820-3476 missing PM_SLP_S4_L

[Fizzl92]

Hello,
this 820-3476 is not turning on. It has an orange light in the charger and there is no visible damage on the board.
PP5V_S4 is not there because PM_SLP_S4_L is also missing. I checked U1900 and i ge a 1V p-p clock signal on pin 12 (on my oszilloskop) so i think this is fine.
U1950 is missing ALL_SYS_PWRGD. Here it gets a bit confusing for me. As far as i understand PM_PCH_SYS_PWROK is needed for the CPU/PCH to work. But this is created by ALL_SYS_PWRGD. This needs a bunch S0 Rails to be created. But as PM_SLP_S4_L is missing on this board it is not in a S0 state. So i am back to PM_SLP_S4_L not beeing created.
Would be really great if somebode could give me a short explanation of how this works, what comes on first and help me fix this.

 

[dukefawks]

First thing to determine is if the board is power cycling. Check if PP5V_S4 is pulsing, best to do with scope. If not then check PPVRTC. Also check all rails for shorts.

 

[Fizzl92]

PP5V_S4 is straight 0V - even if i put the scope to 0.1 us/div. Same thing for PPVRTC_G3H. There was indeed a short on PP5V_S4. I slowly injected voltage in there (starting at 0.9 V - 3.5 V were enough to feel something getting hot) and U8300 got hot. I replced it. The short is gone now (0.377 V to GND in diode mode) but still no PM_SLP_S4_L and still 0V on PP5V_S4 without pulsing. Also PPVRTC is still not pulsing. I checked ll other rails for shorts and two of the readings are a little low i think.

PP1V35_S3_CPUDDR (and the other ram rails are similar) = 0.09V to GND in diode mode. This may still be ok because its a small circuit with low voltage on it...
PPVCC_S0_CPU = 0.031V to GND in diode mode. I think that is the problem.
should i inject voltage in there (up to max. 1.5 V) and see what gets hot? I am always careful if it comes to shorts on VCORE...

 

[G.Beard]

..........Never mind

 

[dukefawks]

PPVRTC should be present. Check for short and replace U1900 as that generates it. U8300 probably got killed by connecting a shorted or damaged LCD cable. Inspect cable and connector.

 

[Fizzl92]

I wanted to say, that PPVRTC is also straight and not pulsing. But it is present - straight 3,3 V. Sorry, my english is not the best. LCD cable and connector look fine and i get 0,38 V to GND in diode mode at pin 28, 29 and 30 (with the LCD cable plugged in).

 

[dukefawks]

3V3_S5 present? PPBUS voltage? I suspect that there is some more history to this board, U8300 does not "just" die like that.

 

[Fizzl92]

PP3V3_S5 is 3.341 V and PPBUS_G3H is 12.61 V. The only thing the customer said is that it just stopped working... And the whole machine looks near perfekt (inside and outside...)

 

[dukefawks]

What are resistance to GND values on those 2 rails with low diode mode reading?

 

[Fizzl92]

PP1V35_S3_CPUDDR : ~255 Ohm (multimeter counting up and down -+10 Ohm)
PPVCC_S0_CPU : <70 Ohm (multimeter slowly counting down)

 

[dukefawks]

That is within what I would call fine. At this point it could be anything and you really need to find some history. No way U8300 just died, no way in hell. See if you can find any other shorted rails too.

 

[Fizzl92]

I contacted the customer today and he said, that he left the macbook copying some files to an external drove and as he came back it was dead... I checked all rails listed in the scematic and they are all not shorted as far as i can tell. Few of them have a readings of 0.13 V in diode mode to GND (others up to 0.6 V).

 

[dukefawks]

The external drive is a hint. U8300 is on 5V so I bet the 5V got hit with some spike though USB. Measure resistance to GND on all 5V_s3 S0 and compare to known good. It is not looking good though
Also ask if they are using China garbage charger as that is what usually causes weird current loops that kill machines like this, due to bad insulation from mains power.

 

[Fizzl92]

I am not sure if i did the right thing but i measured the resistances on all these rails. I attached a document where i compared it to a working 820-3462 (as that is the closest equivalent working board i have) and i also added the components where i took the measurements. I hope that hepls but even if it did not help this time i have the readings to compare other 820-3462 boards in the future to...

 

[dukefawks]

Not seeing anything that is worrying. Is 3v3_sus present?

 

[Fizzl92]

yes it is at pin 8 of the Bios chip 3.34 V

 

[dukefawks]

This is getting shitty. SMC_ADAPTER_EN present?

 

[Fizzl92]

yes... 3.38 V at R5185

 

[dukefawks]

Ugh . Is there any pulse on PM_SLP_S4_L? Check while connecting power, it may happen just once. Use a scope and set "one shot" mode to be able to capture it.

 

[Fizzl92]

Yes, it is pulsing for a second but after that it is back to 0 V. I only have a CRT so i took a photo with my phone as good as i could. Hope that is good enought...