Author Topic: XB1 Controller PCB Scans, Traces and Info - 1537  (Read 79749 times)

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #120 on: March 02, 2016, 04:52:42 PM »
There are currently 3 versions of the top Power board, but they all work exactly the same at their core. The only real difference is the IR LED control line.

1 - 1537 (first version) has IR LEDs.
2 - 1697 (second version) has no IR LEDs, but kept the same board layout so that IR control line does nothing now.
3 - 1697 (Elite) also has no IR LEDs, but that control line is now the 1-2 switch.

The LED will not light up if the bottom board is not present. I could blather on exactly why, but trust me, it will not.

The only way that it could, is if you happen to get the Guide button pad a little bit offset when pushing it down, and then it will make contact on the Cathode side of the LED grounding it, which will make the LED light up, but that's the only way that can happen with no bottom board attached.


Now, for the thing not lighting up when you have the bottom board in place, you'll need to check the soldering connections there first, as it should. Either connected to a PC USB port or some phone charger, you should get some LED flashing after pressing the Guide button down. Post some good pics of the board if you can and I'll have a look see.

Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline Umbra

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #121 on: March 03, 2016, 12:57:19 PM »
There are currently 3 versions of the top Power board, but they all work exactly the same at their core. The only real difference is the IR LED control line.

1 - 1537 (first version) has IR LEDs.
2 - 1697 (second version) has no IR LEDs, but kept the same board layout so that IR control line does nothing now.
3 - 1697 (Elite) also has no IR LEDs, but that control line is now the 1-2 switch.

The LED will not light up if the bottom board is not present. I could blather on exactly why, but trust me, it will not.

The only way that it could, is if you happen to get the Guide button pad a little bit offset when pushing it down, and then it will make contact on the Cathode side of the LED grounding it, which will make the LED light up, but that's the only way that can happen with no bottom board attached.


Now, for the thing not lighting up when you have the bottom board in place, you'll need to check the soldering connections there first, as it should. Either connected to a PC USB port or some phone charger, you should get some LED flashing after pressing the Guide button down. Post some good pics of the board if you can and I'll have a look see.

Thanks for the quick response. That makes sense. The LED solder points must've been a little off-kilter and a the button caused a short allowing it to light up. Thanks for the explanation.

Here are pics of the boards in question. Aside from the obvious damage and tinkering with the controller B top board I don't see anything out of the ordinary. If you need me to zoom in or test anything, please let me know. Thanks.

Controller A bottom board:




Controller B top board (the led is connected to lamp wire and it's floating above the board right now):





Ebay top board:






« Last Edit: March 03, 2016, 12:58:12 PM by Umbra »

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #122 on: March 03, 2016, 01:57:56 PM »
Both of those board versions are the same, and are a match for the MCU board that you are using, but any will really work with any other board, though I wouldn't use an Elite Power board on any of the older MCU boards, but everything you have there is fine.


The MCU board looks
 fine, so if it worked before you started working on the top board, then it should still be alright.

The controller B top board, if the LED polarity is correct, Q3 goes to the Cathode, then R5 goes to the Anode, and that soldering on the LED looks questionable from that pic. You'll want to make sure that wire is not touching the USB connector as it will short to ground.


On that eBay top board, the LED doesn't look at all like it has been soldered on there. It may feel stuck on there, but the flux alone can do that, so it may only have one side half soldered, then the other side insulated from the pad by the flux under it. That doesn't look to me at all like it's been done properly and if that's how you received it I'd let the seller know about it, if nothing else he could have put the LED on straight and not broken J2, unless you did that. I can't make out the polarity from the pic, so if that is correct, Anode + on the left, Cathode - on the right, I'd just take the iron and a very small amount of solder to it and make sure it's actually on there good. Then test the thing out.


Just from what I can see there, it only looks like soldering issues to me, but if you're 100% sure that's all good, or have redone them all and are sure, then you can try this.


First plug up one of the Top boards (no bottom) with a USB cable to some power source, PC or whatever, and set your meter for measuring DC Volts. Flip the board over and place the black lead on the USB connector shield, or any ground spot, then put the Red lead on the top of R5, you should get 3.5v. If not, check on the bottom of R5, if you have 3.5v there, test the top again and make sure. Don't be rough, as you don't want to rip R5 off there, but make sure the lead does gently pierce the solder as sometimes crud can buildup over solder joints and can keep you from getting a good measurement, so you may only get 3.5v on one side from a false test, but as long as you get it on the top side of R5, then that's good and suggests there is an issue with the LED or it's connections, the soldering, polarity, traces or pads. If you have 3.5v on the bottom of R5 and not on the top, then R5 is bad. It is an 80ohms, 0402 SMT Resistor.

If you have no 3.5v at all, then first try some other power source, a USB port on the PC, if there is still no 3.5v, then there is an issue with the regulators and that's a lot more than you'll probably be able to get into diagnosing there.

