NEO-DO, NEO-IO, etc…
Exactly this version: https://www.mvs-scans.com/index.php/MV1-1

At the time of purchase, I tested the boards as they were on the cabinet, and nothing worked well. (If I knew what was inside the cabinets, I would never turned them on. I know it was a big mistake.)
Now, I decided to clean everything and test the boards again to diagnose and try to repair them, if possible. They are both crazy as expected from the visible damage of the battery leakage and animal activity… But I really want to fix these boards and also want to learn a lot more about these amazing systems, so I started working on them.
I have some experience regarding electronics repairs and finding faulty simple components, so I already tested the monitor for good and the power supply is working fine and clean. I have a pandora´s box for testing arcade machines and it works flawlessly on this cabinet, but I dont like it. The ideal solution would be to repair the two boards, and sell one. If not possible, repair one to fit in the cabinet and save another for spare parts.

But here is what I´ve already done to the boards:
Replace all 470uF16v. caps, because they were measuring 250uF capacitance.
Replace all 100uF caps, because they were measuring 25uF capacitance.
Removed batteries and 470ohm resistors. Cleaned all battery leak damage with IPA. Let boards dry in the sun.
Fixed all visible bad traces (dozens), and continuity tested everything found broken on the two boards.
Continuity tested the following list:
MV1-1 (my version of two continuity lists combined, that I´ve found in another thread, and shows continuity in these MV1-1 boards) I adapted some pins that I´ve found a bit different on this version, so if you have any official continuity list for this version of the board, please let me know.

SP1 BIOS (LSPC-A0 version of the board)
1----5V
2----GND
3----68000-12 (54), LSPC (23), NEO B0 (60), NEO G0 (50), 43256 (19) D2
4----68000-12 (55), LSPC (22), NEO B0 (59), NEO G0 (49), 43256 (18) D2
5----68000-12 (56), LSPC (18), NEO B0 (58), NEO G0 (48), 43256 (17) D2
6----68000-12 (57), LSPC (17), NEO B0 (57), NEO G0 (47), 43256 (16) D2
7----68000-12 (58), LSPC (16), NEO B0 (56), NEO G0 (34), 43256 (15) D2
8----68000-12 (59), LSPC (15), NEO B0 (55), NEO G0 (33), 43256 (13) D2
9----68000-12 (60), LSPC (14), NEO B0 (54), NEO G0 (32), 43256 (12) D2
10---68000-12 (61), LSPC (13), NEO B0 (53), NEO G0 (31), 43256 (11) D2
11---GND
12---68000-12 (62), LSPC (12), NEO G0 (18), 43256 (19) Backup RAM
13---68000-12 (63), LSPC (11), NEO G0 (17), 43256 (18) Backup RAM
14---68000-12 (64), LSPC (10), NEO D0 (3), 43256 (17) Backup RAM
15---68000-12 (01), LSPC (9), NEO D0 (2), 43256 (16) Backup RAM
16---68000-12 (02), LSPC (8), NEO D0 (1), 43256 (15) Backup RAM
17---68000-12 (03), LSPC (7), NEO D0 (64), 43256 (13) Backup RAM
18---68000-12 (04), LSPC (6), NEO D0 (63), 43256 (12) Backup RAM
19---68000-12 (05), LSPC (5), NEO D0 (62), 43256 (11) Backup RAM
20---NEO-GO (59)
21---68000-12 (29), PRO C0 (60), NEO E0 (64), 43256 (10) C2 D2, HC259 (1)
22---68000-12 (30), PRO C0 (61), NEO E0 (1), 43256 (09) C2 D2, HC259 (2)
23---68000-12 (31), PRO C0 (62), NEO E0 (2), 43256 (08) C2 D2, HC259 (3)
24---68000-12 (32), PRO C0 (63), NEO E0 (3), 43256 (07) C2 D2, HC259 (13)
25---68000-12 (33), NEO B0 (92), NEO E0 (4), 43256 (06) C2 D2
26---68000-12 (34), NEO B0 (93), NEO E0 (15), 43256 (05) C2 D2
27---68000-12 (35), NEO B0 (94), NEO E0 (16), 43256 (04) C2 D2
28---68000-12 (36), NEO B0 (95), NEO E0 (17), 43256 (03) C2 D2
29---68000-12 (37), NEO B0 (96), NEO E0 (18), 43256 (25) C2 D2
30---GND
31---68000-12 (38), NEO B0 (97), NEO E0 (19), 43256 (24)
32---68000-12 (39), NEO B0 (98), NEO E0 (20), 43256 (21)
33---68000-12 (40), NEO B0 (99), NEO E0 (21), 43256 (23)
34---68000-12 (41), NEO E0 (31), 43256 (2) C2 D2, LS05 (01) (R9)
35---68000-12 (42), NEO E0 (32), 43256 (26) C2 D2, LS05 (03) (R9)
36---68000-12 (43), NEO E0 (33), 43256 (1) C2 D2, LS05 (05) (R9)
37---68000-12 (44), NEO E0 (34), LS05 (11) (Q9)
38---5V
39---5V
40---5V

