(347) 552-2258 [email protected]

MacBook Pro (820-02773 - A2338 M2)

S

szape

New member
Hello, my MacBook Pro (820-02773 - A2338 M2) suddenly died and was not water damaged.

This is my own macbook and this is my first repair attempt (previously repaired ipads, tablets and consoles which had damaged ports or visible faulty capacitors etc.) I have thermal camera, hot air station, microscope and micro soldering station.

I have schematics and board view of the logic board.

The usb-c amp switches to 20v and then reboots in a loop instantly. Both the port have the same behaviour. Pressing power button nothing happens, touchpad is also not responding, but keyboard Works because i can enter DFU mode. Tried to recover/revive with Apple Configurator but after the apple logo it shows error "Unexpected device state 'DFU' expected 'RestoreOS' (4014)". After apple logo disappears backlight is still on for about 30 seconds and then it turns off too.

Charging battery works in DFU mode.

Removed logic board, under thermal camera L5800 is hot (i can upload picture if needed).
PPBUS_AON is fluctuating between 12V-12.4V (diode mode reads 0.440 instead of 0.496)
3V8_AON is present, 3.8V stable. PP3v8AON_PH1, PP3v8AON_PH2, PP3v8AON_PH3, PP3v8AON_SW2, PP3v8AON_SW3 are 3.8V, PP3v8AON_VSW1 is pulsing between 3,4V-3,9V.
5V is present and steady.
PP2V5_AWAKE_NAND 2.58V looping.
Can't find any short.

I've inspected the board under microscope, Q5840 might have a crack on the top but not completely sure. might be a scratch, i can see only at the top but not at the sides (i can upload pictures if needed).

Please help me.

Thank you.
 
2informaticos

2informaticos

Administrator
Staff member
First of all, welcome to the forum!

The special keys used by DFU are controlled by hardware (UT103) to generate RSLOC_RST_L signal.
If you can't access recovery menu, or boot menu, then keyboard doesn't really work.

Just in case, check if you have more luck with DFU Blaster.
 
S

szape

New member
Nevermind. Reassembled the macbook but nothing happens when i press DFU button in DFU blaster. I have an older mac mini without usb c port. Is this a problem?

Still can enter DFU mode via button combination.
 
S

szape

New member
Borrowed an M1 macbook Air, enters DFU mode flawlessly. Tried to restore but nothing happens, error is: "The system cannot be restored on this device. Gave up waiting for device to transform from DFU state to DFU state. ((com.apple.MobileDevice.MobileRestore . 0xFCA (4042))
 
S

szape

New member
Tried to connect charger cable before DFU cable.
Tried with both charger and battery connected.
Tried with another host machine, with newer macOS installed.
 
S

szape

New member
Something has changed... When I enter DFU mode via keyboard, there is no more apple logo at the restore process and the gives the same 4042 error, not 4014 as before.
Disassembled board again and noticed that P3V8AON_VSW1 is fluctuating between 3.8 and -0.2 Volts. PP3v8AON_VSW1 was pulsing between 3,4V-3,9V before. And L5800 is definitely hotter under thermal camera.

Might be something else than NAND? I hope, i won't replace it, i will sell the macbook for parts if it is the problem 100%.
 
S

szape

New member
P3V8AON in diode mode is 0.144V which should be 0.150V according to schematics and PPBUS_AON is 0.450V instead of 0.496V in diode mode.

P3V8AON_VSW1 is fluctuating on charger (just at that side of the L5800 inductor other inductors are stable 3.8V) when no battery installed. P3V8AON_VSW1 at L5800 is stable 3.8V when on battery and not connected to charger.

When battery installed i measure 11.05v on PPVBAT_AON_CHGR_REG without charger.
 
Last edited:
2informaticos

2informaticos

Administrator
Staff member
"0.144V which should be 0.150V"
Do you really think that is a BIG difference?
Checking on your board with different multimeters, you will get different readings.

BTW, stop checking DC voltage on the input side of the coil from buck converters.
Compare input/output with an o-scope and you will understand why!
 
S

szape

New member
2informaticos said:
"0.144V which should be 0.150V"
Do you really think that is a BIG difference?
Checking on your board with different multimeters, you will get different readings.

BTW, stop checking DC voltage on the input side of the coil from buck converters.
Compare input/output with an o-scope and you will understand why!
Click to expand...
No, it's normal. As i wrote earlier i'm a beginner at these type of work so sorry for being noob. Please tell me what to do next, i do not have an oscilloscope (my multimeter is capable but i'm still a beginner). :)

If you are sure that this is the NAND i give up. But previously had different values and i had apple logo in the recovery process, now it does not appear at all even if i reassembled the macbook. Unfortunately i cannot afford to buy an other mac so i must try my best to save it.

From now on i will only do what you tell me to do.

Thank you.
 
2informaticos

2informaticos

Administrator
Staff member
You don't need an o-scope to diagnose the motherboard.
Only to see the signals on mentioned coils and understand why ALWAYS MUST check DC voltage on the output pin!

Where did you get the replacement NANDs?
Are you sure you didn't damage anything in the process?

Please, don't quote an entire post visible few centimeters above.
Write directly into bottom reply box, instead of clicking Reply button.
 
S

szape

New member
The board is not repaired in any way, NAND is original, was not replaced and the logic board is in original state. So i just disassembled the mac and removed the logic board did not want to do anything until you suggest to do so. Just inspected it under microscope for visible faults and measured a few things and tried DFU restore.

Okay, will not quote.

As i already wrote at the first post I've inspected the board under microscope, Q5840 might have a crack on the top but not completely sure. might be a scratch, i can see only at the top but not at the sides. Can this be the problem? I did not scratch it, others are fine, no scratches/cracks.
 

Attachments

  • 0031.jpg
    0031.jpg
    295 KB · Views: 2
Last edited:
2informaticos

2informaticos

Administrator
Staff member
If you get correct 3V8_AON, Q5840 should be good.

Sorry, I understood wrong your mention about NANDs.
But many times sudden death is caused by faulty NANDs; no need to be shorted to ground.
 
S

szape

New member
Thank you.

So you suggest a NAND replacement? I'll try to find someone who can do it. Made it successfully on an iPad Air 2 many years ago but i'm not risking it anymore.
 
S

szape

New member
Took a deep breath, removed NAND carefully (managed to do it perfectly, no damaged pads, etc.), but nothing changed. When i plug the usb-c charger it switches from 5V to 20v and then reboots in a loop instantly.

Is it worth to order a new NAND and solder it to the logic board or this is a different problem?

Thank you.
 
S

szape

New member
Okay, i ask differently. If i remove a NAND from a working board does it make the same the symptom i have? (When i plug the usb-c charger it switches from 5V to 20v and then reboots in a loop instantly.)

Thanks.
 
You must log in or register to reply here.
Top