XB1 Controller PCB Scans, Traces and Info - 1537

[RDC]

The new XB1 controller is a CG (Common Ground) setup for all of the Buttons. The Triggers are also in a setup where they go Lo when pressed, but they are a bit more complex than the 10k Potentiometer style that its 360 predecessor sported, more detail on those later.

1537 BUTTON/POWER BOARD TOP


1537 BUTTON/POWER BOARD BOTTOM


1537 MCU BOARD TOP


1537 MCU BOARD BOTTOM



If any button is wished to be duplicated, it's just a matter of one side of the new button going to any Ground spot, then the other side going to the button line that you want to duplicate. The connectors J5 and J6 are the best place to solder up wires for all of the buttons, with the exceptions of B, LSC and RSC. The B button has a Via, as well as a solder pad at D14 directly beneath it that can be used. Then the LSC and RSC connections can be made right on the solder joints for them.

If a Trigger is wished to be duplicated, it will get done the same way, one side of the new button will go to Ground, the other side to the LT or RT line, but install a 10ohm Resistor on one side or the other of the new Tact switch so it's not a direct short to Ground when the Tact is pressed. There are no really great places to solder up wires for the LT and RT lines, the only options being a Via, the RC filter after the Hall Sensor (R42/C38 for LT, R39/C36 for RT) or an MCU lead.



For any unfamiliar with my naming scheme..

A = A Button
B = B Button
X = X Button
Y = Y Button
DU = D-pad Up
DD = D-pad Down
DL = D-pad Left
DR = D-pad Right
LB = Left Bumper
RB = Right Bumper
LT = Left Trigger
RT = Right Trigger
LSC = Left Stick Click, Left Stick's button
RSC = Right Stick Click, Right Stick's button
VW = View (Back)
GU = Guide
MN = Menu (Start)
SY = Sync
LSX = Left Stick X-Axis (left/right)
LSY = Left Stick Y-Axis (up/down)
RSX = Right Stick X-Axis (left/right)
RSY = Right Stick Y-Axis (up/down)


FUNCTION - TRACE COLOR

DR - Red w/Black
DU - Yellow w/Black
DD - Green w/Black
DL - Blue w/Black
SY - Dark Purple
LB - Pink
RB - Purple
GU - White w/Green
A - Green
B - Red
X - Blue
Y - Yellow
VW - Light Blue
MN - Medium Blue
LSC - Pink w/White
RSC - Purple w/White
LT - Light Blue w/Light Red
RT - Light Red w/Light Blue


J1/J5 Connector Pinout

1 - RUMBLE MOTOR+
2 - DR
3 - DU
4 - DD
5 - DL
6 - SY
7 - LED IR x 2
8 - 5v USB
9 - LB
10 - D+ USB
11 - POWER OK / SHUTDOWN
12 - D- USB
13 - GROUND
14 - GROUND


J2/J6 Connector Pinout NOTE: "Wired" from here on means connected with the microUSB cable.

1 - V+ (3v Wired), (Battery Voltage Wireless)
2 - LED (under GU, does not drive LED directly)
3 - GROUND
4 - 3.3v
5 - RB
6 - GU (thru D27) Only Guide will work if J2/J6 pin 6 is used for new/alternate Guide button. Use Cathode side of D27 on Power/Button board or controller will not power on when it's pressed.
7 - USB Cable LED
8 - A
9 - X
10 - Y
11 - VW
12 - MN
13 - PNC DETECTION
14 - GROUND


TEST POINTS NOTE: If it doesn't say power source, don't use it to power something.

