820-3332 power rails fluctuating

[jgraffis]

I'm working on this 820-3332 thats got me pretty confused. when the machine came in it powered on then the screen would go black at the apple screen and keep running, The tech working on it powered it up on smc bypass and now the board doesn't power on fully. i get the left fan to spin for about 1 second then turn off then back on continuously.

i tested the power rails and this is what i get.

ppubs_g3h- 12.6v
pp3v42_g3h- 3.42v
PPDCIN_G3H- 16.1v pulses to 20.02v
PM_SLP_S3_L- 0.0v pulses to 3.3v
PP0V75_S0_DDRVTT- 0.0v- pulses to .023v
PP1V05_S0- 0.0v pulses to 1.05v
PP1V8_S0- 0.0v pulses to 1.7v
PP3V3_S0- 0.103v pulses to 3.3v
PP3V3_S0GPU- 0v
PP3V3_S3- 0.25v pulses to 3.3
PP3V3_S5- 3.3v
PP3V3_S5_AVREF_SMC- 3.3v
PP5V_S0- 0.26v pulses to 2.6v
PP5V_S3 0.16v drops to 0.098v pulses to 00.8v
PP5V_S5- 5.02v
PPVCORE_GPU- 0.0v
PPVCORE_S0_CPU- 0.0v
PPVTTDDR_S3- 0.012v pulses to 0.5v
pm_slp_s4_L- 0.0v pulses to 3.3v

im so confused on this one. anyone got some ideas as where to start on this one?

 

[dukefawks]

So it is power cycling, could be anything. Did someone stab it with screwdriver or multimeter? Corrosion? U8900 desoldered and checked it's outputs for shorts?

 

[jgraffis]

the power cycling issues happened prior to even being opened up, so i dont think it got stabbed with anything. there is literally no corrosion or signs of being dropped.

diod mode on these pins indicate shorts to ground with u8900 removed.

pin12 0.015v
pin13 0.0v
pin21 0.015v
pin22 0.0v
pin27 0.0v
pin28 0.015v
pin40 0.0v

based on pins 12 & 13 readings im guessing this is not a good sign for the u8000?

 

[dukefawks]

Resistance to GND on the same pins?

 

[jgraffis]

pin12 30.1ohms
pin13 0.5 ohms

 

[dukefawks]

That's fine, put the chip back.

 

[jgraffis]

sure thing. not sure whats going on with this board. i want to start looking at cpu vcore, but get to bios and its not getting what it needs just power cycles. kinda a rabbit hole for me.

 

[jgraffis]

do you think this could be a possible smc issue?

 

[dukefawks]

Wait, you powered the board without U8900 installed?

 

[jgraffis]

no ive already put the u8900 back on.

 

[dukefawks]

CPU Vcore pulsing? ALL_SYS pulsing?

 

[jgraffis]

CPU Vcore stays at 0.0v ALL_SYS pulsing for 0.03v to 2.9v

 

[dukefawks]

So no Vcore, this could be absolutely anything. If this machine came in working then someone fucked something up and it could be any knocked off component or something.

 

[jgraffis]

i would normally agree with you on that, had it been opened up before screwing up but thats not the case. machine came in we always power on to verify the customer issue if possible. Thats where it started going sideways. way before it even got close to a screw driver.

 

[dukefawks]

Doesn't change the fact that it could be anything. Without any history this one will be a rabbit hole.

 

[jgraffis]

agreed. i did some checking around on the board and found that pin 1 of r9840 is shorted to ground. im thinking there is an issue with u1800?

 

[dukefawks]

Resistance to GND there?

 

[jgraffis]

46ohms. so not short to ground. not understanding why diode mode shows 0.004ohms?

 

[dukefawks]

Because diode mode is not resistance and every multimeter is different with diode mode.

 

[jgraffis]

ahhh i see. yea this one is just screwed im thinking. i never had one go all to shit while here and not messing with it. i think this rabbit hole is gonna just be a board replacement.