820-3023 green light, warm cpu

samiam

New member
figured i'd ask before i spend another few hours messing around with it
have a 13" 820-3023 which had liquid damage, cleaned

initially PPBUS_G3H was jumping around between 8.5 and 3.4V, so I replaced R7051, R7052, and R7002

magsafe is green and CPU gets warm within about a minute but doesn't get as far enough for fans to spin

saw visual damage and put a wire between R2822 and Q2820 (PM_MEM_PWRGD)

wasn't sure if I should still be looking around there and whether that voltage divider wasn't working right because they read as 15k in circuit

getting diode reading .000 on R2801 pin 1
my working comparison board is giving .081

or should i be look at other S5 and S0 rails?
 

larossmann

Administrator
Staff member
i'm guessing a CPU mosfet or cap in that area is shorted. Is PPBUS_G3H a steady 8.5v now? Is the machine in an S0 state, meaning S0 rails are present?
 

samiam

New member
g3h is steady now, yes def in S0 state
getting 5.17V on PP5V_S0_CPU_VCCIO probe 371

you even hear the board make a super faint little power squeak as it tries to turn on

not getting the 1.5V on PM_MEM_PWRGD probe 379

you think it's Q2820 shorting it to ground? or something else. not really any caps in that area, just a whole bunch of mosfet gates
 

larossmann

Administrator
Staff member
Let's look at this logically. Q2820 is going to short PM_MEM_PWRGD to ground if voltage on pin 2 is too high.

So what is PM_MEM_PWRGD_L measuring? This means PM_MEM_PWRGD is present when the signal is LOW, so when it's HIGH it will short it to ground.
 
Top