The ‘F’ curse strikes again!

This could be a perfect title for this repair log but let’s start from the beginning

Got this faulty PCB from a friend:

It had sever issues, only most parts of graphics were missing  and colors were clearly wrong while sprites barely visible

After studying the hardware I could identify the tile RAM in two MB8128 (2018/6161 compatible) chips @IC73 and @84.Probing them revealed irregular actitvity on some address lines.Piggybacking them with a good chip didn’t change anything.I could trace these address line back to the outputs of some 74LS157 multiplexers.Testing them with my HP10529A logic comparator gave me faults on three of them which I promptly desoldered and tested out-of-circuit having confirm they were all bad:

This improved graphics so that I could clearly also see the results of the ROM/RAM test:

IC73 and IC84 (tile/backgrounds RAMs) were riported still bad as well as the color RAM @IC85 so was time to investigate deeper.

Data bus of the color RAM (a 2016) was connected to a couple of 74LS175.Piggybacking the one @IC108 restored colors while piggybacking three 74LS245 @IC104, @IC105, @IC106 (involved in tile RAM data bus) restored most all  the  graphics and also the RAM/ROM test passed succesfully:

But there was another issue (see picture below):

To cut the story short I found another faulty 74LS157, one of them that I previously tested as good has gone bad while I was troubleshooting the board!Replaced it fixed completely the PCB.

Ah, I forgot…all faulty TTL ( nine in total : four 7LS157, one 74LS283 -not reported in this log-, one 74LS175 and three 74LS245) were manifactured by Fujitsu! Of course, the curse stroke again but I pushed it back! 🙂

Got this faulty Violent Storm PCB from Ebay (well, actually seller sold me as Metamorphic Force one…)

When I turned it on all I got was a solid black screen but  this time watchdog circuit was not active.As many other Konami boards RESET signal is generated by PIN10 of  the SIL custom ‘051550’.In this case signal was correct (first LOW then HIGH) but not on main CPU (an MC68H000GFN16 in PLCC package) where it was stuck LOW.So was time to investigate.I could trace RESET signal from PIN10 of the ‘051550’ custom to PIN7 of a custom marked ‘053252’ (44PIN in QFP package) while , from other side, RESET (PIN18) of the main 68000 CPU was tied to PIN12 of this ‘053252’ IC through some inverter gates (74F04).This was confirmed also by Metamorphic Force schematics (which runs on same indentical hardware) I could find at a later time:

As I said RESET signal was correctly generated by ‘051550’ custom, routed to ‘053252’ ASIC but then coming out from it stuck LOW.So with this assumption I had no other way than try to replace the ‘053252’ ASIC (in MAME source it’s indicated as CRT and interrupt control unit):

And I was right since board successfully booted and all was properly working:

I was closing the case and declaring the board 100% fixed but the curse of the Konami hybrid sound module struck again!Infact while I was testing the board, the sound suddenly faded.But this time it was not only matter to replace all the SMT caps with tantalum ones but also the LA4705 audio amplifier broken due high ESR of the old caps and we all know that capacitors have the role to stabilize amplifier preventing its oscillation and failure:

That’s all for today, I hope you enjoyed this repair log!