[Solved] GMMK Foam mod attempt. Wrecked PCB I'm afraid.

[problemxyz]

I tried the foam mod on a full size GMMK. After reassembly the PCB seems dead. Nothing happens anymore when plugging in the usb cable. No Led light, or registring keystrokes from the switches.

Each time I plug in the USB cable however, I do however get an error message saying

'unplug the accessory using too much power to re-enable USB devices'. I searched this error and seems to be meaning a short most of the time.

What I did notice is the foam was a little thick (2mm ...)and there was pressure on the plate when I tried to bring it back together. I can't imagine anything else being the cause really.  I was trying to be really carefull when disassabling and disconnecting the PCB, like anyone would really when you watch teardowns.

Visually there is really nothing that looks like a connection got compromised in any way. Except for the pins of the usb connector on the front side of the PCB. Two of them seemed to have hardly any solder around them, looking like this.



And I did try to add some solder on to these pin. Wasn't probably the smartest move. Ik feel like I made it works. It definitely didn't end up looking any better.

Here's a picture of that



Not sure where to go from here. I don't posess the technical knowledge to know which connections to trace with a multimeter. Sure, I understand what the beep means on this particular setting about the resistance. But which of the connections are supposed to connect and which not?

So yeah I do have the feeling that I simply wrecked the pcb beyond repair. Not beyond repair per se, but beyond my personal capacity to repair. Or is there something where someone non-technical like myself can still take a stab at?

FYI, My computer is a 2015 macbook pro running Catalina

[Leslieann]

If it's saying it's drawing too much power, the pins are clearly working, it's probably they are soldered from the other side.

Sounds like a dead short, did you test outside the case?

Before grabbing a multimeter you need to do a real good visual check.
A dead short should can often be relatively easy to find once you start looking for it, metal shaving between controller pins, screw poking the wrong place, RGB knocked loose... Also check all the hot swap sockets and make sure none were pushed off the pcb, one may have come off and is now grounding the RGB power into the controller.

Do the above first and let us know what you find or don't find, from there full pics of the board may be needed, possibly both sides.

Oh, one last thing, the GMMK uses bending tabs to lock parts together, make sure none of those tabs are grounding out where they should not. You may be inclined to not bend them back so as to not stress it and make it easier to disassemble next time but they could be shorting something. Like I said, test outside the case and see what happens. Good luck.

[Tactile]

Not all foam is created equal and some can be electrically conductive, especially if it's formulated using graphite (Is your foam black?) or designed to protect electronic components from static electricity (Here, again, the foam used is electrically conductive). Perhaps the damage isn't permanent. Have you tried it after assembling without foam ?

[problemxyz]

Really awesome great useful feedback I appreciate it.

did you test outside the case?

I did exactly that but forgot to mention in it in the starting post. I took the PCB out of the case entirely and tested it, same problem persists.

Before grabbing a multimeter you need to do a real good visual check.
A dead short should can often be relatively easy to find once you start looking for it, metal shaving between controller pins, screw poking the wrong place, RGB knocked loose... Also check all the hot swap sockets and make sure none were pushed off the pcb, one may have come off and is now grounding the RGB power into the controller.

Yes! Now I know what exactly to look for. Been scanning the board visually for the things you mentioned. I must say everything really still looks so darn unaffected.

bending tabs

I didnt bend those back but that is only because I didn't reassemble the board. And tested the PCB while disassembled and that is the current state as well.

I need to be post pictures now which I will do ASAP.

Cheers
Kevin

[problemxyz]

I tried to get some detailed shots happening with my smartphone. Had to fight a lot glare if the pcb was lit properly, that's why their relatively dark because I had to direct the light away from it.

https://imgur.com/a/Vvjr5Sv

Halfway I switched to macro thinking that would make it better but Looks like i needed to get even closer. Which obviously made me end up making countless pictures of the PCB
Maybe daylight would have been better. And happy to make a new batch later.
But curious to see if this will already get us anywhere.

[Leslieann]

Your problems are going to be between that cable connection and the controller, the rest is most likely irrelevant.
Get better pics of those areas.

[problemxyz]

Here is some more detailed pictures of the pin connector for the usb cable:






https://imgur.com/a/g8gumAP

I should also add that I tested the usb cable with multimeter probes. All of the pins had a corresponding spot in the male USB connector that beeps. Sorry for being an electronics retard.

[Leslieann]

Stop looking at the connector and start looking at or near the controller.
Your problem is between the connector and the controller.

[yui]

some flux are conductive, maybe try cleaning your flux residues around the connector too.

[problemxyz]

Stop looking at the connector and start looking at or near the controller.
Your problem is between the connector and the controller.

