Help with DC Astropad hack?

[pmc]

I decided that I hate my Astropad so it's my pick for hacking. This would be my first controller hack so I'm a little green. I've read several tutorials that were linked to from here but need a little bit more assistance.

The Astropad looks pretty simple inside. There's a daughter-card for the d-pad that has the switch contacts on it. Five leads come from the daughter card and attach to the main board (L, R, U, D, and gnd). I'm assuming I can just cut the daughter card off and solder directly to the wires.

A, B, X, Y is set up the same way. I want to cut the wires and connect leads from my microswitches directly to those wires instead of soldering to the daughter-board copper switch contacts or using the mainboard connection points. Seems easier and less risky to do the soldering off the board.

The two shoulder switches are wierd analog 3-wire switches. I found a thread on this board from fall 2002 that indicates how to wire those switches:

http://www.arcadecontrols.org/yabbse/index.php?board=1;action=display;threadid=3511;start=msg21723#msg21723

The start button has two red wires coming from it and they disappear under the board (which will be difficult to remove so I think I'll leave it in place).

My questions:

1) How can I validate the wires that I think are grounds really are? I own a multimeter, but don't know how to use it for this application. Is there a procedure I can follow to validate all of my wiring choices?

2) Should I validate that the grounds for both daughtercards are the same? How do I do that? My CP is currently all daisy-chained together for ground. I assume that I'll be connecting that daisy-chain to the ground circuit of the DC controller. I'm not sure what the implications of dual grounds are.

3) I assume that one of the two start button wires is a ground.

So this looks pretty easy excepting the wierdness of the shoulder switches. Cut three sets of wires and solder them to the correct leads on the CP.

Anything to worry about? I assume I can blow the controller ports by mixing up grounds, but otherwise, is there any other danger or anything to be concerned with?

Thanks folks! I'll post a pic of the final wiring if I'm successful with this. I won't, of course, if I get scared and bail on this...

 -pmc

[abrannan]

1) How can I validate the wires that I think are grounds really are? I own a multimeter, but don't know how to use it for this application. Is there a procedure I can follow to validate all of my wiring choices?

2) Should I validate that the grounds for both daughtercards are the same? How do I do that? My CP is currently all daisy-chained together for ground. I assume that I'll be connecting that daisy-chain to the ground circuit of the DC controller. I'm not sure what the implications of dual grounds are.

3) I assume that one of the two start button wires is a ground.

The answer to your questions is this:  

Follow the traces.

That will answer all of your questions.  A common ground line will touch all of your controls on the board.  Just follow all of the traces on your daughterboards, the one (and there should be only one) trace that touches all of the contact points (the "meshed Es" where the button pad makes contact) is the ground.  Follow that wire to the main board and over to the other daughterboard, and to the start button.  Now you know which wires are grounds.  

If you're still not sure, post pics, or PM pics to me and I'll help you out.

[pmc]

Thanks. I'll probably post some pics for a second opinion. I'd hate to bust my DC.

The gnd on the duaghter boards is easy to locate as you describe. It's the main board that looks trickier. Sounds like I'll need to free it from it's mount to study the other side.

[TalkingOctopus]

The controller probably only has one ground.  You can trace it using your multimeter by measuring the resistance (ohms) from 1 button's ground to the another button's ground.  If there is indeed like a normal DC controller where there is only 1 ground for the board, there will be close to 0 ohms between the ground wires for each button.

[pmc]

Let's see if I can get a picture to post.

This is a picture of the A/B/X/Y button board. The D-Pad board looks exactly like it. I assume that the ground is the trace that goes around the outside and touches each switch. It's on the yellow wire. You can see the three wires that come from the shoulder buttons.

I have not used a multimeter to validate that the grounds are the same.

[pmc]

Here's the start button. You can see that both wires are the same color. One goes to S11 and the other goes to S12. I assume one is GND and the other is not. How to tell the difference?

[orntar]

have you tried just taking your multimeter ans touching it to one site of the start button and then go around and touch a known ground? if nothing, then try the other side.

[orntar]

i meant to say put it on checking continuity setting (closed circuit). mine beeps when you have a closed circuit. if it beeps on a known ground, then taht side is th ground.  

[pmc]

OK. I'll try that. I tried to a get a few readings and pretty much got nowhere (resistance, voltage, you name it). I tried with it both unplugged and with it plugged in to the DC.