Only the “LS05” don’t give continuity on both boards, but if I follow the traces, visually I can’t find any physical connection.
Now I dont know what else to do, because everything I tested is OK, and I don´t find any more broken traces. I have a small DIY oscilloscope (DSO138) but I dont know what to look for in the TTL comunication, and I dont know if I can use it to test logic circuits. It supports a maximum freq of 200KHz. Can I use this if I don’t have a logic probe?


So, the behavior of the boards is like this at the moment:
Board 1 (worst damage): Does lots of green garbage on the screen, flickers and clicks like the watchdog symptom, but if I let it on for a while (2or3 min) sometimes it stops reseting and shows fixed green trashy squares on the screen. Same happens if I remove the bios Chip. Bios chip makes no difference in this problem.
If I pass my finger gently on the top of some RAM chips it influences the garbage on the screen. Nothing is getting minimally hot, except for the YM2610, and lower 5814...

Board 2: Boots to weird text like:
IIEE AA RROO
DDRRSS IIEERRAADD
00000022 5555 5500

Sometimes, if i let it on for a while, it starts to show white garbage on the screen… If I pass my finger gently on the top of ZMC2 IC it influences the garbage on the screen. If I tap in it with something plastic it has no effect.
Some chips get a little warm but the warmest is the YM2610.
All this tested with no game, or socket board inserted. I have a world heroes 2 cartridge that I tested on a four slot MVH and it works fine!
I tried flexing the board, or pressing on the chips, but nothing changed. IF I leave the board on the sun, and power it on right away, the garbage on the screen changes a bit.

I really want to fix this boards, first of all because I can´t stand the idea of trashing this awesome systems. I´m a console collector, I love gaming systems, and I love to fix stuff and want to learn as much as I can about electronics, specially repairing old classic games machines to keep them running. Home consoles and arcades. Bringing these old things back to life gives me a lot of joy… and headache :P

I´ve read a lot of amazing repair logs from "channelmaniac", but I didn´t find any matching these symptoms...
Is there any fan made schematics on these systems that shows what continuity should be between all pins of all chips? I can’t find schematics for this model, so if I can help build one too, I´m available to contribute!
Is there any signal I can measure with my DIY oscilloscope or multimeter to track down the problem?
I have some videos and images of the problems:
Board 1:https://www.youtube.com/watch?v=pmfjtrPJqmY

Board 2:


Sorry for so many questions!

Thanks for any help, and I´m sorry if my English is not perfect.
Best regards!
T.S.

Could you upload some photos of the boards? That one with the crazy text - looks like it could be a clock issue somewhere, in that it seems to be outputting the same character over and over for a number of cycles maybe. It's interesting how the numbers seem to output OK though - ie. the 5555 5500 part, which indicates that the lower of one of the RAM is bad somewhere. Have you tried the SMK Dan diagnostic BIOS yet - I expect the same results, but worth trying - certainly on the one that seems to be watchdogging.

With regards to scoping signals, you could start by looking at the pinouts for the PRO-B0, PRO-C0, and LSPC-A0 - see the pink clock pins on the diagram, scope those to check those chips have their clocks.

Thank you for your answer.

I dont have an EEPROM programmer and I dont have another BIOS than the original.

You might be right. The little crystal next to the battery got corroded in one leg on the board that is watchdogging, I´ve already swapped crystals between boards with no different result in each board. Maybe both clock crystals are dead. I will try to check with this oscilloscope, but I might need to ask a friend for a decent one.

