Jun 282017

Got some months ago in a trade this pretty mint original R-Type board:

Board was faulty, it was stuck on this static screen:

System is made of three boards:

  • ROM board

  • CPU board

  • Video Board:

Swapping all the boards with a good set I could figure out that none of mine was properly working.So I started to troubleshoot the ROM board since it’s the most simple.The program ROMs were dumped as good but I found a bad 74LS245 @1A:

This gave me a working ROM board and a good starting point so I moved on to troubleshoot the CPU board.Analyzing the NEC V30 (uPD70116) main CPU revealed the interrupt line was asserted (active high on this CPU), here’s a screenshot from my scope triggered with RESET signal

The interrupt is generated by the uPD71059C  Interrupt Control Unit , probing it revealed that input pin 3 (READ STROBE) was stuck low :

This comes from output pin 12 of a 74LS244 @IC72 :

I pulled it :

It failed the out-of-circuit testing:

Finally the board booted but with multiple issues.First noticeable one was a totally missing sound but this was because the analog audio section was modified (by factory) to work in Nintendo cabinets.I simply reverted the mod by installing the missing capacitors @C101, C102 an C108 and removed the jumper wire:

All the other issues concerned the graphics.Some examples:

Honestly it would take too long to explain in details what I did (and maybe I even lost track of it…) before the board worked fine again :

All I can say is that I had to replace  in total 15 TTLs (all from Texas Instruments, are we facing another Fujitsu perhaps?)

But I claimed victory too soon , the board went bad again showing jailbars all over the screen:

I quickly pinpointed this further failure in one of the two ‘KNA6034201’ custom tiles generator on video board:

Testing it on a Kung-Fu Master PCB  (where it’s used for sprites too) gave me confirm it was really bad :

Looking at this PCB I noticed that the other custom ‘KNA6034201’ was replaced by a little sub-board:

Technically speaking the custom acts like a big shift register, it can take up to 24 bit of data from tiles or sprites ROMs, here’s snippet from R-Type schematics  :


As you can see in the sub-board picture above the custom functions have been reproduced using six 74LS166 with common shift load and clock signals.So I wondered myself : why not reproduce it in a more modern way?I fired up my CAD and drawn schematics :

Routed them to a PCB layout :

I sent files to manufacturer and after some time I got the bare PCBs:

Here is the final result compared to orginal custom:

Luckily I made no design mistake and testing was 100% successful.Another board repaired and another custom IC delivered to eternity!


 Posted by at 7:36 pm

  5 Responses to “R-Type repair log (a.k.a. custom ‘KNA6034201’ reproduction)”

  1. Impressive and learn full! Thank you

  2. Thanks.Next project is reproducing another custom, stay tuned…

  3. you should start a new area on the site for collecting internal schematics of customs/hybrids.

    others will donate data i’m sure.
    i intend to reverse engineer the customs on Rygar and Commando – so that can be added when i’m done.

  4. Wow. That’s a big load of bad TIs

 Leave a Reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>