820-00165 - backlight working partially

Atomrepair

Member
This board came in with only slight corrosion on U7701 and no backlight. I replaced U7701 and got backlight back, but not completely. On the left half of the screen only half of the leds are on, resulting in a row of dark spots. I tried different screens first, but they show the same pattern.


LED_RETURN_6​
0,58​
5,4V​
LED_RETURN_5​
0,58​
5,4V​
LED_RETURN_4​
0,58​
0,5V​
LED_RETURN_3​
0,58​
0,5V​
LED_RETURN_2​
0,58​
0,5V​
LED_RETURN_1​
0,58​
0,5V​


I assumed my soldering might be bad and replaced U7701 with a new one. No change.

P5V_S0 and PP3V3_S0 are stable. I checked LED_RETURN_1; it has connectivity to U7701.

LVDS connector looks pristine, inside as well as outside. Pins are all properly fixed in place.

Are these LED_RETURN voltages normal? Whats next? replace U7701 again with lower heat to make sure its not messed up?
 

2informaticos

Administrator
Staff member
Diode mode readings are good, not bad solderings; no need to change U7701.
But you should check if all resistors have 0 ohm, as you have bad voltage on two lines.
You could have bad connections inside J8300.
 

Atomrepair

Member
Resistors are all measuring as 0-2 Ohms, connections from resistors to J8300 are fine.
I've replaced J8300; still have the same problem.
You mentioned I have bad voltage on two lines; which ones? I dont know what normal readings are for these rails and have no comparison (except older pro's which seem to be different).

I measured all LED_RETURN lines after working on the board; they're the same as before.
 

Atomrepair

Member
update: I measured continuity from all 6 LED_RETURN resistors to the corresponding balls on U7701. Had to fix the board upright and use a very small probe point, but I got what I wanted to know.

Backlight stopped working. I replaced U7701 (again) and it worked, but again with the same dark spots on the left half of the screen. Just now, backlight died again.
LCDBKLT_BOOST is 8.4V now.

PP5V_S0 and PP3V3_S0 are fine.

I measured all resistors surrounding U7701 - all are fine. Pads underneath U7701 looked fine as well...

I've been testing with three different full lcd assemblies.
 

Atomrepair

Member
I have now; area just above D7701 and the test point above it are perfectly clean, no corrosion. measuring 0 Ohms from test point to D7701 and also from test point to pad 5 from U7701
 

Atomrepair

Member
I fixed the board by taking a U7701 from a donor board and using that.
I can think of two reasons for my previous replacements not working;
1. bad soldering
2. bad IC's.

I think my soldering was fine; checked all sides under the microscope and could see no dodgy stuff. Also, my three replacement chips all failed in the same way. I'm now doubting whether my LP8550's are dodgy. Here's a side by side pic of the two. Left is the original corroded IC, right is the replacement.

2020-06-30_11-59-29-.jpg

Has anyone had trouble with these IC's before? I got these from alibaba.
 

tlough75

Member
I had to do the same thing on a board I worked on over a year ago. I replaced u7701 on an 820-3437 due to no backlight and had some leds not working. I thought "it couldn't have been my soldering" but sure enough, I pulled the board out of my rabbit hole box a year later, replaced U7701 and all the backlight LEDS came on. It is crazy the things we can miss when we are in a rush.
 
Top