So here are some pics:
Board 1:



Board 2:

If the clock crystal was dead it wouldn't be doing anything I think. So crystal will be OK. I noticed that board has a D0 chip - that makes checking the clocks a bit easier. Just look up the pinout of the Neo-D0 on the neo dev wiki and scope the pins on that side where the clocks come out. Are all the traces around the battery OK?

EDIT: And in fact, the crystal you are looking at there (near the battery) is the RTC crystal for the 4990. If that was faulty you would just get a calendar error.

I spy some corrosion around the ZMC2. if the pictures were higher resolution we could help more. Do you have access to a high resolution camera?
also, as recommended to me, there is a special bios made by Dan for the Neo Geo specially for troubleshooting.

and why are there pen markings on the ram chips? That's kind of weird

several damage to the monitors and game PCBs due to rodent and bird dejections sitting on the boards for the past 20 years…

Click to expand...

wow, you are one brave soul.

Board 2 looks like a vram error.

Thanks everyone who´s trying to help!

I´m sorry that i´m a bit absent, but i´m trying to find someone who could lend me a decent oscilloscope. Mine doesnt read any signal above 1Mhz. I will also need na eeprom programmer, and it will take a while to arrive.

I will try to see if there is any more corrosion around ZMC2, and I will try to take better pictures of the board with better lighting. To make this post I resized the pictures to match a decent file size.

Please dont care for the markings I´ve made on the ram chips. I just did that for orientation purposes when testing continuity with the multimeter.


Thank you all and I will try to come back to these boards and this thread during this week.

ftadeufs said:

and I will try to take better pictures of the board with better lighting. To make this post I resized the pictures to match a decent file size.

Click to expand...

The difficulty with the pictures isn't so much the lighting as it is the size / resolution. You can upload high resolution images very easily these days with sites like imgur.com

Hello Everyone!

Sorry for the delay in my feedback, but i´ve been waiting for eproms since the last time i replied to this post. For two times they failed delivery.
Finally yesterday I received some M27C1024-10F1 in the mail.
I also received an EISTAR LP-1 logic probe (bought one in case I would need it too).


So, I used a TL866 Plus II to program them. One with the unibios 3.3, and one with the SNK Dan diagnostics BIOS.


I tested both in board 1 (watchdoggy): Same result as original bios. Watchdog kept doing its thing in both BIOSes.
I tested both in board 2: I got 2 diferent results. Unibios boots to crosshatch but with errors on screen (see pictures below), and if I press B button it moves forward to a garbage error screen. SNK diag BIOS boots to garbage graphics.


Now probably its time to check for clock signals and comunication between the IC´s. Is this logic probe usefull in any way? I´ve tried it on the clock signals and it doensn´t pulse. lots of the times it just says HI (1) or High impedance.

I might have an 100MHz HP1740A analog oscilloscope by monday, would that be helpfull? I found one Cheap (120€ inc. 2 probes and manuals) near here, and with that I´m sure I can check perfectly all the clock signals.

Here are some board pictures in IMGUR [phone camera ]:
Board1: https://imgur.com/V6JazAm
Board2: https://imgur.com/mbxGHil


So, here are some pictures of the Boards with the diferent BIOSes:
Board 1 - Any BIOS same result (watchdog):


Board 2 - UNIBIOS:


Board 2 - SNK DIAG:



Any ideas?
Thank you all for the help!

Hello Guys,

I´ve checked all the clock signals on all chips of Board 2, and all Clock frequencies are OK. I´m focusing on the board 2 because it seems to be the best one to fix first.
I managed to purchase a 100MHz oscilloscope, and measured all the clock signals, that seem to be OK in frequency, but the waveforms and amplitude are not constant. I really dont know if its OK if the frequency is good but the amplitude is 3vpp... OR if the waveform is not a very square wave, and with some peaks… please see the pictures below. are these normal waveforms?:
Cristal:


Other (dont remember which):



Thank you all for the help.

Olá

Hi

The pictures aren't of the best quality, its difficult to see.

