MVS MV1FT not fully booting. Wrong graphics/letters

[yandritos]

Hi all!

I was reding many threads and I think this is the best place in which I can look for some help with my case. I have 2 MVS boards, the old first one is working like a charm, the new one is a model I was looking for "MV1FT" with control pad ports and volume control.

In this MV1FT I could see that the board and games are not booting, we reach a point with some bad graphics (no matter if a game is plugged or not, the result is the same). This was the result with the original Bios (the BIOS is OK as I plugged in on my old board and all works properly).




Then I found the marvellous neo geo BIOS diag and I burned it with my T48. The result is different but not much better (the BIOS diag works properly as well in the old board):



The first pic is from my TV for local testing and the second with my Naomi (I just wanted to see if a monitor change was showing something else but the result is the same).

Now the only track of failures I found is that the original battery had corrosion and 2 board tracks were broken. I removed the battery and briged the tracks with a wire....but the screen output result is exactly the same. This is the picture of the board with the battery zone:




Any advise about what else should I check will be more than wellcome, as I can´t read a BIOS Diag result I don´t truly know how to start. Apparently the rest of the board is in a very good conditions.

Thanks in advance!

 

[ack]

I would start by replacing the slow/32k vram chips and see if helps with the screen showing the right data characters.

neogeo-diag-bios/docs/ram_locations/mv1.md at master · jwestfall69/neogeo-diag-bios

Disassembly of smkdan's Neo Geo Diagnostics BIOS with added features - jwestfall69/neogeo-diag-bios

github.com

 

[fly_corsair]

I have the same problem with a Mv1fzs, have cheked traces and no Issue found.
No "watchdogging", 68k seems to be working...
Pic with stock bios:


Pic with Diag bios:

Sorry for invade your topic, just looking for a solution... our defect seem to be similar .

 

[ack]

I would guess the error it is complaining about 32K VRAM Upper. It looks like its trying to say

ADDRESS:  000000
ACTUAL:     0100
EXPECTED:   0000

If you validated the traces already then I would try swapping that vram chip.

neogeo-diag-bios/docs/ram_locations/mv1fz.md at master · jwestfall69/neogeo-diag-bios

Disassembly of smkdan's Neo Geo Diagnostics BIOS with added features - jwestfall69/neogeo-diag-bios

github.com

 

[fly_corsair]

Thank you Ack, i will try to replace the vram... i will post the results...

 

[fly_corsair]

I replaced both vram's and no change on screen...
Can work ram cause this?

 

[ack]

work ram wont cause that.

 

[fly_corsair]

I'm checking the signals from 68k and observed that pin a16,a17,a18 dont have activity, only ground sinal. Is this normal?

My "osciloscope" is limited to 200khz.
Example a15 signal:

A16,A17 and A18 signal:

 

[ack]

It would depend on whats being executed on the cpu. If its the diag bios its likely ok because it puts up the memory error then goes into a loop waiting for input.

 

[fly_corsair]

Verified today Pins E0-E15, conecting lspc2 to Vram, found E8 stuck in high, all others have activity.
I've Removed pin 11 of high 32k vram and LSPC2 still with pin 49 in high. Any ideias?

 

[ack]

Try checking the 68k data pins going into the LSPC. Maybe there is something wrong with D8 (floating?)

 

[fly_corsair]

I've checked D0 - D15, all pins with acrivity...
During my troubleshooting i obseved that 2 74LS245 ( output Video i believe ) wich receive D0- D15 are getting warm, this is normal?

 

[ack]

warm is usually ok, to hot to touch is not.

 

[fly_corsair]

Its not too hot to touch.... so probably the LSPC is fault?

 

[ack]

could be. Another thing you can try is pressing down on different chips / parts of the board while its on and see if anything changes. If it does its an indication of a bad solder joint some where.

 

[fly_corsair]

Pressed all chips without any change...