Macbook Pro M1 (A2338) 820-02020 not charging.

[Gueric]

I returned home from college. My MacBook battery was flat. Plugged it in. Nothing. Won’t boot. Waited for an hour. Nothing.

When plugging the charger : 20v at 0.02 amps.
pressing any key will make the « low battery » icon show up. Then switch to 5v and 420mA constant. Trackpad clicks, keyboard works.

Decided to put the board into another M1 body with a charged battery, it’s booting perfect, still 5v and 420mA tho, and battery is not charging.

I diode moded every pin of U5200 and they’re all fine. I know very little about Apple silicon machines so any help would be appreciated.

hey Mon2. I hope you’re doing fine.

edit : basically : in the Mac that still has some juice, board boots fine, works fine, battery percentage is showing, but no charging at all, and board won’t pull more than 5v 420mA.

inside my actual Mac it wont boot at all since the battery is flat and charging system is defective.

[mon2]

Hey

Attach the power adapter. Carefully, confirm the voltage to ground of:

PP1V8_S2 ; @ C5280
PPCHGR_VDDA ; @ C5275
PPCHGR_VDDP ; @ C5277

Post the voltage for each.

Next, remove all power. Meter in resistance mode. Measure the resistance across CHGR_CSI_FILT_P & CHGR_CSI_FILT_N. You can measure these via probes on C5221 & C5222. We are checking the input resistance path for the current sense circuits that power U5200. Post the resistance in ohms.

[Gueric]

PP1V8_S2 is exactly 1.800v.
PPCHGR_VDDA and PPCHGR_VDDP are both 5.175v.

Resistance across C5221 and C5222 is 02.75 ohms.

By the way, if that can help, even though the board is working fine at 5v and no charging, bare board pulls 5v and 190mA, then drops to 0.00A after 2-3 seconds.

and originally this Mac had a defective slave CD3217 (might not be relevant to this charging issue, but I wanted to add this detail)

thanks mon2 :-)

[mon2]

Review the cd3217s: (see details)

https://repair.wiki/w/MacBook_Pro_A2...w_at_5V_repair

Review the battery charge resistance path on the isl9240. With no power.

[Gueric]

mon2 are you sure a CD3217 could be defective if the Mac fully works except for the « battery not charging » part ?

Even though yes, it’s stuck at 5v, but it negotiates 20v at first when plugged in for the first few seconds and drops to 5v when interacting with it (pressing any key or even the trackpad)

[mon2]

I think the CD3217 are ok. You can test all the permutations of the controller (flip by 180 degrees etc) to see if that makes a difference. It is a positive sign that 20v is selected but if it is resetting then some trigger is causing this fall back. The CD3217 tests are to confirm the LDO rails are ok on the pd controller. Do you own a thermal camera? If yes, check if any part is thermal throbbing in the camera.

I recall reformatt observing that Q5230 is a point of failure.

Remove all power. Meter in resistance mode. Measure this DUAL mosfet using resistance mode checks:

source / drain
source / gate
gate / drain

Checking for a low resistance of the mosfet. Note again, this is a DUAL mosfet. No power during the resistance checks.

[Gueric]

Okay well you were right. The board behaves exactly the same no matter the port and USBC orientation EXCEPT for one orientation, when plugged one way into the master CD3217 (the one i didnt replace). The ampmeter doesn’t even turn on and PPBUS is 2.7v, dropping to 1.5v after a few seconds. Tried without the ampmeter and same result. Boards works with 3 out of the 4 possible « connections» Turning the USBC around makes it work fine.

What does this even mean? Is it a CD3217 issue or something else ?

also some details : i never replaced the master CD3217. I replace the slave, which works perfect in either orientation. The board never had water damage and i didnt replace anything else.

thanks mon2

[mon2]

Do not have a whole of experience with the PD controllers (yet) but can state that cross shorts are a real thing. I am also on reddit forums and see daily reports of 'defective motherboards' that will not charge. Such a nasty design.

