I have read the tread and I forgot what wattage the 100 ohm resisters need to be for the triggers and I am about to RadioShack if someone could help me out real fast that would be great
1/4w is fine. Others would work too
Ground from the stock hori pcb to any screw inside the case?
Heres how everything is wires in my case (hopefully it makes sense). I suspect Iām missing something idiotic which is leading to my back button shorting and right direction shorting randomly (increases frequency the longer I play until the point its rapid fire)
http://i1154.photobucket.com/albums/p529/chaoticmonk/wiring.jpg~original
Chaotic monk: You should be nominated for this yearās SRK awards because of your OUT-STANDING recent contributions. Iām not coming to these forums as frequently as I did, but this thread has really sparked interest for doing some more padhacks and dual-mods. Excellent job !
ChaoticMonk
When you have that many PCBs in the mix and are seeing odd issues after youāve verified the wiring, itās probably a grounding problem. Every IC induces some amount of noise on the ground plane and if you are daisy chaining from board to board youāre only making it worse for the downstream boards. Itās like making a photocopy of a photocopy, the more you do it the more it looks like shit (I love me some analogies)
What you want to do is maintain a single point ground and VCC. To do this you start with the incomming USB +5V and GND lines and splice both lines once for every board that needs power (4x in your case). What youāll end up with is a splice that looks like a 4-prong fork for both the Vcc and GND lines and then you hook up a pair to each board in a star pattern. Also make sure you connect the USB shield wire to GND as well.
I canāt post a spiffy photo as my work blocks SRK so Iām posting this from my iPhone. Here is some lame ASCII art to drive home the idea. The VCC spice will look the same. Ignore the periods, needed for proper spacing
I canāt guarantee this will fix your problem but it canāt hurt and it will guarantee that each boardās GND plane is at the same potential.
select is next to GND/KGND on the remora so make sure itās not bridged
Hahaha thanks but Iām not doing the work, just regurgitating it with pictures lol
Kind words appreciated though 
Thanks Grammaton. Since Iām using a 2 usb cable setup does that mean I need to do the above x2?
Yes, personally Iād see if that fixes your problem before going through the effort of Grammatonās solution. It worked for me
Given Phreakās discovery of the sensitive voltage with the Xbone pad, would it be safe to to have one connected to an IMPv2 board along with a 360 TE motherboard + Cerberus dual mod?
My concern is if it is in Xbone mode but is accidentally plugged into a PS3 or 360.
Unfortunately, yes. In that case each PCBās GND and VCC pins will have two wires connected to it, one from each USB cable.
Iām not sure why connecting to a metal screw would work, especially if that screw is encased in plastic ā it would just be floating (potentially.) If that screw was somehow connected to the metal case and the case was grounded to Earth ground (itās not) then I can see that working. Grounding issues are some of the most common and frustrating problems to fix so youāre not alone. If it does work, great, sometimes itās just OK to not ask questions, itās not like we are launching these sticks into space or anything so a few unknowns are OK.
Should be just fine, the systems outputs are nearly 5v on the nose if not lower. Iād need a spare power board to confirm the max input voltage but remember I plugged mine into a USB power box, which doesnāt quite have the same regulations a system, PC, or a similar host will have.
For example, the Energizer box was 5.3v out, whereas an iphone power brick is 5.0v, so itās simply a matter of paying attention. Basically, any game system designed for USB peripherals will be locked to nearly 5v and wonāt deviate on the high end, so you should be plenty safe.
One thing we havenāt addressed and should be cognizant of is that the USB host (the game console) can only provide a max of 2.25W of power. Iām not sure what the power draw is of each of these boards but when you start to get into the 3 or 4 boards being powered I would start to get concerned. A powered USB hub may be necessary at some point.
Thanks dude. Iām going to try it out on a mod Iām doing now then. Iām guessing the theory still stands that with an IMPv2 board installed, there will be saved selection for Xbone mode and a saved selection for the dual mod (which because it is going through the Cerberus it will auto-detect?)
I have plugged my xbone pad into both consoles and it worked fine afterwards.
^ likes
Just done a couple of xbone pad hacks and wired them up with stock xbox 360 pcbs found in the Madcatz fight stick pros.
I decided to use a DPDT switch for a simpler setup and it works perfectly thanks to the schematics in this thread and advice on trigger resistors for which 100ohm worked perfect with mine aswell. LT RT were a little fiddly to solder to but I went and got a finer soldering tip and everything went smoothly. I kind of āpre-tinnedā the leg of the component before I soldered my wire to it. I put my 100hm resistor away from the board as the legs of the resistors arenāt the strongest.
I was abit daunted at trying this padhack as was worried I might damage it as it does look alot finer than the usual xbox 360 pcbs im used to but with abit of patience it worked out great.
I was sure not to put blobs of hot glue on the soldered points incase there was a dodgy connection and needed resoldering so used the J&J method of using super glue to hold the wires down and leaving the soldered points exposed incase I need to get to them with a multi-meter etc.
https://fbcdn-sphotos-e-a.akamaihd.net/hphotos-ak-frc3/1475814_773540362663362_821023814_n.jpg
https://fbcdn-sphotos-b-a.akamaihd.net/hphotos-ak-ash3/1504127_773541372663261_1154260755_n.jpg
Maybe on the next (if any) setups I will be a little braver and try to use Toodles IMPv2 board aswell as I still have a load of them lying around.
