MVS 1-slot and 2-slot woes: Backup Ram errors, resetting etc.

Emby6201

Active member
Joined
Dec 21, 2011
Messages
338
Reaction score
31
Location
Florida
MVS 1-slot and 2-slot woes: Backup Ram errors, resetting etc.

I have THREE bad mvs systems and I'm wondering if any are worth salvaging:

TWO SLOT:
_________________________________________
Ive got a 2-slot that is having a backup ram error. - Everytime says:

"BACKUP RAM ERROR 5555 B255

As I understand, this could be traces bad from the ram to the CPU, or bad ram - the ones near the battery.

I think I found the rams near battery - looks like there might be some corrosion, but not entirely sure - I think the battery is bad. I can just measure on the board, I assume?

Is that right - replace the SRAM's close to the battery?

MV1C - ONE SLOT:
__________________________________________

This one resets with certain games; others - works fine.

I see this with my 161-1 ; and my only other MVS cart (Metal Slug X)

Metal Slug X - resets upon booting toe title screen.

Many of the old games on the 161-1 work fine - I've played Robo Army, Viewpoint, Crossed Swords, etc. - no problem.

Ive tested my carts in my single working MVS - they both work fine.

I've tried cleaning contacts, checking the voltage - using the boards in another machine, etc. - no dice.

MV1-B - ONE SLOT:
__________________________________________

This one doesn't boot, and gives a backup ram error - but it gives a different error each time, it seems.

What does it mean if it changes everytime?


I'm trying to DIY on these, any help greatly appreciated!
 
Last edited:
The biggest help you'll find is to do a google search for channelmaniac's repair logs. There's a few sites that'll show up, arcadecomponents.com (his site, iirc) and newlifegames.com ( I think). One of his repair logs explains the backup ram error and where to look when it's changing within a certain range.
I would strongly recommend that you remove the battery in these boards before you do anything else though.
 
The biggest help you'll find is to do a google search for channelmaniac's repair logs. There's a few sites that'll show up, arcadecomponents.com (his site, iirc) and newlifegames.com ( I think). One of his repair logs explains the backup ram error and where to look when it's changing within a certain range.
I would strongly recommend that you remove the battery in these boards before you do anything else though.

I looked at his posts before I posted and that's what gave me these ideas, but I didn't see a changing value error.

But, a little of it was over my head, and I may have missed it. There were a lot of logs on there!
 
Changing value can usually be a bad trace still - a cut trace could lead a bit sometimes being high, sometimes being low. It could also just be a bad chip outputting random garbage.
 
Here was a post I made about my 2 slot board:

I went to turn on the games in the game room for the kids last night and my Neo Geo came up with the following error:

BACKUP RAM ERROR
ADDRESS WRITE READ
00D00000 5555 0055

It is a 2 slot with only the 161-1 multicart installed.

Hopefully, it's something simple, but any help on this would be appreciated.

Here was the response by channelmaniac:

If it still has the original battery on it then it's most likely a corroded trace going to one of the control lines on one of the backup RAMs. Check the traces and components around the 74HC32 next to the battery.

I can't help past that, but we found a ton of corroded traces on my board exactly where he said to look. :(
 
On the mv1c, there's likely a bad trace from the cartridge slot. Probably those games use some pins that older games didn't.

2-slot, yes, you can measure the battery on the board, but really, just cut the leads and get rid of it till you figure out the issue. Also, if you can see corrosion on a ram chip, it's possible its corroded up through the trace underneath the resist. Double check all the traces from that chip. This goes double if it's anywhere near the battery.

Mv1-b - as I understand it, the ram test tries to write to the ram, then read it back. If the value it reads back isn't what it tried to read, it fails and gives the ram error. If the read is random, it could indicate a bad ram chip, storing the values incorrectly, or a trace could be corroded or cut, corrupting the data.

Good luck and just plug away at them till you get em working.
 
I found this post:

http://www.neo-geo.com/forums/showthread.php?118684-Backup-RAM-errors

With info on the pinouts for the chips to the CPU. Those checked good, but the post looks like its for the four-slot, and not the 2 slot that I have. So I'm not sure if that matters.

The rest of the pins - to the 74hC32 chip, 2 were there, and two weren't (the ones to pin 13).

The reference there is to D10, but mine aren't there - the 74 chip is at J2, so I'm not sure if the pinouts are correct.

Does anyone have the correct pinouts for the 2 slot from the SRAMs there? I have been looking for the last hour or so, and all I keep finding is for the one slot, or the above post.

Heres the pinout info i used, which I'm not sure is for my board:

"RAM___CPU

1______43
2______41
3______36
4______35
5______34
6______33
7______32
8______31
9______30
10_____29
21_____39
23_____40
24_____38
25_____37
26_____42

etc.

There are 3 pins for each backup ram chip that are not connected to the CPU:

lower ram:

20_____74HC32 (13) located at D10
22_____PRO-C0 (16)
27_____74HC32 (8) located at D10

higher ram:

20_____74HC32 (13) located at D10
22_____PRO-C0 (8)
27_____74HC32 (6) located at D10 "
 
have a 100% working 1 slot with a brand new 161 in it for sale if you want
240 shipped to the lower 48
 

Attachments

  • IMG_1716.JPG
    IMG_1716.JPG
    88.9 KB · Views: 15
Last edited:
Follow up - I replaced the battery and now I get:

00D00000 Read 5555 Write 0055 BACKUP RAM ERROR

I am hearing this is a traces issue.

