[SOLVED]820-3462 Working Fine. No Webcam

[bjf]

This board had some minor water damage by the lid sensor causing the computer to randomly sleep. That's repaired. The only issue with it now is that the webcam is not recognized. I installed a known-good board and the webcam itself works fine. Here are my diode measurements for the connector (with webcam attached) for the bad board. 1 0
2 .470
3 .470
4 .455
5 .571
6 .502

The webcam connections seem low but there doesn't seem to be much in the circuit besides U1800. L3507 has continuity through it on both sides. I replaced the connector on the board. I'm stuck with this. Any ideas for this would be appreciated.

 

[dukefawks]

Camera is a simple USB device. Is there 5V on pin 4?

 

[bjf]

Yes, sorry, should have mentioned that. 5V is present at pin 4. I've also measured continuity from the connector pins to the actual webcam on a test screen that has the glass off.

 

[dukefawks]

Are other USB devices working? Are those values with camera plugged in?

 

[bjf]

Values are with camera plugged in. USB is working elsewhere.

 

[dukefawks]

Well, what about with camera disconnected? Only pin 2 and 3 are of interest.

 

[bjf]

2 and 3 are both are at .490 with the camera disconnected.

 

[dukefawks]

I have no idea. Are 2 and 3 shorted together maybe? There is nothing to this thing, 5V, GND and 2 data lines. It could be that this pair of USB data lines is fucked, but .490 seems ok.
Get a camera cable from a scrap LCD and see if all the pins have continuity to the actual camera module.
I assume you did a PRAM reset and tried booting from a USB device to rule out software.

 

[bjf]

I checked the continuity with a test screen with the bezel removed. Continuity is good on all lines. However, the measured resistance between these two lines is around 167Ω. On a good board I'm seeing something around 2.5MΩ.

 

[dukefawks]

There is your problem! Could be the connector, L3507 or crap under U1800. The last suspect would be dead U1800, but I think it is corrosion.

 

[bjf]

Unbelievable. Despite measuring continuity through it on both lines, L3507 was bad. Lessoned learned. Thanks for sending me in the right direction to test the resistance between the two data lines. For whatever reason, I wasn't going there. Board is working perfectly now.