My friend Luca sent me this original Konami The Main Event PCB (a decent wrestling game):

After booted up game showed a clear sprite issue:

In these Konami boards sprites DATA are stored in 40 PIN 4Mbit MASK ROM so I went with my logic probe on the two ones @K4 and @H4 and found that the one ‘799B05’ @K4 has some DATA lines inactive.This was confirmed also by the MASK ROM test:

Reprogrammed the DATA into a 27C400 EPROM replacement gave me correct sprites:

Next issue to troubleshoot was the absolute lack of sound.Using my HP10529A logic comparator I found a 74LS74 @G5 with bad outputs but this was not enough to restore sound.Probing the audio Z80 CPU I found that all the control lines were silent so this convinced me to replace it (I had confirm it was really bad as I tried it in another board).At this point I had sound but only some speech samples randomly played.Analyzed the 2018 SRAM @F5 and found weird activity on DATA/ADDRESS lines.Desoldered and tested it confirmed it as bad:

Sound was finally fully restored.Last issue I had to solve was that some Player 1 inputs were not working, for example COIN 1 were stuck LOW as shown in TEST MODE:

As show in schematics inputs are handled by some 74LS253:

Probing the one @E12 with a logic comparator gave me problem on the two outputs, this was confirmed by out-of-circuit testing:

 

After replacing it TEST mode reported all the inputs working again:

So the board was 100% fixed.

Bought this mint original Kaneko Air Buster PCB from my friend Corrado:

PCB was working fine but sound samples (drums, bass line, etc..) were played sometimes randomly other times in a bad way:

In this PCB (and a lot of other  similar audio hardware) samples are played by the OKI MSM6296 chip which take DATA directly from the ROM that I dumped as good.Connecting an external AMP to pin 36 of the OKI6295 (this pin is the digital-to-analog converted signal fed into the UPC324 OP-AMP, you can download its datasheet here in the downloads section ) I had the same issue so this convinced me to replace the chip (with a rebadged one but it’s the same):

This restored fully music/FXs.Job done.

Yes, yet another Double Dragon repair log..

Found this PCB in the pile, honestly I can’t remember when and where I got it :

Once powered ON I was greeted by this title screen:

Also sprites in game were all blocky with some parts floating on the screen:

I also noticed that some sound FXs (especially speeches) were missing so I decided to first troubleshoot this issue.Probing one of the OKI MSM5205 @IC81 I found that all four data outputs were not toggling.The chip read data from the two samples EPROMs not directly but through two 74LS157 multiplexers.I found that select line (PIN1) of the one @IC96 was stuck HIGH so data inputs were not selected at all.I traced it back to an output  (PIN3) of the 74LS393 @IC62 so I desoldered it but it was good.Probing the its CLEAR pin I found that it was stuck HIGH.Tracing it back lead me to an output (PIN6) of a 74LS74 @IC75. I desoldered it and tested it out-of-circuit  having confirm that it was bad:

With sound FXs fully restored I decided to come back to the sprites issue.Fault was located in the VIDEO board since I swapped a good one and issue sprites came back normal.Luckily we have schematics so I started to check for parts of circuits involved in the sprite generation and handling but after two days of troubleshooting and a lot of suspected ICs replaced in vain I came to a dead end.I was nearly to give up and declare the board as not repairable but checking the last part of the object generation schematics I found a missing signal input (called ‘OCGL’) on PIN9 of a 74LS08 @IC36.This signal is generated by an output (PIN5) of a 74LS174 @IC69 always on VIDEO  board.

As I said this ‘OCGL’ signal was present as output of the 74LS174 but not as input of the 74LS08 so it was lost somewhere on the PCB.Following on solderside the traces between these two ICs I came across a via under a 74SL157 @IC34.I desoldered it and found this:

Oxid literally had corroded and eaten the pad and part of trace thus preventing the ‘OCGL’ signal reaching the input of the 74LS08.A tiny piece of AWG30 wire:

and Billy e Jimmy Lee bros could fight again!