TP1 - D- USB
TP2 - D+ USB
TP3 - 3.3v (Wired),
TP4 - 0v (Wired),
TP5 - 5v USB
TP6 - 3.3v (Wired),
TP7 - 3.5v, RUMBLE MOTOR+ (switched power source)
TP8 - 3.3v (Wired),
TP9 - 3.3v (switched power source on battery, constant when Wired)
TP10 - 3.3v (Wired),
TP11 - V+ (3v Wired), (Battery Voltage Wireless)
TP12 - 0v (Wired),
TP13 - 0v (Wired),
TP14 - PNC DETECTION
TP15 - 0v (Wired),
TP16 - GUIDE (2.6v not pressed, 0.5v pressed)
TP17 - 3.3v (Wired),
TP18 - USB 5v DETECT (2.95v)
TP19 - 0v (Wired),
TP20 - 0v (Wired),
TP21 - GROUND
TP22 - GROUND
TP23 - 0v (Wired)
TP24 - GROUND
TP25 - 3.3v (Wired),
TP26 - GROUND
TP27 - LT MOTOR+
TP28 - HEAVY MOTOR+
TP29 - RT MOTOR+
TP30 - LIGHT MOTOR+
TP31 - 0v (Wired),
TP32 - 0v (Wired),
TP33 - 0v (Wired),
TP34 - 0v (Wired),
TP35 - 0v (Wired),


My Schematic* of the Button/Power Board - https://www.acidmods.com/RDC/XB1/Schematics/XB1_1537_TOP_BOARD.pdf


My Schematic** of the MCU Board - https://www.acidmods.com/RDC/XB1/Schematics/XB1%201537%20BOTTOM%20BOARD.pdf

* Information will be updated when/if I get them scoped and/or figured out.
** Some of the MCU pinouts are guesses for now, I'd need to put the Logic Analyzer on them to see exactly what they are doing.




STICKS - The Sticks are the same 10k POT style as previous controllers, but are new and smaller than the 360 versions. They have a 1.8v AN+ that comes from a Regulator (U8) and it's also the vRef for U1 on the MCU board.


TRIGGERS - The Triggers use Linear Hall Sensors now, U10 (RT) and U11 (LT). They have 3 pins, 1 - VDD (power), 2 - Output, 3 - VSS (ground). The VDD for them comes from U9, which is turned on/off by U1 at 125Hz, 8ms period, On ~1.4ms, Off ~6.6ms (~18% Duty Cycle). This is done mainly to save power, but it also makes them a little harder to tinker with, but not impossible. (more later)


POWER - The Power/Button board has 3 separate Buck/Boost circuits for power management.

U1 is for when the USB cable is attached, it takes the 5v and knocks it down to ~3v for the other two circuits.
U2 is for a 3.5v source that powers the LED, IR LEDs and the Rumble motors. It's TP7 on the MCU board.
U3 is for the 3.3v source that powers pretty much everything on the MCU board. It's TP9 on the MCU board.

(more on the PnC pack/cable when I have them to tear down)


Clean PCB Scans.

BUTTON/POWER BOARD TOP CLEAN



BUTTON/POWER BOARD BOTTOM CLEAN



MCU BOARD TOP CLEAN



MCU BOARD BOTTOM CLEAN




PLAY & CHARGE PACK

The PnC battery pack this time around is a 3v Lipo, 1400mAH, 4.2Wh. The actual cell is 3.7v and 2.33Wh, so where does the 4.2Wh come from..?



..from the other 2.33Wh cell in there that is in parallel with the first one.



The PnC pack uses the same contacts as the AA batteries for sending power to the controller. The 4 pin connector is for 5v and I2C communication between the charging circuit and the MCU in the controller.



The charging circuit this time around is built into the PnC pack, and the IC responsible for the battery charging is a Texas Instruments BQ24250.



PnC CHARGE BOARD TOP


PNC CHARGE BOARD BOTTOM


PnC CHARGE BOARD TOP CLEAN


PnC CHARGE BOARD BOTTOM CLEAN


My Schematic of the 1556 PnC Battery Pack ROUGH DRAFT - https://www.acidmods.com/RDC/XB1/Schematics/XB1%201556%20PnC%20Pack%20DRAFT.pdf

[Rodent]

Very nice work as always thank you RDC 

[KingMike_OS]

Thank you RDC   

[GhoSt]

very saucy, I'm going to have to pick one of these up at some stage now.
Out of curiosity what do you use to take such level scans?

[RDC]

Welcome guys.


@ GhoSt - Mainly I use a level PCB, because that's really the only way to get a level scan of one.

Being flat on the scan bed isn't as important as being parallel to it, as long as you're not too far away from it.

