Macbook Pro 820-00840-a not powering on

[Simplebeans]

Hey guys i just got myself a Macbook Pro 820-00840-a 2017, the seller said they had spilt liquid over it and since then it has never turned on.

I am using a Apple 61W USB-C Power Adapter and USB-C Charge Cable and on my USB tester i can see when i plug into the power i get around 5.23V for about 2 seconds and then it jumps up to 20.43V and stays at 0.00amps. I am currently testing with the board out of the chassis disconnected from the battery and only plugged into the USB-C Charge Cable.

I was greeted with some corrosion on the bottom of the board, 3 areas in particular that looked concerning to me.

First this J6951 connector with the cable detached and the Q7065 mosfet looking a bit worse for wear as well. I have cleaned them up they were much worse and this is the best i could get them for now, please see the images below:





The good news is the top side of the logic board there is no corrosion or any signs of liquid ingress however when I was digging around and investigating why PPBUS_G3H is showing me 0V on U7000 (ISL9239) chip I noticed in the top right hand corner it has slightly blackened in colour and broken off, see the image below at the red arrow:



I went ahead and measured capacitors C7065, C7066, C7067, C7068 and C7069 (Meter in diode mode, red probe on ground point and black probe to each side of the capacitors) and both sides are beeping (shorted), the capacitors are in circuit with Q7065.

My thinking on this is it could be the Q7065 or the U7000, i am leaning on the side of the Q7065 for now because PP3V3_G3H is being produced so the U7000 has PM_EN_P3V3_G3H signal. But i realise anything’s possible here (liquid).

What are your thoughts on this? I am thinking first to reflow/replace the Q7065?

[mon2]

Re: Macbook Pro 820-00840-a not powering on

1) Clean up the FPC cable contacts with alcohol.

2) The mosfet is sharing pins 5-6-7-8 and looks to be ok but again, use IPA to clean it up as much as possible.

3) The ISL9239 looks to be chipped and may be damaged but ignore that for now.

Meter in diode mode, black probe on ground point and red probe

No. In diode mode, RED probe (yes RED probe) goes to ground & Black probe to the point under test (DUT).

[mon2]

Re: Macbook Pro 820-00840-a not powering on

Not enough coffee for me. Ignore this post.

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

No. In diode mode, RED probe (yes RED probe) goes to ground & Black probe to the point under test (DUT).

Hi i realised after i had typed it that i made the typing error concerning testing the capacitors in my first post. So I have tested it this way already (I have just edited my first post to correct this). Not enough coffee for me sorry about that mon2! Ha.

1) Clean up the FPC cable contacts with alcohol.

2) The mosfet is sharing pins 5-6-7-8 and looks to be ok but again, use IPA to clean it up as much as possible.

I have cleaned up these about 3 times with cotton buds/ISOPROPYL ALCOHOL 99.9%. This was the best i could get it at the time.

I will try again one more time for good measure.

3) The ISL9239 looks to be chipped and may be damaged but ignore that for now.

Ok.

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

I have cleaned up these about 3 times with cotton buds/ISOPROPYL ALCOHOL 99.9%. This was the best i could get it at the time.

I will try again one more time for good measure.

mon2 just redone the cleaning and tested still the same results showing - On my USB tester when i plug into the power I get around 5.23V for about 2 seconds and then it jumps up to 20.43V and stays at 0.00amps.

Also to add to my first post i measured R5380 and R5381 (Meter in voltage mode black probe on ground point and red probe to point 1 on each resister) and got a reading gof 3.40V for both so SMBUS_SMC_5_G3H_SCL/SMBUS_SMC_5_G3H_SDA are receiving power on the respective pull up resistors.

[mon2]

Re: Macbook Pro 820-00840-a not powering on

Post the voltage of PPBUS_G3H @ fuse F7000, pin #1 and pin # 2.

Then, remove all power / no battery. Meter in resistance mode (2k scale).

Check the resistance to ground of F7000, pin # 1 and pin # 2.

Have to leave for a while to move 2 stoves with muscles I don't have - will check back later. It is Arnold / Wolfgang Puck time. (I'll be back??).

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

Post the voltage of PPBUS_G3H @ fuse F7000, pin #1 and pin # 2.

Then, remove all power / no battery. Meter in resistance mode (2k scale).

Check the resistance to ground of F7000, pin # 1 and pin # 2.

Have to leave for a while to move 2 stoves with muscles I don't have - will check back later. It is Arnold / Wolfgang Puck time. (I'll be back??).

HAHA Arnold!

Ok meter in voltage mode (20V) black probe to ground and red probe on pin 1/2 F7000 both measuring at 0V.

