What's new

frsj8112

Enthusiast
Joined
Jul 21, 2015
Messages
187
Reaction score
108
Location
Sweden
I made a newbie mistake. I flipped the JAMMA harness connector on to the edgeconnector on my Thunderforce AC conversion and toasted som DAN803 diode circuits.

I desoldered these (2 out of 4 that went bad) and replaced them with the non broken ones and now I have some stuck inputs.
Namely the shot buttons on both player 1 and 2.

Im guessing that the 315-5296 Sega custom I/O went bad :(
Anyone with any experince with these?
The PCB itself is running the game and so on, but some stuck inputs.
 
This was my first arcade mistake at the age of 15. Thunderforce up in smoke!!!

The DAN resistor packs just replace them all and you should be good to go. They are also on CPS1 i think grab from there.
I think there may be a LS157 in the circuit also but this is YEARS ago ;)

Edit:
Nope, no 157, that's the regular MD. So all the DAN's & RA's need changing.
They glow orange and burn to a crisp when plugged in backward did you see that ;)
 
Last edited:
I will check the junk room for you tomorrow. They are impossible to find NOS (at least back then)
 
You probably damaged the custom '315-5296' which handles I/O.Are there external pull-up resistors/arrays?Now I can't remember.I ask this because usually these I/O customs can have internal pull-up resistors so it could also be you only burned them.
 
There is a row of DAN's and a row of RA's on each. I got lucky with the custom but needed to change almost all of them to get it perfect.
 
ive desoldered the three components connected to this input, one DAN803, one DAP803 and a resistor ladder (8x1kOhm) and that single connection to the 315-5296 has no measurement when probing it with diode test against ground on the multimeter. All other inputs on the same pin group does have standard values on diode test.

So I'm guessing that the 315-5296 is bad :(
 
ive desoldered the three components connected to this input, one DAN803, one DAP803 and a resistor ladder (8x1kOhm) and that single connection to the 315-5296 has no measurement when probing it with diode test against ground on the multimeter. All other inputs on the same pin group does have standard values on diode test.

So I'm guessing that the 315-5296 is bad :(
Don't worry if there's not one in the junk room i know who's got a spare...

Edit: it's on system 18 so yeah got plenty :) i will send you one next time.
 
just as a last chance, try adding an external 10k pullup resistor to the bad chip input.

as for the DAN packs, forget them - just use 8 1N4148 diodes - for people in the u.s. the equiv number would be the 1N914
 
as for the DAN packs, forget them - just use 8 1N4148 diodes - for people in the u.s. the equiv number would be the 1N914
Nice to know, there's nothing like making a search easier ;)
 
as for the DAN packs, forget them - just use 8 1N4148 diodes - for people in the u.s. the equiv number would be the 1N914
Correct, that's what we've used for the Sega I/O boards repros since those diode arrays are impossible to find nowadays.

P.S.: System C is even more sensitive since it's lacking the diodes.
 
I've had a situation where stuck inputs were not resolved by replacing the diode arrays. In that case, I had to replace the IO custom, unfortunately.
 
I've had a situation where stuck inputs were not resolved by replacing the diode arrays. In that case, I had to replace the IO custom, unfortunately.
I can also confirm that, unfortunately I've never had the chance to swap mine.
 
Maybe you can do some external hack with 74LS245s or a GAL to spit out the inputs on the data bus. I did that to fix a Master System with a faulty custom chip. It's on my blog.
 
It's megadrive h/w so easy to reverse the inputs, (or totally change the h/w!)

Let's get the altera part dumped/reversed first before worrying about the easy customs. ;)
 
The megadrive input hardware is not used on C2, though. This uses a custom IO chip.
 
The megadrive input hardware is not used on C2, though. This uses a custom IO chip.
Indeed but it's easy to work something out.
Depending on how far you go from the original design a small code patch can have it running with any input hardware.
 
Indeed but it's easy to work something out.Depending on how far you go from the original design a small code patch can have it running with any input hardware.
Isn't it just reading some specific ports/addresses?
Then you just need to spy on the adress bus, maybe you also need few vital signals (R/W, UDS, LDS).
 
Indeed but it's easy to work something out.Depending on how far you go from the original design a small code patch can have it running with any input hardware.
Isn't it just reading some specific ports/addresses?Then you just need to spy on the adress bus, maybe you also need few vital signals (R/W, UDS, LDS).
I think so, it's not a pseudo security chip like the jaleco etc where other information is inside.

Another thing... <for a repair> Maybe it's possible to re map the controls to NON DEAD pins on the I/o connector? (and again shift the code to the new bits)
Or another one, re-use the dipswitch lines and patch the code ;)
 
I think so, it's not a pseudo security chip like the jaleco etc where other information is inside.
Another thing... <for a repair> Maybe it's possible to re map the controls to NON DEAD pins on the I/o connector? (and again shift the code to the new bits)
Or another one, re-use the dipswitch lines and patch the code ;)
Hacking code is a possibility but you'd have to do the job every time you want to change the game and also it would be tailored for the working inputs you have left. I'd rather design a solution that could be reused in the future.
I'm sure it's as simple as spying the right addresses coupled with few other signals like mentioned before.
 
Back
Top