A2338 M1 Not turning on - Repaired slightly but still having issues

[Stephen]

Repaired a customers computer, had a faulty APL 1097 chip and replaced, all well, not shorted anymore, however I am finally getting 12.28Volts on PPBUS_AON, however it stays at 12.28 and doesn't increase to the normal 12.60 Volts. I read that this is the cause

• If the voltage is approximately 12.28-12.35V, it may indicate a communication issue between the SOC and PMICs, or corrupt firmware.

Can someone point to me what would be the next step to fix this. I can feel we are getting close to fixing this board, however this is a data recovery and we need to fix the computer to recover the data. Any advice would be great. This was a previous coffee spill board, thank god no pads are destroyed.

[Stephen]

Update found the issue.... another APL chip, but this time the APL 1096 chip is faulty and shorted to ground. UGHHHHHHHHH

[Stephen]

Update: Replaced the APL 1096 chip, that short was removed it seems the short went away but now I am seeing a short on two areas next to the APL 1096 chip. I'll send the picture of my issue from my thermal imager, I replaced these areas and the blown CAP next to it as well, but they keep shorting. I cannot seem to understand what is going on since the APL 1096 short was removed. APL 1097 short was repaired but my god. I am still stuck at 12.28Volts on PPBUS_AON. I know this is the cause. Can someone point me in the right direction, I feel I am getting closer to resolving this issue but my God lol

[Stephen]

Update: Replaced with a new APL 1096 IC chip, short is now gone. Now I do not have a short on the board but still no power. Still getting 12.28volts on ppbus_aon and 5volts .01 amps reading. Then it set itself to 5volts .28amps jumping continually to .39A and fluctuating. I am in awe lol

FOUND ANOTHER short...lol this board is crazy...just got rid of that short. Now I'm stuck at 5volts .1 amp

[mon2]

Hey Stephen.

1) what was the outcome of the case where you joined the Rosmann group for support? Or did you post your resolution here in the forum?

2)

https://repair.wiki/w/MacBook_Pro_A2...a_at_5V_repair

[Stephen]

I am not getting .45 amps anymore I found that short so far. So now it's stuck at 5volts .01amps PPBUS_AON is stuck at 12.28volts. Still figuring it out. This board had some major spill damage, board has no corrosion anymore since we cleaned it and then took pics of the damaged areas before hand, however, with that said board looks brand new now, and fixed all shorts so far, but I know there is still a short somewhere. Going to take a break from this again and go further into repair Monday and pull up schematics and see what I find.

UPDATE: Checking voltage on PPBUS_AON I am getting 12.28volts but then it drops to 11.85 volts and goes back and repeats. So it drops to 11.85volts then climbs back up to 12.28volts. Does it like twice, then holds at 12.28 then 2 seconds later does it again. Loops like this.

[Stephen]

Hey Stephen.

1) what was the outcome of the case where you joined the Rosmann group for support? Or did you post your resolution here in the forum?

2)

https://repair.wiki/w/MacBook_Pro_A2...a_at_5V_repair

On 1) it was a shorted IC chip that corrupted the SSD data, had to repair the board and it was fixed after. Customer wasn't too happy but they were grateful to save the board since it was a maxed out 8TB 64GB i9 board.

[Stephen]

Side note: injected voltage 1volt 5 amps into board...NO SHORT is pulling ?? what is going on MANNNN lol This board has kicked my butt all week.

[mon2]

Try a DFU. One option will keep your data and rebuild the corrupt firmware.

[Stephen]

When attempting to try to connect into DFU (doesn't show yet) I see the USB C meter reading jump from .02 amps to .35 amps and not going higher. When I release the keys the amps drop back down to .02 amps. It stays at 5Volts. Any suggestions?

[Stephen]

Checked for shorts but this computer is not showing any shorts now so I am pretty in awe. This is a major repair and important but I can't seem to pin point this issue anymore. Any ideas MON?

When it is plugged into meter alone with power it is showing 5.14V and .02A to .08A Doesn't go higher than .08A

[mon2]

The board is not switching into 20v power delivery contract so review the basics of the CD3217 power delivery controllers. Confirm the presence of the LDO rails on each of these devices.

Confirm:

PP1V5_UPC0_LDO_CORE ; will be active only for the connector being used
PP3V3_UPC0_LDO
PP1V25_S2


CD3217 and T2 Power on Sequence - LogiWiki

Do note that the CD3217s are specific to the M1/M2 - older versions cannot be used. Are the CD3217s onboard the original from Apple / Foxconn or have they been replaced?

[Stephen]

The CD3217's have not been replaced and are the original on the board.

I do however have a spare donor M1 board with a two good CD3217 chips. same exact model that works (activation locked used for parts)

[mon2]

Check the LDO / power rails on each of these devices. Be sure they are present.

