820-00165 (Backlight Rabbit Hole from HELL)

This is by far the worst backlight I have ever fought!
it came in for no backlight water damage. Very slight water damage around C7793 and WX7720. Neither appeared to require rework.

Failure mode 1 Comes on then fades away after a 1 second. Voltage on Cathode of D7701 just slowly fades and screen is black. 22v, 17v, 16v after several seconds.
Failure mode 2 Back light does not even come on. 8v on D7701 cathode.
Failure mode 3 (works fine. Not a failure at all)

Testing / repairs done so far.
Diode test to D7701 cathode is .576 "perfect"
Flex board test "no change"
Thermal board test (hot / cold) "no change"

Feedback XW7720 was reworked. (and tested by soldering on to pad 5 on U7701 while it was removed and resistance testing while flexing board).

U7701 ( 2 times) "all resistors in this area were checked for proper values"
J830 LVDS connector (R7717-R7722 values verified)
C7796, C7797, C7799, C7712, C7713
D7701
Q7706
Q7707

Ran jumper to VIA and it fixed nothing! Failure mode 1 persists.
isolated VIA on both sides from D7701 & U7701. Worked perfect the first try then entered failure mode 1!
Ran jumper from U7701 LCDBKLT_BOOST to anode of D7701 failure mode 1.
 
Last edited:

2informaticos

Administrator
Staff member
Post BKL_EN/PWM.
Something similar happened to me with bad pad 2 of R7704; you can solder tiny wire directly (instead of 33 ohm resistor).
 

JohnB8812

New member
Sounds like an SMC_LID issue to me...tried known good DC-in board and flex? How do the pins on the trackpad connector and J9500 look? Any chance of corrosion near SMC?
 
The Hell has ended. It was a compound problem. Bad board & bad DC-IN. The board had a bad backlight. I tested it on the bench with a test monitor and my DC-IN. it was bad. Not sure what fixed it. U7701, or the VIA. When I put it in the machine the DC-IN was bad on top of it so I was chasing my tail around. Originally I had tested with my test DC-IN and the back light didn't work. When It worked on the bench it failed in the unit. When I threw in my test board it worked but I didn't hook up the ribbon. I never put two and two together and was chasing my tail round and round. It wasn't until I gave up on the machine and swapped the board, changed it's SN and then discovered it was doing the same thing that I realized I was making assumptions about the DC-IN.

What made this so hard to figure out was how intermittent the DC-IN failure was on top of it all! Because it worked on the bench I was not surprised because it was very intermittent. I never once thought the DC-IN. I couldn't get it to fail long enough to measure!

I sure feel stupid!

Thanks John and 2informaticos for the guidance.
 
Top