What's new
By:

Weird boot-loop issue with MV1C

Kujako

Kujako

Champion
Joined
Jun 3, 2020
Messages
1,409
Reaction score
852
Location
USA, California
So, I've had this little MV1C in a Lions case sitting on my workbench for years quietly chugging along testing various carts. Recently however, it will boot loop. But not right away. Games will boot, let me start playing but within a few seconds it will reboot. Any ideas? Not done a BIOS modification so no option currently to run the NeoDIAG BIOS.
1670791253257.jpeg

1670791274309.jpeg
 
So... noticed that some of the 3D printed posts were delaminating. So went ahead and pulled the board out and replaced it with another MV1C I had around with some vinyl posts instead. This at leasts gives me something to compare against. I also hit the cart slot with some deoxidizer and the issue got worse. Which does suggest that it is the slot itself, which would make sense given this was my test rig and has had a lot of games cycled through it over the years. Also checked the board with a thermal camera, but see no hot spots that would indicate bad RAM etc.

[Working board in the Lions case, bad board now loose]
1670960530255.jpeg

1670960494480.png
 
and now it won't even post... so something is majorly wrong. Just get the initial garbled graphics screen, never gets to the BIOS or self test. Which feels like a watchdog issue, but I would love some opinions.

This is with no cart installed.
1670961646997.jpeg
 
Man... the more I look the worse it gets.

A) Seems that at some point some yahoo bypassed the fuse. That glob of solder bridges each side of the fuse holder.
1670962634627.png


B) Back-up RAM was at some point replaced. Replacement chip looks fake and is in so crooked it may be shorted.

[Original chip on left, second replacement chip on right]
1670962703453.png


[RAM 2 looks suspect, and is in crooked]
1670962905898.png



So... one thing I did before it stopped working, is I cleared the backup RAM. Shouldn't have mattered, no battery. But, thinking that is now stopping it from booting.

Thoughts?
 
IMO, the garbled screen would indicate a BIOS failure. Ram errors would actually display a failed POST code like “VRAM ERROR: WROTE 8888 READ 8800” on boot. I wonder if the board warped and a trace is loose? I would reflow the BIOS and see if anything changes.
 
avramce said:
IMO, the garbled screen would indicate a BIOS failure. Ram errors would actually display a failed POST code like “VRAM ERROR: WROTE 8888 READ 8800” on boot. I wonder if the board warped and a trace is loose? I would reflow the BIOS and see if anything changes.
Click to expand...

That'd be simple enough. Fair possibility that the 3D printed posts the board has been on have put pressure on it. That's the one thing I really dislike about the Lions cases, the posts are garbage. When I do a build myself I always swap them out with nylon or metal posts, but this one was already cased when I got it.

No luck, did fix the fuse while I had it under the soldering iron. Could try installing a new BIOS.... will think about it. Not sure I have a BIOS Masta around.
 
Last edited:
Found an old (2016) NeoBIOSMasta in my odds and ends pile. No idea what I pulled it off of. Think I'll remove the MV1C BIOS altogether since the fuse is in the way of lifting the pins. This'll let me run the NeoDIAG BIOS.
 
Last edited:
And nothing... but only 75% sure I have the thing installed right. Both the UniBIOS and DiagBIOS show the same screen.

1671225140403.jpeg
 
1B and 1C are a pain, too much stuff got consolidated into unreliable big chips.
Is it a watchdog loop, with screen jumping and audio click at 2-3 times per s ? Or is it stuck. Is the watchdog jumper free or linked ?
I don't remember if 1C has a discrete 74hc259 or if it's integrated in a chip. On other boards it can go bad and cause watchdog (among other things)
 
Ryoandr said:
1B and 1C are a pain, too much stuff got consolidated into unreliable big chips.
Is it a watchdog loop, with screen jumping and audio click at 2-3 times per s ? Or is it stuck. Is the watchdog jumper free or linked ?
I don't remember if 1C has a discrete 74hc259 or if it's integrated in a chip. On other boards it can go bad and cause watchdog (among other things)
Click to expand...

I dont think it does, I believe everything is integrated into the custom ICs.

HOWEVER, I reseated the NeoBiosMasta a few times, and it posted (I really dislike these adapters sometimes). Currently getting a BIOS ADDRESS error, which I have no idea what it means.

1671244530834.jpeg
 
Raph_friend said:
Your adapter make the bios in directly conctact with 68K pins, right?
Click to expand...

Correct. Checking pins 44-46 on the 68k (areas of potential causes according to the NeoDiagnositcs BIOS wiki) and I think everything is making contact as expected. But trying to double check (hard to do with the adapter installed).
 
Last edited:
Raph_friend said:
First of all do a complete continuity test.
1671249599998.png
Click to expand...
I’ll give it a shot, but the address error is specifically on pins 44-46 of the 68k. However, it may be that there’s a fault in the BIOS adapter (it was in my parts pile). So I’ll recheck everything as best I can.

If it does look to be the CPU failing I’ll see if I can find a replacement. I’d be able to remove and replace the one that’s there, but I doubt my ability to cleanly remove a known working one from another board.
 
The original diag bios only has 16KB of code, but the rom is 128KB. The remaining 116KB of space is filled with 7 copies of the bios. Towards the end of each bios is a byte to indicate which one it is. The first bios (the one that should be running) is 0x00, first copy is 0x01, .. 7th copy is 0x07.

So the error
Code: 
BIOS ADDRESS (A13-A15)

EXPECTED 00
ACTUAL   01

Is saying it expected to be running the first bios (0x00) but instead the first copy (0x01) is running. This would likely indicate A13 on the bios (pin 35) is floating or maybe something is forcing it high. I would bet its just a bad connection between the cpu and NeoBiosMasta.
 
Last edited:
You must log in or register to reply here.
Back
Top