I'll try the continuity trick. Is it safe to do when the pad is connected to a powered DC? I imagine that ground isn't ground without being connected.

[orntar]

I'll try the continuity trick. Is it safe to do when the pad is connected to a powered DC? I imagine that ground isn't ground without being connected.

yea, ground is only ground when there is power present, but all you are checking for is "connection" between two points. so it does not matter.
it does not matter if it is powered up, but i recommend unplugged.

[pmc]

yea, ground is only ground when there is power present, but all you are checking for is "connection" between two points. so it does not matter.
it does not matter if it is powered up, but i recommend unplugged.

Works like a champ! Very cool trick. Unfortunately, I didn't get the answer I wanted.

The A/B/X/Y ground connects to the should button grounds. The D-pad ground connects to the Start button ground (S11) and the analog stick ground.

I don't think the two grounds are connected together although the two solder points for each are right next to each other.

How do I wire this? Can I tie the grounds together? Why in hell would there be two grounds in the first place?!

[orntar]

there could be something going on in the intricuit chips on the board. but since it runs on one voltage (correct me if im wrong on voltage), the grounds should be able to be wired together.

you can check this by bridging the two grounds with an aligator wire. but i am not completely sure if this is ok. if you dont mind the miniscule chance that it will fry your pad, check it. (i would try it if i was doing this mod)

otherwise, wire it just like the pad. you will have to find another solder point for the second ground, and daisy chain the groups you mentioned. so instead of 12 connections (7 buttons, 4 dirs, 1 ground), you would have 13 (7 buttons, 4 dirs, 2 grounds) no big deal.

let me know if you understand that.  

[pmc]

I do, except for two things:

1) I don't care if I fry the pad. They are $0.90 each at Toys R Us and they have at least 10 more there in case something goes wrong. I have no fear of hurting the pad. In fact, I'd like to hurt it! Frying the DC, however, will really suck since I may not easily be able to get another one anymore. And I don't want to fry it even if they were plentiful.

2) If I maintain two grounds, where do they wire back to? My current CP has daisy-chained grounds that wire back to a IPAC. Are you suggesting that I break the daisy chains to stay consistent with the pad's dual grounds? I could do that pretty easily by creating two daisy chains on the CP that are either combined at the IPAC or maintained separately when connected to the Astropad. But I'd have to rewire the CP a bit.

All that said, if you look at that link I provided in the first post, it implies an EASY hack excepting the trigger buttons. That implies a single ground. And given those little daughterboards, it does seem like a straight-forward wiring splice and solder job.

I'm itchin' to just try it, but I'm worried about frying the DC. If you can give me any assurance that it's safe for the DC ("is it safe?"), I'll go for it. I think I'll jumper the grounds first and if that works, I'll go to town and wire the whole thing up because I can't see any questions remaining to be answered. Like you said, it's no big deal.

Thanks for the help.

[orntar]

1) I don't care if I fry the pad. They are $0.90 each at Toys R Us and they have at least 10 more there in case something goes wrong. I have no fear of hurting the pad. In fact, I'd like to hurt it! Frying the DC, however, will really suck since I may not easily be able to get another one anymore. And I don't want to fry it even if they were plentiful.

i don believe there to be any problem at all, let alone to the dc.

2) If I maintain two grounds, where do they wire back to? My current CP has daisy-chained grounds that wire back to a IPAC. Are you suggesting that I break the daisy chains to stay consistent with the pad's dual grounds? I could do that pretty easily by creating two daisy chains on the CP that are either combined at the IPAC or maintained separately when connected to the Astropad. But I'd have to rewire the CP a bit.

yes that is exactly what i meant. but...

All that said, if you look at that link I provided in the first post, it implies an EASY hack excepting the trigger buttons. That implies a single ground. And given those little daughterboards, it does seem like a straight-forward wiring splice and solder job.

ok i forgot about that. it should be straight forward. the grounds are probably combined in a chip on the pad, after all it only has five wires going out. so i am confident that they are the same ground.

I'm itchin' to just try it, but I'm worried about frying the DC. If you can give me any assurance that it's safe for the DC ("is it safe?"), I'll go for it. I think I'll jumper the grounds first and if that works, I'll go to town and wire the whole thing up because I can't see any questions remaining to be answered. Like you said, it's no big deal.

go for combining, i would almost guaranty that all will be well.

