Dead 2018 Macbook Pro A1990 820-01041-07 stuck at 5V 0A

[pcash0007]

First time looking at a logic board this closely, learnt enough from videos and forums here to be dangerous

I was doing a OS update when the screen went blank at some point. Does not seem to be in DFU mode (checked using apple configurator).

PPDCIN_G3H is 5.16V

PPBUS_G3H is 0.045V

Resistance between TP_PPBUS_G3H and ground is 23.8 Ohms (which seems low)

F7000 and F7001 seems fine.

From what I learnt seems like there might be short on the down side on PPBUS_G3H.

Is injecting low voltage and see what heats up the next step(I have access to a friend's thermal camera)? On that note couple of questions...Why do we have to inject voltage directly at PPBUS_G3H to find the blown capacitor, If the capacitor is shorted, why doesn't it get hot with the USB-C plugged in, does something else sense this and shut the power(that might explain 0A draw)?

I'm thinking of getting a variable power supply, in the meantime can I use low voltage adapter 2/3V to apply power to PPBUS_G3H to see what heats up?

Thank you.

[reformatt]

Check edges of the board near fan intakes first before any voltage injection. Usually corroded capacitors, most common one on this model is C9081 going short.

23 ohms is a worry though and usually indicates a shorted DRMOS near the CPU. Check resistance to ground on the coils near the CPU. You can also measure between PPBUS_G3H and each CPU coil to see if you have 0 ohms to eliminate whether any of the DRMOS have gone short.

Edit: Also disconnect external things on the board like trackpad. Have had a bad trackpad cause the same issue, read about 10 ohms to ground on PPBUS IIRC.

[pcash0007]

Thanks reformatt.
Visually everything looks good including the C9081.
Few questions...I'm having tough time finding where the CPU coils are, help please, a part# or TP# or something I can lookup in FlexBV.

[mon2]

Coils in the boardview files will have a "L" prefix.

[pcash0007]

Thank you mon2.

Looks like the coils around the CPU are L72xx, L73xx, is that right? If so, the resistance between PPBUS_G3H and these coils is between 6 - 14 ohms

LA340, LA640, LA650 - 6.3 onms
L7211, L7221 and L7231 - 8.2 ohms
L7410, L7420 - 13.9 ohms

Also the measurements in my previous post and this one are with the board removed with no other peripherals attached.

Thank you.

[mon2]

Yes. No way to sugar coat this so need to rip the bandaid off. The CPU is very likely dead. The high side mosfet in this low voltage rail is leaking so the low voltage device (ie. CPU) has been spiked by the PPBUS_G3H voltage which would have been ~12-13 volts. The proof is in the low resistance between the ISL PPBUS_G3H voltage producer leg to these multi-phase power supply rails.

[pcash0007]

Oouch, that does hurt, but it is what it is. Thank you for helping me put this to rest. I'll look around for a replacement logic board.

[pcash0007]

Another question....PP1V8_SLPS2R seems to be the enable for U7000 which makes PPBUS_G3H. PP1V8_SLPS2R has 0V. Is it possible that's why PPBUS_G3H is zero?

The resistance between ground and the coils near CPU is also very low even lower than PPBUS_G3H and the coils.

[anhbanxoi]

You don't have PPBUS_G3H mean U7000 is not enabled or defective. But first check U6960 area, you will need that power rail up before getting PPBUS_G3H

[reformatt]

Thank you mon2.

Looks like the coils around the CPU are L72xx, L73xx, is that right? If so, the resistance between PPBUS_G3H and these coils is between 6 - 14 ohms

LA340, LA640, LA650 - 6.3 onms
L7211, L7221 and L7231 - 8.2 ohms
L7410, L7420 - 13.9 ohms

Also the measurements in my previous post and this one are with the board removed with no other peripherals attached.

Thank you.

A non zero value doesn't necessarily mean the CPU is dead. Under normal circumstances, you have a low ohms value to ground on the CPU coils. So with a short or low impedance on PPBUS, total resistance should exceed both in series with each other (as they are linked via ground and assuming you have no DRMOS failure).

23 ohms seems too high for a CPU failure. I would expect 0 ohms between a coil like L7211 or L7221 and C7262 if a DRMOS had shorted. Make sure everything else on the board is disconnected when you do these measurements. And any low resistance on PPBUS_G3H like that will cause the power rail to shut down.

