A2442 MacBook Pro 14" 5V restart

[Stephen]

Got a 14" MBP constantly restarting at 5V .04 Amps. Meter turns on, then off and repeats. I am very familiar with this issue in A1990, A1707 or A2141 etc boards however the power sequence for this board is different and I am trying to figure out if we have a Power Sequence Diagram for these boards yet? Also if you have came across this issue and fixed it please point me to the right direction. Also the board is new, no water damage nothing no drop damage and in perfect condition, no visible blown caps and is in PERFECT condition from what we see, it has never been worked on either and looks to me it has an issue somewhere.

[Stephen]

Also from my experience usually this would be a case of a bad USB C Chip on the board. There are 4 on this board and I do not see any pin holes in them, I could be wrong though for sure.

[mon2]

Remove all power. Meter in resistance mode.

Measure the resistance to ground of PPBUS_AON.

Post the reading.

[Stephen]

When measuring I’m getting 10 M Ohms and dropping.

[Stephen]

Now it’s down to 4 M OHms

[Stephen]

Mesuared again it rose to 15 M Ohms and dropping.

[mon2]

So it is definitely NOT shorted.

When you power up, what is the voltage to ground of this rail?

Remove all power. Measure the resistance to ground of each inductor on the board. Someone is shorted on the logic board and that is why you are power cycling.

It could be one of the ACE controllers (CD321x) - as per @Piernov's articles, each and every controller must be functional for the 20V power delivery contract to be selected.

Do you have a thermal cam? If yes, apply the power adapter and check the thermal view of which component is throbbing with the hot / cold cycling.

The inductor resistance check is a good start.

On the ACE controllers - each LDO power rail on the IC must be checked.

[Stephen]

VOltage to ground 12Volts then drops to 3 when it restarts.

[Stephen]

As suspected it could be a faulty CD321... I don't see any visible issues but that was my assumption ill check further. Crazy how this is a brand new board and its dead lol

[mon2]

USB Type C power delivery is crapware. We started some designs with PD 1.0 spec - I almost burnt my hands in testing out with some rather expensive tools - PD bus analyzer + cables from Taiwan (an approved supplier of the spec and was recommended) + Cypress development tools. We tested to sink the current they advertised.

It was like I was touching a working heater element. Dropped the cable and the design like a hot potato.

Reported the mishap to VIA Semiconductor in Taiwan and they claimed they knew the fault and would correct it. Apparently it has been corrected now.

The fault is often due to the cross shorting of the delicate USB C pads inside the connector shell while removing the adapter is 'hot' enabled with 20V delivery. The high voltage gets injected onto the lower voltage connector pins and kills the ACE controller(s).

TI has a whole article on this but do not believe this protection is onboard the devices, yet.


This short video is a must watch:
https://www.ti.com/video/60052597440...RT%20TO%20VBUS

Oh the irony - TI builds an external device to protect against such damage yet these TI ACE controllers get killed by the same event.

To locate the offending / damaged CD321x controller, remove all power.

Meter in DIODE mode.

Check the diode mode reading of each USB C connector pin - start with the CC1 / CC2 pins and compare the readings to the other ports onboard.

Red meter probe to ground (yes RED). Black meter probe to the contact to test.

[Stephen]

Will report my findings Monday. Went a bought a glorified heat camera. Been meaning to do it so decided to pull the trigger.

[mon2]

Seek Thermal recently announced a low cost thermal module for which they offer a development kit. We are reviewing a low cost solution. The resolution is lower than others but it should work fine for board repairs.

FYI:
https://shop.thermal.com/S104SP

[Stephen]

HAHAH the USB C shorts happening at an angle is so true. That 20V being inputted into the wrong line....BYE BYE USB C CHIP haha. God that video proves what I always thought what a normal user would do when removing the cable, an angle is common and will result in blown chip.

[Stephen]

Replaced all 4 USB C chips nothing. Still turning off and off like the same result. Put a thermal gun to it, no heating spikes. Any suggestions?

[mon2]

What is the current draw of the Type C port? Still as per post #1 ?

Check the voltage to ground of each power rail, on each CD3217.

[Stephen]

Will post my findings tomorrow thanks mon2

[Stephen]

Ok back on this topic. Couldnt seem to find the issue for some reason and did the unthinkable, i said fuck it. Removed all 4 USB C chips. Tested the board to make sure it is not turning off and on again. Viola it stopped. So one of the chips or a few were bad. Ordering 4 new USB C CD3217 B12 chips hopefully we are good. I am kind of exhausted and do not feel like reballing these chips with our stencil so just going to feel better with new chips. Wish me luck.

[Stephen]

Sad news nothing lol

[mon2]

What is the current draw of the logic board?

Apply the power adapter and check the power rails for each ACE controller against the schematic.

Also consider to remove all power -> meter in diode mode -> check the diode reading of each USB Type C connector pin (red probe to ground; black probe to pin to test) -> compare the measurements against the other type C connector(s)).