Jump to content
Due to a large amount of spamers, accounts will now have to be approved by the Admins so please be patient. ×
IGNORED

Neo Geo MV-4 Repair


Recommended Posts

Hi guys,

 

One day I'll make one of these and actually finish it but for now here's another one I'm working on and it'll get there soon :)

 

Symptoms: The board displays a white/grey screen when powered on

 

Video of the symptoms:

 

Some backstory - I have had this 4 slot for quite a while and looked at it multiple times but I could never get anywhere with it, it had fairly substantial battery corrosion and I had other Neo motherboards so I have left it on the backburner until I had the skill to be able to fix it properly. This weekend I've been looking at a few boards including a Neo 1 slot (RAM fault, cleaned the cart connector and interconnects, Z80 error, replaced Z80 and it's now working perfectly) and my 1 slot bootleg (stuck inputs). The bootleg now had a new issue where it is reporting a colour RAM fault at Bank 0 when dip switch 1 is turned on, otherwise it is just going to a black screen and not syncing properly. This was the exact fault the 4 slot had the last time I'd looked at it, so I pulled out the 4 slot to see if I had a similar fault that came up on it when dip switch 1 was turned on.

 

Picture of the damage from the battery on the 4 slot

 

File 25-2-17, 20 25 21.jpg

 

The 4 slot now gave me a white/greyish screen with some colours bleeding into the screen before it would go back to the same grey/white screen again. Since I now suspected a colour RAM fault, I decided to fire up the bootleg board without it's RAMs installed at all (the bootleg is a 1:1 clone of a 1FZ with reproduced ASICs and EPROMs where the custom BIOS mask ROMs would normally have been, but the logic/RAM is exactly the same as the original)

 

Without the colour RAMs installed on the bootleg I got exactly the same screen as I did on the 4 slot! Without any of the bleeding and flickering, which I assumed meant that the 4 slot RAM wasn't completely dead, but it was certainly not working.

 

I connected up the logic probe and probed both the RAM ICs, and these were the results:

 

4364 @ F4

 

Pin 1 - Hi

Pin 2 - Hi

Pin 3 - Lo Pulse

Pin 4 - Lo Pulse

Pin 5 - Lo Pulse

Pin 6 - Hi/Lo Pulse

Pin 7 - Lo Pulse

Pin 8 - Lo Pulse

Pin 9 - Hi Lo Pulse

Pin 10 - Hi/Lo Pulse

Pin 11 - Lo

Pin 12 - x

Pin 13 - x

Pin 14 - Lo

Pin 15 - x

Pin 16 - x

Pin 17 - x

Pin 18 - x

Pin 19 - x

Pin 20 - Lo

Pin 21 - Lo Pulse

Pin 22 - Lo

Pin 23 - Lo Pulse

Pin 24 - Lo Pulse

Pin 25 - Lo Pulse

Pin 26 - Hi

Pin 27 - Hi

Pin 28 - Hi

 

4364 @ H4

 

Pin 1 - Hi

Pin 2 - Hi

Pin 3 - Lo Pulse

Pin 4 - Lo Pulse

Pin 5 - Lo Pulse

Pin 6 - Hi/Lo Pulse

Pin 7 - Lo Pulse

Pin 8 - Lo Pulse

Pin 9 - Hi/Lo Pulse

Pin 10 - Hi/Lo Pulse

Pin 11 - Hi

Pin 12 - Hi

Pin 13 - Hi (Weak)

Pin 14 - Lo

Pin 15 - Hi

Pin 16 - Hi

Pin 17 - Hi

Pin 18 - Hi

Pin 19 - Hi

Pin 20 - Lo

Pin 21 - Lo Pulse

Pin 22 - Lo

Pin 23 - Lo Pulse

Pin 24 - Lo Pulse

Pin 25 - Lo Pulse

Pin 26 - Hi

Pin 27 - Hi

Pin 28 - Hi

 

Oh dear. The data lines on the RAM at F4 are almost all completely dead and on the other RAM they are stuck. The way that the Neo colour system works (as I understand it) is a custom Neo PAL feeds into the RAMs via a HC32, this data is then fed through a pair of LS273 into a group of resistors which converts the data into it's correct colours. It is then sent through to the edge connector via an LS05. I logic probed the logic chips driving the RAM and most of them looked normal, the pins that weren't talking I assumed was probably due to the dead data lines on the RAMs.

 

Since I had the bootleg board sitting out and the RAMs on it were socketed, I decided to use these to see if I could get anything out of the 4 slot. At the very least the bootleg RAMs were able to tell me there was a colour fault, so I knew they worked enough to give me something. The RAMs on the 4 slot were very corroded so I cleaned them up and piggybacked the RAMs from the bootleg:

 

File 25-2-17, 20 39 55.jpg

File 25-2-17, 20 44 40.jpg

 

And then..... progress! A crosshatch screen! When I got the board it was missing it's original BIOS ROM so I replaced it with a Universe BIOS since this is what I'm going to use on it once it's working.

 

File 25-2-17, 20 35 59.jpg

 

It looks ugly, the colours are completely wrong but this shows me 2 things - it confirms that the RAMs are at fault and need to be replaced and it also confirms that I have sync and that the BIOS is being run, which is fantastic news.

 

So I removed the RAMs from the board

 

File 25-2-17, 20 45 14.jpg

 

Reinstalled replacement 6264 RAMs from a 1 slot Neo scrap motherboard (battery damage was too bad on that board to save it sadly)

 

And we get this:

 

File 25-2-17, 20 45 48.jpg

 

Success! A normal looking crosshatch screen!