I use voltage injection/thermal camera to verify it is a DRMOS failure, and remove the DRMOS to confirm the short on PPBUS_G3H is removed. Then board is a donor at that point.

[pcash0007]

Thank you reformatt.

I think the board is toast. I did not get to check with thermal camera yet, I plan to do it just to experiment and learn.
I see that resistance between TP_PPBUS_HS_CPU and TP_PPVCC_S0_CPU is ~9ohms which leads to me to think one of the DRMOS U7210 - U7230 is shorted.

To determine which DRMOS is shorted, should I introduce voltage @ one of the PPBUS_G3H fuses?

[pcash0007]

I'm trying to find a replacement board for this laptop. The number of the board is 08-01041-07, I see a lot of 08-01041-A on ebay, are these interchangeable? The year(2018) and model(A1990) match on both the boards. Please Help. Thank you.

[anhbanxoi]

I'm trying to find a replacement board for this laptop. The number of the board is 08-01041-07, I see a lot of 08-01041-A on ebay, are these interchangeable? The year(2018) and model(A1990) match on both the boards. Please Help. Thank you.

Yes you can install 2018 A1990 to 2019 A1990.

[pcash0007]

Cool. Thank you.

[gbuniva]

Sorry I posted my concern here because was the only way.
I have a 820-01041-A that works only when battery is disconnected.
No signs of liquid damage, board looks good.
The only way that turns on and boots normally into the system is w/o battery and conecting the USB C power cable to the bottom left side connector.
Otherway doesn´t turn on.

Measures connected to Magsafe and w/ battery :

PPBUS_G3H = 12.50v
PP3V3_G3H_T = 3.35v
P5VG3S_VSW = 0v
CHGR_BGATE = 12.43v


PP1V2_AWAKE = 1.21v
PP1V8_AWAKE = 1.81v
PPVDDCPU_AWAKE = 0.76v
PPVDDCPUSRAM_AWAKE = 0.81v
PP0V8_SLPS2R = 0.81v
PP1V8_SLPS2R = 1.81v
PP1V1_SLPS2R = 1.11v

SMBUS_3V3_BATT_SCL = 3.02v
SMBUS_3V3_BATT_SDA = 3.02v
I2C_PWR_SCL = 1.86v
I2C_PWR_SDA = 1.86V

Any sugestions will be appreciated.

[mon2]

SMBUS_3V3_BATT_SCL = 3.02v
SMBUS_3V3_BATT_SDA = 3.02v

These look low. Should be ~3v3 for each.

Confirm the voltage to ground @ L6960 / L6961. Then remove all power and place the meter into resistance mode (not diode). Then check the resistance across R6934 and again for R6935. Each should be ~0 ohms.

P5VG3S_VSW =0v

remove all power and place the meter into resistance mode (not diode). Then check the resistance of this rail. Something wrong here. Need to confirm the resistance and also that the ENABLE pin for this regulator is present (P5VG3S_EN_R = pin # 4)

[pcash0007]

Starting a new thread, the concern on the original thread changed and did not want to mix them up, continuation of
https://www.badcaps.net/forum/troubl...0a#post3256393.

I applied 2v(1v did nothing) to F7000 and noticed 2.5amp draw. The hotspot was C6553. I did not see any increased heat at any of the DRMOS near the CPU, should I be applying voltage at a different place to test the DRMOS?

Thank you.

[mon2]

Excellent. Flux and remove C6553 off the board. Then measure the resistance to ground again on F7000. Is the resistance much higher now? If yes, attempt to power up the board again if this is the only fault.

[gbuniva]

Hi mon2, thank you for your assistance:

L6960 = 3.42V constant
L6961 = 2.4 - 2.8 v fluctuating all time

R6934 = 0 ohm
R6935 = 0 ohm

P5VG3S_VSW rail in resistante mode gets 350ohm and infinity like a diode

Now something strange just happened, when a was about to measure P5VG3S_EN_R = pin # 4 with the battery connected, the Mac turned on (never did before)... now I have it running and charging the battery from 0%.... I´ll keep it ON and charging until 100% to see what happens and report you back.
Thanks

[pcash0007]

My apologies for the cross posting, I created a new post with a link to this but the two seem to have gotten merged somehow

I remove C6553 and the resistance between F7000 and ground increased significantly into mega Ohms. Now when the power is connected after negotiation, it jumps up to 20v , 1A. I'm waiting for the replacement capacitor, once that is in place, I'll put the board back and see if it boots up.

Thanks for all your help so far.