Unplug the USB cable and set the meter to Ohms. Check R10, it should be around 1.5k (1500ohms).

I have yet to see Q3 fail, as it's just an NPN Transistor, let alone 2 of them fail, so for now I'd leave that of the testing, but that there is everything in the LED circuit. You have 3.5v that is always on when the USB cable is connected. That goes thru R5 to keep the LED from frying. Then the signal for the LED comes thru R10 to Q3 turning it on, and it connects Ground to the LED Cathode and lights it up.

If you have 3.5v on the top of R5, and if R10 is 1.5k, and if the LED connections are good, and if the LED is installed correctly, and it still doesn't work connected to the bottom board, then that MCU board is suspect.


Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline Umbra

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #123 on: March 03, 2016, 03:08:13 PM »
Both of those board versions are the same, and are a match for the MCU board that you are using, but any will really work with any other board, though I wouldn't use an Elite Power board on any of the older MCU boards, but everything you have there is fine.


The MCU board looks
 fine, so if it worked before you started working on the top board, then it should still be alright.

The controller B top board, if the LED polarity is correct, Q3 goes to the Cathode, then R5 goes to the Anode, and that soldering on the LED looks questionable from that pic. You'll want to make sure that wire is not touching the USB connector as it will short to ground.


On that eBay top board, the LED doesn't look at all like it has been soldered on there. It may feel stuck on there, but the flux alone can do that, so it may only have one side half soldered, then the other side insulated from the pad by the flux under it. That doesn't look to me at all like it's been done properly and if that's how you received it I'd let the seller know about it, if nothing else he could have put the LED on straight and not broken J2, unless you did that. I can't make out the polarity from the pic, so if that is correct, Anode + on the left, Cathode - on the right, I'd just take the iron and a very small amount of solder to it and make sure it's actually on there good. Then test the thing out.


Just from what I can see there, it only looks like soldering issues to me, but if you're 100% sure that's all good, or have redone them all and are sure, then you can try this.


First plug up one of the Top boards (no bottom) with a USB cable to some power source, PC or whatever, and set your meter for measuring DC Volts. Flip the board over and place the black lead on the USB connector shield, or any ground spot, then put the Red lead on the top of R5, you should get 3.5v. If not, check on the bottom of R5, if you have 3.5v there, test the top again and make sure. Don't be rough, as you don't want to rip R5 off there, but make sure the lead does gently pierce the solder as sometimes crud can buildup over solder joints and can keep you from getting a good measurement, so you may only get 3.5v on one side from a false test, but as long as you get it on the top side of R5, then that's good and suggests there is an issue with the LED or it's connections, the soldering, polarity, traces or pads. If you have 3.5v on the bottom of R5 and not on the top, then R5 is bad. It is an 80ohms, 0402 SMT Resistor.

If you have no 3.5v at all, then first try some other power source, a USB port on the PC, if there is still no 3.5v, then there is an issue with the regulators and that's a lot more than you'll probably be able to get into diagnosing there.

Unplug the USB cable and set the meter to Ohms. Check R10, it should be around 1.5k (1500ohms).

I have yet to see Q3 fail, as it's just an NPN Transistor, let alone 2 of them fail, so for now I'd leave that of the testing, but that there is everything in the LED circuit. You have 3.5v that is always on when the USB cable is connected. That goes thru R5 to keep the LED from frying. Then the signal for the LED comes thru R10 to Q3 turning it on, and it connects Ground to the LED Cathode and lights it up.

If you have 3.5v on the top of R5, and if R10 is 1.5k, and if the LED connections are good, and if the LED is installed correctly, and it still doesn't work connected to the bottom board, then that MCU board is suspect.

Again, thank you so much for a quick and detailed response. Controller B's top board is a mess from experimentation, but the wires connecting the LED to the back aren't touching anything else (when the negative side touches the USB connector when plugged in it'll short and light up.


The LEDs are wired with he correct polarity. the multimeter lights them both up with the red probe on the left, but not vice versa.

The LED on the ebay one is positioned a little low, but the connections look good. If I press the multimeter into the spots above Q3 and R5 on the back the light comes on, so I think there's connectivity through the board.

On both top boards the voltage at either side of R5 is 3.52V, this extends all the way to the positive side of the LED. If R5 is a resistor, should the voltage drop here? The R10 on both is 1.5K ohms.

One thing I did note is that when putting the red probe on the negative side of the LED (and the black at the USB ground) gives me 1.91V on one and 1.95V on the other (the ebay one is more of a yellow). So there appears to be a voltage drop across the LED, but no light (this is both with and without a bottom board). So maybe the amperage isn't there or these LEDS don't want to play nicely with the voltages.

The bottom board worked when I tried the blue LED in controller A on it and I've been careful with it since. If I had other controllers to work with I could do substitution. So you think both of these top boards should work with either revision bottom board? I might have to buy a used one.

