820-3115 boots halfway then shuts off

ytesfay80

Member
I have another 820-3115 that only turns on when I plug the charger in(There was small liquid damage by the LP8550 but it was mostly covered in a bunch of dust.).
The BKL_PWM was missing so I ran the 3.3v wire to it. so the backlight is back on.
It will not turn on or off with the power button.
When I plug the charger in it will chime and boot to the apple screen with the loading bar. The bar goes very slow and stops at around 1/3 the way and the cursor will appear and the spinning wheel then it just shuts off and the fan stops spinning. When I remove the charger the fan will start spinning for like another 5 seconds but no bootup then it will shut off again.

I don't even know what I should check for this issue. I tried with another HDD and the same thing happened.
 

larossmann

Administrator
Staff member
Corrosion or something under the SMC. If it wasn't turning off at boot I would guess SMC_LID is low, either from corrosion on the board you can't see or a bad battery indicator(which I hope you are unplugging for these tests). If small liquid got by LP8550 it isn't a stretch to imagine it got under the SMC as well which is right next door.
 

ytesfay80

Member
Yes I tested without the battery indicator and I forgot to add that I already ultrasonic'd the board. I see a cap that is near the smc and it has signs of liquid damage so there might still be some under the SMC. Should I remove the edge bonding then add some flux and heat the SMC a little before throwing back in the UC?
 

larossmann

Administrator
Staff member
only if you know how to do that without ruining the SMC. You can fuck the board entirely if you don't know how to clean. I would remove edge bonding(ALL OF IT, NO EDGE BONDING SHOULD BE UNDER THE SMC BROKEN OFF), and do a light reflow. duke advises against reflow because people who lack experience will fuck the entire board(which is usually the case). i've never had experience with cleaning go well without a full reflow of the SMC but my SMC reflows have never fucked a board so I don't worry about minor cleaning vs. full reflow. so i'd practice reflowing a chip like this and seeing it "move" on a board that is totally dead before doing it to this one.
 

ytesfay80

Member
Okay I have a Weller WHA900. what temperature and air would you recommend this reflow so I don't kill the chip or blow it off. When I am replacing other QFN's or other components I usually have the heat around 7 and the air around 8.
 

dukefawks

Administrator
The power button does not work....Didn't it occur to you that this may be the issue here? Measure the voltage on C5710, disconnect keyboard maybe? Boot ASD and see if there are errors?
Maybe it goes into sleep, SMC_LID high?
 

ytesfay80

Member
I was able to hold the option key to get the machine to the boot option screen with my usb on it. I could move cursor but it would not click so I had to use the arrow keys to select the efi version of asd but in order to start the test I need to click so I will just return this machine back since it already needs a new keyboard and ram and now potentially a trackpad.

I was able to get to the login screen only when doing a PRAM reset and still the click didn't work and it randomly shut off one time while on the login screen but still this is not worth its trouble.

Thanks for the help though duke and louis.
 

dukefawks

Administrator
BKL_PWM is missing because SMC_LID was low. This also disables the trackpad click button. This would also explain the going into sleep mode when OSX was loading.
All this thing needed was a keyboard and a BIL or BIL connector.....it doesn't get easier then this.
 

ytesfay80

Member
Okay. I have keyboards for this and I have a donor board to get the bil connector from. Should I replace the BIL cable as well because I have a machine for parts I can take from.
 

ytesfay80

Member
sorry about that I am too eager to just replace things.
anyways I first measured SMC_LID_R on J6955 and I got 0v. The only thing I had plugged in was the HDD cable, trackpad cable and the screen and for some reason I actually got to the login screen but the trackpad didn't click when I hit shutdown.

Next I tested with only an individual screen bering the only thing plugged in and when I went to measure SMC_LID on R5071 the tiny resistor was covered in blue corrosion. I put some flux and heat to get rid of it then tested the resistance and it was 100k. I plugged the charger in and it was showing around 0.26v - 0.30v on both sides of the resistor.
 

larossmann

Administrator
Staff member
You need 3.4v on SMC_LID. If you unplug the BIL(which can drag down SMC_LID), you should have 3.4v on it. Fix that before anything else.

Also, if i mention two things in a post, please read entire post. I said to check SMC_LID, then consider reflowing SMC. You jumped to reflowing SMC. If you check things out of order it will fuck up the entire process.
 

ytesfay80

Member
I decided to remove R5071 and when I did the left pad(pp3v42 side) was completely gone so I put a new resistor on there and ran a wire to the cap next to it.
Now I am getting 3.4v on both sides, the machine booted up fine and asd passed. Now I just need to replace the keyboard.

Once again thanks duke and louis. Glad I didnt reflow the smc and replace the bil connector for no reason.

Update: The power button works on the machine so I won't have to replace the keyboard.
Could you explain how SMC_LID caused the power button to not work?
 
Last edited:
Top