If you have components that are in the way that make the PCB sit at too much of an angle, a little bit of an angle is usually alright, but if it's too much then you have a couple of options..

1- Remove some components so they allow the PCB to sit more level. In the case of this Button/Power board the Bumper switches are removed. Then the height differences between the remaining components are level enough.

2- Use some brace or counter weights on the back side so the PCB will 'appear' level to the scan bed. This was done with some Quarters on the CG2 scans here. https://www.acidmods.com/forum/index.php/topic,34543.0.html  They can be seen on the front scan at the Headset connector being used to hold that edge of the PCB up so it's level to the scan bed. Then on the back side scan, you can see them thru the center mounting hole, where they are being used to balance the PCB on the Rumble, PnC and Headset connectors so the board stays level and doesn't sit all lop sided like it would if the counter weight wasn't back there.

If you meant the hardware being used, it's a dinosaur Scanjet 3570c, and they are scanned at 1200dpi.

[rafaliyo86]

RDC you are the best

[Modded Matt]

b button trace is not on the id list?

[RDC]

PLAY & CHARGE CABLE

The PnC cable has a Bi-Color (Amber/White) LED in it for showing the charging/charged state of the PnC Pack, along with a few other components and a 4-layer PCB.

Pin 4 of the microUSB connector is used for changing the LED, 0v = Amber, 3.3v = White.

PnC CABLE





HEADSET

The Headset connector this go around uses a 12 pin connector, that will more than likely do plenty more before all is said and done as it's overkill. The connector has the Audio Codec chip, a Texas Instruments TLV320AIC3204 and a PICLF26K22 inside it along with other support components for doing signal processing.

HEADSET AUDIO BOARD TOP


HEADSET AUDIO BOARD BOTTOM


The Microphone and Speaker connections, wire colors, at least for now, are..

GND = Ground, Copper
MIC = Microphone, White
HPR = Headphone Right, Black, Speaker -
HPL = Headphone Left, Blue, Speaker +

[BattleBeaverCustoms]

RDC - Great scans and traces as always. I know there are tons of guys doing some form of a 2.5mm mod to the Mic Puck to use 3rd party headsets. I actually do one myself by installing a small 2.5mm port where the wire stop sits. Ive seen guys say to drop out the black wire (HPR) to essentially let the Speaker and Mic Share a ground since the aftermarket headsets use a 3 pole plug and not a 4 to allow true stereo from the puck to the headset.

question, can anything be changed to alter the voltage and increase speaker volume? Can a resistor be taken off and bypassed?

[RDC]

No, not like that there isn't.

The TLV chip is using the HPR and HPL in a BTL configuration, like bridging the Left and Right outputs of a car amp so you get more power, but also mono sound. Since the XB1 headset only has one speaker, this is why that was most likely done. Using only one side or the other is a Single Ended setup, and naturally has less power, so less volume, but it's also incorrect to do that as the TLV has not been reconfigured to output that way.

Everything on that TLV chip is done Digitally, from how it gets it's signal to how it's setup internally to use that signal as well as the max gain of said signal. It's not just a simple amplifier, it's a pretty complex DAC, AMP and mess of things. The only part that's Analog where changing some passive component's value might do anything is at it's outputs, but even then you're not going to get any kind of noticeable difference for the work involved there.

Until some 3rd party company gets up to speed, or some crazy with a Logic Analyzer and nothing better to do for a couple weeks figures it all out and codes up some chip to reconfigure the TLV (not likely), the only real options are to use the stock headset, deal with the lower volume or run a 4th wire so the headset speaker gets driven properly like the stock one is, and at that point just using the stock one is the better option.

[FOOKz™]

Thanks again man. Also Glad to see they are using lithium batteries instead of Nickel-Cadmium.


Note: Like RDC says a lot of the stuff is overkill on the XB1 system hardware. Us modders can exploit the extra capabilities of the hardware.

[light-of-jesus]

Great work as always RDC.

RDC I was wondering if you knew at what voltage, as the batteries are depleted from use, does the controller shut off?

