Mv1ACH no sound issue + vertical sprites [Solved/Neo-Do + Neo-zmc2]

[Retro Mecano]

Working on a ACH slot with np sound issue. It has also vertical lines on game screen which appear on small area like game title. Board is very clean, no corroded traces and bettery removed.
I cleaned and reckeaned it and cart port also.

Diag bios with cart gave the following error:

Checked few traces between Z80,YM2610, and RAM near it, no damage.
Some old thread with same error suggested swapping Z80, which i did, and it gave the exact same error.
Slot boots fine with original bios, no errors.

What to look at?

 

[ack]

The error is indicating a data path issue between the Z80 and ym2610 or perhaps the ym2610 is bad. Your ym2610 seems to have a lot of rust on the legs. When you checked the traces was that will a multi-meter?

 

[Retro Mecano]

The error is indicating a data path issue between the Z80 and ym2610 or perhaps the ym2610 is bad. Your ym2610 seems to have a lot of rust on the legs. When you checked the traces was that will a multi-meter?

Yes that was with a multimeter. I do not have an ossilloscope and don't know how to use one.
Recently i grabbed this probe hoping it will help me how to diagnose, but honestly i have no idea how to use it .
Can this help me find out if the YM2610 is dead or not?

 

[Nic034]

Hello

Check trace below z80, ym2610 and 68k. Remove them using Chipquik or low melt solder to avoid damage on pcb and check then put sockets and replace them if possible to go ahead quickly (when performing repair spare parts is determinent to quickly solve issue). If corrosion is present on the board, it can be present below some chip and is often the case on AES. Use a microscope 20x or higher to inspect the complete surface, inspection is 50% of the job ! Sometimes, corrosion is present below the original mask on traces (look for dark spot or dark area on traces). First steps : inspection and cleaning !

Z80 have high rate failure on SNK MVS boards.

 

[Retro Mecano]

Hello

Check trace below z80, ym2610 and 68k. Remove them using Chipquik or low melt solder to avoid damage on pcb and check then put sockets and replace them if possible to go ahead quickly (when performing repair spare parts is determinent to quickly solve issue). If corrosion is present on the board, it can be present below some chip and is often the case on AES. Use a microscope 20x or higher to inspect the complete surface, inspection is 50% of the job ! Sometimes, corrosion is present below the original mask on traces (look for dark spot or dark area on traces). First steps : inspection and cleaning !

Z80 have high rate failure on SNK MVS boards.

Z80 is already removed, checked under it, and replaced with a new one.
For YM2610 to be honest is a pain to remove. I tried it once before and it was a real mess, with cut traces and broken legs. That's why i want to know if the probe will lead me to know of it's dead or not.
Need to learn how to use it..

 

[Nic034]

No so hard to remove using Chipquik and hot air. If corrosion is not removed, it will move on traces later. Good luck

 

[Retro Mecano]

No so hard to remove using Chipquik and hot air. If corrosion is not removed, it will move on traces later. Good luck

I took your words and removed (horrified lol) the YM2610. It went very well surprisingly. So i checked the 64 connection with a multimiter, from both sides. Continuity was ok. No corrosion.

Then i soldered on the new YM2610 and... sadly the same error..

 

[Retro Mecano]

The error is indicating a data path issue between the Z80 and ym2610 or perhaps the ym2610 is bad. Your ym2610 seems to have a lot of rust on the legs. When you checked the traces was that will a multi-meter?

As update, i replaced the YM2610, checked the traces and yet, having the same error.
Checked also continuity between Z80, YM2610, SM1 and 6216. This slot does not have NeoC1, so i don't how trace the pads to the NeoBufs.
I still don't get why with the Original bios, it works, i get no errors.

 

[ack]

Is there a NEO-BUF thats used in the sound section?

The original bios doesn't check the health of the ym2610.

 

[bodgit]

According to https://wiki.neogeodev.org/index.php?title=MV1A GA9 is noted as the NEO-BUF involved.

(Ironically I ordered some of the Furrtek NEO-BUF replacements last night)

 

[Retro Mecano]

Is there a NEO-BUF thats used in the sound section?

The original bios doesn't check the health of the ym2610.

YEs they are. Watxhed few videos of GadgetUK repairing some slots with sound issues, finds out that few were due to defective Bufs.
But i don't know really how he finds out. The guy must some diagramm printed in his memory or something...

 

[Retro Mecano]

Is there a NEO-BUF thats used in the sound section?

The original bios doesn't check the health of the ym2610.

UPDATE

I gambled on swapping Neo-D0 as it was the chip on the audio section that i didn't check. And spund is back!!

Thanks ack and nico34 for the suggestions!

No the only thnig remaing is the vertical sprites. Besides cleaning and pin check, what area is responsible for this?

 

[Nic034]

Great man, Looks like I’m starting to get a nose for it!

If this were AES, I’d tell you to check the traces on the right side pins of the LSPC2 but since it is AES, I don’t have as much experience with it haha.

If the tests pass OK with the diagnostic BIOS and you’re seeing vertical lines on the screen on the NEOGEO logo and some sprites, it’s usually an issue on the input or output side of the LSPC2. I often see this happen on AES coming from Japan..