[orntar]

so......

did you get it?

[pmc]

so......

did you get it?

Yes. I get it and I'll do it. But I gotta go hunt colored eggs and eat ham for the weekend. I won't be able to play with this until next week.

I  had a beer with an electronics guy last night and I asked him about the separated grounds and he had the same explanation that you've provided.

Anyway, I'll let you know how it works. And I'll take pictures for the next guy if it does.

[pmc]

Got around to wiring tonight and ran into some trouble.

I was pretty confident, so I wired everything up with solder and shrink-tubing and expected everything to work on the first run. It didn't.

Start button doesn't work and isn't incorrectly mapped to any other buttons. So I popped open the DC lid and tried the directionals. It's all messed up. L-R-U-D-X-Y-A-B are all mixed up. I don't see why given how straight forward the wiring is but I also don't know what would go wrong if I mixed up a ground. The good news is that the buttons and joystick register an effect on the DC when pressed. But it's the wrong effect.

I checked connections with a multimeter and everything seems OK.

Then I figured that I could force the START button to work by temporarily closing the circuit by jumpering S11 and S12 (assuming that I mixed up which was ground). No effect.

Sooooo..... I'm going to swap S11 and S12 and cross my fingers.

Also, I tied all three grounds together and tied them into the ground loop on my CP. One big 'ole ground loop. Oh... and I skipped the shoulder buttons given their wierd wiring requirements. I just cut them and left them so I could wire them after I got the easy part of the pad working.

Anything obvious I missing?

- pmc

P.S. Orntar: I didn't mail that package yet. Sorry!

[orntar]

Start button doesn't work and isn't incorrectly mapped to any other buttons. So I popped open the DC lid and tried the directionals. It's all messed up. L-R-U-D-X-Y-A-B are all mixed up. I don't see why given how straight forward the wiring is but I also don't know what would go wrong if I mixed up a ground. The good news is that the buttons and joystick register an effect on the DC when pressed. But it's the wrong effect.

I checked connections with a multimeter and everything seems OK.

Then I figured that I could force the START button to work by temporarily closing the circuit by jumpering S11 and S12 (assuming that I mixed up which was ground). No effect.

Sooooo..... I'm going to swap S11 and S12 and cross my fingers.

Also, I tied all three grounds together and tied them into the ground loop on my CP. One big 'ole ground loop. Oh... and I skipped the shoulder buttons given their wierd wiring requirements. I just cut them and left them so I could wire them after I got the easy part of the pad working.

i would take care of the ground loop. i think they are undesireable. it would be better if everything went to one spot, as opposed to around in a circle. mysterious, i know, but sometimes that could be a prob. but if your buttons are working, just incorrectly, might not be a prob.

Anything obvious I missing?

just go over all the connection and make sure thay are to the correct side of the switch (on the board) but if all is wel, its hard to say what kinda voodoo you have going on there.

- pmc

P.S. Orntar: I didn't mail that package yet. Sorry!

no prob.  

[orntar]

Start button doesn't work and isn't incorrectly mapped to any other buttons. So I popped open the DC lid and tried the directionals. It's all messed up. L-R-U-D-X-Y-A-B are all mixed up. I don't see why given how straight forward the wiring is but I also don't know what would go wrong if I mixed up a ground. The good news is that the buttons and joystick register an effect on the DC when pressed. But it's the wrong effect.

i just had an idea. try grounding it like the pad was originally. that my be the problem, the ground has to be received by that chip in the right place or else it give the wrong signal, messing up the button. you could just try unhooking start and checking the rest of the buttons.

[pmc]

Start button doesn't work and isn't incorrectly mapped to any other buttons. So I popped open the DC lid and tried the directionals. It's all messed up. L-R-U-D-X-Y-A-B are all mixed up. I don't see why given how straight forward the wiring is but I also don't know what would go wrong if I mixed up a ground. The good news is that the buttons and joystick register an effect on the DC when pressed. But it's the wrong effect.

i just had an idea. try grounding it like the pad was originally. that my be the problem, the ground has to be received by that chip in the right place or else it give the wrong signal, messing up the button. you could just try unhooking start and checking the rest of the buttons.