Then removed all power meter in resistance mode (2KΩ) black probe on ground point and red probe at each pin, now both measuring .002Ω which tells me its a dead short? The .002Ω is the resistance through the probe wires.

Just another note my USB tester reaches 20.43V but does keep moving between 20.43V to 20.46V so it does fluctuate albeit not by too much.

I went ahead and tested the Q7065 -multimeter into continuity mode red probe at PIN 5 (Drain) and black probe to any PIN 1 or 2 or 3 (Source) all measuring around .400 no beeping. So mosfet must be a good un.

[mon2]

Re: Macbook Pro 820-00840-a not powering on

Am back - kicking the can till next week. Need to eat some spinach in the meantime.

Then removed all power meter in resistance mode (2KΩ) black probe on ground point and red probe at each pin, now both measuring .002Ω which tells me its a dead short? The .002Ω is the resistance through the probe wires.

Whoa! Not good. Yes, there must be a shorted cap on this rail.

Remove F7000 off the board and repeat these measurements (with no power) for both pads # 1 and # 2 of F7000.

You should see the short on one of these 2 sides - is it the provider side (F7000, pad #1) same side as the ISL9239 (U7000) or the consumer side = pad # 2 of F7000 ?

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

Am back - kicking the can till next week. Need to eat some spinach in the meantime.


Whoa! Not good. Yes, there must be a shorted cap on this rail.

Remove F7000 off the board and repeat these measurements (with no power) for both pads # 1 and # 2 of F7000.

You should see the short on one of these 2 sides - is it the provider side (F7000, pad #1) same side as the ISL9239 (U7000) or the consumer side = pad # 2 of F7000 ?

F7000 removed. And measured again but this time multimeter in resistance mode (20KΩ), PIN 1 is 5.3KΩ (upstream) and PIN 2 is showing 0Ω (downstream).

[mon2]

Re: Macbook Pro 820-00840-a not powering on

Ok so you definitely have a short circuit on the downstream (consumer) side of the logic board. Check for any obvious shorts or damaged parts. Perhaps the color will be off, etc.

Moving forward, if you attach an adapter (no battery) -> what voltage do you see @ pin # 1 of F7000 pad ? (No fuse in this test).

You are checking to see if U7000 is able to create the PPBUS_G3H rail now that the short circuit to ground is removed due to F7000 being off the logic board.

Would you have an adjustable power supply to use for voltage injection ? First need to check if there are any obvious signs of a damaged part but then can move to voltage injection.

For us, we like to use 0v8 on the power supply. Red probe of the adjustable power supply on pad # 2 of F7000. Ground to ground. Need to keep the voltage down to a safe level - just in case the same rail is shorted to the CPU rail. Too high of an injection can cause death to parts like the CPU.

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

Ok so you definitely have a short circuit on the downstream (consumer) side of the logic board. Check for any obvious shorts or damaged parts. Perhaps the color will be off, etc.

Moving forward, if you attach an adapter (no battery) -> what voltage do you see @ pin # 1 of F7000 pad ? (No fuse in this test).

You are checking to see if U7000 is able to create the PPBUS_G3H rail now that the short circuit to ground is removed due to F7000 being off the logic board.

Would you have an adjustable power supply to use for voltage injection ? First need to check if there are any obvious signs of a damaged part but then can move to voltage injection.

For us, we like to use 0v8 on the power supply. Red probe of the adjustable power supply on pad # 2 of F7000. Ground to ground. Need to keep the voltage down to a safe level - just in case the same rail is shorted to the CPU rail. Too high of an injection can cause death to parts like the CPU.

Ok will have a look at the board again tommorrow, give it another once over. Whilst removing the F7000 fuse i knocked off the C7055 , gotta love these hot air stations or is it my soldering skills haha! I will also solder that back on tommorrow.

Attaching the USB-C Charge Cable (no battery) now i measured again at F7000 PIN 1 and now i am getting a reading of 13.05V so it looks like U7000 is a good un also! That's awesome.

Yes i have a bench power supply. So its best to set that to 0.8V and is there a recommendation on amp setting as well? Is there anything you recommend you could do as a cursory check/action with your comment about the CPU so I can give the CPU the best chance of survival before injecting? (aka the CPU sugeon haha!)

[mon2]

Re: Macbook Pro 820-00840-a not powering on

That is good news for the U7000 operation. To be honest, was originally scared of those. Now after a few repairs (well more than a few) - confident I can replace one with a hot air tool in my left toe. The secret is flux + high heat + low air but enough to melt it and nudge it with tweezers. Then clean it up with low temp solder + IPA to clean the site + flux again and apply a fresh unit. Make it dance once the solder balls start to melt. Be patient and use enough heat and air to make it dance and bounce the chip back on its own to clear up any hidden solder balls.