If you have any other ideas please let me know. You've be extraordinary helpful. Thanks again
« Last Edit: March 03, 2016, 03:09:58 PM by Umbra »

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #124 on: March 03, 2016, 03:51:37 PM »
If your meter is making the LEDs light up then the board can do it.

The Resistor is in there as a current limiter, and you're not going to measure any voltage drop at it on an open circuit.

That 1.91v and 1.95v is the measurement from both boards? If so then that sounds right, as you will measure a drop there because it's a semiconductor.

Try this, just the top board again, plug up the USB cable and flip it over, take some piece of wire, few inches or so, and jumper the bottom of R5, to pin 2 of J2, the left side connector. Pin 2 is at the bottom right hand corner of J2 and is marked with a 2. You don't have to stick the wire inside the connector, and I don't recommend that in case something shorts, you can touch the lead on the board where it's soldered to the left of the 2 marking.

What should happen is the LED comes on, if it doesn't turn on, then Q3 is bad. That will test 100% of the LED circuit from the control line to the power supply for it, which we already know is good at 3.52v, so if the LED comes on, then the bottom board might have some issue, or, there could be some other power issue up on that top board, as it's doing quite a few different things.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline Umbra

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #125 on: March 03, 2016, 04:48:37 PM »
If your meter is making the LEDs light up then the board can do it.

The Resistor is in there as a current limiter, and you're not going to measure any voltage drop at it on an open circuit.

That 1.91v and 1.95v is the measurement from both boards? If so then that sounds right, as you will measure a drop there because it's a semiconductor.

Try this, just the top board again, plug up the USB cable and flip it over, take some piece of wire, few inches or so, and jumper the bottom of R5, to pin 2 of J2, the left side connector. Pin 2 is at the bottom right hand corner of J2 and is marked with a 2. You don't have to stick the wire inside the connector, and I don't recommend that in case something shorts, you can touch the lead on the board where it's soldered to the left of the 2 marking.

What should happen is the LED comes on, if it doesn't turn on, then Q3 is bad. That will test 100% of the LED circuit from the control line to the power supply for it, which we already know is good at 3.52v, so if the LED comes on, then the bottom board might have some issue, or, there could be some other power issue up on that top board, as it's doing quite a few different things.

Thanks for the information.

The 1.91V is for one board and 1.95V is for the other. I attribute this to the different color of the LEDs. I don't recall which reading was for which, if it matters I can find out.

Is pin 2 on the top or bottom row. It sounds like it's the bottom, but I don't see a 2. Is this the connection you'd like me to establish?



I made this connection, from the bottom right pin on J2 to the lower connection on R5 and on both controllers the LED lit up.

It sounds like we're running out of things to test on the top boards and it seems unlikely (though not impossible) that both top boards are bad.

I've looked over the bottom board and nothing looks out of place, but some of the SMD components are remarkably tiny. Maybe with all the plugging and unplugging a tiny crack developed in the solder on one of the pins securing the connectors to the board (J5 and J6?). I could try reflowing that. My other thought was maybe something came loose on the opposite side when I used my fingers to press the boards together. One connector has nothing on opposite side, but J6 has a ton going on on the other side.

On a somewhat related note: if I get another controller to substitute parts with - do you know if I could use a board from a wired controller?

I can't thank you enough for helping me. I had no idea anyone has this much insight into these boards, let alone would be willing to help.



« Last Edit: March 03, 2016, 04:53:15 PM by Umbra »

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #126 on: March 03, 2016, 05:09:59 PM »
Yeah that's pin 2, I was looking on the board I have here versus your pics, but that's it.

That mess under JJ6 on the other side of the MCU board is the RF module. It's power would have to be shorted out for it to keep the controller from turning on, and you'd almost have to do that on purpose as removing any of the parts on it isn't going to do that.

I''s possible that some of the joints on J1, J2, J5 or J6 could ahve come loose, but you should be able to see that with a magnifying glass and moving them around a bit.

So that pretty much confirms everything in the LED circuit and the LED is good. So after a look at the connectors, if you see nothing iffy, lets plug the boards together and start poking around.

Plug the USB cable back up and check the voltages on TP7 and TP9. They are both on the right hand side, looking at the bottom of the MCU board.

There are no official wired XB1 controllers. The wireless one works as a wired with the USB cable, but so far that's it.  Any XB1 wired controller will be 3rd party and for sure not have anything even remotely compatible inside of it.
« Last Edit: March 03, 2016, 05:13:11 PM by RDC »
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline Umbra

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #127 on: March 03, 2016, 06:24:23 PM »
Yeah that's pin 2, I was looking on the board I have here versus your pics, but that's it.

That mess under JJ6 on the other side of the MCU board is the RF module. It's power would have to be shorted out for it to keep the controller from turning on, and you'd almost have to do that on purpose as removing any of the parts on it isn't going to do that.

