What's new
By:

Neo Geo MV4F restoration - some community help appreciated

DeWitt

DeWitt

Enthusiast
Joined
Jan 26, 2020
Messages
186
Reaction score
181
Location
Germany
Hello everyone,

last week I helped some nice unknown guy from the internet who lived 200 Km away to fix his cabinet. I just asked for money for gasoline and something to eat :-) 8 hours later everything worked fine and he was really really happy and I had a lot of fun! We played too long and I sleeped in my car. At the next day he gifted me an non working 4 slot board and samurai showdown:

IMG_8995.jpg


Wow, what a nice guy!! He even donated me a little bit money :-)

The board itself has an calendar error, you will see why :-D I'll document my work, so maybe it helps someone else. I'm by far no expert! I started with electronics at the end of last year. Just tell me if I'm doing something wrong!!

IMG_8994.jpg

The PCB itself was really dirty...
IMG_8998.jpg
IMG_9003.jpg

If you look at the batterie area you can see the problem... so I removed the batterie.
IMG_9007.jpg

Oh boy, that was a mess. Now I started cleaning a bit using some isopropyl alcohol and noticed gloves are necessary. My skin started burning. Don't do my mistake, ware gloves from the beginning on!! Then I used a fiber pen to remove all this mess and I desoldered every component in the batterie area. Some of them felt to dust when I touched them...
IMG_9032.jpg

I ordered some new parts and cleaned the area using vinegar and waited 30 minutes to let the vinegar do its work. After that I used isopropyl alcohol to remove the vinegar, cleaned the board with distilled water, heated my oven to 70° C, turned it off and let my boards in there over night. I also covered the solder holes with some tape I cut out. Here is the result:
IMG_9051.jpg
 

Attachments

  • IMG_9060.jpg
    IMG_9060.jpg
    712.2 KB · Views: 323
  • IMG_8999.jpg
    IMG_8999.jpg
    857.6 KB · Views: 305
  • IMG_9061.jpg
    IMG_9061.jpg
    889.2 KB · Views: 370
Last edited:
I covered the outer are with tape and paper so I can spray paint it. So no conductor tracks are exposed.
IMG_90541.jpg

I used some two component clear coat spray paint. Worked nicely:
IMG_9062.jpg

Now I would like to do a batterie mod, but I have no idea how. I have a MV4F board and found no infos what todo.

Does any one know it and can help me?

Thank you very much for your help :-)
 
So, if figured it out myself:-) I had to remove the 470 Ohm resistor. Here is the result:

IMG_9064.jpg

The solder pints on the top aren't that nice. I think it has something todo with the spray paint I used. But oll of the are nice at the bottom and all have good contact. I'm happy :-)

The battery and the holder will arrive tomorrow.

The good thing is, everything works as before. So I still get and calendar error. That will be my next task.

As always: I have no idea what I'll need todo but I'll figure it out. Any tips are welcome:-)
 
Last edited:
Here a short update. I fixed the calendar error and use the Diagnose Bioas to test everything (except the sound...)

I use the following resources to find the issue:
https://wiki.neogeodev.org/index.php?title=CALENDAR_ERROR
https://wiki.neogeodev.org/index.php?title=UPD4990
calendar chip.jpg

This way I tested every trace and noticed that pin 6 which is connected to Neo-F0 pin 3 had no contact. This sounds easy, but it was not. I hat to find and understand the informations listed above...

Then I also soldered the wire to pin 6 of D4990AC on the bottom side and I used a left over resistor leg and soldered a wire to it. I put it in one of the holes of the trace running to pin 3 of the Neo-F0 and soldered it into it.
IMG_9079.jpg


I don't now if this is the way you fix a broken trace... I did it and it worked! Pin 3 was just a little bit to small and on top of the board, so no :-D

IMG_9077.jpg

I started the Diagnose Bios with the buttons ABCD pressed and started the calendar test, perfect. Every other test parsed too.