I tried lots of combinations last night. For one, I found a broken wire and it was the ground for the d-pad. After fixing that, things improved a little. I then got out the alligator clips and got experimenting.

Start maps to "X" and "Down" and most of the other buttons and joystick moves result in combo moves I've never seen before. So I assume a button is being held down. So I tied the GND and NC pins closed on the should buttons together assuming that they are interfering (I left them disconnected and twisting in the wind). Or maybe it was NO. Nothing worked after that.

What I'm going to do is start to hack the second pad but I'll do it one step at a time to see where things go wrong. Personally, I think that since the shoulder buttons are reported to require a connection to an NC pin, I may have an open circuit affecting the whole pad.

My last thought.... is it possible that START does not have a wire labeled GND because it has no GND? Perhaps neither wire are meant to be directly grounded.

Anyway... I had figured that this was a slam dunk. I was wrong. All the wires were soldered and tied. Looks like I still have some experimentation to do. I hope to make some time for this early next week...

[orntar]

well, i got my controllers. thanks pmc, by the way.  

i was looking at em and now have a much better idea of what has to happen in order for it to work. i have made some drawings, but the solution for you hinges on this question..

will you have more than one interface in your cab? ie, keyboard encoder and dc?

if no, i think i have it all solved. if yes, then with some diodes you should be able to do it. you are experiencing some bleedthrough of the grounds. let me know what you are going to be doing with your cab and i'll see if i can help. it will still be a while before i get around to mine.

ps. your package will be on its way soon. thanks.  

[pmc]

you are experiencing some bleedthrough of the grounds.

That's what I thought was happening. I was figuring I'd just wire the original shoulders back up and just use the d-pad and buttons because I have proven that works and it suffices for most cabinet-oriented games anyway. Well.... acually, I have the 4-buttons and start-button working together. I have not proven that I can get the D-pad ground to play well with the other two grounds yet.

To answer your questions:

I have swapable CPs so I have a standardized interface for it. The current CP is connected to an IPAC. I disconnect that interface to connect to the DC hacked controller interface. More CPs will come down the road, and I'd like to also be able to maintain the ability add new consoles in the future (but there's nothing on the radar now).

One other thing to note that may be important: my IPAC has two ground connections on it. So on my CP I ran two ground loops: one for player 1 and one for player 2. I ended up tying them together (at the IPAC), but can separate them if I need to. So f I need two gounds, I can rewire the CP for two separate grounds and leave them separate when using the DC, but tie them together when using the IPAC. I thought that was a great idea until I remembered that controller number 2 would also (presumably) require it's own set of grounds. I could wire for that too, but I'd loose my ability to switch back to the IPAC without also tying the two extra grounds together each time. Too much of a pain. Better to go get original DC controllers and do the proven hack. But I'm not giving up on these $.090 contollers yet!

So where would the diodes go?

ps. your package will be on its way soon. thanks.  

New toys! Yeah! Thanks man. And thanks for the help with the controllers. They're just sitting on the floor half-assembled right now and I sneer at them while booting up Pacman on the MAME box...  

[orntar]

ok here is a new idea. does not need diodes. it uses the same grounding system as the original astro does. you would need one db25 connector for each player tho. so who knows if its usefull. let me know if you have any questions.

http://www.orntar.net/dc_wiring.jpg

this would also allow you to add more systems as you needed/wanted.

[pmc]

So I've been brooding over this for a couple of days now. What you propose is what was originally posted to wrongcrowd in 2001 (Matt pointed me to the archive of the tutorial, sans photos):

http://web.archive.org/web/20010215023922/http://wrongcrowd.com/~arcade/dcastropad.shtml

Basically, maintain the grounds. Combine 'em back together at the IPAC for use with MAME. I didn't realize that this pad had separate grounds (and I had heard that it was unlikely) so I confidently wired everything up...

While I've been beginning to like the maintain-separate-grounds approach, I've decided that I don't think I want to do it for several reasons:

• My CP is already wired and I have to rewire it to support multiple grounds. I use a hard-wired SCSI ribbon cable for an interface (50-pin) so I don't have the flexibility to playing with alternate DB25 arrangements. I've been testing with alligator clips.
• I have two controllers, so I'd have to support 8-grounds!
• I do have one controller working with shared grounds, but so far, only START and A/B/X/Y. If I can get the DPAD to work too, I'll bail on the shoulder pads and use as-is. Except that I haven't tried both controllers at the same time!
• I can probably eBay some original controllers (single ground) and hack them easier than continuing to muck with this. I've been having fun, but eventually I'd like to be able to use the DC in the cab.

