MV1A won't boot

I got a MV1A board, and it won't boot to a game or the cross-hatch screen test. Every time I turn it on I get an arrangement of different color blocks. The screen does not flash, and it appears to be the same arrangement of blocks every time. I'm powering the board using an old AXT12v desktop power supply. Here is a photo of the screen I'm getting.

But the image is static, it is not flickering or anything. This talks about an image that is flickering because the console is constantly resetting.

I'm not sure that matters. You still wanna check for bad traces, then probe outputs, etc.

I already checked the traces connecting the cpu to the bios socket using a multimeter and I can't find any broken connections. But I do not know how to probe outputs, or what that even means?

Have you verified the 5V rail is 5V? I have seen that behavior one time and was the result of the 5V rail only being around 4.5V.

ack said:

Have you verified the 5V rail is 5V? I have seen that behavior one time and was the result of the 5V rail only being around 4.5V.

Click to expand...

OK I checked the 5v rail from my PSU and you are right, it isn't reaching 5v. So I decided to hook the 12v rail to a voltage regulator to provide 5v. Now I do get a different image and it's flashing like the description. IDK if it is clicking because I haven't tried to hook up speakers.


As far as a logic probe goes, how would I go about doing that? I am not an engineer or anything so this feels a little over my head. But if someone could provide me with a basic tutorial I am willing to try if it can help me diagnose the problem.

Also I found this on the board, maybe it is the source of the problem?

As you can see one of the contacts for the IC1 is missing. Also the trace that connects pin 2 to pin 5 of the U4 has been ripped out. You probably can't see it in this photo, but there is a very fine wire connecting those pins. Whoever owned this before me tried to make repairs at the backup ram area, but it does not look like the pin with the missing contact for the IC1 is actually meant to connect to anything, and the ripped out trace has been repaired already. Do you think this could be the problem area, or is it just a red herring?

Your picture of the video isnt displaying.

Its possible what you pointed out is causing the problem. If it is, unibios should be able to bypass the issue. The board just wont be able to keep date/time.

You may also have other problems and it wouldn't surprise me cause that board is from china. They do some crazy hack up jobs on them to get them (semi)working.

Sorry, I fixed the photo. I already tried Unibios 4.0 and it's still in watchdog. What would be my next step in diagnosing this? Daskrabs mentioned a logic probe, would this do the job?

Edit: I just tried the Unibios again, this time I am able to get to the crosshatch. So I guess this means that the backup ram area is the problem? I don't know why I was in watchdog with the unibios last time though, maybe it was not seated well with the contacts in the bios socket?

Thats the logic probe I have and seems to work fine for me.

That fixed photo actually is what usually happens if not seeing the bios. It would be displaying stuff like that and resetting over and over. Perhaps the bios wasnt seated all the way or was in backwards.

I would try plugging in a stock bios and see what happens. If it goes to the crosshatch it may just mean your only problem was the low 5v rail and the bios not being seated right. If you get a mostly green screen and resets it would point to the RTC having issues, which are the 2 ICs you mentioned plus the crystal above them.

I tested the stock bios again, it will not boot with the stock bios. I haven't hooked any joysticks or controllers up to the board yet, but it seems to be fine with the unibios, although the colors are off. I am unsure if this is because of the board itself or the RGB-to-s-video encoder that I built using a AD725 chip and Ben Heck's schematic here. The whole screen seems to have a sort of yellowish tint. Here is a screenshot I took by hooking it up to an s-video to hdmi upscaler and hooking that up to my elgato hd game capture.



Update:
Upon cleaning the pins of the stock bios along with the pins of the bios socket using a combination of brass cleaner and rubbing alcohol I was able to get the stock bios to boot past watchdog mode, but it still gives me a message on a black screen that says "Calendar Error," as ack suspected. Would simply changing the battery fix this issue, do I really need to change out the crystal and the two chips? Would any old CR2430 do the job.

Also after hooking up my Undamned USB Decoder I was able to navigate to the color test menu, I see that the colors are really off, it's making me worry that something might be wrong with the GPU. I am awaiting a Din-8 connector to test the RGB via a framemeister, I'm hoping that it's just the s-video encoder, but IDK, this looks really bad.

Bump:

I think I broke the board and it might be beyond saving at this point. When turning it on I accidentally forgot to connect the ground from the psu, having gotten lost in the mess of wires on the work bench. So for a minute 5.3v was being sent to the the 5v rail and the board was not turning on and I couldn't figure out why. I used 5.3v because it actually gets reduced to 4.9v when the board is on and drawing power. But only when it's on and voltage and ground are properly connected. I think I may have fried something but IDK what. The board will no longer boot, even into watchdog mode. It only outputs single color screens, like all blue, or all red, or gray, sometimes if I'm lucky I get different colored tiles like in the first photo, but it does not reset like watchdog does. Furthermore the tiles, unlike watchdog, are all just solid colors, there are no random bits of kanji or alphabet letters mixed in there.

Is it still possible to repair the board? Or is it a lost cause? Any ideas on what exactly I may have fried? I purchased the logic probe hastily after this out of panic, hoping that it can help me pin point what I damaged. I do not know how to use it though. I'm just going to look for a tutorial online.

Single color screen, as told to me by some more knowledgeable people on here probably means something important like the chip that handles got fried. I had a CMVS that someone accidentally mixed up power supplies and plugged a 12v ac->dc instead of the intended 5v, from that point all you ever get is a red or blue screen....and sometimes it will kind of change from red to blue. These were MV-1c boards where the video processor is built into one of the big surface mount ICs and the only possible fix would be to replace that. Older boards I guess have it separate, so you might be able to replace it....but when it comes to stuff like this, who knows what else got fried.

Thanks, I guess I'm just going to give up on this board then.

well wait for someone other than me to reply though, As I said, this might be reparable on your board, but not on the MV-1c, so I'd wait for someone else to give your board the death sentence.