I''s possible that some of the joints on J1, J2, J5 or J6 could ahve come loose, but you should be able to see that with a magnifying glass and moving them around a bit.

So that pretty much confirms everything in the LED circuit and the LED is good. So after a look at the connectors, if you see nothing iffy, lets plug the boards together and start poking around.

Plug the USB cable back up and check the voltages on TP7 and TP9. They are both on the right hand side, looking at the bottom of the MCU board.

There are no official wired XB1 controllers. The wireless one works as a wired with the USB cable, but so far that's it.  Any XB1 wired controller will be 3rd party and for sure not have anything even remotely compatible inside of it.

I tried both boards on the MCU board and TP7 reads ~3.53V and TP9 reads ~3.34V (give or take 0.01V).

I went ran some liquid flux over the board connections for the pins and ran a flat tipped soldering iron over them. While the iron was hot I also desoldered the rumble motors and what I think are the RB/LB buttons. I don't think removing those would matter for testing. I pushed the boards together and nada. Then I put the rubber/carbon button over it and pressed and to my great surprise I got a blinking orange light! Swapped the board and got a blinking yellow-orange light! I think that was it. I couldn't see any cracks, but it was either that or the removal of one of those wired components and those hadn't been tampered with yet.

Thank you so very much for helping me through that! Can I paypal you a few bucks for your trouble?

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #128 on: March 03, 2016, 06:47:59 PM »
Have you not been pressing the Guide button after plugging it up? I mentioned that quite awhile back. ;) Either way, long as they work now.

The LB/RB are mounted to the Top board. Those Gray and Black wires are for the Trigger Rumble motors.

Thanks but no worries, I was here working on some FPC design anyway, and your pics made up for it, versus some that I get for troubleshooting, which could be a PCB or Nessy for all I can make out.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline Umbra

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #129 on: March 03, 2016, 07:27:02 PM »
Have you not been pressing the Guide button after plugging it up? I mentioned that quite awhile back. ;) Either way, long as they work now.

The LB/RB are mounted to the Top board. Those Gray and Black wires are for the Trigger Rumble motors.

Thanks but no worries, I was here working on some FPC design anyway, and your pics made up for it, versus some that I get for troubleshooting, which could be a PCB or Nessy for all I can make out.

Yes, I had been pressing the guide button (or in the case of the butchered top board -  the wires I soldered in when it looked like continuity was being severed on the lower carbon pad). I sometimes overlook stupid liitle things like that and kick myself later, but luckily not this time.

Thanks so much for all your help! You saved me from having to buy and dismantle another controller for a console I don't have.  :#1:




Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #130 on: March 20, 2016, 03:03:48 PM »
This should have been posted in the DS4 section.

Not really getting the exact issue you're having there, it doesn't work, then you do it again and it does? Then test it while it is working so you can see what's going on there. If you mean with your wire on there it doesn't work, then removing the wire and it does, then whatever the other end of your wire is connected to is causing it, as a wire alone will not do that, or your solder joint could be hitting something else, or so big it's wrecking the connection to the Daughter board when it's in place, the FPC with the button contacts on it, as it needs to be connected for the Triggers to work correctly.

Trace the daughter board out and you'll see some of what's going on there with the Triggers, but you'll need to scope that line to really see it. Then trace back from that pad 18 on the motherboard and solder there instead. Every version of controller has a better place to use for making a test spot than that carbon pad.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline reeso3000

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #131 on: April 11, 2016, 03:44:13 PM »
Hi Rdc,
 Firstly great job and thanks for the time and effort you put in creating this thread! Secondly - I have a 1697 controller that will not power on  - the light will pulse once when a usb cable is attached(sometimes) but there is no life with the controller. From the info you have provided I have checked various voltages while the usb was attached as seen below:-
 

U1

Left to right with usb socket at top of board
Pin
1 - 0v                  5 - 0.58v
2 - 0.62v             6 - 0v
3 - 4.95v             7 - 0.09v
4 - 0v                  8 - 4.65v

U2

Top to bottom left side first usb socket at top of board

1 - 2.95v               5 - 2.88v
2 - 3.53v               6 - 1.22v
3 - 2.96v               7 - 0v
4 - 0v                    8 - 0v

U3

Top to bottom left side first usb socket at top of board

1 - 0.55v              5 - 0.01v
2 - 0.16v              6 - 0.03v
3 - 0.56v              7 - 0v
4 - 0v                   8 - 0v


TP6,TP8,TP9,TP11 - 0v TP5- 5v

J1
8 - 5v

J2
1 - 2.97V
4 - 0.21V

Just wondering if you could shed any sort of light on where the issue might lie - It appears there is no 3.3v getting through to the secondary board. Wondering if u3 might be the culprit but I would imagine it would at least receive a voltage on one of the pins which doesn't seem to be the case. My electronics knowledge is lacking but im ok with soldering. if you (or anyone else!) has any suggestions I would really appreciate it. Thanks in advance!!
« Last Edit: April 11, 2016, 04:04:45 PM by reeso3000 »

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #132 on: April 11, 2016, 10:22:14 PM »
Right off I see there is no 5v on pin 2 of U1, but you do have it on pin 8, this is assuming you have those pin numbers all correct?

