820-00138-A got image but no backlight

[Salem Aboshala]

Dears, this is my first post here and first try to fix my own laptop Apple MacBook Pro Retina 15-inch Mid 2015.
I have some general electronics troubleshooting experience and still as my hoppy but this is my first laptop trying to fix.

History: my laptop backlight stopped working without any reason.
Symptoms:
I have image on screen using my flashlight.
I have fully working laptop with the help of my external SAMSUNG LED tv via HDMI cable.
Keyboard backlight still working and can be adjusted normally.

Since two days I tried to follow all similar issues here this forum and still in the same place.

Tried to test with another "used" 820-00138 logicboard brought from ebay but still no backlight. Could be the same issue (faulty U7701)??
From first visual inspection there is no any sign of liquid damage nor corrosion nor any sign of component failure.

I followed the schematics carefully and there was nothing wrong with all measurements in LCD Backlight driver circuit.

Here are the volt readings as I still remember on U7701 :

1,2=0v
21=0v
5,18=5v
9,10,11=12v
17=3.3v
PPVOUT_S0_LCDBKLT=0v
PPBUS_S0_LCDBKLT_PWR_SW=0v
PPBUS_S0_LCDBKLT_FUSED=12v
PBUS_SW_LCDBKLT_PWR=12v

I followed all measurements on LVDS connector (J8300) and was fine I guess except this pin #8 (LCD_BKLT_PWM_R) which I think not normal:
When I power on the laptop there is another chip pulls down this pin to 52Ohms, but it is open circuit in power off mode. I checked with / without LVDS cable and the same behavior.

From my own primitive experience, I think (U7701) still needs some information from another chip (i2c I guess) to start boosting.
FYI, main LCD is normally detected by OS and tried SMC and PRMA reset but nothing changed.


Could anybody help in this case ?

[mon2]

1) Welcome.

2)

When I power on the laptop there is another chip pulls down this pin to 52Ohms

Never power the board and measure in resistance or DIODE mode with your meter. Always remove all power before using either resistance or diode mode of testing.

3) With power to the board, meter in DC volts mode (10 volt or higher scale is ok) - measure the voltage to ground of:

I2C_BKLT_SCL ; as below
I2C_BKLT_SDA ; used to configure the internal registers of the backlight controller

EDP_IG_BKL_ON ; this is the enable signal for the backlight

Post each measurement.

Adding to your workload:

4) Remove all power. Meter in resistance mode (not diode mode). Measure the resistance to ground of PPVOUT_S0_LCDBKLT. We are wanting to check the resistance to confirm if there is a shorted component on this boosted power rail. It is possible that the backlight IC attempted to boost the voltage but due to a shorted component, there was an excessive current draw and then the charger IC turned off the backlight.

5) The P-channel mosfet @ Q7706 will enable if the gate pin # 3 is negative or lower than the voltage being passed through source & drain pins. That is, if R7702 is pulled to ground through the charger pin # 11 then a voltage divider will be built for pin # 3. Then R7702 & R7701 will work to lower the voltage that is incoming @ PPBUS_SW_LCDBKLT_PWR and this new lower voltage is enough to enable the Q7706 mosfet to pass the voltage onto the LCD backlight circuit. However, the backlight IC must pulse the use of this voltage on / off to boost to a voltage of 25-45 volts to light up the LED array that resides inside the LCD display. It is possible that the backlight IC is defective but perform some tests and post your results.

[Salem Aboshala]

1. Thanks alot for your reply and greetings mon2
2. My little multimeter from Uni-T and it has auto-mode feature and it starts with sensing ac/dc voltages first, and then it starts to look for resistance if there was no voltage detected (UT123D).
3. 4.78v , 5.03v and EDP_IG_BKL_ON=3.33v coming via R7742 (0Ohms)
4. It gives 14Kohms and then goes to 500Kohms repeatedly, I even injected external power (50v) to insure no leakage and there was 0mamps.
5. Q7706: 12.66v@pin#3 , 12.76v@pin#4.

I think Q7706 will stay off till BKLT_SD goes down right ?

Thanks alot for your contribution
Appreciated

[Salem Aboshala]

Correction, I2C_BKLT_SCL = I2C_BKLT_SDA = 5.03v now, or maybe it is different slightly from time to time as it is a communication signals right ?

[mon2]

Yes on Q7706.

The measurements look good and the backlight is being enabled.

Remove all power and confirm diode D7701 is ok. Test in diode mode. With no display attached, the diode should conduct in one direction (not the band) and block in the other.

Your smbus / i2c lines look ok and not stuck.

Update: If you wish to try, take a piece of wire and mate one side to ground (shield is ok) and the other side to pin # 3 (gate pin) of the mosfet @ Q7706. This will force the Q7706 mosfet to enable. If indeed, the backlight IC is being enabled and ok, it should be able to boost this 12v input voltage.

Assuming that the LED return signals are connecting to this backlight IC OK + D7701 is ok + Q7701 is ok -> the voltage should be boosted.

[Salem Aboshala]

Thanks alot agin mon2 for your follow-ups,

Remove all power and confirm diode D7701 is ok. Test in diode mode. With no display attached, the diode should conduct in one direction (not the band) and block in the other.

Already checked = OK

Update: If you wish to try, take a piece of wire and mate one side to ground (shield is ok) and the other side to pin # 3 (gate pin) of the mosfet @ Q7706. This will force the Q7706 mosfet to enable. If indeed, the backlight IC is being enabled and ok, it should be able to boost this 12v input voltage.

I think you meant to short BKLT_SD to ground, but I shorted pin#4 & pin#6 instead to bypass 12v to PPBUS_SW_BKL, and I got what I expected at PPVOUT_S0_LCDBKLT = 12v, which means the booster still wont work.

Assuming that the LED return signals are connecting to this backlight IC OK + D7701 is ok + Q7701 is ok -> the voltage should be boosted.

I think U7701 still get LCD not ready for LED power on, otherwise pin#11 (BKLT_SD) should start signaling right?

[Salem Aboshala]

I did tested the following:

1. Connected external 48v LED strip the PPVOUT_S0_LCDBKLT line and the booster didnt work.
2. Injected LED tester @ PPVOUT_S0_LCDBKLT with 10mAmp and there was no current drawn from Retina LED.

Here I started to think the Retina display has a defective LED circuit and asked U7701 not to start boosting.

Is there anyway to test the Retina LED from external source ?

Any thoughts would be appreciated
Thanks

[Salem Aboshala]

Good news, I started to disassemble the LCD and I discovered one cable disconnected (CNT201), just reconnected it and the backlight is working fine now.



So many thanks mon2 for your help
Have a great day