CPS1 Street Fighter II HF sprite glitches (solved)

[ItsBobDudes]

Hi all,

I have a CPS1 Street Fighter II HF. I am getting artifacts on some characters but not on every frame. I have isolated the issue to the B board using a known good board set. I checked the Mask ROMs 1-8, 10-13, and Program ROMs against MAME all verified OK. Traded all the GAL chips with the known good board. The B board itself is in great shape, I can't find any broken traces or suspicious solder joints. Looking for suggestions where to look next. Pic of the sprite artifacts below:

 

[ShootTheCore]

First off, make sure your power supply is delivering 5.0V to the JAMMA edge. If your PSU is adjustable, try adjusting it a little lower or higher to see if the behavior changes. I wouldn't go below 4.85V or above 5.15V. Some boards run better with a little more or a little less voltage.

If that doesn't help, I suspect one of your EPROM sockets or A to B board connectors has a faulty pin or a cold solder joint.

Try pressing firmly down on each Graphics ROM (Sockets 1 - 8 ) while the game is running and see if the graphics glitches change or go away when applying pressure. Try that with the A to B board connectors as well. If you find that applying pressure to one spot fixes the issue then inspect the connector / socket and reflow the solder joints on that component.

If the pressure test doesn't help, I suggest picking up a logic probe and checking the Address and Data lines on the Graphics ROMs - Sockets 1 through 8 - for inactivity.

I made a tutorial video on how to probe the EPROMs for sprite corruption on a Fire Shark board.

View: https://www.youtube.com/watch?v=2PAtTIAijeA


The same principles apply here - just use the data sheet for a 27C4000 EPROM if your B Board is model 91634B-2 or a 27C4096 EPROM if your B board is model 91635B-2.

Look for an Address or Data line on one of the Graphics ROMs that doesn't pulse constantly between High and Low on the logic probe while the glitches are showing - that'll be your culprit. Watch the game while you're probing the lines - you may see some changes in the glitching that could be a hint.

Note that it is possible for a ROM to verify good in a chip programmer properly but malfunction when running on a game board. This is because a chip programmer only reads the ROM once sequentially with steady voltage while a game board addresses the ROMs randomly, constantly and sometimes with variable voltage if caps are old. You may want to try replacing the mask ROMs with EPROMs and see if the behavior changes.

 

[ItsBobDudes]

Note that it is possible for a ROM to verify good in a chip programmer properly but malfunction when running on a game board. This is because a chip programmer only reads the ROM once sequentially with steady voltage while a game board addresses the ROMs randomly, constantly and sometimes with variable voltage if caps are old. You may want to try replacing the mask ROMs with EPROMs and see if the behavior changes.

5V was bang on. I pulled out the logic probe but I was unable to find any dead address or data lines on ROMs 1-8. Your comment on the programmer reading the chips sequentially did give me hope it was a ROM issue though. I replaced every ROM chip on 1-8 with chips from my known good board and when I got to 6 the issue went away. Thanks for the detailed post that lead the way!

 

[ShootTheCore]

Glad you got it figured out!