The dot on the board at the corner indicates pin 1, then it goes..
Code: [Select]
O < dot on PCB
  1 --    -- 8
  2 --    -- 7
  3 --    -- 6
  4 --    -- 5

Regardless though, pin 2 should have 5v on it (~4.6v) when the microUSB cable is connected, so that suggests an issue with R1 or D23, then worst case U1.

Since you have no voltage on pin 2, first check R1 and make sure it is 0ohms. Then D23 could have been knocked off the board or opened up internally, but that's where I'd start.
« Last Edit: April 12, 2016, 06:19:05 AM by RDC »
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline reeso3000

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #133 on: April 12, 2016, 01:44:33 PM »
Hi Rdc,

Sorry I didn't realise the white dot indicated pin 1 so the correct voltages are are as follows: -


U1
Pin 1- 0v             Pin 5 - 4.65
Pin 2 - 4.88v       Pin 6 - 0.09
Pin 3 - 0.60v       Pin 7 - 0
Pin 4 - 0v            Pin 8 - 0.57

D23 5.07v and 4.88v

U2

Pin 1 - 0.53    Pin 5 - 0.53
Pin 2 - 0.73    Pin 6 - 0.25
Pin 3 - 0.50    Pin 7 - 0
Pin 4 - 0         Pin 8 - 0

U3

Pin 1 - 0.56    Pin 5 - 0.01
Pin 2 - 0.12    Pin 6 - 0.03
Pin 3 - 0.56    Pin 7 - 0
Pin 4 - 0         Pin 8 - 0

For some reason having re-checked I am no longer get 3v on pin 1 J2 (0.56) and also the reading on u2 have all dropped well below 3v

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #134 on: April 12, 2016, 02:24:21 PM »
That still looks wrong, pin 2 and pin 8 should have the 5v on them, not pin 2 and pin 5. The pin numbers go in a U shape, not in rows.

O < dot
  1               8
  2               7
  3               6
  4 ->->->- 5



U1 should measure more like this..

1 - 4.3v          8 - 4.3v
2 - ~4.6v        7 - 0.6v
3 - 3v             6 - 0v
4 - 0v             5 - 3v


Post a few pics of your board if you can.

You for sure have something up with that U1 circuit, as pin 1 is PG (Power Good) and it's not on, nor is there any 3v out of it anywhere, so it's not working, or it's something after that shorted, but we need to start there first.

Is the L2 component on your board? It should be on the other side of the PCB right under U1. There is one close to each of those ICs, U1 has L2, U2 has L3 and U3 has L4. I've seen L2 get knocked off the board before and if it's missing that would cause the incorrect voltage readings you're getting there.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline reeso3000

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #135 on: April 12, 2016, 03:00:32 PM »
Hi RDC,
 Please see pictures attached. The L2,L3 and L4 components are all present. I am embarrassed to say I read your first post on my phone and missed the part where you show the pins going in a U! My apologies, the correct reading for U1 is:-

Pin 1 - 0
Pin 2 - 4.88
Pin 3 - 0.60
Pin 4 - 0
Pin 5 - 0.57
Pin 6 - 0
Pin 7 - 0
Pin 8 - 4.65





Thanks

Ps The purple wire you see in the pictures is just a wire I soldered to ground to make testing easier
« Last Edit: April 12, 2016, 03:12:56 PM by reeso3000 »

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #136 on: April 12, 2016, 03:32:09 PM »
No biggie, just need to get on the same page is all.

Unplug the microUSB cable and measure the Resistance at U1 from pin 3 to pin 5. Never have any power in there when checking Resistance.

Also measure the Resistance from pin 5 of U1 to Ground. Put the Black lead on pin 5 an Red lead on Ground.


Where exactly is that wire soldered to for your ground? Pin 14 of J2? If so that's fine.

Also, be aware that this board will not power up on it's own, it must be connected tot he MCU board for the Guide button to turn it on, but you don't need it connected for any of these tests so far.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline reeso3000

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #137 on: April 12, 2016, 04:08:35 PM »
Thanks for the prompt responses
The readings I got are:-

3 to 5 -  0
5 to ground - 000.7/8

As for the purple wire it is indeed soldered to 14

Thanks


Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #138 on: April 12, 2016, 04:58:26 PM »
Alright, that pin 3 to pin 5 measurement is good. But, that pin 5 to Ground measurement is far too low, it should be around 2k, so something is most likely shorted on that rail. That's the good news, the bad news is that it goes to a lot of places, so there are many, many, many things that it could be, including even U1 as the cause, so....