How complex is the diode method? Worth playing with? From what I know about electronics (not a hell of a lot), I'd guess that it works like this: I put a diode inline with each signal wire (or on the ground?). Current is blocked until a button is pushed and the voltage pushes the diode open so the circuit works.  Each circuit is otherwise isolated because the diodes block current until required. How's that for a guess? What happens when you press all the buttons at once!?  

I hope that you didn't do that beautiful drawing just for my benefit!

[gnateye]

those pads stink, sorry, but you are gonna have a hard time no matter what, your best bet is to go find a madkatz 6 button dreampad, the shoulder buttons are mirrored on the top as regular buttons so you can have all 6 wired up normal. plus a lot of the olde4r mad katz pads have holes in the pcb that you can just stick your wires thru and solder. sadly though if you plan to have more than one dc controller you have to run seperate ground for each controller or it wont work. but the regu;lar hacking is so much easier on a madkatz pad i have done quite few and those are the ones i look for now when i need to do a hack. if you had aone of those pads yo uwould have been done weeks ago. they shouldnt be that much more then the cheapies you got now, a few bucks a t most.

you can see one in this auction:

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&category=62054&item=8107546408&rd=1&ssPageName=WDVW

[orntar]

Basically, maintain the grounds. Combine 'em back together at the IPAC for use with MAME. I didn't realize that this pad had separate grounds (and I had heard that it was unlikely) so I confidently wired everything up...

yea me neither, i thought you'd be fine too. apparently they found a better way to do stuff  << sarcasm.

While I've been beginning to like the maintain-separate-grounds approach, I've decided that I don't think I want to do it for several reasons:

• My CP is already wired and I have to rewire it to support multiple grounds. I use a hard-wired SCSI ribbon cable for an interface (50-pin) so I don't have the flexibility to playing with alternate DB25 arrangements. I've been testing with alligator clips.
• I have two controllers, so I'd have to support 8-grounds!
• I do have one controller working with shared grounds, but so far, only START and A/B/X/Y. If I can get the DPAD to work too, I'll bail on the shoulder pads and use as-is. Except that I haven't tried both controllers at the same time!
• I can probably eBay some original controllers (single ground) and hack them easier than continuing to muck with this. I've been having fun, but eventually I'd like to be able to use the DC in the cab.

yea, that is what i thought, that is why i said im not sure if it is usefull. damn, and i was gonna do the same thing as you. i might still, i guess, but i will be cursing it all the way.

How complex is the diode method? Worth playing with? From what I know about electronics (not a hell of a lot), I'd guess that it works like this: I put a diode inline with each signal wire (or on the ground?). Current is blocked until a button is pushed and the voltage pushes the diode open so the circuit works.  Each circuit is otherwise isolated because the diodes block current until required. How's that for a guess? What happens when you press all the buttons at once!?  

thats not quite it. you would put a diode in between the start button common and n.o., then it would not get ground to both sides when pushed, cause the way it works now, it sends ground to a combo of buttons, that is why there is no real ground on the start, it is doing something else. i thought i had it all planned, but i don't think i do now. i am no electronic expert either.

I hope that you didn't do that beautiful drawing just for my benefit!

actually, i did. but it is what i do (for work and fun) i enjoy it.

[Grasshopper]

those pads stink, sorry, but you are gonna have a hard time no matter what, your best bet is to go find a madkatz 6 button dreampad, the shoulder buttons are mirrored on the top as regular buttons so you can have all 6 wired up normal. plus a lot of the olde4r mad katz pads have holes in the pcb that you can just stick your wires thru and solder. sadly though if you plan to have more than one dc controller you have to run seperate ground for each controller or it wont work.

I was a bit alarmed to read this. My plans are based on the assumption that only one common ground is needed for all the buttons even when using two or more DC controllers.

I was under that impression that as long as the controller did not use some some of scanning matrix (like most PC keyboards) then ground would simply be a voltage and they could all be safely tied together.