I've traced from the rams to the CPU - good.

Possibly the 74HC32 connections to the RAM like I said before?

I'll post some pics up later, but I am not sure how to trace all these down as there are tons of tiny little connections leaving this area.
 
Replaced the RAMS - Now I get a:

"READ 5555 WRITE 0051" error.

Can't seem to find the answer to this.
 
So the error went from b255 to 0051 after you installed new ram?
Double check your soldering job. If the ram was known good, you're looking at a connection issue or maybe a bad logic chip (74hc32?)
 
MV1-B - ONE SLOT:
__________________________________________

This one doesn't boot, and gives a backup ram error - but it gives a different error each time, it seems.

REPAIRED :
- Replaced backup rams and the coin-battery.
 
So the error went from b255 to 0051 after you installed new ram?
Double check your soldering job. If the ram was known good, you're looking at a connection issue or maybe a bad logic chip (74hc32?)

Replaced the logic chip (74hc32) - no change.

I have the list of pinout / from RAM to CPU i posted earlier in this chain of posts, but I don't know the rest. I've checked those, and they are there, continuity-wise. But that list does not have all the pins.

When I look at the RAM chip pinout, and the CPU pinout from spec sheets, none of it makes sense to me. I have looked to the pinouts on each, and I don't understand how they matchup.

For instance, in that post I referenced below, it says pin 1 matches to pin 43. I get continuity there - great.

But when I lookup pinouts on 62256, and the 6800hc toshiba CPU, I see pin 1 says "A14", and pin 43 on the CPU is "A15" These do not match, so I must be missing something.

Is there somewhere that has an easy to read / understand?

I have been told to check data pins 11-13 and 15-19, but from what I have found online, (listed below) I don't know where these lines go to the CPU.

1______43
2______41
3______36
4______35
5______34
6______33
7______32
8______31
9______30
10_____29
21_____39
23_____40
24_____38
25_____37
26_____42
 
*UPDATE*

Replaced 74hc32 - no change.

I struggled figuring out the Data lines from the CPU to the backup ram, as its not explicitly stated anywhere that I could find.

I figured it out after a couple hours messing with schematics and pinouts and saw that D2 from the CPU to the SRAM Upper Ram was NOT there, even though it looked good. Reflowed it and then it was good.

Tried again and finally booted! Though, booted to cross hatch even though carts in both slots.

Figured it was a dirty slot; cleaned both with alcohol and contact cleaner. Waited a few minutes -turned it on and now - WATCHDOG OF DEATH.

Clicks quickly (about 1x a second) with garbage screen, so I think this matches the watch dog error.

NOTE - the backup ram lower ram had NONE of the data lines reaching the cpu from my testing for continuity. Though, I'm not sure if I am correct. Here's the testing I did:

DATA_CPU_RAM Pinouts:

D0_5_11

D1_4_12

D2_3_13

D3_2_15

D4_1_16

D5_64_17

D6_63_18

D7_62_19
 
"But when I lookup pinouts on 62256, and the 6800hc toshiba CPU, I see pin 1 says "A14", and pin 43 on the CPU is "A15" These do not match, so I must be missing something."

The 68000 is a 16bit CPU, the RAM you talk about is 8 bit. So the two 8 bit chips are byte interleaved to form a 16 bit stream. That is why the address lines are off by 1.

"the backup ram lower ram had NONE of the data lines reaching the cpu from my testing for continuity"

So you've tested D0-D7 (CPU) going to D0-D7 for one RAM? That means the other ram should be D8-D15 (CPU) going to D0-D7. Again, because of the interleave, 1 RAM fills in the lower 8 bits, one RAM fills in the upper 8 bits.
 
I think then the backup ram issues are okay now, but something else is wrong or maybe I accidentally cut a trace or something?

I've got a diagnostic ROM coming from a fellow member. Going to see if maybe that can point me in the right direction.


"But when I lookup pinouts on 62256, and the 6800hc toshiba CPU, I see pin 1 says "A14", and pin 43 on the CPU is "A15" These do not match, so I must be missing something."

The 68000 is a 16bit CPU, the RAM you talk about is 8 bit. So the two 8 bit chips are byte interleaved to form a 16 bit stream. That is why the address lines are off by 1.

"the backup ram lower ram had NONE of the data lines reaching the cpu from my testing for continuity"

So you've tested D0-D7 (CPU) going to D0-D7 for one RAM? That means the other ram should be D8-D15 (CPU) going to D0-D7. Again, because of the interleave, 1 RAM fills in the lower 8 bits, one RAM fills in the upper 8 bits.
 
Click of death (watchdog) issues on the Neo Geo are usually a bad RAM or bad control lines on the Backup RAM. You have a stuck or missing signal on pins 20, 22, or 27. Check them on both of the Backup RAMs.

Also make sure you have 5v on pin 28 when the power is on.

B255 errors are usually in the battery backup circuit. Last time I saw that was a bad transistor over next to the HC32 chip. Check the resistors and the traces that go through the board and under the chips themselves to make sure they aren't open because of battery damage.
 
Back
Top Bottom