A1706 short PP3V3_S5

[madjid911]

Hi friends,

I have a macbook pro A1706 820-00923-A that won't start no signs of life, no liquid, checking the voltages I found PP3V3_S5 was shorted, I removed all the processor coils, but when I 'injects a voltage it is the processor which reacts, can you confirm to me that this is definitely coming from there please?

I have 20V on the usb-c connector, and 13v stable on ppbus_g3h.
I saw that a track had burned out on PPVIN_2V7NAND_LB after resistor R9350
I injected voltage on L7660 pin 1 in short but only the processor reacts

Thanks for your help

[mon2]

With no power, meter in resistance mode, measure the resistance from this coil where you are injecting and to the CPU rails. Could be the PCH component of the CPU. 38C = ~100 F and does not sound crazy hot for a CPU (with no heatsink).

What is the resistance to ground of the CPU / PCH rails ?

[madjid911]

With no power, meter in resistance mode, measure the resistance from this coil where you are injecting and to the CPU rails. Could be the PCH component of the CPU. 38C = ~100 F and does not sound crazy hot for a CPU (with no heatsink).

What is the resistance to ground of the CPU / PCH rails ?

C8154 pin 2 (PPVCCPRIMCORE_SUS_PCH) to L7410 pin 2 (CPU) : 3,9 ohms
C8154 pin 2 to GND : 2,9 ohms
L7410 pin 2 (CPU) to gnd : 3 oms
L7660 pin 2 (PP3V3_S5) to gnd : 1,2 oms

[mon2]

Flux and lift up one leg of L7660. Then with no power, measure the resistance to ground on each side of the PCB pads for L7660.

What is the resistance to ground measurements for each PCB pad? This is to confirm if the short is on the producer (U7650) side or the consumer / downstream side of this power rail.

[madjid911]

Flux and lift up one leg of L7660. Then with no power, measure the resistance to ground on each side of the PCB pads for L7660.

What is the resistance to ground measurements for each PCB pad? This is to confirm if the short is on the producer (U7650) side or the consumer / downstream side of this power rail.

L7660 pin is short 0,2 OHMS, pin 2 is 5,43KOHMS

PP3V3_S5 is produced by who please? by the processor? In this case it would be the processor which is short

[mon2]

PP3V3_S5 is produced by who please?

U7650.

By isolating this rail by removing inductor L7660, you have confirmed that the downstream or consumer side of this power rail is shorted. The upstream or aka producer side is ok. This power rail is likely shutting off naturally after sensing a higher than normal current draw due to the hard short to ground. The current sense pins for this rail exist @ P3V3S5_CSP2 and P3V3S5_CSN2.

Noting that the CPU / PCH is heating up with a voltage injection (should be 1 volt with 3A - 5A typically) then it is very possible that the CPU is defective. The PCH component does use the same PP3V3_S5 power rail.

Do also inspect the heat signature on the other side of this logic board for perhaps a capacitor that is hard shorted and causing this night glow on your thermal camera.

For your injection, consider to use 1 volt but keep increasing the current to confirm the shorted component. A capacitor (ceramic or tantalum) will crumble and split open under the pressure. Very possible the CPU / PCH will do the same.

[madjid911]

U7650.

By isolating this rail by removing inductor L7660, you have confirmed that the downstream or consumer side of this power rail is shorted. The upstream or aka producer side is ok. This power rail is likely shutting off naturally after sensing a higher than normal current draw due to the hard short to ground. The current sense pins for this rail exist @ P3V3S5_CSP2 and P3V3S5_CSN2.

Noting that the CPU / PCH is heating up with a voltage injection (should be 1 volt with 3A - 5A typically) then it is very possible that the CPU is defective. The PCH component does use the same PP3V3_S5 power rail.

Do also inspect the heat signature on the other side of this logic board for perhaps a capacitor that is hard shorted and causing this night glow on your thermal camera.

For your injection, consider to use 1 volt but keep increasing the current to confirm the shorted component. A capacitor (ceramic or tantalum) will crumble and split open under the pressure. Very possible the CPU / PCH will do the same.

I checked behind I didn't see any condenser heating, So the processor is really dead.
The client absolutely wants to recover his data, a swap seems complicated to me with the exchange of nand, the change of processor seems easier to me with the reballing machine, what do you think if I change the processor and clean the region me for the bios? I have any chance of finding all the data?

[reformatt]

A1706 has a lifeboat connector to isolate the SSD from the CPU. The intended method for data recovery is to plug in the Apple Data Migration Toolkit to that connector so that it will mount the SSD on another MAC.

It's no longer made (and Apple don't publicly acknowledge it's existence). Any that are around are really expensive and only usable on the MBP's from 2016/2017. Not worth trying to track one down for one data recovery, but most data recovery places will at least have access to one. Only problem is last time I queried a local mob for it, they quoted AUD $700 for recovery.

The NAND's are underfilled so you'd have to transplant to a donor board. Doable but these A1706 boards the copper they use is ultrathin and pads lift super easy. If you can do a CPU then go for it, but be sure to quote appropriately. Typically customers desire for the data is inversely proportional to the cost. The higher the cost, the less important it is!

[mon2]

New Apple Customer Data Migration Tool Kit 076-00236 for MacBook Pro 2016-2017 | eBay

[madjid911]

This seems like a waste of time, thank you for your help and advice.