 

Next is to put the 4 slot top board back on and see what happens with a cartridge installed. I'm not hopeful of this since the battery has eaten through multiple traces in the area and as expected I got the crosshatch screen of death.

 

So I've set to work repairing the traces from the CN10 connector which is right next to where the battery is. Thankfully MKL from the Neo Geo development wiki has mapped out all the pins from this connector and where they go to on the board, the vias are so small on this board it is excruciatingly difficult to map these out so a HUGE thankyou to the guys who made this information public

 

https://wiki.neogeodev.org/index.php?title=MVS_board_connectors_pinouts

 

Currently I've replaced 13 traces.

 

A6 - Pin 3 @ B7

A8 - Pin 5 @ B7

A9 - Pin 6 @ B7

A10 - Pin 7 @ B7

A11 - Pin 8 @ B7

A12 - Pin 9 @ B7

 

B7 - Pin 16 @ B9

B8 - Pin 5 @ B9

B9 - Pin 14 @ B9

B10 - Pin 7 @ B9

B11 - Pin 12 @ B9

B12 - Pin 9 @ B9

 

Pin 9 @ B8 reconnected to Pin 3 @ C11

 

Work in progress photo:

 

File 25-2-17, 20 46 44.jpg

 

And this is the photo of all the trace repairs on the CN10 connector I've completed so far:

 

File 25-2-17, 21 44 23.jpg

 

You can see in the photo of the battery damage there is a 2003 with some very corroded pins. I've actually checked the diodes, capacitors, resistors and the traces other than those going to the CN10 connector and they're all reading ok. The 2003 and LS05 need to be replaced along with the LS259 above the 2003 and potentially the LS244 underneath the LS05. I'm unsure as to whether these are causing the issues with my top board not being read properly, but they need to be replaced anyway so I am hoping that replacing these will help.

 

At this stage I have confirmed continuity on all pins on the CN10 connector on the bottom board. There is however corrosion on the pins on this connector on the top board and unfortunately this is not mapped out at least as far as I can find. So the next step is to work out what should be where and reconnect pins as required on the top board. But I'm definitely happy with the progress so far. Hopefully I can get it running and have a 4 slot setup.

Link to comment
Share on other sites

For now I am focusing my attention on the top board as I believe the bottom board now has all connections intact which should mean that other than a logic fault, everything is functioning as it should on the bottom board. I measured a few components on the far side of the board for +5v and I'm getting a perfect 5.00v even on the opposite side of the board to the CN10 connector.

 

Moving to the top board the first IC I look at is the F138 @ 2A - this is the IC that tells the bottom board which of the 4 cartridges is loaded and enables use of that slot/the logic driving that particular slot. channelmanic in his repair logs has had multiple failures of this IC, so I stuck it with the logic probe. It was not at fault, but I thought for the interest of those attempting diagnosis I would put the results here anyway.

 

The function table found in the datasheet for the IC determines that if the G2A or G2B line is high, or if the G1 line is low, then regardless of what the select inputs and other enable inputs are doing, all outputs are logic high. In my case this was correct as while the G2B line was low and the G1 line was high, the G2A line was also high which meant that regardless of all other input lines, the outputs would all be logic high (as they are) - if one of the output lines had been in a low logic state, this IC would have been faulty as in order for this to happen, G1 must be high, G2A and G2B must be low. In order to do this test I installed a cartridge on the top board and probed the same IC both with and without the cartridge in order to determine if the board is even attempting to locate a cart BIOS.

 

Pin 1 (Select Input A) - Low

Pin 2 (Select Input B) - Low

Pin 3 (Select Input C) - Low

Pin 4 (Enable Input G2A) - High

Pin 5 (Enable Input G2B) - Low

Pin 6 (Enable Input G1) - High

Pin 7 (Output Y7) - High

Pin 8 (Ground) - Low

Pin 9 (Output Y6) - High

Pin 10 (Output Y5) - High

Pin 11 (Output Y4) - High

Pin 12 (Output Y3) - High

Pin 13 (Output Y2) - High

Pin 14 (Output Y1) - High

Pin 15 (Output Y0) - High

Pin 16 (Vcc) - High

 

I also probed the LS05 next to the battery on the bottom board since it is very corroded and probably damaged and got the following results - an LS05 is an inverter, so whatever goes in on an input line should be the reverse on the output line.... this was definitely not the case here. This particular IC requires pullup resistors in order to function (to pull high input lines low) which are doing their job a bit too well.

 

Pin 1 (Input) - Low Pulse

Pin 2 (Output) - Low Pulse

Pin 3 (Input) - Low Pulse

Pin 4 (Output) - Low Pulse

Pin 5 (Input) - x

Pin 6 (Output) - Low

Pin 7 - Low (Ground)

Pin 8 (Output) - Low Pulse

Pin 9 (Input) - Low

Pin 10 (Output) - Low Pulse

Pin 11 (Input) - High/Low Pulse

Pin 12 (Output) - Low Pulse

Pin 13 (Input) - Low Pulse

Pin 14 - High (Vcc)

Edited by xodaraP
Link to comment
Share on other sites

I have now removed the 2003 (this is a 7 channel transistor array) as well as the offending LS05 as both were heavily corroded.

 

Unfortunately the news isn't good, the holes are also badly corroded along with almost every trace in the area of these 2 ICs being completely rotted out.

 

Here is a picture - I'll still see if I can get it going by replacing the traces, but I'm just not sure if the through holes will carry enough conductivity to perform their duties with this much corrosion involved and I don't really want to run 34 traces for these 2 chips when I'm not 100% sure it's going to solve the problem.

 

File 26-2-17, 22 22 08.jpg

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...