820-000165 Liquid damage

[estebanquito356]

Hello,


i recently got a 2015 13'' MBA 820-000165 that had a liquid split on it


It's got clear signs of liquid, but not big signs of corrosion under the microscope.

When plugged in the light turns green and amber. Power consumption goes from 0.01A to 0.29A to 0.01A again, and repeats.
Fan sometimes does quarter fan spin, sometimes 5v is so briefly that it doesn't.

Already tried with a good know working IO board.


Here are the voltages i measured:


PPBUS_G3H: 8.6V steady

PP3V42_G3H: 3.4V steady

PP5V_S5: 5V steady

PP5V_S4RS3: Goes from 0.09 to 5v to 0.09 in a circle

PP5V_S0: Goes from 0.09 to 5v to 0.09 in a circle

PP3V3_S5: 3.32V

PP3V3_S4: 0.03 to 3.3 in a circle

PP3V3_SUS: 3.32 steady

PP3V3_S3: 0.00 to 3,3 in a circle

PP3V3_S0: 0,00 to 3,3 in a circle

PP1V8_S3: 0,00 to 2,28 in a circle very briefly

PP1V2_S3: 0,00 to 1,2 in a circle

PP1V05_SUS: 1,05 steady

PP1V5_S0: 0,00 to 1,49 in a circle

PP1V05_S0: 0,00 to 1,05 in a circle



Any help in how to continue will be much appreciated

Thanks

 

[2informaticos]

First of all, welcome to the forum!

Check for short all secondary power rails.
Start with big coils, but do not forget pages 57758.
If no short, the power rail must pulse at least, as the other above mentioned.

Test the board starting in SMC bypass.
Reball SMC, as basic step for liquid affected boards.

 

[estebanquito356]

Thank you very much for the welcoming

I quickly checked resistance in every big inductor, everything looks fine. Only the cpu lines where lower at around 18ohm but im sure that's very normal.

I'm yet to check pages 57/58.

What i did do was a SMC bypass, and now the pulses are way longer, for example the fan was barely trying to spin, now it goes to full speed before it turns down.

Thanks very much for the help, i'm going to inspect pages 57/58 (any advice in what to look first at?) and i'll let you know.



Cheers!

 

[2informaticos]

If stays on longer time in SMC bypass mode, check all sensors from pages 41-43.

I still recommend to reball SMC!

 

[estebanquito356]

Hello again,

since i'm not very familiar with page 41-43, and it's my first time working on them, anything i need to check up first? Is there anything that is most common to fail or do i need to check every component on every circuit?
It is my first time working with current/temp sensors so any tips are well appreciated. Thanks!

 

[2informaticos]

You must check output voltage for all sensor circuits.
Reball SMC and maybe no need for more tests...

 

[estebanquito356]

Hello again,

I checked all of the sensor's voltges:


C5455 pulses from 0,15 - 0,11V
C5465 0v
C5433 0,12v
C5475 0v
C5445 0v
C5485 0v
C5425 0v
C5495 0v


(this voltages dont make much sense to me)


Unfortunalelly i dont have a stencil to do a SMC reballing, best i can do is a reflow. I did put some good flux on it, heat it up slowly until it floated, tapped it with the tweezers, made it dance around with the hot air, and let it cool down. No change was observe to the behaviour of the voltages' pulses.


I've found a little bit of corrosion arround U6100 area. Not color spots or formations, but dark resistors. r6115, r6128, r6126, r6125 have all one end too dark. I've measured them all and the traces and they seem to be good, don't know if trouble there can relate to the symptoms of the logic board.

Again any help is much appreciated. I understand this board can be on the hard one's to fix, so i may put it on the side for a bit.
I will be posting others that might be way easier to fix soon.

Thanks in advance!

 

[2informaticos]

You must check ALL SPI Bus Series Termination resistors (aprox 20) and traces; some of them are placed underneath PCH.
Also clean/remove J6100 (some of mentioned resistors near it too).
Finally, try known good BIOS.