What's new
By:

FIXED - Konami Mystic Warriors BAD "2D" sound prom - Help!

TheDeath

TheDeath

Professional
Joined
Jul 2, 2017
Messages
431
Reaction score
400
Location
Italy
Can you help me troubleshoot this issue with my board? I can't get it to work... sorry if i made some english errors :D
Mystic Warrior has 2x 27c160 (2MB x8bit) sound proms, located at 1D and 2D. The mask rom check gave me a bad "2D", since those proms are built like this it's very hard they get broken, i did infact piggyback it and tested with mame set, all good.
Then i have taken the schematics for the sound area:
2ypdt6w.jpg


I have checked anything you can see here. There are no Fujitsu TTLs but all my reports (74LS08,74LS157,74LS139) are ok. To be sure, i have changed them with new ones.
If i did understand what's going on here, those 2 proms are treated like a 1 unit (they both have CE Jointed) but since the rom is splitted in 2 parts, the OE is driven by the LS139 who "talks" wich of the 2 proms should output when is needed by the code.
Like most konami boards i have a bad Lm358 Amplifier underneath the 054986A, sound is distorted and very loud when it gets hot after a while. Custom is recapped and the DAC is ok, but since this is an output i can leave it until i fix the bad 2D. My concern is (IF) the custom underneath the module could affect this, i realy do not know what is supposed to do (marked 054321)
Any idea about what i should check?
thanks
 
No one can help me? :( now i also have 3F sram BAD in boot test.. the hell is going on this pcb?
 
If you're sure that your ROM chips are good then start checking the traces on the ROMs for continuity with a continuity tester. Also, use a voltage tester to check that each ROM and the memory chip each have 5V of power on the Vcc pins.

It's also quite likely that the memory chip has gone bad. You'd use a logic probe or oscilloscope to probe the address and data lines to know for sure.
 
Thanks for the help mate.
So i was still troubleshooting why the hell 3F was going bad suddently (62256 sram) after probing around i have noticed that was literally cut off the circuit by a HIGH state on the chip select.
I have routed it back to an AND gate until i was able to get back in the circuit but i haven't find anything till the schematics were showing connections, after that i could not find that "DAA" and "USEMODE" cause i do not have that page and who knows if i will be able to find it somewhere.

Anyway, i have turned off and on again after a while i was probing and then i noticed that if the board gets a little hotter the game boot ups. When it gets cold again, it doesn't.
I don't know if this has to come because of the damn 054986 module (recapped and new lm358) or that big chip with the heatsink that i've no clue what it realy does.
I also noticed that i have used mostly 25V/50V caps instead of 16V cause i did not have them around, could they be the problem and due to the voltage capacity they take more time to charge?

As you can see my knowledge is pretty much generic on this stuff :D

oh, also the 2D prom was changed with a new 27c160. Before soldering i have dumped and compared with the mame file, it's ok, so i have no doubdt i made it wrong. Still, not reconized..
 
Damn voltage :)
I wasn't make any progress so i decided to step back to the beginning, starting from voltages.
My 5V rail was going for 4.90V and apparently it was not enought. I did some voltage regulations a week ago so i've overlooked thinking was fine, it wasn't. This board really requires you to have stable 5.0V otherwise you will get a bunch of errors in the audio compartment.

So now that i have figure it out i can go on with the 2D prom.
I decided to change the right custom 054539 since it was connected here and there, also CE of the prom goes there, but sadly nothing.
I'll try to change the left one and the 4464ram underneath but i'm not sure if this is a good move

I have checked continuity with proms and both are connected, exept for CE that goes into 2 separate ways
 
Have you already tested the address and data lines on the 2D ROM with a logic tester or oscilloscope to make sure they're all showing activity?
 
ShootTheCore said:
Have you already tested the address and data lines on the 2D ROM with a logic tester or oscilloscope to make sure they're all showing activity?
Click to expand...
Yes, i just did that. They all look ok with my logic tester (unfortunatly i do not have an oscilloscope). I might just concentrate with CE and OE, tracking back and see if i can find something
 
Fixed.. finally.

Can i go with the "F" word? Because who draw that schematic was an asshole. And it's twice that, the same error is in the maskrom check.
1D and 2D... are swapped.

I was checking the proms in the sound test menu, you can hear a scale note that is coming from both proms, one tone was obviously missing.
I did a couple of times watching activity on CE and OE, then i found that OE in 1D was not pulsing sometimes and was stuck HI randomly.
Since the mask check told me 2D bad, i obviously overlooked it, thinking was fine.
That gave me the right idea to check 1D, I pulled it off and checked with mame, the prom was 90% blank... andress and data were showing good activity but they were just flowing bunch of zero's and nothing else...

So even if the mask check told me about 2D be bad, was actually 1D. Even schematics are swapped...
Dumps are correct, but when you look at schematics and mask check you have to think in reverse. What is 1D is 2D and what is 2D is 1D.
Konami black humor
 
Last edited:
You must log in or register to reply here.
Back
Top