Friday, June 22, 2012

Fixing a PS3 Controller that Wont Charge

Recently I found a cheap used PS3 at a local Game and Trade that was closing its doors. It's the older "fat" 80GB model. Everything was in pretty good condition except for the controller. It was obviously the store demo controller and thus was unbelievably disgusting. I had to handle it with gloves on. I entirely disassembled it and disinfected every part of it. There were a few things wrong with it:

1) I never eat or drink while playing for good reason (there was dried soda and chip fragments inside) along with mass amounts of gunk amassed from every hand that touched it.
2) There was a screw missing and the bottom tab snapped off so it was obviously opened before.
3) The analog nubs were missing their rubber tops.
4) I didn't realize it until the controller had low battery, but it refuses to charge.

I easily remedied the first two with a thorough cleaning and a screw from my screw collection. The third will be repaired with some help from my friend EBAY. Finally the last one was a little tricky. I think the charging issue might be due to some liquid damage. I measure the voltage when plugged into a USB adapter and noticed that instead of a nice 5V, I was getting a unstable 2V. This explains why the PS3 complains about too much current being drawn when the controller is plugged in. I figured I had nothing to lose so I went about trying to fix this problem.

Ultimately I figured that it would be far easier and cheaper to add in my own charging circuitry than track down the short circuit culprit and replace parts I likely did not have on hand. I started with a chip I am very familiar with and had in my parts box, the max1555. This tiny five pin smt chip handles safely charging the lithium ion battery and even has an open collector output that is on when charging and off when done. It accepts USB 5V or DC up to 7V or so. I etched my own board and decided to make things difficult for my self by making it tiny (it ended up being not much larger than the sot23 package).

I simply opted for a quick and dirty approach by covering a single sided copper clad pcb with permanent marker and scrapped the areas I wanted etched with a scalpel.

Size comparison  to a US quarter.
Man its hard taking pictures of something so small.
All etched and ready to have marker removed with the help of some acetone.


Close ups after soldering in the components with a penny as a reference for size.

I proceeded to test the circuit by hooking her up to a lithium ion battery and measuring voltage and current. Crossed my fingers and held my breath just waiting for a small poof of failure which luckily never came.
Just shy of 4V, looks good.
Charging at around 100mA looks safe.

Finally I needed to desolder (used a hot air gun, Warning: HOT) the USB female jack from the controller board and lift the V+ power pin so it would no longer power the broken internal charge circuitry and instead hooked into my own charger. I had to be quick but gentle with the hot air so I wouldn't damage anything.
 Pulled socket and used desoldering braid and some alcohol to clean up a bit.
 Cleaned up the socket as well and lifted the V+ pin.
Replaced and soldered down.
Now all I needed to do was solder three wires: ground, V+ in, and battery+.
 Battery+ is orange and ground is brown. I used the gnd pad conveniently silk screened for me and plugged the battery in to find the battery+ (red wire on the battery).
 V+ in from USB is white and soldered to the lifted pin
Quick test to make sure everything still works. No magic smoke released here.
 Taping wires so they wont interfere with the battery. I used some more tape to attach the charging board to the front under the SONY emblem.
 Battery inserted.
 All put back together and charging status led visible as an orange glow on the front.

Outcome: Success! Now I just need to replace the nubs and I'll have a good as new controller without having to shell out another fifty or sixty bucks. I already had all the parts so it cost nothing and only took around 2 hours in all. When not charging, the controller looks stock. I feel accomplished and saved a controller from being scrapped. Now I can get back to gaming once it charges up.

15 comments:

  1. Great job! Is the controller still able to work over usb?

    ReplyDelete
    Replies
    1. I left the data and ground connections but rerouted the +5V power so I'm not sure. I'll have to test.

      Delete
  2. damn it this post makes it harder to choose between hardware and software engineering

    ReplyDelete
    Replies
    1. I was faced with that same problem. So in the end I didn't chose one over the other: I chose both. That is why I opted for embedded microcontroller design. I get to fiddle with some hardware and design the software that runs on that hardware. For me it's a win-win situation.

      Delete
    2. may be i will go that way...but i like to become a game creator...and a hardware hacker...

      Delete
  3. Rerouting Vcc will render USB useless. The controller and PS3 won't detect any USB Signal without the Vcc (Vbus) line.

    See: http://electronics.stackexchange.com/questions/7256/usb-without-vcc

    ReplyDelete
    Replies
    1. I figured that might happen but found it odd that usb still worked before my modification even though the Vcc line was being pulled low by a short somewhere within the circuit. The only problem with my current setup which would prevent usb communication would be that the usb controller chip in the controller would not receive the +5V (necessary only for power) from the bus although the differential data lines and ground reference are still intact and operational. An option would be sourcing voltage locally (of course through a step up to feed +5V) from the controller to the usb controller chip, but that wont work well if there is a local short. I might try tracking down what is pulling the power line down to see if I can repair that. Only then can I patch Vcc back in like normal.

      Delete
  4. More important than being able to play with usb is if the controler is able to sync/resync with the machine, I mean, it's recogniced when the PS3 ask you to press the PS button. If the controller was able to connect via USB before changing the V+, just by adding a switch to re-route current from and to the old charging circuitry should be enough. Let me know and good work!

    ReplyDelete
    Replies
    1. The controller did indeed work when connected via USB before modification (it just didn't charge) so I think that I'll add a small jumper or switch to route Vcc the way it was wired originally just in case I need to resync or what not. I intend to eventually pick up another controller for multiplayer action, but for now this mod has kept this one out of the trash.

      Delete
    2. You're right. Nice and tidy work!

      Delete
  5. In picture 15 of 20 I’m measuring the two spots that you soldered and it gives me an output of 0.28 when measuring tool is at 20V, is that a correct voltage?

    ReplyDelete
    Replies
    1. Do you mean .28V? Is the battery connected when you are measuring? The lithium ion battery should have a voltage of 3-4.1V.

      Delete
  6. So what's the follow up on this does the controller work? ty

    ReplyDelete
    Replies
    1. The controller works when already paired but isn't recognized by the PS3 over usb in cases where I would need to pair or access diagnostic modes. This is simple to fix by reconnecting the lifted V+ pin from the usb socket back to the circuit board. I plan on adding a tiny switch to do this in case I ever need to repair the controller in the future. Other than that the battery now charges perfectly.

      Delete
    2. Hi,
      I'm about to follow your step concerning this fix but as the PS3 controller is not mine, I have to keep the possibility to pair the device. Do you think that i can keep the original V+ pin USB in place and solder the V+ from the fix on it ? Will it allow me to pair and charge the controller ?
      thank you very much
      Have a nice day

      Delete