I'm not seeing anything from your pics that stands out as a cause for that, so now you're left with having to pull component after component that is connected to pin 5 of U1 until you find the offending one, and even then you may not as it could be U1 that's gone and shorted.

Pin 5 of U1 is the output, there should be 3v on there when the microUSB cable is connected, but you have a short somewhere on that rail killing that. Now that could be..

U1, U2 or U3 causing it, as well as 6 or 7 different capacitors, a solder bridge or internal damage between pins 3 and 4 of U1 or between 5 and 6 of U1, a solder bridge or internal damage between pins 3 and 4 of U2, a solder bridge or internal damage between pins 3 and 4 of U3, a solder bridge between pins 1 and 3 of J2, and those are just things that are right on the rail. The more I'm going to look into this the more I'll find that would need to be checked.


What is a little perplexing is that you said you had 3v on pin 1 of J2 at some point, then at that point U1 was working correctly as pin 5 of U1 (which didn't measure 3v at that time) is connected to pin 1 of J2, that did measure 3v, which either means something was tested wrong there or something is intermittent on that board and it has finally decided to stay in the 'bad' spot.

That's an interesting one for sure, and if you're not up for laying into it, then PM me, because I for sure will. I was able to pretty much duplicate your board measurements putting a short on that rail to ground, but that does nothing for narrowing it down there, it's kind of a mole hunt now.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline reeso3000

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #139 on: April 12, 2016, 05:32:56 PM »

Thank you again for your prompt replies and details on tests I can try. As I mentioned previously u2 was getting too hot to touch so I thought I would start there. Anyway by removing u2 I now have the following voltages -

U1

Pin 1 - 4.76
pin 2 - 5.07
Pin 3 - 2.98
Pin 4 - 0
Pin 5 - 2.98
Pin 6 - 0
Pin 7 - 0
Pin 8 -4.63

U2 (Pads the pins were attached to)
Pin 1- 2.98
Pin 2 - 0
Pin 3 - 2.98
Pin 4 - 0
Pin 5 - 2.98
Pin 6 - 0
Pin 7 - 0
Pin 8 - 0


U3

Pin 1 - 2.96
Pin 2 - 0
Pin 3 - 2.98
Pin 4 - 0
pin 5 - 0
Pin 6 - 0
Pin 7 - 0
Pin 8 - 0

Also now have 3v on pin 1 of J2

Does this point to it being u2 as the fault or could I have just taken the faulty component out of the circuit if its attached to u2? If it is u2 do you know if these dc converters can be purchased anywhere?

Thanks again :D

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #140 on: April 12, 2016, 06:01:16 PM »
You never mentioned anywhere that U2 or anything at all was getting hot before now, or I'd have suggested removing that awhile ago.

That could mean that U2 is bad, or that something on it's output rail is bad.

The closest thing I've found to a replacement for U2/U3 is the PM3414DJ by MPS, and you can get them from DigiKey, Mouser or the like.
« Last Edit: April 12, 2016, 06:09:51 PM by RDC »
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline reeso3000

  • Guppy
  • *
  • Posts: 6
  • Post quality +0/-0
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #141 on: April 13, 2016, 01:28:50 AM »
Hi Rdc, I only noticed it last night while doing the extra checks apologies once again for any time of yours I wasted! I thought I had edited and added a post or 2 ago. Found the chip you mentioned so thank you for that.

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #142 on: April 13, 2016, 02:22:19 AM »
Welcome. Not wasted perse, but it would have sped things up some as we could have started there, but it's good to know that everything before there works now, as that can come back and get you later on anyway, not checking things that you think are good.
« Last Edit: April 13, 2016, 02:23:16 AM by RDC »
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline MDave

  • Guppy
  • *
  • Posts: 5
  • Post quality +0/-0
  • Gender: Male
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #143 on: April 19, 2016, 10:10:04 AM »
Hello! On my quest to search for Xbox One analog module component specifications (specifically to figure out the resistor value use) I came across this board.

Now the mod I am trying to do is not for an Xbox one controller, but I do use the Alps analog modules to replace some cheap lesser quality ones in another kind of controller entirely. I got it to work, but I figure the ohms or something in the resistor values are different and that's why I get full value sensitivity from pushing the analog stick even just a little bit.

I did a youtube video of the controller mod and to show it's results (before and after): https://www.youtube.com/watch?v=_rVVGzepVck

If anyone could guide me on the right track to fixing the sensitivity, that would be great! I apologise, I am very new to electronic engineering and I don't know all the correct terminologies yet.

I think I understand that I would have to replace the tiny surface mount resistors (two of them) that are near the analog stick modules on the board to ones that match the Xbox One's resistors?

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #144 on: April 19, 2016, 12:16:41 PM »
Strap in, this ended up a bit longer than I figured...