The boards still look a bit dirty , and if I'm seeing alright, there seems to be a lot of corroded pins in some of the mem's / ICS.
( mainly near the crystal, close to the battery ),

I would first and foremost: remove all the socketed parts ( bios, etc)

And then go and just wash the boards , with slightly worm water and soap ( not joking )

I've bought an 2 slot some years ago ( OLX ) and it was filthy with dirt, rats poop, etc, and that's what I did ...

After letting it dry at the sun for some days, I reassembled the memories and.bios ( sockets) removed the battery that had leaked already, but no damage done in the PCB luckily, and the only fault it had was just a bad backup ram error ( that gets ignore if you own an unibios) couple weeks later I replaced all the backup rams and color as well, since this Sony rams tend to cause Much trouble

Cheers

Cts

Obrigado! Thanks!

I already left them in the sun for some hours and the watchdog sometimes stops. But it only happens when it´s warm. I tried slighly heating with a heat gun while powered on, without success on stoping the watchdog.
The second board starts showing garbage on the screen when warm.

I will try to remove all the thru hole components next to the battery and wash the boards in distilled water with detergent.
I already removed all those ICs before, and tested in a TL866PlusII and it gave a pass test on everyone, but I´ve seen some cases of false positive results.
I will also try to reflow all SMD solder joints.
All the continuity tests I´ve done in the list above were OK, but I might be missing some other lines/vias, etc…

What are the cheapest/best ram chips for these boards?



Thank you!
Abraço

The Sony rams I replaced where CXK5814 but I replaced them with some nice Toshiba TMM2018 ( video ram) and while i was there i put the rams on sockets to make things easier in the future , these Toshiba ones are much much better

The backup rams I don't recall, I think I replaced with some generic ones ( I also put them on sockets )

Cheers

Hello Guys! Thank you all for the help provided! I´ve fixed the second board!

After trying the Unibios I could boot into the games, but with lots of graphical glitches. Got crazy with it and replaced all the Work/Video/Backup RAM chips (with cheap SEC KM62256CLG-7L) , and the graphics behaviour changed a lot for better. Now the on screen text was readable! Still some glitches, and with the DAN diagnostic bios, a VRAM error showed up. I replaced both the fast vídeo RAM chips with TMM2018 as sugested (One was faulty, but replaced the two), and now everything works flawlessly with the second board!

I almost can´t believe that after all the junk that was on this board, that it now works!! I´m happy Spent some money on spare ram chips, but it was worth it! … the only regret I have is that I popcorned it a little with the hot air gun when removing the old ram chips … idiot!
But the fact that nothing functional was harmed, makes me happy anyway!


Now to the watchdoggy one!
So, when a board is watchdogging, where do you start to trace the problems? CPU Reset lines? Or you go randomly until you find a floating pin or something?
First thing I´ll do is recheck all the clock signals again.


Thank you all!

I´ll post pictures later this week

Hello guys,

Replaced all the ram and fast ram chips, to see if the first board was watchdoging due to a defective ram chip, but nothing changed. now it is really time to trace signals with the logic probe or oscilloscope. I checked all the clock signals in NEO-D0 and they are all fine.

I powered both boards without bios chip installed (That way both are stuck in watchdog). The working one has the bios ROMOE* pin 20 High pulsing low, but the one that still doesn´t work is stuck HI without any pulsing. I tracked signals from NEO-G0(59) and its output is stuck HI. I checked NEO-G0 HC04 10(pin 51), in the working board it pulses but on this one its stuck HI, but I traced everything back to PRO-C0, and theres no signs of pulsing. Could the PRO-C0 be dead?

thanks!

I don´t know where to go from here, because i checked a lot of traces and everything is good like in the working board.



I´ve been checking signals on the RAM chips, and I get stuck Hi on all the ram WE\ And OE\ , and CE\ is low as expected.


On the 68K all the addres lines are pulsing together with the watchdog, but there is no activity on the data lines. they are all stuck: Some high, some low. Is this any indication?

On the board I fixed, both data and address lines are pulsing with the watchdog without BIOS inserted.

I wish I could learn more about all these signals and trace down the problem. I can´t find any more broken traces…


thank you all.

In a last attempt to fix this board, i dessoldered all of the logic IC´s and tested all good. I Socketed everything.

I´m starting to think I have a bad 68K or PRO-C0.

Anyone have any ideas what to look for next?

thanks.