On the current, 3A / 5A is good. Honestly, we are using a 20A / 30A injector tool from Aliexpress but it is not of concern as the 3A / 5A draw will be ample to heat up the bad boy on the board. Mist the caps with alcohol (use Q-tip) and watch for the evaporation to locate the bad part.

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

That is good news for the U7000 operation. To be honest, was originally scared of those. Now after a few repairs (well more than a few) - confident I can replace one with a hot air tool in my left toe. The secret is flux + high heat + low air but enough to melt it and nudge it with tweezers. Then clean it up with low temp solder + IPA to clean the site + flux again and apply a fresh unit. Make it dance once the solder balls start to melt. Be patient and use enough heat and air to make it dance and bounce the chip back on its own to clear up any hidden solder balls.

On the current, 3A / 5A is good. Honestly, we are using a 20A / 30A injector tool from Aliexpress but it is not of concern as the 3A / 5A draw will be ample to heat up the bad boy on the board. Mist the caps with alcohol (use Q-tip) and watch for the evaporation to locate the bad part.

mon2 hahaha that’s never really been any of my concern because I do actually enjoy soldering even if they are BGA chips onto boards and that dance makes it all worth while, while you wait and hope it goes on! Ha.

My reservation on soldering is what happened to me today when other components are taken as collateral damage when you are not even having that in the cross hairs. Looking at the U7000 that’s magnified ten fold I just seen how many close proximity components it has. Do you have any advice on this? Any advice would be appreciated.

I have some 22AWG solid core hook up wire is this wire good enough for injecting voltage you reckon or do you recommend any other? I have used this wire previously with no issues but I always wanted to ask a seasoned technician on this. Better quality wire for voltage injection if there is such a thing.

[mon2]

Re: Macbook Pro 820-00840-a not powering on

In hindsight, recommend:

a) low air pressure if possible
b) if not possible - when using higher air pressure, keep the nozzle inside or near the inside border of the ISL9239 / ISL9240 device. That is, limit where the air nozzle leaks its air for the surrounding parts.
c) best if you can apply kapton (aka polymide) high temp tape onto the small parts you do not wish to have flying around. I totally agree on these tiny 0201 parts are easy to disturb. Still hunting for a nice microscope for soldering. Undecided if it will be the style that is used by Alex (Northridge Fix) or the common trinocular style with a nice HD camera. So far, finding the resolution is higher grade with the style that Alex is using.

For the voltage injection, our 'short killer' tool is fitted with what look like normal meter probes - granted that they are short (no pun) in length.

I think you will be fine with 22AWG but more concerned about how flexible is this wire? You do not want to rip off any SMD pads in the process. Even multi-strand wire should be ok. When we do this process, we hit the pad and then turn it off to just pulse and view the results. So far, this has worked for us.

On Friday, had a short on PPBUS_G3H and used 0v8 with our 'short killer'. After just one jolt, the short was removed and allowed us to have PPBUS_G3H once again on this defective board.

This is what we have - only small complaint is that the display is quite dim. I think the vendor just dumped the dim display tool onto us.

https://www.aliexpress.com/item/4000...rch-mainSearch

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

Yeah that may be a good idea especially applying the kapton tape to help minimize the movement of the other components. I will have to get myself some Kapton tape and try that next time.

Thanks for mentioning about the 22AWG solid core hook up wire and taking off the pads, I never even thought about the flexibility. It is not completely rigid but it could be more then the typical multi-strand wire so I will have to get some of that then ordered also.

The other issue i am mindful of is when applying heat with the hot air station especially near to components where there is a FPC connector, there always that one with the plastic on the connector to be weary of incase of melting it and heating it up. I considered Kapton tape before for this scenario but i could not find one with a high temp resilience, thinking along the lines of blowing 450°C.

Thanks for your help so far mon2 will be hopefully at logic board later on tonight.

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

In hindsight, recommend:

a) low air pressure if possible
b) if not possible - when using higher air pressure, keep the nozzle inside or near the inside border of the ISL9239 / ISL9240 device. That is, limit where the air nozzle leaks its air for the surrounding parts.
c) best if you can apply kapton (aka polymide) high temp tape onto the small parts you do not wish to have flying around. I totally agree on these tiny 0201 parts are easy to disturb. Still hunting for a nice microscope for soldering. Undecided if it will be the style that is used by Alex (Northridge Fix) or the common trinocular style with a nice HD camera. So far, finding the resolution is higher grade with the style that Alex is using.

For the voltage injection, our 'short killer' tool is fitted with what look like normal meter probes - granted that they are short (no pun) in length.

I think you will be fine with 22AWG but more concerned about how flexible is this wire? You do not want to rip off any SMD pads in the process. Even multi-strand wire should be ok. When we do this process, we hit the pad and then turn it off to just pulse and view the results. So far, this has worked for us.

