Macbook Air 802-00165-A Good backlight fuse, bad wled?

brooke.j

Member
Hi, I have been looking for a schematic for 802-00165-A and I'm not happy with the last place I bought from so can you suggest somewhere or is that not permitted here? Here is what I know: Minor liquid damage board looks great. -Powers on with no backlight but works with external monitor. -The backlight fuse is good (tested and open) but I don't have a schematic to test the wled. - Does anyone know the pin voltages?

Thanks
 

brooke.j

Member
Ok now I have the board view and schematic. So I think I'm on the right path. I have another board here (same one) for parts and I'm hoping I can take the WLED driver off of it, if that is what the issue is. Any suggestions or comments are greatly appreciated.
 

brooke.j

Member
Perfect, New here so didn't know where to look. Of course there is a sticky on the subject. I'll start probing asap.
 

brooke.j

Member
Ok I just want to say that if you tell me once I won't make the same mistake again. I measured the fuse by putting red probe on power coming into fuse and black on ground and blew the fuse. Probably a noob mistake. I'll replace the fuse and move on. I have a basic understanding of measuring voltage and I"m here to learn. I'm probably stupid because for the life of me I can't google anything that says to test differently for voltage. Also great sticky on voltages but where do I test for the 39v-49v, at the lvds connector right? Total dumbass I know but thx for the help.
 

brooke.j

Member
Ok so I blew a perfectly good fuse, no prob I've replaced those before. Without changing the fuse I notice that Pin 1 of the Backlight fuse Reads 8v. Should I be getting12.6v?
 
Last edited:

brooke.j

Member
I'm an idiot. For some reason I had my probes connected to my multimeter wrong. Gotta get some fluke full proof gear. OK lesson learned.
 

brooke.j

Member
Hi again, should I be testing the PP3V3 on the LCD connector to post voltages outlined in Louis' sticky post? Or or should I be testing PPHV_S0SW_LCDBKLT? Where is the best place to test the backlight voltage?
 

brooke.j

Member
Hi guys. Can you tell me what component I need to test so that I can give you the voltage readings you asked for? I'm a little lost but not dumb so that's why I joined this forum in hopes of filling in the missing pieces. I'm really grateful to have any help you can spare. Thank you
 
Last edited:

larossmann

Administrator
Staff member
Q7706 is a transistor that will let the power from the backlight fuse(PPBUS_G3H) through to the system during conditions that power should be let through to the system(system is on, not sleeping, etc). This transistor is a P channel MOSFET that will allow power to flow from the source(pin 4) to the drain(pin 1/2/5/6) if the voltage on the gate(pin 3) is less than the voltage on the source(pin 4).

So look at how this is set up. R7788 takes the 8.5v from pin 4(source) and brings it to the gate(pin 3), so that the voltage on both pins is equal, which means it'll never open. Fuck. :( This is what R7789 & Q7707 are for. R7788 by itself just takes 8.5v from the source and puts it on the gate of the transistor. However, when R7789 is given a path to ground, R7788 and R7789 become a voltage divider that lowers the voltage on the middle point between them. R7789 will send some power to ground from R7788, which will lower the voltage at pin 3 of Q7706 to 3v, but only if pin 2 of R7789 is given a path to ground.

So we know in order for Q7706 to allow power through to the backlight circuit, its gate(3) has to be lower voltage than its source(4). This means that R7789 has to be given a path to ground. Q7707 sits between R7789 and ground, though! Q7707 is a set of two transistors, but these are a little different than Q7706. These are N channel mosfets. These will let power flow from source to drain when the voltage on the gate is around equal or higher than the voltage on the source.

So let's check the source here... source is ground, aka 0v. So anything on the gate at all should cause it to open.

So BKLT_PLT_RST_L is 3.3v, which is great, but EDP_BKLT_EN is 0v, which is bad.. this means it isn't going to open, which means R7789 doesn't get to send some of the 8.5v from pin 3(gate) of Q7706 to ground, which means Q7706 will never open. aka no backlight.

Now, we have to think about what would cause this. EDP_BKLT_EN comes from the PCH/CPU chip, U0500. Under what circumstances would the CPU say "hey don't make a backlight!" ??

Maybe if it thinks the computer is closed?