[Stephen]

PP1V5_UPC0_LDO_CORE = 1.5Volts
PP3V3_UPC0_LDO = 3.4Volts
PP1V25_S2 = 0.14Volts constant around .14volts dropping to .12 volts up and down. So nothing higher than 0.14volts

PP1V5_UPC1_LDO_CORE = 1.5Volts
PP3v3_UPC1_LDO = 3.4Volts
PP1V25_S2 = .14 volts as well


Also Thanks MON! you been a great help. I feel I am so close to getting out of the water here.

[obelix24]

I had simillar issue on a2338, macbook was taking only 5V 0,2A ... PPBUS_AON was only 0.07V (diode mode). When i was injecting voltage here (PPBUS_AON pad on F5200) 1V it doesn't show anything so i slightly increased voltage to approx 2,3V and on thermal camera i noticed that UR730 was getting warm after replacing macbook starts with 20V.

maybe this will help.

[Stephen]

Isn't it kind of risky injecting voltage higher than 2volts into the line for the CPU? That concerns me a little.

[Stephen]

I raised the voltage on injection and still no short. So I am leaning towards mon2 on this one. I went ahead and swapped out UR730 just for the hell of it just in case and still no power. same result. so most likely CD3217 issue but I will wait on his response to confirm.

[mon2]

PP1V5_UPC0_LDO_CORE = 1.5Volts OK
PP3V3_UPC0_LDO = 3.4Volts OK
PP1V25_S2 = 0.14Volts constant around .14volts dropping to .12 volts up and down. So nothing higher than 0.14volts NOT OK

PP1V5_UPC1_LDO_CORE = 1.5Volts OK
PP3v3_UPC1_LDO = 3.4Volts OK
PP1V25_S2 = .14 volts as well NOT OK

PP1V25_S2 are local LDO power rails that are used to power the single wire BMC circuit (aka CCx pin) for the power delivery contract negotiations. Based on these measurements, if you were to apply a PD bus analyzer, suspecting that you would not be able to see any quality communication between the external Type C power adapter and this logic board so the power up default of 5 volt remains static.

USB Type C connectors are reversible by design. Rotate the power head by 180 degrees and test again. If these LDO rails continue to fail - to me this means they are defective. From my understanding, each such LDO rail is used to power the internal BMC circuit and no other external load exists on the component. Just a bulk filter cap on each.

Do review these non-working LDO rails (1v25) by removing all power -> meter in diode mode -> red meter probe to ground; black meter probe onto this local PP1V25_S2. What is the diode mode measurement?

Repeat for the other PD controller. USB Type C is an evil design. We reviewed the 1.0 spec years ago and I almost burnt my hands after perhaps an angular removal of the connector cable. Reported our findings to VIA Semiconductor (Taiwan) who built the e-marker IC that was present inside of the active cable. They said they would improve the silicon. In summary, imagine when the PD contract is @ 20v and the cable head is removed while it is 'hot'. This 20v can swing onto the dainty fine pitched contacts nearby and cause the low voltage signal lines to be hit with 20v. Zap goes the PD controller.

Do wait on other comments but suspecting this may be a fault. Post each measurement before we can conclude this to be the root cause. Would you have a thermal camera? What does it show for these CD3217 when you add the power adapter? With the very low current draw, doubt there will be anyone in heat.

[Stephen]

PP1V25_S2 are local LDO power rails that are used to power the single wire BMC circuit (aka CCx pin) for the power delivery contract negotiations. Based on these measurements, if you were to apply a PD bus analyzer, suspecting that you would not be able to see any quality communication between the external Type C power adapter and this logic board so the power up default of 5 volt remains static.

USB Type C connectors are reversible by design. Rotate the power head by 180 degrees and test again. If these LDO rails continue to fail - to me this means they are defective. From my understanding, each such LDO rail is used to power the internal BMC circuit and no other external load exists on the component. Just a bulk filter cap on each.

Do review these non-working LDO rails (1v25) by removing all power -> meter in diode mode -> red meter probe to ground; black meter probe onto this local PP1V25_S2. What is the diode mode measurement?

Repeat for the other PD controller. USB Type C is an evil design. We reviewed the 1.0 spec years ago and I almost burnt my hands after perhaps an angular removal of the connector cable. Reported our findings to VIA Semiconductor (Taiwan) who built the e-marker IC that was present inside of the active cable. They said they would improve the silicon. In summary, imagine when the PD contract is @ 20v and the cable head is removed while it is 'hot'. This 20v can swing onto the dainty fine pitched contacts nearby and cause the low voltage signal lines to be hit with 20v. Zap goes the PD controller.

Do wait on other comments but suspecting this may be a fault. Post each measurement before we can conclude this to be the root cause. Would you have a thermal camera? What does it show for these CD3217 when you add the power adapter? With the very low current draw, doubt there will be anyone in heat.

Thermal camera shows no heat on the CD3217 chips.

DIODE mode, removed power checking rail .25 volts on PP1V25_S2

Flipped 180 on usb meter, all voltage is same. when having input power.