After replacing the Diagnose Bios with a normal one and I started it... then I noticed the sound was too loud and I cloud not change the volume. Some isopropyl alcohol and sliding the volume slider a couple of times back and forth fixed it (board was turned off and waited until it was try before turning it on again).

I was so happy that I tested every slot. The colors of this board are very good. My 6 slot board picture isn't that good and sharp, i don't know why?! After playing some games I noticed the board played sometimes a wrong sound snipped and some games crashed the sound. It turned the sound completely off until I rebooted the device... I think there is an other broken trace which flips some bits...

Maybe I can use the Multi MVS cart form @Darksoft to run the sound test rom in combination with the Diagnose Bios. I need to find this out tomorrow!

Thanks to @Frank_fjs who build the minigun I can use it now in combination with my monitor. This way I don't have to run back and force between living room whit my cabinet in it and my workbench which is located at an other floor :-) Thank you!!!
 

Attachments

  • IMG_9075.jpg
    IMG_9075.jpg
    780.1 KB · Views: 352
Last edited:
I followed the instruction @Darksoft provided, thank you very much :-) Sadly it did not work to use the Multi MVS for testing... I think the reason is the following: the Diagnostics Bios does not wait until the Multi load it's ROM into the Flash... so it did not work :-/

It worked on my normal board, I heard some strange testing sounds after loading the ROM into the flash of the multi. So in theory it should work :-D

Now I need someone to burn the Diagnostics Bios M1 ROM to a chip or a Z80 testing cart (would be even better) or I have to use my eyes to look for a broken trace, but I think I need a break. Maybe I wait until I receive the UniBios chips from @Sp33dFr34k, a really nice guy :-)
 
DeWitt said:
I think the reason is the following: the Diagnostics Bios does not wait until the Multi load it's ROM into the Flash... so it did not work :-/
Click to expand...
That sounds like the most probable reason.
 
ack said:
Wanna give this version of the diag bios a try.

https://www.mvs-scans.com/misc/19a00-t1.zip

One of the things I modified in it a couple months ago was to not be so picky about the bios/m1 interaction. It will allow up to 5 seconds for the initial hello message from the diag m1.
Click to expand...
Sound promising! Thank you very much for your help, I appreciate it:-) Am I right that I need an new Diagnostics Bios chip to try this? Since I don't have an flash device I need to ask @Sp33dFr34k again if he could burn me one :S I'm not sure, maybe the was today at the post office... He is very very friendly but I'm not sure if I could ask him for an other favor... I don't want to take advantage of him either!!!!
 
Do you guys even think I can fix this? I'm not that good with replacing chips and I don't have any spare chips. I also have not found any thing like this sound issue in the world wide web :-D
 
DeWitt said:
ack said:
Wanna give this version of the diag bios a try.

https://www.mvs-scans.com/misc/19a00-t1.zip

One of the things I modified in it a couple months ago was to not be so picky about the bios/m1 interaction. It will allow up to 5 seconds for the initial hello message from the diag m1.
Click to expand...
Sound promising! Thank you very much for your help, I appreciate it:-) Am I right that I need an new Diagnostics Bios chip to try this? Since I don't have an flash device I need to ask @Sp33dFr34k again if he could burn me one :S I'm not sure, maybe the was today at the post office... He is very very friendly but I'm not sure if I could ask him for an other favor... I don't want to take advantage of him either!!!!
Click to expand...
Yeah, you would need a new bios chip for it. Just note that this is a work in progress update to the diag bios/m1. This is the full list of changes I've made over the stock version

SP1: Re-implement color bars so they work on AES
SP1: Add SMPTE color bars
SP1: Go back to main menu after manual work/backup ram test
SP1: Send error code to credit leds on MVS hardware
SP1: Attempt to auto-detect if M1 is active (ie: on AES & MV1-B/C)
SP1: Display when M1 is active
SP1: Display slot number if a slot switch was done
SP1: Display when SM1 tests were run
SP1: Pressing ABCD after automatic tests will goto the main menu
SP1+M1: Make 68k <=> Z80 communication test less finicky about timings
SP1+M1: Split 68k <=> Z80 communication into 2 error codes (HELLO vs ACK)
SP1+M1: SM1 output enable test
SP1+M1: SM1 checksum test
M1: Fix broken ram address test
M1: Add ram output enable test
M1: Add ram write enable test
M1: Use rogue YM2610 IRQ as a way to recover from a failed slot switch
M1: Fix looping forever waiting on unset of YM2610's busy bit

Code for all this is at https://github.com/jwestfall69/neogeo-diag-bios if someone wants to mess with it.
 