Do you know whether this characteristic (i.e. the need for each controller to have its own separate ground) is just a quirk of the Madkatz controllers or whether standard DC controllers behave in the same way?

Thanks in advance

[gnateye]

nope sadly for some reason on the dreamcast all of hte controller ports use seperate grounds, and not wiring them seperately could fry the dc, or at the very least the dc controller board, which is very touchy about voltages and stuff. i was bummed when i first had my dc in my mame cab cause i had to rewire my whole ground loop and add a seperate wire to my db25 connecter as i now needed 26 wires. so sorry to the bearer of bad news, but i guess its better to find out now than after frying your dreamcast.

so in closing, one ground per dreamcast controller.

[pmc]

so in closing, one ground per dreamcast controller.

Wait wait wait. You mean, 1-ground per controller is a DC thing and not a controller thing? While I'm likely to bail on the Astropads and replace them with Sega controllers (or Mad Katz as you sugest), you are saying that I'll need separate ground loops for each controller? Let me rephrase one more way just to make sure I get this: It's a DC thing and not a Mad Katz thing. Is that what you are saying?

Meantime I've been happily using the original controllers for gameplay... while standing in front of the cabinet. I'm halfway there.  

And hey... the Astropad thing wasn't a total loss. I've learned a boat load through this process. For a $0.90 contoller... I'm happy as pie to spend the time learning. I'm just glad you mentioned the above before I went and wired both controllers into the same ground loop!

Thanks!

[Dave_K.]

You do not have to run seperate grounds for each DC controller (unless you are using a joystick which uses seperate grounds like the astropad).

If you use the original Sega pads or Madcats pads, you can daisy chain a single ground around your control panel for both joysticks.

The issue of frying the DC joystick port happens when you share a ground wire from the DC with another power source (say from a PC or another console).  Using a switcher or molex connector isolates these grounds to their respective power supplies.  

[gnateye]

sorry, but thats not true at all, there is a low voltage capacitor on the controller board, if more voltage goes thru it than the dc expects for each seperate controller port, it blows and the controllers dont work.

dave, have you tried the method you are stating?

in my experience, and i have hacked quite a few dc pads madkatz and sega brand. i have always had to make seperate grounds for each controller, which was a big hassle the first time, as the dc was sharing a cab with the mame pc. but now the dc has its own cab so no worries on the wiring. basically teh dc is a touchy bastage when it comes to voltage, the capacitor is made to blow so the dc wont fry, but rreplacing the controller boatrd cost more then replacing hte whole dc nowadays.

so let me restate, in my experience, it is a dc thing not a controller brand thing. one ground for each player to the dc from the controls.

does anyone else with dc hacking experience want to jump in and settle this? snaaake? anyone?

[Grasshopper]

sorry, but thats not true at all, there is a low voltage capacitor on the controller board, if more voltage goes thru it than the dc expects for each seperate controller port, it blows and the controllers dont work.

dave, have you tried the method you are stating?

in my experience, and i have hacked quite a few dc pads madkatz and sega brand. i have always had to make seperate grounds for each controller, which was a big hassle the first time, as the dc was sharing a cab with the mame pc. but now the dc has its own cab so no worries on the wiring. basically teh dc is a touchy bastage when it comes to voltage, the capacitor is made to blow so the dc wont fry, but rreplacing the controller boatrd cost more then replacing hte whole dc nowadays.

so let me restate, in my experience, it is a dc thing not a controller brand thing. one ground for each player to the dc from the controls.

does anyone else with dc hacking experience want to jump in and settle this? snaaake? anyone?

I'm more confused than ever!

Admittedly I'm no electronics expert, but I don't really understand what you're saying. Linking the grounds together shouldn't change the voltage.

I can sort of understand one of the DC capacitors blowing if too many amps flow through one of the ports. However, presumably if you link two grounds together you are doubling the total amps that can flow through the circuit without causing a problem.

Incidentally, Is there one capacitor per port?

[pmc]

I'd definitely like to see a ruling on this because no matter what controller I use, I'll be connecting them to a CP with a common ground. I can rewire to separate Player 1 and Player 2 grounds but won't bother to if I don't have to.

-p

Orntar: got the package in the mail today. Thanks!

[Grasshopper]

I've just been doing some googling on this. It looks like it's a resistor that blows not a capacitor.

