What's new
Jamma extenders were $24 for postage as I shipped them registered. $2 for the shipping box. The ATX cables were over $20 shipped. $4 for the jamma connector. Add up all the other connectors, IDC cables, cost of the two PCBs etc and it all adds up.
 
I was looking for this since long time ago, thank you!

How would you integrate the kick harness into this setup? I'm currently using two 3-pin molex connectors for A-B-C and X-Y-Z.

supergun-mando.jpg
 
Last edited:
The cons is that i use HSS-0136 and wish to put start for coins.

HSS-0136.jpg
 
Some test PCBs arrived. Working perfectly! Switches for remapping are difficult to access so a small flathead screwdriver or similar will be required to slide them into position. Can now assign the top row of Saturn buttons as punches for Street Fighter. :)
58419654_10217114447066343_3442944792059707392_n.jpg
 
There's another project here based upon the Atmega328. Not tried it myself but may be worth a look.

https://github.com/Arthrimus/Saturn_Controller_Demux
Be careful with that code, I wouldn't use it in that form, since it doesn't configure the outputs as open drain, this means that when buttons aren't pressed, it outputs high - you don't want to accidentally touch the GND with those pins :)

Also, this may be just me, but if possible, I prefer to set the microcontroller's outputs all at once, not each time for every output.
 
Looks great. I'm definitely more interested in this than I was originally now, as I was leaning towards the SNES adapter before. The one idea I keep coming back to myself for it is, how about one switch to swap the top and bottom button rows, and another switch to turn the top row into turbo buttons for SHOT 1-3? =O
 
Last edited:
I could change it to use a 3 pole switch, so that the top and bottom rows are swapped over with a single switch. The downside is the PCB size would need to be increased to accommodate the larger switch and most 3 pole switches are those ugly, bulky toggle switches. You'd also lose the ability to remap between individual buttons.
 
How about using a multiplexer instead of a large switch to change the rows? Maybe the pcb size wouldn't increase if PIC and multiplexer are used as smd parts?

And maybe one smd chip at the top and the other at the bottom of the pcb would be another solution, too?
 
Last edited:
It looks great. I'm waiting for three OSHPark pcbs and I'm planning to order the other components.

Could you confirm this circuit configures the outputs as open drain not outputting high when buttons aren't pressed as RGB says?
 
I don't believe so, since in the original application I was tying into other logic ICs so an open drain was not desirable.

I'll put the code up on github later so if there are any changes anyone wants to make they can do so.
 
Been working great for me @twistedsymphony. No damaged controllers, arcade boards etc :)

It does exhibit odd behavior if the adapter is connected without a controller being connected, in that all inputs are locked on. Not sure why that is?
 
@Frank_fjs, you can easily check that, just plug in the adapter to the supergun, power it on, but without the PCB game connected, and then using a multimeter check what you get on the button outputs.

Edit: this is regarding your/rsignal's previous question.

 
I'll try that in a moment.

What should be happening, either way?
 
@RGB

Did the test.

5V with no button pressed, 0 volt with button press.

Test done with no game connected, positive probe of multimeter to player 1 button 1, negative probe to ground.

Is this the right way to test and is this normal / desirable behaviour?

Edit: Neo-Geo controller exhibits exact same behaviour.
 
Is this the right way to test and is this normal / desirable behaviour?
That is normal, as originally designed behavior but it is NOT open collector as you want for these adapters.

Hmm I think no button pressed should be no connection. Button pressed, continuity with GND.
That is what you would see if the outputs were set as open collectors.

I could be wrong but I don't believe the 16F690 supports open collector output. You could however find another chip that does, and move the code there.

Edit: Neo-Geo controller exhibits exact same behaviour.
are you checking that with the controller output pins disconnected?
 
Back
Top