What's new

frsj8112

Enthusiast
Joined
Jul 21, 2015
Messages
187
Reaction score
108
Location
Sweden
Hi guys, I’ve been working on some Namco System 1 boards and also Pac-Land and on one of the boards, the 40-pon custom 27 is not working. This acts both as a clock divider and as sync genratoe. Is it possible to reverse engineer this into a new DIP 40 chip? :)

I’ve seen JROKs page about his work on custom 51xx which is used on Galaga

http://www.jrok.com/hardware/custom_ic/custom_51_install.html
 
I'm going to give you my universal answer: with time anyone can achieve anything.
;)
 
It's partially a serious answer.

You think you can't reverse the custom chip by yourself?
You can use your free time to learn electronic basics, programming, etc.
It might take a while but at the end you'll be so proud of yourself.
 
Here is a video that might show the depth of knowledge required:

 
Thanks :)

I've been looking at the Namco custom 07xx which is used on some of the older boards. It looks like a predecessor to custom27.

cust07xx.PNG


And I've also been making a pinout of the Custom27 based on the schematics from Pac-Land, Rolling Thunder and Pac-Mania.

cust27.PNG


I've also found that JROK and Cambridge Arcade Repair Professionals have made CPLD versions of the custom 07xx.
 
The most important thing when you try to do reverse-engineering is figuring out the direction of the pins, if output, input or bi-directional (with or without high Z capability)
As for the Namco custom '27' it most likely contains counters/flip-flop for clock dividing (and more obviously), you can clearly see it on 24-29.
 
Yes, I'm already aware of this but it's only an attempt and not a full reverse-engineering of this 'CUS27', it doesn't work well.
If you want to fix your faulty board, the only option for now is to take this custom from another board, sorry.
 
Last edited:
I prefer not to do that and wait for a proper solution.
OK but keep in mind it may never come out :)


Some people mentioned that it's very similar to this one: arcadeshop.com/i/949/07xx-custom-ic-replacement.htm
Functions are similar because they act both as clock divider but they are diferent (diffrent pin counting, etc) probably the 'CUS27' is an evoution of the' 07xx'
Anyway, I've reproduced the '07xx' with simple gates time ago :

https://caiusarcade.blogspot.com/2020/06/namco-07xx-reproduction.html

I made also a smaller version with surface mounted devices

 
I just got a Namco System 1 boardset from @Darksoft which has a faulty 27 custom chip.
It seems quite simple to reproduce, aside from the very basic clocks generation, it's just a typical video timings generator.
It could almost be replaced with circuits from other manufacturers.

I'll try to reverse it.
 
Hi guys, I’ve been working on some Namco System 1 boards and also Pac-Land and on one of the boards, the 40-pon custom 27 is not working. This acts both as a clock divider and as sync genratoe. Is it possible to reverse engineer this into a new DIP 40 chip? :)
Funny I end up here again :D
BTW it's a DIP42 chip.
 
just to add,
you can save a lot of time sometimes by looking for bootlegs that look original but have sub-boards like a lot of sega / techmo games.
though they arent always correct.
i looked at 2 "chips" similar to what your after from rygar.
both work fine normally, but if you try to invert the screen(cocktail mode) then both of them go to shit because one messed up the design and forgot a gate, the other just didnt try to support it.
.
 
I just got a Namco System 1 boardset from @Darksoft which has a faulty 27 custom chip.
It seems quite simple to reproduce, aside from the very basic clocks generation, it's just a typical video timings generator.
It could almost be replaced with circuits from other manufacturers.

I'll try to reverse it.
Fantastic news 🙏🏻😍
 
Already a lot of progress!
Not bad at all for a first attempt (and I only got the board yesterday afternoon) given my work is theorical, based on my knowledge of video timings on different pieces of hardware as I don't have access to a known good '27' chip.
To me problem at the top is Vsync staying low for too long. I'll make an other try when I'll have time (playing with Vx signals in the equation generating Vsync).

20210220_105105.jpg


20210220_105154.jpg


20210220_105158.jpg
 
I can't guarantee timings are 100% correct, all I know is I can play my game now, that's good enough to me.

Also a few things remain to be implemented:
- support FLIP in the equations (easy)
- figure out OTEN input function
- figure out MODE0 and MODE1 inputs function

To me MODE0 and MODE1 are used to select resolution:

MODE1MODE0RES
0​
0​
288x224
0​
1​
288x244
1​
0​
?
1​
1​
?

Again just a guess, I have no way to confirm it as for now.
 
Last edited by a moderator:
Back
Top