So with the new trigger spot you solder both of those little points or just one? And you dont need resistors then?
So Iāve been thinking a lot about the best way to hook up the trigger buttons so we donāt push the hall sensor out of limits. I figured the best way to get a feel is to simulate the circuit, so thatās what I did. With a bare XB1 pad with no magnets in close proximity, the Vout pin of the Hall effect sensor will essentially be at Vcc/2 or 1.65V. For simulation purposes, we can model this pin as a Voltage source with output of 1.65V. The trick is to get the line voltage to drop low enough to register a button press without drawing too much current from the sensor. Iāve poured through some Linear Hall-Effect sensor datasheets and 10mA seems to be the absolute max rating for output current, so thatās our do-not-cross line.
Here are my findings:
[list]
[]You most definitely do NOT want to tie your signal to the pin itself. As I mentioned, the Hall sensor is essentially voltage source and itās going to do its best to maintain Vout at 1.65V by sourcing as much current as it needs to accomplish this. If you tie directly to the Vout pin (even with a series 100ohm resistor) all youāre doing is loading down the sensor. In this configuration, the only way the line voltage drops below 1.65V, and low enough to register a button press, is if youāre demanding more current than the sensor can source; If youāre doing that youāre most definitely operating it WAY out of spec and I donāt expect it to last too long.
[]You most definitely DO want to tie your signal after the RC filter (as noted in the acidmods forums.) Doing this puts the 100ohm filter resistor in series with your signal. Now when you press the button and pull the line down to 0v we can calculate the sensor output current using ohmās law: V=IR and solve for I. We can see that I = (1.65V - 0V)/(100 ohm) = 16.5mA. But wait a minute, didnāt I just say that 10mA is the do-not-cross line? Why yes I did. So as it turns out, you still need a series resistor even if you tie your signal after the RC filter
[]So how much series resistance do we need? The non-specific answer is as much as possible while still allowing the MCU to register a button press reliably. Too much resistance and the voltage will not drop low enough to register a hit, too little resistance and we risk damage to the sensor by drawing too much current.
[]The series resistor creates a Voltage divider with the 100ohm filter resistor. If we use a 100ohm resistor we will drop the voltage to exactly half or 800mV. The equation for a voltage divider is Vout = Vin * Rseries/(Rseries + Rfilter). If you plug in 100 for both those resistors you see we get a factor of 1/2. Now we can use I = V/R to calculate the current and we get I = 800mV/100 = 8mA. This should do the trick but I think we can go higher.
[*]I donāt have an O-scope at home so I donāt know the exact line voltage that the XB1 requires before it says you pushed the button, but I can tell you Killer Instinct does not require much trigger pull at all; I estimate about 1/5 of the total travel distance to register a hit. Thanks to the acidmods forums we know that an unpulled trigger produces a voltage of ~1.4V (itās not the full 1.65V because the trigger magnet has a small amount of influence even at rest.) Pulling the trigger 1/5 of its travel distance means the line voltage will be at 4/5 itās resting value or (4/5)1.45v = 1.1V. To be conservative lets see what resistance we would need to pull the voltage down to 1V. We now solve for Rseries using the voltage divider equation and set Vout to 1V. 1V = 1.65 * Rseries/(Rseries + 100). If you do the algebra you get Rseries = 166ohm which isnāt a standard value so lets go with 150ohm. With Rseries set to 150 ohm, the Hall sensor will draw roughly 6.6mA which I think is the sweet spot.
[]The best way to find out what the max series resistance we can tolerate is if somebody wires in a trim pot that goes from 0 - 1000ohm and finds the resistance that just barely registers a hit. Take that value and reduce it by 20% and you have your optimal resistance.
Since everybody loves pictures, here is my Hall simulator schematic done in LTSpice. I simulated the button press to happen at the 40ms mark
http://i.imgur.com/LeVqUcQ.png
Here is the Vout and source current for the hall sensor with no Rseries resistor
http://i.imgur.com/6I3b5B2.png
Here is the Vout and source current for the hall sensor with a 100ohm Rseries resistor
http://i.imgur.com/bs8EBcD.png
TLDR
Wire up your RT/LT buttons AFTER the RC filter and STILL use at least a 100ohm series resistor (possibly 150 ohm.) The former will drop the line voltage to 800mV @ 8.5mA and the latter to 1V @ 6.6mA. The goal is to drop the voltage low enough to register a hit with minimum current draw. I speculate that 1.1V is the point where KI registers the hit so both resistor values should work but the 150 ohm is easier on the sensor.
So I am totally tripped up right now. Working on a Q4RAF and have an IMP board installed and XB1 pad installed with the current configuration.
USB > IMP > D1+/D1- to Q4RAF PCB
> D2+/D2- to XB1PCB
>Switch = Home/Guide Button
So testing on XB1, everything is smooth. Switches easily with home, buttons all work,directions all work.
Plugging it into a XBOX/PS3, the stick isnāt seen.
Now I remember earlier days when using an IMP, it sometimes needed to have the new addition be D1 while the original PCB was D2. Is the same needed for a Qanba?
I used 330 ohm because I didnāt have 100s handy. At least in KI, it works just fine.