They are just using the motors to give the controller some heft, so it doesn't feel like the complete POS that it really is, and most likely they're using only 1 now to make it even cheaper for them to make them. Less parts, less time spent making them, less weight for shipping, that all adds up. 

The 5v 500mAh rating on the back is for when it is connected with a USB cable, hence the 'Input' before the DC5V 500mAh spec, that's not the battery rating.

The XB1 controllers have no Resistors in the Stick circuits, and the Capacitor values in there will make little to no difference at all.

The value of the Stick makes 0 difference. They are all typically 10k, but if they are 1k, 10k or 100k they would work exactly the same as they are just used as voltage dividers, I'll get to that in a bit.

Without seeing the guts of the thing it's hard to tell what your options are, but the actual Stick doesn't account for as much of the overall performance of the controller as much as the way that it has been coded to work, and there is really jack all you can do about that. If that thing has any kind of calibration routine on startup or waits until you have moved the Sticks to their max ranges, then it becomes a real mess to get around, as you have to let it calibrate to the crap values, then switch up and input your own.

Being that it is too sensitive, the only thing you can try to do is add a couple of Resistors to each POT (Potentiometer) to kind of deaden it's sensitivity.

So a little background on how these things work. They are just a voltage divider. Imagine a scale from 0 to 10.

0-1-2-3-4-5-6-7-8-9-10

In the very middle is 5, so at dead center you would read 5. If you move more to the left, the numbers decrease, more to the right and the numbers increase.

The same thing happens with the Stick and it's Voltage on that center lead of the POT, as it's just a Variable Resistor.

Code: [Select]
GROUND = 0v
VOLTAGE = 10v

GROUND ---/\/\/\/\/\/\/\/\/\/\--- VOLTAGE
                   ^
                   |
                   |
                 WIPER

Sitting in the middle, you would measure 5v on the Wiper. The more you move it in one direction the close it gets to Ground, and the Wiper voltage goes down, then the more you move in the other direction, the closer it gets to Voltage and the Wiper voltage goes up. The value of the POT has no real bearing here, well technically it does as far as current is concerned, but you will measure the exact same voltage on a 10k or a 100k POT in the same positions as the difference between them is the same.

If it was a 10K POT, you' have 5k on each side. For a 100k POT, you'd have 50k on each side. No matter what it's the same, again only talking about voltage and that's what is measured there to tell where you have the Stick positioned.

http://www.raltron.com/cust/tools/voltage_divider.asp

The issue there is, it's not really using the full range of that Stick, and seems like only half or so is being used, so what you can try and do is to 'pad' the outside leads with some Resistance, so you then have to move the Stick farther to get the same voltage on the Wiper.

On the Stick, each POT has 3 leads. The Outside leads are power and ground, and these are the ones you need to have the Resistors on. They need to be between the Stick and PCB (Printed Circuit Board, the controller board) and without seeing the PCB I can't even guess at the best way for you to implement that as it could be dual layer or who knows what in there. But the Resistors must go between the Stick and the POT leads.

PCB to Resistor to POT (outside lead, power)

PCB to POT (center lead)

PCB to POT (outside lead, ground)

As far as the Resistor value goes, I'd start with 1k (1000ohms) as it's easy to get and will make enough difference to tell if it's doing anything, but not so much that it will be way off and not work. You would be best off using some high tolerance Resistors, 1% or lower, versus the standard 5% ones, as you could end up with some drift (stick moving on it's own while at rest) if the values between the Resistors used is too great. You ideally want them to be identical, so the lower the tolerance % the better.

What that does is change the way the Voltage Divider works, and comparing that to the number scale from before, it would look like this now.
Code: [Select]
          0-1-2-3-4-5-6-7-8-9-10               < how it was
0---1---2---3---4---5---6---7---8---9---10     < how it is with the added Resistors
At the very center you still have 5, but you have to move farther now to get to any other number.

With the Stick, the Voltage Divider now looks like this.

Code: [Select]
GROUND = 0v
VOLTAGE = 10v

GROUND ---/\/\---/\/\/\/\/\/\/\/\/\/\---/\/\--- VOLTAGE
                         ^
                         |
                         |
                       WIPER

At the center you still get 5v on the Wiper, but you have to move the Stick more now to get to the lower and higher voltages.

The catch with this is, you can never get to the very end of either side now, as the added Resistance you're putting in there is fixed. All that you'll be able to change is the value in the middle from moving the Stick. This doesn't really matter though as you're not needing it to be variable, you're just wanting to widen that range some so you can decrease the sensitivity.

Being how close the POTs of the XB1 Sticks sit against the board, you're not likely to get them installed without having to do some trace cutting on the PCB (Printed Circuit Board) and I'd for sure need to see both sides of the PCB, with the stick removed, before I could suggest the best way of going about that.

