MacBook A2338 M2 (820-02773) Not Charging

[rounin]

Post:Hello everyone,I'm troubleshooting a MacBook A2338 M2 (820-02773) motherboard with a charging issue. Here’s the situation:Initial Fault:

• L8430 was burnt. After replacing it, the system worked normally for about two weeks, then failed to power on.

Current Symptoms:

1. Without Battery:
   • When connected to power, voltage ramps from 5V to 20V, then drops back to 5V, cycling repeatedly.
   • After disconnecting F5200, USB-C voltage stabilizes at 5V, and Pin 1 of F5200 measures 12V.
2. After Replacing U8100:
   • With the battery connected, the system powers on normally, and the battery icon displays correctly, but it does not charge.
   • Without the battery, the same 5V-20V cycling issue persists.
3. Replaced U5200:
   • Although I didn’t suspect U5200, I replaced it anyway. The issue remains unchanged.
4. Battery Test:
   • Tested with a known-good battery (verified on another working board), but the fault persists (no charging).

Measurements Taken:

• With power + battery connected:
  • I2C_SMC_PWR_3V3_SCL ≈ 2.78V
  • I2C_SMC_PWR_3V3_SDA ≈ 2.78V
  • SYS_DETECT_L = 0V
  • C5220 resistance ≈ 2.2Ω

Additional Info:

• No signs of water damage.
• I’m unsure where to check next.

Can anyone with experience offer suggestions? Could this be an SMC or other circuit issue? What key points or signals should I check? Any help is greatly appreciated!

[mon2]

Has the display assembly been replaced? Do you have True Tone? There was an issue earlier this year where the lcd assembly replacement will raise this fault.

[rounin]

Thank you for the help!

the display has not been replaced, and I don’t believe this issue is related to the screen.

When testing in a bare-board state (no display or other components connected), the power supply ramps up to around 19.5V (not quite 20V) and then resets, cycling back。

[mon2]

If available, inspect entire board with a thermal camera. Search for hot spots.

Remove all power. Meter in resistance mode. Measure the resistance to ground of each inductor. Suspecting you have a shorted part onboard. Post any low resistance inductors.

This is power cycling.

[rounin]

Thank you mon2 for your continued support!

Accidental Fix for Charging Issue:
While re-measuring points around U5200, I made a mistake and accidentally shorted Pin 1 of R5215 to Pin 2 of C5216 with my probe,
causing smoke! I thought I had completely ruined the board and felt terrible.

I quickly disconnected the power and battery, checked the main circuits, and found no shorts.
To my surprise, when I reconnected the power, the voltage stabilized at 19.7V—no more cycling!

I then connected the battery, and the charging current reached over 1A. The battery started charging normally!

It’s unbelievable—could this be some kind of “defibrillator” fix for the board?

New Issue—Touch Bar Not Working:
After reassembling the board, the system powered on fine, and everything seemed perfect… until I noticed the Touch Bar wasn’t displaying.

Previously, after replacing L8430, the Touch Bar worked perfectly.
I suspect the voltage spike from the short may have exposed or worsened an underlying issue, causing the Touch Bar circuit to fail completely.
This might be the root cause of the original problem.


Next Steps:
I’m now troubleshooting the Touch Bar issue and will continue to update with my progress.
Any suggestions on where to start with the Touch Bar failure would be greatly appreciated!

[mon2]

Good that the board is working. The accident may have smoked out a partially shorted capacitor due to the high current surge. Now the part is no longer on the rail after this event.

The touchbar is a complex design and still studying its operation. From what I know so far, the touchbars are available in different lengths for different macbook models.

Each is based on a MIPI DSI 1 lane spec. According to Wentin, the interface operates in the low speed COMMAND mode of the MIPI DSI spec. The interface uses low voltage CMOS interface @ 1v2 swings.

He goes very fast in his discovery process but is well documented here:

https://www.youtube.com/watch?v=5TPB...WentingChannel

PS: Working on some hardware to test the same. We killed far too much time in diagnosing a touchbar last week that was in the shop. In the end, there was nothing wrong with the touchbar. The bar only lights up after login using the user password. It is software based so be sure that your OS is not corrupted.

Suggestions:

1) boot using internet recovery with your network connection to boot a known good OS into RAM. Does this allow for your touchbar to work? If yes, issue is s/w based.

2) If touchbar is still dark - seek out a known good touchbar for your model for quick testing. We did this and found even the new M1 / M2 touchbar we brought in for ~90$ (Canadian) was still dark. That is when we suspected that there is a logic board problem. Rabbit hole to hell.

3) The SMD connectors used by the touchbar on these new models are @ 0.4mm pitch. Waaay too small pitched for my Fluke meter or scope. Respectively, took an old ATA ribbon cable in our junk pile and extracted a SINGLE naked wire out of the many conductors and wrapped it around our BLACK meter probe to create a needle like probe.

Removed all power. Meter in DIODE mode.

Red meter probe, yes red to ground on the logic board.

Black wire tip to each pin to test on the touchbar connector. DO NOT, repeat DO NOT fall the mistake where we thought every pin was shorted to ground. The issue for us was that we were touching the METAL SHROUND (GROUND) around this custom HIROSE 26 pin connector. A rookie mistake.

Be careful and touch only the signal contacts with a steady hand (no caffeine before this exercise). You are testing for valid DIODE measurements. If you see OL then it is very possible that line is OPEN or the EMI filter used on the MIPI lines is not soldered / damaged from a fall.

Test:

MIPI_DFR_CLK_CONN_FILT_P
MIPI_DFR_CLK_CONN_FILT_N

MIPI_DFR_DATA_CONN_FILT_P
MIPI_DFR_DATA_CONN_FILT_N

DFR_DISP_VSYNC

DFR_1V8_DISP_INT

DFR_1V8_DISP_RESET_L ; this line should NOT be in low state - if stuck @ 0v then it is stuck in RESET mode (you will have to measure in DC volts mode here)

From what I know, if there are CLK and DATA lines on the MIPI DSI interface then the target display bar is capable of operating in MIPI DSI VIDEO mode = 80 Mbps or higher data rates.

According to Wentin, the interface is being used in COMMAND mode = 10 Mbps.

Either way, the DIODE mode readings for the above should NOT be OL on your meter. No power during DIODE mode testing.


4) Test the voltage to ground of the I2C lines used by the DFR TOUCH = touchbar interface:

I2C_DFR_1V8_SCL
I2C_DFR_1V8_SDA

5) Test the voltage to ground of:

DFR_LID_OPEN_L ; if LOW, the lid is considered to be OPEN so the TB should enable

PP5V_S2_DFR_FILT ; should be @ 5v

PP1V8_AWAKE_DFR ; should be @ 1v8

Post the details of the above and be patient. For us, it was a life sucking exercise because the client did not want to release her password. Understand but we killed hours in trying to fix something that was not broken. The root cause was the family dropping their macbook and creating a shorted cap on the LCD backlight voltage rail. A bit time consuming because our Qianli camera is too bulky to allow us to power up this logic board and take thermals.