Thanks for clearing that out. I saw the controller and connector as being in the same location for some reason. Ill look up what a controller really is and where it is on the board, check whats up in my case and report back.

some flux are conductive, maybe try cleaning your flux residues around the connector too.

The problem was there before the flux was in the state where they are now in the pictures. So the flux couldn't have caused the issue. On the other hand, it probably wont hurt to clean it up. For the record it's the board's own flux that started moving when I wanted to resolder the pins as stated earlier in the topic.

[yui]

Thanks for clearing that out. I saw the controller and connector as being in the same location for some reason. Ill look up what a controller really is and where it is on the board, check whats up in my case and report back.

the controller should be a black plastic square device, larger than the LEDs but very likely smaller than your thumb, with lasered in markings
(photo from anandtech tear-down https://images.anandtech.com/doci/10665/GMMKBRN_11.jpg)

[problemxyz]

Thank you very much yui

I found the controller:

Im probably way out of my depth here. Especially considering, as Leslieann stated I need to look for a problem between the connector and the controller.
So, what do I even look for basically. I don't see anything broken or blown in this area. It looks "normal" to me.

Here is the area between the USB pins and the controller. Fwiw, if worth anything.

[Leslieann]

So, what do I even look for basically. I don't see anything broken or blown in this area. It looks "normal" to me.

Sorry I should have specified a bit more.


What to look for (you will probably want a magnifying glass):
Damaged parts
Pins on the controller for anything shorting between them, same for traces
It's not likely to be something missing, it's going to be something bridging something it shouldn't so look for things not aligned.

Other things,
Did this ever actuakll work and if do did you try updating the firmware before this happened? Did you change the cable or mess with the cable or extension in any way? Not all cables are equal. Make sure you're connecting the extension properly and not reversed. If you messed with the cables you may have swapped some wires.


One last thing, while the computer can see something is pulling too much power it may not be this that's pulling too much, it may just be the straw that broke the camel's back. Try another computer or completely different port, if using a hub or cable try connecting direct, if connecting direct try a different group of ports or try front vs rear.  Maybe you have something else plugged in or your power supply is dying. It could also be an update by Apple causing problems, Apple has a history of messing up USB. You may want to try this before you keep looking at the board which may be completely fine.

[problemxyz]

Make sure you're connecting the extension properly and not reversed.

A little embarrassing, but bingo.

Crazy.
Now the USB ports or the computer or anything they don't matter. The PCB just fires up beautifully.

Thank you for sticking with me all the way through. It is very much appreciated.

[Leslieann]

A little embarrassing, but bingo.

Thank you for sticking with me all the way through. It is very much appreciated.

You're welcome, I'm just sorry it took so long to figure out.

If it makes you feel better, anyone who's rewired a usb cable has probably done similar, I certainly have on more than one occasion.

[problemxyz]

If it makes you feel better, anyone who's rewired a usb cable has probably done similar, I certainly have on more than one occasion.

Yes!
Overlooking something silly can still mean you're giving it a lot can't it :-)

[yui]

good thing that usb has short protection and that you are using a recent enough mac. some generations did forgot the protection, with the result being dead motherboard in the case of a short on usb

[problemxyz]

I can't even image wrecking my computer because of only such a thing like reversing a usb pin connector on a pcb  .
So those protections were there before, then forgotten, then came back ... ?

[Leslieann]

Most USB devices lack the power to do much system damage, especially a flipped wire like this
Flipping the 5v power wires is not an issue most of the time, it's when you flip a power and data wire that things go south really fast.

That's one of the great things about older USB, it was really simple and hard to do much damage.

[yui]

I can't even image wrecking my computer because of only such a thing like reversing a usb pin connector on a pcb  .
So those protections were there before, then forgotten, then came back ... ?

when i said forgotten i meant that apple thought they could not include the protection to save a few pennies and get away with it, did not work, so they went back to having the protections.

[problemxyz]

Long live old USB! :-)

As a special thank you for the help, here is a picture of how the GMMK is back from ultimately not having been dead ever. But having been treated as a bit of a reject nonetheless.
Speaking of ugly ducks, behold the its latest iteration with its hideous native chrome rim and knockoff carbon PBT keycaps. The Box Jades don't really make the board more approachable either...

[Leslieann]

Long live old USB! :-)

As a special thank you for the help, here is a picture of how the GMMK is back from ultimately not having been dead ever. But having been treated as a bit of a reject nonetheless.
Speaking of ugly ducks, behold the its latest iteration with its hideous native chrome rim and knockoff carbon PBT keycaps. The Box Jades don't really make the board more approachable either...

Looks and switches are fixable, meanwhile it still works, be it for primary or use as a spare.
It also served as a learning experience and one day may save a much more valuable board.

I'd say it's a win.

[problemxyz]

Win for sure.
Ugly duck but a fun ugly duck :-D