[SOLVED]MBAir 820-00165 -- no post

jdwhatnot

New member
A machine came in, said the screen is dark. Confirmed no backlight, 7V on the BKLT supply after the diode. Checked U7701 and there was slight off-white corrosion along the side of the IC. Brushed it away and fluxed the chip hoping for quick fix. It did not solve the issue. So I put it aside until I could get back to it. About a week later when I had more time I went to check it out again, but it's now decided to not post at all. I went ahead and just replaced U7701 anyway.

We have orange light, fans don't stop spinning. Usually I just give up when I see this but I need to figure this problem out because it's not the first time I've seen it, plus it was actually booting to OSX when it came in. So....

We are in S0, all power rails present, chipset has proper signals. One weird thing is that sometimes if I measure PM_RSMRST_L on R8133 it will appear to try post; fan will die, and it sometimes brings the fan back up for the next post cycle, but usually 3v3_SUS dies and doesn't come back up. So I replaced R8133, same symptom, replaced U8130, same symptom. There was no other corrosion seen, and the board has been ultrasonicked.

Any ideas? Unfortunately chipset replacement is out of the question and my hot air station doesn't throw out enough heat to desolder an SMC, but that's where I'm leaning. Also, on a whim with another board I swapped out the ROM with another of the same model and it did not fix the issue, so there's also that.
 

TCRScircuit

Super Moderator
Staff member
This could be many things... Could be a random corroded resistor somewhere, could be spi resistors or traces, could be SMC, could even be the CPU but let's not go there yet --

Please measure the following --

PM_PCH_PWROK -

PPVCC_S0_CPU Both voltage and Resistance to gnd

PLT_RESET_L

What exactly do you mean by it "trys to post" ?
 

jdwhatnot

New member
This could be many things... Could be a random corroded resistor somewhere, could be spi resistors or traces, could be SMC, could even be the CPU but let's not go there yet --

Please measure the following --

PM_PCH_PWROK -

PPVCC_S0_CPU Both voltage and Resistance to gnd

PLT_RESET_L

What exactly do you mean by it "trys to post" ?

High
1.8V on both rails
High

"Trying to post" being S0 state, power rails present, fan is spinning; the intended result being three complete posts and then a nice Mac chime.


So, for an update... it looks like I've leaned something new. So on all other boards I can recall having worked on over the past few years, Air boards have been unique in that when power is applied the board will do three complete posts and then chime and become operational. This board spins up the fans once, waits about 5 seconds, and then chimes. Never seen this behavior before, but I've not worked on a lot of 00165's with hardware issues yet.

But the new U7701 works and backlight is operational. So yay for science!?

EDIT: I need to mark as solved, how can I do this?
 
Last edited:
Top