[WIP] Street Fighter 2 Hyper Fighting Repair

Hey guys,

I bought this SF2 Hyper Fighting board from a member here and it looked very bad from the outset. But through some research I managed to figure out the main problem. It's listed as a Work in Progress only because there is still more to be done - but I thought I'd start with what the board played like when I received it and what I've done so far.

So when I first received the board - I had the below graphics glitching, it was bought as faulty so I knew what I was in for, and it looked pretty bad. The game was running perfectly in the mess, sound was working perfectly and throwing my Final Fight B/C boardset onto the A board told me that while there was some graphics glitching on the A board (some corrupted text), it was nowhere near as bad as what I was seeing here. This meant that the fault was either with the ROMs on the B board or the custom on the C board.

CPS1 is notorious for graphics glitches and these are usually a result of a RAM failure on the A board (the main board where the JAMMA edge connector is) or a failure of the custom ASIC on the C board (the small board on the top of the stack)

This is what my Final Fight B/C boardset looked like on the SF2 A board. Notice the text glitching on the first image, and the graphics glitching on the second image, the map is all over the place. Please excuse crap photography :P

This is the work in progress part - as yet I have not fixed the A board issue with this game. But the thread will be updated once I fix it.

So to fix the major glitching in the SF2, the issue was confined to either the B or C board. I looked at the C board custom under a magnifying glass and did not see any lifted/bent pins so I was prepared to assume that the issue was not there. I then dumped all the mask ROMs and found that #3 mask ROM is not reading correctly against the MAME set, the rest are fine. While I was testing the ROMs I was doing some reading into the issue and came across this information that Womble had given someone about CPS1 graphics faults:

It is also worth checking that the PALs are correct for this board' date=' and are in the correct locations, I have seen faulty boards that had random chips installed into the sockets so it could be sold as "complete". From memory the PALs on the A board are standard across all CPS1 boards, called IO, SND, PRG. The one PAL on the B board at location A1 is game specific, it is usually involved with arranging the gfx tiles but your game sounds like the screen is correctly assembled, but in the wrong place, if A1 is bad, or missing you usually get a black screen or the game broken up in to 2cm square blocks. My money is still on RAM tho, this is all on the A board, simplest method is to swap the A boards, or put the BC boards on a known good A board, that will help you isolate the fault.[/quote']

 

This gave me some hope that in fact the C board custom was ok, but perhaps it was a PAL issue at play. I ordered some GAL16V8 chips and then waited a month for them to arrive from China. They arrived today.

 

There are 4 main PAL chips on SF2 CE B/C boards. These are the S9263B at position A1 (top left of the B board), IOB1 middle bottom of the B board and IOC1 and C632 on the C board next to the custom. I programmed replacements for all of these so as to be able to identify the issue. Thank you very much to porchy and the guys at jammarcade.net for the dumps they have posted on there.

 

I put all 4 replacement chips on the board so I could see what I ended up with - and I made it both better and worse at the same time. Again please excuse the terrible photography and my reflection in the monitor. But basically what I was left with was a very dark blue no longer blocky picture. I got no boot screen and the game would watchdog after the faceoff screen in the attract mode.

 

 

This meant I was on the right track - so I decided to replace the older chips back on the board one at a time to see where the fault was now. Granted this is backwards, but I was hoping to replace all the older chips to hopefully help prevent future faults. I first replaced the S9263B with the original one - the ground pin actually snapped off when it was removed from the board, so I ran a wire from what was left of the pin to a ground off board and put it back in and fired it up. This gave me the same issue as I was having already but worse, I now had the dark blue screen and it was blocky. This meant that this was in fact the bad chip in the first place, and would need to be replaced. So I put my newly programmed GAL16V8 in place.

 

Next I tried the C632 on the C board - and then I got this

 

 

Success!

 

As you can see there is still some glitching in the game, which I believe will be rectified by replacing the bad mask ROM at position 3 (will update once this is done) and fixing the issue on the A board causing the text glitching in Final Fight and on the boot screen of SF2 (says EPC instead of ETC) - which I will have to leave for another day. For what it's worth - the RAM on the board is almost entirely Fujitsu branded, so this is the most likely culprit. It may also be the custom ASIC on the A board, I'll look at it again soon.

 

Update: I have replaced all the RAM chips on the A board and this had no effect on the glitching, so sadly the custom is at fault and will need to be replaced. I've put this on hold for now until I come across a replacement. The B and C boards are otherwise functioning perfectly.

 

Thanks for reading guys :D

[WIP] Street Fighter 2 Hyper Fighting Repair

Hi Womble :D thanks very much for the help with your repair threads, what do you suspect is the most likely culprit here?

Unfortunately, the big custom on the A board is generally the fault. The only sources for them would be another A board (with fingers crossed) or a CPS 2 A board which I've never seen the same custom fail on.

If you can find a CPS 1 or 2 board that has the video output fried from a monitor that needed, but didn't have an ISO installed, then you struck gold for a replacement custom.

Thanks CM, I was afraid of that. I do have a non working CPS2 A board but the only fault is it's currently only showing green so I don't think it's worth scrapping. I can still play the game on another A board so I'll put it aside until I come across a scrap A board for the custom :)

Out of interest is it worth reflowing the custom? It doesn't seem to be coming loose but never know I guess

Sent from my iPhone using Tapatalk

[WIP] Street Fighter 2 Hyper Fighting Repair

Fair enough - I've had really good luck with CPS1 so far so it was bound to end sooner or later :P I'll replace the dead mask ROM and see how bad what I'm left with is and then hold out for a dead board to steal the custom from

Sent from my iPhone using Tapatalk

So to fix the major glitching in the SF2, the issue was confined to either the B or C board. I looked at the C board custom under a magnifying glass and did not see any lifted/bent pins so I was prepared to assume that the issue was not there. I then dumped all the mask ROMs and found that #3 mask ROM is not reading correctly against the MAME set, the rest are fine. While I was testing the ROMs I was doing some reading into the issue and came across this information that Womble had given someone about CPS1 graphics faults:

Is a dump when you copy program of a chip and how do you do it?

So I put my newly programmed GAL16V8 in place.

Umm how do you program GAL16V8?

I think I have the same issue as you with what looks like a Champion edition cps-1 board.

Video in link https://drive.google.com/file/d/1iU-voIC6Y0nxSkxiUnLzE8lzd3G-1ytJ/view?usp=sharing

[WIP] Street Fighter 2 Hyper Fighting Repair

A dump is reading a ROM and saving it to a file on a PC which can be compared to another file, burned to another chip or even submitted to MAME if it's an unknown version

A GAL16V8 can be burned by a programmer in a similar fashion to an EPROM, not all but most programmers can burn PAL and GAL chips. I'd be happy to burn one for you if you can't

Looking at your video you're running one of the Chinese hacks like Red Wave or Rainbow edition - the error you see there may be a bad PAL, but more likely is the A custom as with most of the CPS1 issues :(

Do you have another A board you can try it on? Or could you borrow/get access to one?

Unfortunately no, If I had another cps-1 system to test with it would isolate the problem to a specific board.

I have read that alot about the custom on the A board failing but it looks like a clean pcb on inspection. I might be one of the lucky ones.

No visual problems on A board and the game works as it should with only graphical errors.

I saw another post by ArcadeKing regarding vertical bold lines and how the mask roms are responsible for that. Im hoping so anyways.

I do have a MiniPro TL866A USB Programmer coming tomorrow which I can use to try to dump mask roms and compare.