DeWitt said:
Do you guys even think I can fix this? I'm not that good with replacing chips and I don't have any spare chips. I also have not found any thing like this sound issue in the world wide web :-D
Click to expand...
I would try cleaning the cart slots if you haven't already. @XianXi has a good video about how to do it.
 
ack said:
DeWitt said:
Do you guys even think I can fix this? I'm not that good with replacing chips and I don't have any spare chips. I also have not found any thing like this sound issue in the world wide web :-D
Click to expand...
I would try cleaning the cart slots if you haven't already. @XianXi has a good video about how to do it.
Click to expand...
I use the geto version of an old Costco membership card with a torn up t-shirt wrapped around it soaked in 90% isopropyl.
 
@ack Okay I understand! That sounds cool, I'll give it a try.

I will also try to clean my cart slots! I think I'll try the ghetto version from@'Kujako' :-D Since I have that stuff at home^^

Thank you all for your help :-)
 
I cleaned the slots like you suggested @ack using the method from @Kujako :-D Here a my tools:
IMG_9084.jpg

The triangle hat a perfect tight fit. Sadly no changes at all... I tested every slot and nothing changed!

I also noticed that the sound test does not output any sound... here is a video of it:
https://www.dropbox.com/s/c0dow407k07s9rt/img_9085.mp4?dl=0

Here is also an video of the mixed sound problem:
https://www.dropbox.com/s/745r6wvesomjcq8/img_9086.mp4?dl=0

I also noticed an cut trace on the sound slider in the analog area, I don't know how I missed that:
IMG_9087.jpg
I would say that can't be the problem, it's in the analog are.

I also noticed something strange: the last days the right sound channel hat bad sound (some disturbing sounds) and the actual audio was quiet in the background. Today that disappeared when I plugin in my headphones via the 4 pin audio output. Now it is normal all the time. I don't complain, but thats not normal.

I have also news about the new Diagnostics Bios @ack provided. @Sp33dFr34k burned one for me :thumbsup: I'll report how it will work with the Multi from @Darksoft. I'm looking forward to this!!

Thank you all for helping me out!!
 
Last edited:
Thanks to @Sp33dFr34k, the Diagnostics Bios arrived this weekend, thank you very much:)

I soldered a batterie onto the MVS board (for config saving) and replaced the m1rom from joy joy kid: small ROM = fast loading : D Then I started the board, thanks to @ack the bios it’s much easier to navigate. I enabled "boot to game" with only one game configured and waited a minute or two to make sure the Multi loads the rom. Then I soft restarted the MVS board and pressed D to start the Z80 test. Thanks to @ack the bios waited some time and then started testing. This was the result:
IMG_9110.jpg

I also tried an random MVS cart, same result, different address… Since this is not an listed error, I think the bios can’t find it M1 rom for testing. Just a guess. What do you think @Darksoft? I think it's not possible to use the Multi for this test.

Btw when I use the M1 Rom without the Diagnostics Bios I get the same sounds on this board and an other board I tested.

I also ordered my self a Logic Probe and found out that the YM2610 does not output the bios testing sound. But I haven't tested further... I just found that out because I compared it with the YM2610 of an other board ;) Every testing tips are welcome!
 
looks like z80 and 68k cant communicate properly. I would test that first with an original cart. If the error is the same, most probably this is coming from the circuitry/shared ram between those 2. Could also be a bad z80.
 
You must log in or register to reply here.
Back
Top