ReproductionsComments Off on Konami ‘082’ reproduction
Sep242019
This is my own implementation of the Konami ‘082’ custom IC with the use of simple logic gates.Original part is a 28 pin 600mil IC found on many PCBs of 80’s.As often happens with most of Konami custom ICs it comes with scratched-off part name:
I was able to shrink my board layout to almost same dimensions of original part by using TSSOP devices :
Perhaps not the neatest design, you may call it a ‘poor man’s’ version of the existing CPLD based solution but it works fine and I’m happy with it.Here’s final testing on a Track & Field PCB :
Board looks very clean and plays but has graphical issues
Initial visual inspection showed nothing. All voltages seemed fine too.
Next I started gently pressing down on the custom chips which made a lot of changes when I pressed on 053250
Found a bunch of pins on that were lifted.
I reflowed these and checked the rest of the chip which all seemed fine.
Next boot up test gave me a lot better results but there were jailbars down the whole screen sometimes.
Continuing pressing the custom chips I could make everything change again by pressing on custom 054157.
Found most of this row lifted
and few a few here
Reflowed those little guys and all is fixed in Xexex land.
Its a really nice game that I’ve never played before.
Received from Austria this Heavy Barrel PCB, a run and gun arcade game released by Data East in 1987 :
Board was in really good shape and game was fully playable too but lacked of sound :
Using an audio probe revealed no sound came out from analog circuit hence the fault was of digital nature.Audio system is ruled by a 6502 CPU located on top board :
Its pinout :
Checking it with a logic probe revealed that the /NMI line (pin 6) was asserted as well as two address lines were stuck low (A12 and A14)
A non-maskable interrupt triggered is usually due to some hardware failure hence for first I dumped the ROM containing the audio code (located on bottom board), it turned out to be good.Then I focused on the RAM accessed by the 6502, a 8K x 8-bit device (6116 compatible)
As soon as I piggybacked it with a good chip I got some random sounds played.The scope revealed weak signals on some data lines of the RAM (a good signal on the left of below picture for comparison)
I removed the IC :
And sure enough, it failed the out-of-circuit testing :
Not really much of a repair log but its just something I’ve been up to recently among other things.
Saw this recently on eBay for a cheap cheap price.
I didn’t necessarily want the unit itself but more the CRT adapters it came with but the price was right.
First pass on the visual inspection showed that something wasn’t quite right.
Removing the 4 screws on top lets you get into this thing and there is a spacer post and screw lying in the bottom which is clearly from the PCB housing all those buttons.
While fixing this in place I had a look over everything else and found something a little worrying.
The crispiness aside this is not standard. There are 2 diodes missing and replaced with a transistor with the middle leg cut off, a lot of the traces are gone due to the charring and the 10K resistors have been replaced for two in series and are also the complete wrong value.
I just opted to remove all of this and replace with what it should be.
There is a schematic for this in the back of the manual so could check the values and the connections but I also have the BMR 95 which I took a look at as well
Here is what I ended up with
Its by no means a neat job but short of reproducing that little PCB its pretty much the best I could do given the amount of damage.
I checked all the other components on this and they tested good.
I have now tested this on one of my monitors and compared the readings against my other unit and they are all good so Ill consider this one fixed.
After done it I thought it was good to take into account its “companion” too, the ‘CUS98’ which most times come along with the first.Package is pretty the same but with two further pins than ‘CUS99’
But functions are totally different and, as Rolling Thunder schematics suggest, the main one is to generate the master reset :
Secondary functions are coin counters and lockouts handling as well as lamps driving, it seems only few games (like Pac-Land and Gaplus) use this last feature :
I was obviously interested in reproducing at least the reset generation function which is essential for correct operation of the PCB (but it seems Rolling Thunder can boot into game also with a missing ‘CUS98’).Therefore, as usual, I studied earlier Namco hardware where the custom had not yet been used and I came to a prototype developed on a breadboard:
The protoytpe worked pretty good :
Hence I developed a proper board :
Final testing on a Pac-Land PCB:
As said some functions have been intentionally omitted favoring the needed ones (watchdog circuit ad reset switch were implemented too).This will at least allow to save PCBs with a faulty/broken or missing ‘CUS98’.