The good news is that it appears to be fairly easy and cheap to replace. Check out this link.

http://groups.google.com/groups?q=%2Bresistor+%2Bdreamcast+%2B%22controller+port%22&hl=en&lr=&ie=UTF-8&scoring=d&selm=MPG.18f7f59a816bf7e198a2e1%40news.cis.dfn.de&rnum=3

[Dave_K.]

Sorry gnateye, I believe you are correct.  In my cab, I do have seperate grounds for each player side.  So each hacked DC pad I have uses a ground pin in my molex connectors, which then daisy chain through just one half of the control panel.

I think the reason I got confused was that I checked the joystick port board with a multimeter and found each port uses the same ground (they are all linked on the board).  I then assumed it would be safe to use this ground reference anywhere in your control panel (as long as its hooked up to DC controls).  In theory I think this may still work fine, but as you mentioned, the fuse (or rather low wattage resistor) is quite a touchy bastage.    In fact I've herd of some people blowing the joystick port by simply unplugging, and re-plugging a joystick into the DC while it is on.

So in theory it may be ok to use the same ground reference, but in practice its probably safer to use seperate grounds.  And this is not just Dreamcast specific, this would apply to all consoles.

I'm sure you guys have seen this pic before, but here is an example of my control panel (duh, I still don't know why I thought I shared the grounds...its pretty apparent from the pic).

[Grasshopper]

gnateye, have you actually damaged a Dreamcast by combining the controller grounds, and if you did then were you using the same brand of controller in each port?

Dave_K, have you done a continuity test between each controller's button ground with both controllers connected to the DC (obviously with the DC switched off)? If there is no resistance then I would be pretty certain that the grounds can be combined safely. However I'm not prepared to bet my Dreamcast on this until I've heard a few more opinions and done some testing myself! Nice looking panel by the way. Would it be possible to see a picture of it with the lid closed?

Thank in advance

[Dave_K.]

Grasshopper, I did the continuity test, and as expected, there is no resistance.  The grounds match.  I then decided to get daring and try my first test.  I wired the ground from joystick 2 up to joystick1's start button on my control panel.  Result: button worked just fine.  I then got a little more daring and decided to bridge the grounds from both joysticks (this would surely blow the resistor/fuse if gnateye was correct).  Result: both player sides/joysticks worked fine.

So, I guess the theory was true, you can run a single common ground for all DC joysticks.  But I'd like to point out that even though this may be a bit easier (and you save one pin in your connector and some wire), I don't think its worth it.  It seems like a much cleaner implementation to use seperate grounds. This way you do the same hack for all joysticks, and use the same molex pin configuration for each connector.   I use this setup for both DC and Playstation.  Would suck to get each controller mixed up not knowing which one had the ground pin connected and which didn't.

Anyway, thanks for the feedback on the panel, here is another pic.   I have black pica vinyl on it.  Pretty plain and simple (and little dusty).

[gnateye]

i stand corrected. in the past i had major funkiness on my first hack and by seperating the grounds it fixed the problems, so maybe it was something else. but im still gonna do it this way on my stuff, of course everyone is gonna do it however they like. but i stay in the "better safe than sorry" school.



good luck all.

[Grasshopper]

Thanks for the replys guys, and a particular thanks to Dave_K for risking frying his Dreamcast in order to answer my query!

Although I'm now reasonable happy that the DC's grounds can be combined I can see the logic in having separate grounds because it's best not to make too many asumptions about how future consoles/gamepads may behave.

I'm trying to come up with an arrangement that is as future proof as possible. However, if I do decide to use separate grounds then I'll have to make a few compromises. I think I'll start a new thread to expain what I'm trying to do.

Dave_K, I like the look of your Dreamcast cab, but I'm even more jealous of the Robotron machine in the background! And is that a Joust even further along?

 

[Dave_K.]

Dave_K, I like the look of your Dreamcast cab, but I'm even more jealous of the Robotron machine in the background! And is that a Joust even further along?

Wow, name that game in 10 pixels!   Yeah thats a Joust behind Robotron.  I also have a mint Stargate which completes my classic Williams trio of games.  Other games you don't see in that pic are my Playstation2 cab (an NBA on NBC originally) and a sit down SpyHunter which I've been in the process of restoring for the past 5 years (ok maybe neglecting is a better word).