Help me with Mad Catz DC Hack

[pmc]

I finally got around to dragging out my soldering mess and to take another whack at a controller hack (my last two tries turned out to be multiple-ground pads).

I hit a snag that I can't figure out and maybe you can help.

Seems that my D-pad maps to my joystick in a strange fashion:

Up maps to L which maps to D which maps to R which maps back to Up.

[Hiub1]

Er... Can you post a pic of the pad? Are you actually soldering to the pad contacts or holes elsewhere in the PCB?

[pmc]

Yup. Here you go. If the pics don't post, I'll follow-up (I've had problems with that in the past).

I'm pushing the wires through the holes and soldering on the back side. My wire is too big for all of it to fit through, so I pulled off two strands and wound them around the wire before sticking the rest through. My soldering skills are not great, as you can see.... but I get connectivity and that's all I'm asking for.

I have triple checked (by eye and with a multimeter) to ensure that I have the correct holes.

- pmc

[pmc]

Here's the backside which shows my awesome soldering skills. 

This is specifically showing ground, down, left, and right (if I recall).

- pmc

[Hiub1]

Ok, from what I can gather in the picture, this is the following:

brown: up
black: left
red: down
yellow (in cluster):right
yellow (in middle) ground

Is this correct? Could there be a problem with the actual joystick you are wiring everything to? I am not very familiar with the J-Stick, is there anything odd about it that could cause the problem?

[pmc]

Is this correct? Could there be a problem with the actual joystick you are wiring everything to? I am not very familiar with the J-Stick, is there anything odd about it that could cause the problem?

Yup. that's correct. And if I jump from the CP ground to each of the signal solder-points on the board, it works OK.

I'm going to try the joystick with the IPAC one more time to make sure I'm not missing something obvious.

The j-stick has four cherry-switch like connections on it (for u, d, l, and r respectively). It's wired like any other j/s I believe. It has a switchable 4/8-way restrictor plate and I've ensured that it's in 4-way mode to simplify the troubleshoot.

I'll keep mucking with it, but unless someone points something out, I'll resolder in a way that the behavior seems to require (even though it'd be wrong!).

- pmc

[Dave_K.]

Do you have both the DC pad and IPAC hooked up to the same controls at the same time?

[owenowen]

Totally off topic but your camera takes some pretty quality pics!

[pmc]

Do you have both the DC pad and IPAC hooked up to the same controls at the same time?

Nope. I have a swappable control panel and use a SCSI ribbon cable as a connector. I disconnected the panel from the cabinet and connected it to the DC controller to test. I swapped it back to verify that the j/s is OK. All of the grounds are tied together.

I still can't imagine what's wrong. Why would a j/s "up" return a "left" but when I short between common and signal on the j/s contacts, it goes "up"?

One other thing I should note for you guys: when I test for continuity, I get a strong beep for a good connection. But I also often get a weak beep for connections that should not be connected. i.e., if I test for common between say, the "A" button and the j/s "up" button, I get a strong beep. But then if I test the signal contact and "up"'s common I get a weak signal. I have assumed that this simply because the board components are not 100% isolated and there is bleed through. Could this be related to my problem?

Also, not only have I not yet wired the triggers, but one has been removed to get me access to the bottom of the board.

If I have time today, I'll swap all wires to the left and see what happens.

Totally off topic but your camera takes some pretty quality pics!

It's the Canon Digital Rebel (the 300D). http://www.dpreview.com/reviews/canoneos300d/

Those are no-flash (tungsten room lamp for illumination) 3072x2048 images at "best" quality JPG compression shot at ISO1600. I reduced them to 640x480 for posting and compressed with Q=5 or 6. You oughta see the 3072x2048 images at ISO100 in RAW! This camera is waaay better than my now unused AE-1P (http://www.mir.com.my/rb/photography/companies/canon/fdresources/SLRs/ae1pgrm/).

Cost is coming down on the 300D. If you are in the market for an entry level pro-sumer digital SLR, I think that this remains the best deal out there.

- pmc

[Hiub1]

You know, some pads act funny when you remove all the components. I remember when I tried to hack a Quantum Fighter Pad and I removed the triggers it would do stupid stuff. Also, are you sure your SCSI wiring is 100% correct? You might have made a mistake in the plug part of it and not have realized it. It is easy to mix up wiring the connector so I suggest you check that too.

[pmc]

OK. fixed. I don't understand it, but when I rotated every wire counter clockwise, it all works fine (up is wired to left, etc.). The missing trigger was a red herring as all it did was rotate a plastic cog that turned a wheel on the board. The trigger itself is entirely mechanical.

Common sense says that it was wired wrong and now it's right. I stand by my assertion that it was wired right and didn't work, and now it's wired wrong and it works.

[pmc]

I have both of my MadCatz controllers hacked and tested now (yay!). I only have the triggers remaining. You can see in the first photo above that there are three contacts for each trigger. One is obviously the ground. When I short the 2nd one to ground, nothing happens. When I short the 3rd to ground, the DC (or the game anyway) complains that the controller has been disconnected.

So how do I wire the triggers to work correctly? Can I map them to the C and Z buttons and then wire them from there?

TIA.

- pmc

[pmc]

So how do I wire the triggers to work correctly? Can I map them to the C and Z buttons and then wire them from there?

Ummm... I just realized that C and Z need to default to something and it would be reasonable for them to be digital versions of the shoulder triggers. I googled on that hunch and it seems to be true...

So I think I'm golden so long as I didn't blow the controller ports during my tinkering!

 

[Hiub1]

I believe they should be the trigger buttons, if not you are screwed my man. The reason why I hack only Sega pads is because of how reliable hacking the shoulder buttons is. On every other pad they use weird pots and stuff that have to be grounded individually and becomes a pain in the butt.

[pmc]

I believe they should be the trigger buttons, if not you are screwed my man. The reason why I hack only Sega pads is because of how reliable hacking the shoulder buttons is. On every other pad they use weird pots and stuff that have to be grounded individually and becomes a pain in the butt.

Yup. They are. I've got both controllers wired via the digital equivalents and working great. I had forgotten that one reason the MadCatz are favorites to hack is because you don't have to deal with the triggers. I know you can wire the AstroPad triggers easily, but that contoller has two grounds as I painfully discovered.

I somehow managed to fry a VMU in the process (or, perhaps the PCB's ability to "see" the VMU").

Turns out I did run two grounds on my CP after all (one for each player) but I tied them together at the IPAC and had forgotten that I ran two. So for the DC controllers, I kept the grounds separate which gives me a little better piece of mind.

I'll post a pic of my completed wiring if anyone wants to see it. Thanks for all the help from everyone.

- pmc

[Hiub1]

Glad to see you got everything working.