Now, the kick in the shorts, as I already mentioned, is if that thing is doing some kind of weird arse calibration on startup, or after you move the sticks around, then all it's going to do is calibrate off of the new values and then work exactly the same, or worse, than it already is. That's when you'd have to get far more technical to correct the issue. I'd only try this on 1 of the Stick's Axis before doing the entire thing.
« Last Edit: April 19, 2016, 12:22:19 PM by RDC »
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline MDave

  • Guppy
  • *
  • Posts: 5
  • Post quality +0/-0
  • Gender: Male
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #145 on: April 19, 2016, 01:26:48 PM »
Ahh thanks for the extremely informative reply! Glad to know how these things work now. I will answer as best I can:

Yes it seems like it gets calibrated after I tilt the sticks all the way on first push after powering it up / plugging the controller in. You might be able to see this in the video upon first push of the sticks, the axis red bar in Window's game controller setting shows it going to full sensitivity.

I took some shots of the unmodded one:






It looks like there is some pads on the left edge of the main board that I could possibly use, to bypass the ribbon cable connection and analog stick board traces ...

And yeah, it looks like its a simple single sided board. There is nothing underneath the analog stick module, no visible traces.

I will see about getting some 1k resistors and doing some sort of test. Resistors between the Stick and Pot leads, got it!

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #146 on: April 19, 2016, 01:57:07 PM »
That could just be Windows doing that, no way to exactly tell until it's modified, then tested again.

The PCB is dual sided. You can't see the power trace for one of the Axis on the Right Stick board, as it's on the other side under the Stick.

That won't be very difficult at all to get a couple Resistors in there, and the way they have that setup you could even do it with only using 2 Resistor per Stick board.


I don't see a voltage pad in that group, so you'll need to find that first with a DMM so you know what the Sticks are using for power.

Then pick a side and remove the FPC cable to the Stick board and the main board, set it aside as you'll not be needing it for this.

Install the first 1k from the voltage pad/source to one of the POTs outside leads. It doesn't matter which POT you choose, as they are connected on that board, but you must make sure that it is the Voltage side, which is the outside lead that has the thin trace connected to it.

The other outside lead is the Ground side and is connected directly to the ground plane, no thin traces going to it, and again either one can be used. That side of the POT you need to connect to the GND pad on the main board thru the second 1k Resistor.

The _ from here on is for L or R depending on which side you choose to test on.

The _X and _Y will each go to the center lead of the POTs. The _X is for left/right, _Y is for up/down.

The _T pad will go to that _2 Tact there on the bottom of the board, just follow the trace from the connector to it and use that solder joint.

The _3 pad will go to the Stick button lead, again just follow the trace from the connector to the solder joint and use it.

The PWR_RESET pad goes to the RESET Tact (if you decide to test on that board)

All total, you'll have 4 wires (5 if the Reset side is used) and 2 Resistors connecting the Stick board to the main board instead of the Flex cable.

This will make the buttons also run thru that 1k to ground when pressed, but it's a common ground setup and that value should not really mess with them.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline MDave

  • Guppy
  • *
  • Posts: 5
  • Post quality +0/-0
  • Gender: Male
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #147 on: April 19, 2016, 02:23:30 PM »
Thank you very much for the corrections and detailed instructions! I suspect that the top pin on the flex cable (above the PWR_RESET pin) is GND and the bottom pin (below the _T pin) is the voltage/power so I will follow that trace on the underside of the main board.

I need to get hold of a DMM and resistors before I proceed then. Guessing these would do? 1/2W 0.5W Metal Film Resistor 1% Tolerance 1 Ohm

Offline RDC

  • Acid Modder
  • *
  • Posts: 748
  • Post quality +78/-2
  • Gender: Male
  • The CGnome Project
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #148 on: April 19, 2016, 03:25:21 PM »
1/2W would work, but are kind of large. You only need 1/4W at most, and even 1/8W will do here, but the 1/2W will work just the same.
Screwing up is one of the best learning tools, so long as the only thing you're not learning is how to screw up.

Offline MDave

  • Guppy
  • *
  • Posts: 5
  • Post quality +0/-0
  • Gender: Male
  • Acidmods User
Re: XB1 Controller PCB Scans, Traces and Info - 1537
« Reply #149 on: April 23, 2016, 09:50:32 AM »
Alright progress update! Got the resistors through the post, and I managed to test with the resistors, and upon testing ... it's even more sensitive! However, if I detach the resistors and connect them back, all while the controller is still connected, it does the trick of lowering the sensitivity almost perfectly. It doesn't quite reach full sensitivity when pushed all the way (about 90%), but its miles better then before. It centres perfectly, and no stick drift. I might be mistaken on it being more sensitive than before upon start up with the added resistors, but it looks like it does do the auto calibration you mentioned. Ahh well, not much I can do about that I think. It was a fun project seeing how far I can go modding this controller, but I think I have hit a dead end. Thanks for all the help though!

Unless I somehow mod in some switches that unplugs and reconnects the resistor leads, hah. A bit too extreme for something like this though.