Macbook Air 820-3437 Dead: no s5_pwrgd: MISTERY

[phantomEVO]

3,3v

[mon2]

Measure the voltage to ground (not diode mode) directly at S5_PWRGD (Pin #20) on the U7501 IC.

Next, remove all power -> meter now in DIODE mode -> red meter probe to ground; black meter probe onto S5_PWRGD directly on U7501 IC.

Finally, still with no power -> meter in DIODE mode -> short the meter probes and you will hear a tone beep -> one meter probe directly at S5_PWRGD (Pin #20) on U7501 IC and the other meter probe on the pull-up resistor (R8141) on the same line. Do you hear a tone beep?

Confirm again that R8141 has 3v42 on one side of this pull-up resistor. This 3v42 must be stable.

This signal must be HIGH to denote that this power rail is OK unless U7501 is defective / not soldered in properly / U7501 replace is not the same as the original (confirm the topside markings - are the latest replacements for U7501 new or from a donor board?)

[phantomEVO]

I have Tried all those things before: pwrgoodline is not shorted, 3v42 is stable, each pin of u7501 is ok and goes to the right Point… so even pin 20 beeps to r8141

at this point, i will try to solder another u7501

[phantomEVO]

tried another u7501, still the same

[mon2]

This same power good signal enters the SMC. It is possible that the SMC is defective and causing this line to become stuck at a logic low. The only way to rule out the SMC is to locate a sweet location on this trace to cut to isolate the side that goes into the SMC. Respectively, this trace will need to be repaired after this review. You mentioned that the power good line is not shorted but no measured values. Assuming the line is not shorted, then it must be the SMC who is tanking this line.

[phantomEVO]

i will go crazy...i've soldered another smc..now i have 3,2 on the other side of r8141 but no power led on magsafe... i have even pwrbtn_l 3,4V!!!!!




ppbus_g3h is 8,16, PP3V3_S5_SMC_VDDA 3,4 sysonewire 3v (should be 3,4?!) PP1V2_S5_SMC_VDD 0,9 v!?!?!?!? should be 1,2? (line is not shorted)

[mon2]

Yes, PP1V2_S5_SMC_VDDC should be 1v2.

Remove all power -> meter in DIODE mode -> check the diode mode readings for this rail. What is the measurement?

[phantomEVO]

0,510

[mon2]

Good reading.

Follow the guidelines here (same case with the PP1V2_S5_SMC_VDDC):

820-00165 PPBUS_G3H : 0V | Rossmanngroup Forums

[phantomEVO]

SMC_RESET_L : 3.4V
SMC_WAKE_L : 3 V
SMC_CLK32K : 0V
SMC_EXTAL : 0V
SMC_XTAL : 0.4V
PP3V3_S5_SMC_VDDA : 3.4 V
PP1V2_S5_SMC_VDDC : 0,9V
PP3V3_S5_AVREF_SMC : 3.28V
GND_SMC_AVSS : 0V
SMC_TCK : 3.4V
SMC_TMS : 3.4V
SMC_TDO : 3.4V
SMC_TDI : 3.4V

i'm thinking that this smc is not good (i've bought for new)

[mon2]

Flux it well and try a reflow but otherwise, yes, could be a defective SMC. They should not be new but pulled from a donor board. This is a power rail created by the SMC itself to operate internal logic. Juts be gentle as to not tear any of the BGA balls / pads. Tearing balls is bound to inflct some level of pain.

[phantomEVO]

i will do a lot of soldering on mobile phone....even cpu! so is not a problem to solder an smc..i will solder again the old smc...


unbelievible that i have s5 pwrgood with this smc -.-'

[mon2]

Aha moment!!! As noted that the SMC can pull down this line. Now it is confirmed.

[phantomEVO]

Problem is to find a donor board?

Could I take from 820-2796 ??

[mon2]

https://logi.wiki/index.php/SMC_Compatibility

[phantomEVO]

SOLVED!!! I've resoldered the originaL SMC!!!!!