Just a quick fix for a Shadow Warriors PCB I had since many years but never looked at:
Gameplay was fine but all the backgrounds GFX were scrambled and messy:
This is the part of circuit which generates the backgrounds, located on video board:
Luckily EPROMs and RAMs were socketed so I first tested the latter and found a bad one @3D (a Toshiba T5563 pin to pin compatible to a generic 6264) :
Replacing the bad chip was enough to clear the issue.
I got this Major Title 2 PCB for cheap from ‘Smitdogg’ of MAME team:
X i
He said me board had some GFX issues.Actually when I powered it up I got no sprites at all:
For first I checked the sprites EPROMs with my logic probe and found no activity on all pins.These four sprites MASK ROMs on video board are addressed by the ASIC “GA22” on CPU board :
I probed it and all its pins were silent too.I tried to reflow it without luck, sprites were still absent.Then, looking at the board, I noticed an open jumper ‘J3’ right hear the GA22 ASIC:
Looking at other M92 CPU boards I could see that this jumper was always closed to position ‘B’ (so with central and right pins shorted together).So I did it on my board too:
and sprites were magically restored!
Now the technical explanation.The jumper is actually for selecting which clock signal must be delivered to the sprites generator ASIC “GA22”.If closed in ‘B’ position, ASIC will receive a clock signal of 26.6666 Mhz (which is the correct value for most of M92 games) :
If closed in ‘W’ postition, the ASIC will get a 18.000MHz signal (no sprites with this configuration) :
Last issue to fix was some audio background noise:
I quickly traced it to a missing 47uF 16V electrolytic capacitor @C213 in analog sound section:
End of job.
Got this NebulasRay PCB ( converted from another Namco NB-1 game) from my friend Josef for a repair:
For the uninitiated the game is a vertical shoot ’em up released by Namco in the 1994.It runs on Namco NB-1 platform.Here are the specs:
Board actually booted but didn’t pass the disclaimer screen:
As said, main CPU is a 68EC020.Probing it, revelead a good activity until it got stuck on the above screen, at this point most of control lines were going to high impedance state terminating the BUS activity.So, there had to be some trouble in the main code execution but, given the hardware complexity and lack of documentation, I was pretty lost.
I started to do some tracing on PCB so I could locate the WORK RAMs ( two Toshiba TC551001 @5C and 6C):
I figured out that they were not addressed directly by main CPU but through the near custom C351 (which is actually the custom sound CPU).I replaced both RAM but without luck.So, I removed the custom C351 and checking traces underneath it I found two breaks which I patched with some tiny wires:
With this fix the board passed the initial disclaimer screen but went straight into test mode although dipswitches were set to off.I quickly traced this to a missing 74HC253 multiplexer @4F:
Finally the board entered in game but all sprites were wrong :
Object RAMs are two Mitsubishi M5M5256 @20M and 21M:
A closer inspection revealed some loose pins that I promptly resoldered.In this way all sprites were restored except for jailbairs on some of them (like game title ):
All sprites ROMs are on the small daughterboard in form of 8 MX29F1610 FLASH devices :
Since the board was a converted one, they were hand soldered not in professional way (let’s say so..) so I reflowed them.Graphics and sound were perfect now but some inputs were missing or stuck like reported in switch test:
The inputs circuitry is so designed on this kind of hardware : JAMMA connector pin is connected to some EMI filter arrays , then to some custom CUS93 resistor arrays, then to some 74HC253 which combine inputs into outputs that go to the ASIC C160 (which is the I/O chip):
After some tracing I could pinpoint the faults in two bad CUS93 resistor arrays @2L and 1F:
Besides, there was intermittent contact between some pins of JAMMA connector and EMI network filter array @1K:
Board 100% fixed.End of (long..) job.
I got an Edward Randy with a black screen (but partial sound).
After a few checks with my scope, I quickly found a PAL @ location N5 with no signals on all of its outputs (BUT signals on its inputs).
As PALs from this game are not available yet online, I looked for other games that possibly shared the same hardware/system in order to try using a similar PAL.
And yes, that system is listed as “Data East Caveman Ninja Hardware” in MAME and some of these games got their PALs dumped. It was the case of Caveman Ninja and Robocop 2 that shares an almost identical PCB layout than Edward Randy (there are only a few differences in the GFX ROMs part).
So I burned the PAL at location N5 from Robocop 2 and plugged it on my Ed Randy board and here is what I got:
Well, at that time I was thinking it was due to an incompatibility between Robocop 2 and Edward Randy PALs and I temporarily gave up, waiting to get a dump from a working Edward Randy to be sure…
Then Shoestring confirmed me Caveman Ninja and Robocop 2 PALs @ N5 were strictly identical. It leaded him to believe that perhaps they are all common to each other and maybe there is a different problem with my board, which pushed me to have a look back at my board for possible other faulty chips. And he did well…
I started looking for other issues on the PCB and noticed that bending the board made the garbled graphics changing, even making them better looking at some point.
So I suspected the two SMC Data East customs labeled “55” having bad solder joints and reflowed the solder on them.
It then went way better. I had clean backgrounds in game, full intro with clean texts and pictures, title and Data East logo appearing (didn’t have all of that before) but no sprites in game. I noticed a square on the bottom-right corner that seemed containing garbled parts of sprites.
I then managed to find where the sprites part was located on the board and found two 6116 RAMs @ locations N9 and M9 that had suspicious signals on their data lines (pulsing but weakly and at low voltage). Piggybacking a known good working compatible RAM made parts of garbled sprites randomly appearing on screen.
I then desoldered one of the two RAMs (at location N9). With no big surprise it was tested bad on my programmer. I soldered a socket and put a new RAM in place. There was still no sprites but the garbled square on the bottom-right corner was not there anymore. Data lines on the new RAM were looking pretty much better though with strong pulses at 5V. The RAM located at M9 still showed weak outputs so I replaced it with another good known RAM:
Then sprites magically appeared !
Only one thing was remaining: the voice generating chips (2x OKI M6295) were missing so I replaced them and got the sound fully working.
I played the game several times since that and it is working perfectly.
Got this board as fully working
I immediately noticed some problems on the sprites
Transparent objects and shadows were solid and some sprites had glitches
Konami maskroms are known to be unreliable so I ran immediately the maskrom test reveling that rom @25g was faulty.
Before desoldering I checked also the custom underneath which sometims has pins lifted but it was good.
Upon a close inspection I saw some rust on 3 pins of the maskrom @25g which in my opinion could be the reason of the fault
Therefore I decided to desolder it and resolder a 27c160 programmed with the correct file
Board 100% fixed