On Friday, had a short on PPBUS_G3H and used 0v8 with our 'short killer'. After just one jolt, the short was removed and allowed us to have PPBUS_G3H once again on this defective board.

This is what we have - only small complaint is that the display is quite dim. I think the vendor just dumped the dim display tool onto us.

https://www.aliexpress.com/item/4000...rch-mainSearch

Voltage injection session completed, i started with 1V and increased from 3A-5A, when i had the amperage at 5A, i felt around and the PCH/CPU seemed cool. There was one area of the board that i kept coming back to the capacitor C6405 and J6602 next to it. When i first received the logic board there was also some liquid spill around this area with some signs of corrosion on the connector points of J6602 and the cable attached to it. I cleaned this area up, see the image below for how it looks now:



This area plus the area directly on the top side of the board when you flip it over was the areas heating up noticibly in perticular U6400 and U6420 are heating up also. These seem to be the drivers for '4X MONO SPEAKER AMPLIFIERS'. I tested alot of the capacitors C6405, C6404, C6403, C6424, C6423 etc on the PPBUS_G3H_SPKRL line and all the ones i tested (meter in diode mode red probe to ground point and black probe to each side of the capacitors) were giving me a beep on both sides? It seems the PPBUS_G3H_SPKRL line is created from PPBUS_G3H line.

I sprayed on the ISOPROPYL ALCOHOL 99.9% and i could see the evaporation occuring on the C6405 then some around and on the J6602 and then also it looked like on the U6400 and U6420 chips.

What are your thoughts?

[mon2]

Re: Macbook Pro 820-00840-a not powering on

A few thoughts:

1) Pour some alcohol over one group -> apply the voltage injection again and see if any of the wet block of caps evaporates the alcohol due to the heat. Then repeat with the 2nd block of caps on the same power rail.

This should allow you to locate the bad cap.

2) Another idea is to flux & remove the current sense resistor @ R6499. With this part out of the circuit, you will know for sure that the above caps have a bad participant on the logic board. With the current sense resistor removed, meter in resistance mode -> check the resistance to ground for any of the caps that you are suspecting. Guessing that you will see a very low resistance to ground within this group and now a higher resistance to ground for the other (isolated) side of the removed current sense resistor.

If you decide to do this, then the point of injection of course will shift to these caps that are showing the low resistance to ground.

If you are confident the heat is from these parts then you could dial up the voltage a bit but personally am afraid that a downstream short that is not visible could be crossing over to the lower voltage rails (ie. CPU / PCH).

Also with a heatsink on the CPU, it is difficult to gauge if the CPU is heating up but perhaps you have already found the area causing this fault.

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

A few thoughts:

1) Pour some alcohol over one group -> apply the voltage injection again and see if any of the wet block of caps evaporates the alcohol due to the heat. Then repeat with the 2nd block of caps on the same power rail.

This should allow you to locate the bad cap.

2) Another idea is to flux & remove the current sense resistor @ R6499. With this part out of the circuit, you will know for sure that the above caps have a bad participant on the logic board. With the current sense resistor removed, meter in resistance mode -> check the resistance to ground for any of the caps that you are suspecting. Guessing that you will see a very low resistance to ground within this group and now a higher resistance to ground for the other (isolated) side of the removed current sense resistor.

If you decide to do this, then the point of injection of course will shift to these caps that are showing the low resistance to ground.

If you are confident the heat is from these parts then you could dial up the voltage a bit but personally am afraid that a downstream short that is not visible could be crossing over to the lower voltage rails (ie. CPU / PCH).

Also with a heatsink on the CPU, it is difficult to gauge if the CPU is heating up but perhaps you have already found the area causing this fault.

1) I tried this on the areas around U6400 and U6420 and it seems apart from the chips themselves heating up the same caps on both chips are evaporating so becuase of this uniformity I am thinking maybe it is not related to these two chips then. I could be wrong.

2) And this sounds a good set up for the future, if required. Triangulate and isolate haha! See below.

So i tested again and i am confident that C6405 is heating up significatly, it seems when i was pressing down on it with my finger it was burning at one stage. You reckon replace first?

[mon2]

Re: Macbook Pro 820-00840-a not powering on

Pluck C6405 (with flux & hot air) off the board and check the resistance again for the group.

[Simplebeans]

Re: Macbook Pro 820-00840-a not powering on

Pluck C6405 (with flux & hot air) off the board and check the resistance again for the group.

Ok will do, how about the J6602 connector what do you suggest the best way to protect while i am heating up the capacitor, because it has plastic. Whilst i am pointing the hot air at it i wanted to try and make sure it does not heat up to much or melt. Is Kapton tape the best solution for this?