Thanks

[RDC]

The way the XB1 controller's power setup works isn't quite like the 360 controller, this one has been done a bit better. The power comes from either 2x AA or the PnC pack that uses 2 x 3.7v Lipo batteries in parallel, similar to the 360 deal, but here these are used to power a couple of DC-DC converters that power everything inside the controller, so you always have 3.3v and 3.5v to the controller. So no matter what voltage the batteries are at, within the DC-DC useable range as well as how M$ has the battery voltage detection setup it's always 3.3v. There is also a 3rd DC-DC converter, but it's just to knock down the USB 5v to 3v for the other 2 DC-DC to use when the PnC cable is connected, just mentioning it for the sake of completeness. 

With the PnC pack, you'll never see anything lower than 3v from it, as it has a 3v internal Reg, but the battery protection circuit that's built into it kills it before that to protect the cells from under voltage issues. All of that battery monitoring is done digitally with that thing, so it's always 3v until it croaks. I'd have to poke around in there for a bit to figure that one out exactly.

With the AA batteries, they work down to around 2v or so before the controller doesn't care for them any more, but because of the DC-DC converters the controller gets 3.3v the entire time, so a fresh set of AA that give around 3.3v or ones run down until they're around 2v or so, the controller always gets 3.3v to it.

[GrammatonKlerik]

RDC. So in light of the trigger fix thread being deleted shortly after I presented my reasonings for why your trigger replacement circuit is likely causing damage to the sensors leading to their premature failure, I thought I would offer you the opportunity to explain it here.  If this forum's policy is ignorance is bliss at the expense of its forum members then so be it. But as somebody who I believe is an engineer you know data is data and I hope you will have the confidence to hear me out and let you decide for yourself if my findings are accurate.

So in light of the thread being restored, I won't repeat myself here other than to say I believe the recommended load resistor should be as high as possible in terms of resistance while still allowing the MCU to register a trigger pull.  Through the use of a trim-pot, I was able to empirically discover that this ideal resistance is roughly 330 ohms.  With a 330 ohm load resistor, there will be no functional difference from the user's perspective, but it will put much less stress on the sensor which I believe is a very wise thing to do given the fact that nobody here knows the actual part# and thus it's output current capability.

On a completely different note.  I would be interested in your take on how to best disable wireless functionality.  Do you know if the Y1 oscillator is required by the Wifi/Security chip on the castellated board to function properly or is it only used for upconverting the baseband to the 2.4Ghz carrier frequency? 

[simonmx23]

So am I right in saying that If I wanted to duplicate the B button I would take it to the small round black pin, remove the  carbon to brass and then solder to that point. For the ground point will I take it to tp22 like   rafaliyo86 has done. Or will i scrap some of the carbon of the actual button and solder to both point like on the you tube video Xbox One Scuf Controler DIY . I hope some one can help

[RDC]

@ simonmx23 - It's solder mask to copper, not carbon to brass.

I don't know how you will do it, but either way you're going to get the same result.



@ GrammatonKlerik - I sent you a PM awhile back with a test to try on the Hall sensors to see if you're results there were the same/different and never heard back.

To recap, I've removed one, powered it from a 3.3v source (not pulsed like it is in the controller) and driven the output of that thing to ground thru a 100ohm (just the RC filter) and at most it draws 2mA on the output, which is also what it draws with your 330ohm on there after the RC filter. I left it with the 100ohm all night and it still works. So 330ohm, 10ohm or straight to ground, there's really no difference and I do not see my method causing the issue that so many have supposedly had with it. Pull one off the board and test it to confirm/modify/deny the test results there.

It's a wireless controller, I don't really see the point in disabling it's wireless capability, but I'll have a look at it when I get a chance.

[BattleBeaverCustoms]

@simon I have duplicated buttons by scraping the pads, and it works perfectly if done right. But since we have RDC's scans as a resource I say use them. I wound up just using the points right at the connectors. If a customer of mine doesn't want an LED swapped, I can leave the front board screwed into the controller thus saving time. Heres a couple examples of my early ones, now i can do most all buttons on just the longer rear board. Hope it helps.

