Possible Cause For Dodgy Usb Connection?


DaveC

Internal Development
Joined
Mar 4, 2004
Messages
9,208
After looking at ED's inside GP2X pictures I noticed something. It appears that there are two big blobs of solder that look to be touching other components. If there is a short here it may be causing it to not work. Could this be the problem with these? I would imagine some of the others have this solder problem too.

gp2xsolder.jpg
 
I thought it was confirmed to be a firmware problem? Connecting my gp2x works perfectly under linux but doesnt work under windows so that might point to it being a firmware problem?
 
I don't have USB problems, though I thought the same while looking at the board.

The rest of the board seems pretty well-soldered.
 
I don't have USB problems, though I thought the same while looking at the board.

The rest of the board seems pretty well-soldered.


The solder blob appears to be touching those little capacitors. Maybe it doesn't cause it to fail completely but just makes it flakey.
 
Last edited by a moderator:
Looks like the work of a 10 year old

ugh0ka.jpg


I'm not expert though. Maybe they put on so much to make it very stable? :huh:
 
After looking at ED's inside GP2X pictures I noticed something. It appears that there are two big blobs of solder that look to be touching other components. If there is a short here it may be causing it to not work. Could this be the problem with these? I would imagine some of the others have this solder problem too.

Looking at the picture it doesn't look like the components that are being touched with the USB connector are actually receiving enough current (in any) to cause interference. After all, they're basically grounded to the connector, not the wires inside. And looking at a few USB connectors here it seems the resistance is enough to prevent any harm or data jumps.

If it works on one OS reliably and not on another, then its all software.
 
Last edited by a moderator:
After looking at ED's inside GP2X pictures I noticed something.  It appears that there are two big blobs of solder that look to be touching other components.  If there is a short here it may be causing it to not work.  Could this be the problem with these?  I would imagine some of the others have this solder problem too.

Looking at the picture it doesn't look like the components that are being touched with the USB connector are actually receiving enough current (in any) to cause interference. After all, they're basically grounded to the connector, not the wires inside. And looking at a few USB connectors here it seems the resistance is enough to prevent any harm or data jumps.

If it works on one OS reliably and not on another, then its all software.


If you look at it the blobs are making contact with 2 surface mount components. Also that usb metal thing is grounded. If those other components are not meant to be pulled to ground it could cause a problem. Or maybe it just drains more power as there is a short there. It is something to look at anyway, it doesn't look right.
 
Last edited by a moderator:
After looking at ED's inside GP2X pictures I noticed something.  It appears that there are two big blobs of solder that look to be touching other components.  If there is a short here it may be causing it to not work.   Could this be the problem with these?  I would imagine some of the others have this solder problem too.

Looking at the picture it doesn't look like the components that are being touched with the USB connector are actually receiving enough current (in any) to cause interference. After all, they're basically grounded to the connector, not the wires inside. And looking at a few USB connectors here it seems the resistance is enough to prevent any harm or data jumps.

If it works on one OS reliably and not on another, then its all software.


If you look at it the blobs are making contact with 2 surface mount components. Also that usb metal thing is grounded. If those other components are not meant to be pulled to ground it could cause a problem. Or maybe it just drains more power as there is a short there. It is something to look at anyway, it doesn't look right.

Maybe that's why some people get insanely short battery life while others get 3-4 hours...I'd really be interested to see inside another unit though (could be worse, could be nothing). This could just as well be a one time fluke and as ED says, it doesn't seem to affect his unit.
 
Last edited by a moderator:
I'm pretty sure most battery problems are the type of batteries.

Use special Photo batteries, and you should be fine.

If there was a short circuit somewhere, the machine wouldn't work properly, very unlikely that only battery consumption is affected.
 
Yeah, I have to agree with EvilDragon on this one.. I solder all the time, and most of the boards I have to work on have to be laid out EXACTLY so all the circuits get to various components and all that. A crossed circuit or shorted circuit would either fizzle the whole unit or it just wouldn't work.
 
I dont think that shorts on the usb-chip would affect the rest of the unit.
It may realy be very possible that this could cause faults, i did read about a short on a ide conenctor wich coused the hdd to work wiredly, it could be formated but it wouldnt store anything and stuff alike.
But of course, that connection could also be perfecktly ok.
 
DijiTao referred me to this thread from my report of this problem. I must say that the pictures looks pretty similar to mine.

I think we need somebody who has a working USB port to verify two questions:
  • Is the anonymous capacitor to the left of CON2 connected to the shroud?
  • Is the right hand side of D2 connected to the shroud?
I am particularly suspicious of the second: if it's meant to be connected then why the hell isn't there tracking for it?!

EvilDragon and Winterkid, I would say that it's quite possible for a fault in one part of the board to stop signals getting through without causing a short-circuit, and that's definitely the symptom I'm seeing. This would be caused if the manufacturer isn't testing the USB port.
 
Last edited by a moderator:


Here is a shot of mine. It's a poor photo but you can just about make out the pieces of wire that have been used to aid joining the connector body to the two components. I think this really proves that it is intended for the connections to be made.

I can't get usb on mine to work. Just to see what would happen I removed the two pieces of wire so there was no connection between the connector body and the two components. Usb still didn't work.

I'm a little unsure of why the connector body would even need to be connected to anything. Every other device I have seen it has just been connected to ground. It's just connected to the shield of the usb cable. Usb should work fine even if the connector body isn't connected to anything.
 
Have none of you a multimeter ????

Sheesh, pull it out and test the continuity if you want to know if there's a solder bridge. If it looks flakey, you can always scrape the blob a little until you're happy with it.

I know that you shouldn't really be having to fix up solder jobs, but c'est la vie.
 
Oh, the solder bridge is there alright. The question is whether it's meant to be there! I'll have to figure out how to post some pictures...

It's intruiging that many people are seeing much the same thing around this connector. What we don't seem to have yet is any good statistics on how many USB ports actually just plain don't work. I'll test my machine against Linux and a modern Windows tomorrow.
 
Just took mine apart to look.

The bridges are there on mine too, it looks like its done on purpose too.

All the pads between the usb gnd and the bridged components are solder masked so theres no risk of a spurious connection.

I call red herring on this.
 
Well, I can confirm that the USB port in question is dead on arrival. Not the slightest peep from any USB diagnostic tool when plugged into any of
  • OSX 10.3
  • Windows 98 SE (well, I know that was a pretty worthless test, but at least it notices other devices trying to connect)
  • Windows XP Pro. If this shows nothing then I think we can be pretty confident it's dead
  • Linux (RH Enterprise 4).
 
My 2X arrived today. I can't get a squeak out of the USB. Tried every port and cable I could find. Even tried via an old USB1 powered hub I had hanging about. I've also been running the "USB Monitor" app, again, no sign of my new toy there either.

I'm using Win2000, SP4.

I also have this issue described elsewhere, that in my "settings -> USB" it always says "USB Connected" no matter whether I have the cable plugged in or not. Pressing X to disconnect has no visible effect. Is this just unique to people who are having 'major' USB problems, or is this just a 'feature' of all of these early 2X's?

Anyone have a clue?

Thanks.
 
Back
Top