MBA Backlight Repair diagnosis

@livfe start on F7700 to make sure that you have a good fuse. Check for continuity of the fuse and that you have the PPBUS_G3H 8.6V on that fuse. Replace the fuse if no continuity. Next you will need to know the actual voltage on the LCD connector. You can measure that on pin 3 and 4 of the connector.

Measure the backlight voltage (PPHV_S0SW_LCDBKLT) on pin 3 and 4 of the display connector. If that voltage is ~50V or higher that means there is no load on on that line which means a complete open circuit without any load (No LCD). Check the connector and check with a known good screen. Remember that you got two pins on the connector for a reason, so make sure to check your connector on both pins and on your cable if you can. If backlight voltage (PPHV_S0SW_LCDBKLT) is 25V-40V that means the boost circuit is working. Check the connector, cable, and LCD.

If backlight is PPBus voltage (8.6V) your boost circuit isn't working. Check the feedback trace, LCD driver, and enables. In particular the Bkl_En. Measure that on pin 2 of R7331 or pin 1 of R7715. BTW, those two voltage divider resistors are also a common failure point. Make sure those are okay. R7731 is a 200K resistor and R7715 is a 100K resistor.

Verify BKL_EN (~3.3V) and BKL_PWM (~3.3V at full brightness). Check to make sure that you have power to the LP8550 backlight driver. Check for PP5V_S0 on C7110 (5V) and C7711 for PP3V3_S0 (3.3V)

If backlight is 0V check for a short circuit then work your way back toward PPBus. The LCDBKLT_FET, fuse, and enable transistors Q7706 are all suspect at this point. Also check the voltage dividers R7331/R7715.

Most common problem with having no backlight is LP8550 or the display connector.

You could try the following as a quick check for that. Get your multimeter and put it in diode mode. Now it get's weird :-) Red probe on ground and Black probe on pins 3 or 4 of LCD connector.

If you get 0.0 or 0.007 that indicates a short to ground, commonly inside the connector or LCD cable.

If you get 0.200-0.300 it is bad LP8550 driver.

If you get 0.459-0.511 it is bad feedback via or corrosion of the feedback ball on LP8550

If you get 0.565 it is corrosion of the solder balls or a internal failure of LP8550

If you get OL on pins 3 or 4 of the LCD connector, the D7701 (Diode) or the trace to the connector is physically broken/blown.

As you can see the most likely candidate is LP8550 followed by a bad connector and failure of the Bkl_EN voltage divider. This is how I would proceed. All of this is academics until you actually get to check the board.

Who knows, it may just be a corroded board with some trace error or some weird error that we never figure out :-)