Start with removing the USB-C paddle board(s). Clean off the contacts with IPA to be sure they are without any debris. Do inspect the side that is posing the fault you noted. Does it look dirty? Damaged contacts? A good tool we use is the Qianli tail tester which is under $20 USD and is a quick way to test the USB-C ports (both sides) + acts like a USB-C power meter. Nice design of a product.

From the articles by piernov and others - if one pd controller is not working, the group will fail to function. Suspecting that the side with the failing USB-C port is the root cause of your problems. Since you are working with a M1 - the sourcing of a valid replacement may be an issue. I do believe you will need to source a replacement that is from the same logic board and from the same location for the donor component. This is because each such controller has its own address and different firmware in rom. Certainly correct me if you find otherwise.

[Gueric]

Hey mon2 im back. I did a few measurements, and tried measuring voltage with the misbehavingCD3217 plugged in. When plugged in the way that doesnt even turn on the ampmeter, measuring PPDCIN_USBC_AON gives me 0v. PPBUS_AON gives me around 2,7v then slowly drops below 1v overtime. PPDCIN_AON_CHGR_R give me a few milivolts at most.

I think I’m with you on this one. I tried cleaning the port as best as I could, try with a different USBC flex cable, nothing worked.

honestly I don’t have enough experience to point at anything except this master CD3217 chip. Thankfully I have one laying around from this same position and same board number. I don’t really know what to try or measure other than replacing the chip itself.

This whole experience allowed me to discover that M series machines can boot from a 5v input and a potential defective USBC controller.

what do you think ? Is there any other option or location to measure voltages or resistance from before attempting a replacement ?

mon2

[Gueric]

Also, it’s a standard M1 Pro from 2020. Only 2 ports and thus 2 CD3217s.

[Gueric]

Sorry important correction : PPDCIN_CHGR_CHGR_R is around the same voltage as PPBUS_AON when plugged in the bad way (around 2v, dropping fast)
measuring on the fuse just next to the CD3217s never gives me more than 100mV. (When plugged in the bad way into the master CD32)

okay some news… what the hell.
Both ports randomly decided to negotiate 20v again?
still, on the master CD3217 ports, it won’t work when plugged the « wrong » way. But the other way it does negotiate 20v aswell. Bruh. All I did was heating the master CD3217 for a few seconds at 150c then waited for a few minutes.

[Gueric]

Some news again… mon2
So the CD3217 port i replaced originally is now rebooting sometimes, and the other port works but only in 1 orientation and doesn’t reboot.

so i decided to put the machine back together. It is pulling 20v and 2.4Amps, so way better.

tried powering it on, it works, but acts strangely on the login screen. After a few seconds on the login screen, it reboots, boots to the login screen, and 5 seconds after reboots again. I really don’t know what to do at this point this whole thing is about to make me quit

right now i turned the machine off and closed the lid and it appears to be charging, 20v at 2.4A.

[mon2]

Allow it to charge for a few hours then test again. Also try to boot from internet to rule out a corrupt OS at this time.

Steps to Boot from Internet Recovery

1. Restart the MacBook: Click the Apple logo and select Restart.
2. Hold Command + Option + R: Immediately press and hold these keys until you see a spinning globe.
3. Connect to Wi-Fi: If prompted, select a Wi-Fi network and enter the password.
4. Wait for the Recovery Mode: The Mac will connect to Apple’s servers and load the recovery tools.

[Gueric]

Okay so mon2 i let it sit for around 3 hours, until the current draw dropped to 1A. Upon opening it it seemed to have been completely corrupt somehow, my password wasn’t recognized, keyboard was misbehaving, and after going into the startup options it couldnt speak with apple Activation servers. I had to manually erase the mac to get it to activate, and now it’s currently reinstalling MacOS. Oh my freaking god.

and still one of the 2 ports isn’t working at all now. Crazy. If the Mac fully works ill stick to just 1 port.