So now we have to figure out when it thinks the computer is closed. find SMC_LID and measure it. This should be 3.4v when the laptop is open, and 0v-0.5v when the laptop is closed. In practice, a voltage as low as 3.1v-3.2v can fuck with the backlight circuit in my experience. So measure SMC_LID, while the computer is on, and open, to confirm that the board thinks the machine is open for use.

EDP_BKLT_EN could not be being created because of something like that. However, it could also be getting created, but be shorted to ground by something that is malfunctioning. A short circuit to ground means that some fucked up component or pathway is sending a signal to ground(the black hole for electricity for the purposes of our discussion where if it goes it will never to be seen again)

Corrosion can also short EDP_BKLT_EN to ground as well. You have to find places where EDP_BKLT_EN has a path to ground. Q7707 is one of those places, the CPU is one of those places, R1430 is one of those places.. I'm not saying the latter two are at ALL common, but anything is possible.

In terms of making measurements, here's where we will just be hardasses and say you just have to figure out some of this shit on your own. We can point you in the right direction but as duke always says "engaging brains" is necessary here. In the sticky he linked you to there are literally colored lines and arrows pointing to specific points. I say you should measure voltage at backlight output, and when you look at the page, the last point is PPHV_S0SW_LCDBKLT. This also shows up on the LCD connector on pins 3/4 and when you look at the LCD connector page, is the only logical signal that could have anything to do with backlight. You will have to be able to close these small gaps in logic on your own and be a detective in some ways or you'll be fucked, even with our advice.

Best of luck!
 

brooke.j

Member
HOLY SHIT!! Thank you so much for that!! You totally made it make sense now. I was testing at the lcd connector on pin 3 (no voltage btw) and doing most of what is described but I just needed someone to put it all into context. I'm a very visual learner and the way you explained it made a great picture in my head now that I have been getting familiar with the schematics and .brd....I don't feel like a tool anymore. WOW Thank you thank you thank you. And please tell me to get my head out of my ass if I ask a stupid question and correct me or tell me I have my head up my ass because of X,Y, and Z. One last thing. If I do need to replace the LP8550 is should I be getting 926-LP8550TLE/NOPB or 926-LP8550TLX/NOPB
from mouser.com? OH I SEE THAT YOU CAN ORDER MIN 1 OR MINIMUM 3,000 IN RESPECT TO THE PART NUMBER (is that the only difference?)
 
Last edited:

brooke.j

Member
Hi, again. I've taken this two weeks to really study and make sure I am able to diagnose in the correct way that will help the forum. (mostly you and Louis so far) to help me without talking to a total moron. Check out my new posts. I hope I've narrowed it down to the Q7707. Thx
 

brooke.j

Member
OK I've been studying and I think I'm ready for some advice that won't be wasted on a total n00b!
Here is what I did.
-I checked the SMC_LID = 3.42v
-I changed the WLED DRIVER (JUST TO SEE IF I COULD DO IT) I did it twice and didn't muck up any pads. I used this part from mouser.com:

926-LP8550TLE/NOPB
MFG Part No:LP8550TLE/NOPB
LP8550TLE/NOPB
TI LED Lighting Drivers
US HTS:8542390000 ECCN:EAR99 COO:MY

The part went on great and I didn't burn it and I also saw it float around on top to make sure all balls were liquid.

I was getting 3.4 on the third pin of LCD connector on and off. Which is weird. Then I decided to make sure that the LVDS connector was soldered well and had no corrosion. After that I plugged in the display and it blew the backlight fuse.
-Replaced backlight fuse. Now no 3.4v on LCD connector.

Here are my readings now.

R7788 PIN 1 = 7.88V
PIN 2 = 7.66V

R7789 PIN 1 = 7.66V
PIN 2 = 7.56V



Q7707 PIN 3 = 7.58v ( should it be lower? )
PIN 2 (BKLT_PLT_RSTL) = 3.34v
PIN 5 (EDP_BKLT_EN) was giving me 3.4v and now is 0v after unplugging LCD connector.

So it looks like the voltage is getting lower from Q7707 but is it now low enough? I know that there could be a short somewhere but if you think replacing Q7707 might be the right direction?

Thank you again for all your help. I have put in mad hours to learn more and not have your time wasted.
 

aprendiz

Moderator
You have a bad display and/or cable and now maybe a bad connector again (hope not), EDP_BKLT_EN only will be present if display is detected, when you get EDP_BKLT_EN and BKLT_PLT_RST_L you will have the right voltage at gate of Q7706...

Try with a known good display assembly..
 
Top