For me, check LSPC2 and CROM data lines traces https://wiki.neogeodev.org/index.php?title=Graphic_glitches

Check for bad trace from cartridge port, MVS have nasty bad pcb traces coating..

If I were to keep going, I’d remove the LSPC2 using Chipquik (secure and easy) to check the traces underneath, then put it back. Given the state of the other chips, there’s a 50/50 chance the whole board is infested with corrosion.

Then, check data lines between LSPC2 to CHA slot.

Keep in mind that corrosion might still be hiding in areas you’ve already treated — that’s often the case

 

[Retro Mecano]

Great man, Looks like I’m starting to get a nose for it!

If this were AES, I’d tell you to check the traces on the right side pins of the LSPC2 but since it is AES, I don’t have as much experience with it haha.

If the tests pass OK with the diagnostic BIOS and you’re seeing vertical lines on the screen on the NEOGEO logo and some sprites, it’s usually an issue on the input or output side of the LSPC2. I often see this happen on AES coming from Japan..

For me, check LSPC2 and CROM data lines traces https://wiki.neogeodev.org/index.php?title=Graphic_glitches

Check for bad trace from cartridge port, MVS have nasty bad pcb traces coating..

If I were to keep going, I’d remove the LSPC2 using Chipquik (secure and easy) to check the traces underneath, then put it back. Given the state of the other chips, there’s a 50/50 chance the whole board is infested with corrosion.

Then, check data lines between LSPC2 to CHA slot.

Keep in mind that corrosion might still be hiding in areas you’ve already treated — that’s often the case

For this revision MV1AX, equivalent for LSPC2 might be the neo-GRC and the neo-zmc2. I am not sure.

But i would suspect either of them become traces to the cartridge port are fine.

I already swapped the cartridge board with another one and its the same glitch. So fault is defenitly in the main board.

I Have seen people 'simulate' some faults using emulators to track down the exact dataline for it. It's a very smart and time saving way to identify the culprit. Do you have an idea how is this done? Coudn't find clear tutorials on the net.

View: https://youtu.be/moZjqVcp5I4?sizpaJ9rVqp6O61rC

 

[Nic034]

Never see this ! Awesome !

 

[nauman]

I would recommend starting to examine the pins around the neo-zmc2. Had pretty much the same issue where one pin was not giving a correct output, which showed up as a solid line in sprites. Also had this on consoles where cartridge slot was dirty, loose and in general in poor shape, so testing continuity from cartridge slot -> neo-zmc2 is key to find the problem. Sometimes you can gently wiggle the cartridge and see the effect come and go if the slot is dirty or if solder points are cracking, although you changed the daughter board so it would then lay in the board connector or in the traces to neo-zmc2.

What the video is trying to say is that each pin would affect a line in the sprite. If you count the lines you would end up with which pin is faulty.
Personally, I would do a continuity test for all the pins (yes, they are many and it sucks), and if they are all working, I would then change the chip if possible. Have changed a few before so I would not be surprised if the neo-zmc2 gone bad.

You have come so far so don't give up now

 

[Retro Mecano]

I would recommend starting to examine the pins around the neo-zmc2. Had pretty much the same issue where one pin was not giving a correct output, which showed up as a solid line in sprites. Also had this on consoles where cartridge slot was dirty, loose and in general in poor shape, so testing continuity from cartridge slot -> neo-zmc2 is key to find the problem. Sometimes you can gently wiggle the cartridge and see the effect come and go if the slot is dirty or if solder points are cracking, although you changed the daughter board so it would then lay in the board connector or in the traces to neo-zmc2.

What the video is trying to say is that each pin would affect a line in the sprite. If you count the lines you would end up with which pin is faulty.
Personally, I would do a continuity test for all the pins (yes, they are many and it sucks), and if they are all working, I would then change the chip if possible. Have changed a few before so I would not be surprised if the neo-zmc2 gone bad.

You have come so far so don't give up now

I suspect also the zmc2. I had a surprising week where most of my repairs are swapping unusula ICs to fail. Would check that tomorrow and keep you updated.

 

[Retro Mecano]

I would recommend starting to examine the pins around the neo-zmc2. Had pretty much the same issue where one pin was not giving a correct output, which showed up as a solid line in sprites. Also had this on consoles where cartridge slot was dirty, loose and in general in poor shape, so testing continuity from cartridge slot -> neo-zmc2 is key to find the problem. Sometimes you can gently wiggle the cartridge and see the effect come and go if the slot is dirty or if solder points are cracking, although you changed the daughter board so it would then lay in the board connector or in the traces to neo-zmc2.

What the video is trying to say is that each pin would affect a line in the sprite. If you count the lines you would end up with which pin is faulty.
Personally, I would do a continuity test for all the pins (yes, they are many and it sucks), and if they are all working, I would then change the chip if possible. Have changed a few before so I would not be surprised if the neo-zmc2 gone bad.

You have come so far so don't give up now

Well iwas right. Swapped the zmc2 after checking the whole traces, and it was the cause of the sprites.

It's the second swap of zmc2 this week lol.

Thanks all for the support.

 

[nauman]

Good job! And hurray for another working board!