[chevy2nova68]

@RDC So in my first attempt to mod and Xbox One controller, I think that i may have made a mistake while de-soldering my power LED. I either messed up the solder point or??? I have no idea. Is there a way to fix the solder point, or solder a small wire to another point for power? Everything functions on the controller EXCEPT for the power LED. It's not a big deal to me just a small annoyance that i would like to fix if it is possible. If needed I can include some pictures of the board if needed.

Thanks for the help!

[RDC]

A picture of the damage is the only way I'll be able to see what was done.

The Vias to the left and right of the pads for the LED are the first place you can use. After that it's onto the bottom side of the board.

[chevy2nova68]

So here it is.... hopefully I didn't goof it up too much...
Hopefully others will learn from my mistakes.

[RDC]

It looks like the pad on the left is still there, just prep the Via on the right of the LED and solder a small wire from it to the LED.

[Methsman]

Hello Sir.

Thank you very much for this PCB Scans.
I used this to made a LED Controller Mod.

I have just one question. I wanted to use TP9 point for 3.3V and TP24 for Ground. The problem is, that the Pad on TP9 falled of by mistake, so i cant use this anymore. Instead of this, i used TP25, but at the end i saw that the LEDs shines, just when i put in the Batteries. I dont want to use that method...I want them to light up when i press the Guide Button and turn the Controller on, just like the Guide Button Led (also this pads are off) :/+)+&&8

So my question is, is there any other Pad i could solder a LED to?

[RDC]

Welcome.

TP25 is not a power source. There is 3.3v on it, but it's not for sourcing anything.

Use TP7, it's a switched 3.5v power source.

Use less heat and/or time to solder your joints. You shouldn't be removing those pads that easily. If you've torn them off, then after soldering them up put a bit of hot glue on the wire, not on top of the solder joint, but back form it a little bit to hold the wire in place.

[Methsman]

thank you for the fast response :-)

So TP7 will work like it should, right? Perfect, im gonna try it later. Can you suggest some SMD Leds, which would fit perfectly for this? Im going to put 3 of them in the mittlepiece (got a transparent one) of the controller. I used some of my broken LED Stripes, and i find the are too big for things lile that.

Thanks in advance :-)

greetz

[RDC]

TP7 will work fine, as long as you don't nuke it as well.

I have no idea what level of soldering experience you have, aside from lifting some TP spot, so the smaller LEDs you might not get along with too well without some practice on them. The smaller SMT components get the more difficult they are to solder, so get you about 10 spares to practice on, and some masking tape, as you'll not find a better third hand for holding that tiny stuff in place while you solder on it with an iron.

Any 0603 or the bit larger 0805 will work and be more than bright enough for what you're wanting to do. You'll see more wire than LED, so use some 30awg Kynar wire if you can. If those are too small, then the 3mm thru hole type LEDs might be what you're needing as they can be made to fit in most places as well.

[Modded Matt]

Well said RDC.

Thanks,

[chiel1987]

hey hello guys,

How many 0603 smd leds can i use on ( TP7 ) and i use (TP21 as Ground).

I hope someone can help me.

Thanks.

[RDC]

Depends on the LEDs you plan to use, in what wiring configuration and what current they need to work. I personally wouldn't pull more than 100mA or so from the spot. That's 10 LEDs at 10mA each in parallel. If you need more light than that there's a switch over on the wall you should use, or you need to change the LEDs you're wanting to use or be powering them with something other than a game controller.

[xboxonehelp]

So I completely messed up and ripped up a button trace by D14, and wanted to know where the best place to connect the trace to, in order to fix it without soldering directly to the access port. Here's my mess up.

[RDC]

No idea what an access port is.

If you're referring to the actual B button pad, then it looks like it was already scrapped on, and you should have just went with that and been done. If you mean the Via, then you're out of luck, as that's exactly where it will have to go unless you want to solder to the MCU lead.

You'll have to cut off what's left of that dangling trace and either prep it to solder a wire onto the bit of it that's still on the board, then jumper down the the Via where it originally went, or from the left of the B button pad to the Via. That trace and Via (red) can be seen in the PCB scans in the first post.