820-00850 A1989 Water Damage No Power Repair – PP3V3_G3H_RTC Short

Fault Description

This 13-inch MacBook Pro 2019, model A1989, logic board 820-00850, came from Epping VIC 3076 after a bottle leaked inside a bag.

The customer reported:

Water damage
No power
Data recovery required
Repair the Mac if the cost is not too high

Because this was liquid damage from a leaking bottle inside a bag, we did not power it on first. The risk of applying power to a wet or corroded board was too high, so the first step was ultrasonic cleaning.

Case Summary

Item	Details
Device	13-inch MacBook Pro 2019
Model	A1989
Logic board	820-00850
Customer area	Epping VIC 3076
Fault	Bottle leaked in bag, water damage, no power
Customer goal	Data recovery first, repair if cost acceptable
Initial USB-C test	Too risky before cleaning
First repair stage	Ultrasonic cleaning
Corroded area	Around UB400 / CD3215
Parts replaced first	9 corroded resistors and capacitors
USB-C behaviour after cleaning	Meter flashes for about 1 second, then no display on all 4 ports
Multimeter PPBUS_G3H	0V
Scope PPBUS_G3H	Rises to 8.9V for about 2ms, then drops to 1.2V
Scope PPDCIN_G3H	Rises to 5V, then drops to 0V
Next checked rail	PP3V3_G3H_RTC at L6900
PP3V3_G3H_RTC	0V
Diode mode	0.03, shorted
Voltage injection	Started at 1V, slowly raised to 2.1V
Thermal result	CB300 hot
Repair	CB300 removed, short gone, then replaced
Final result	MacBook working again, data recovery successful

Why We Did Ultrasonic Cleaning First

The MacBook had water damage from a bottle leaking inside a bag. This type of damage is risky because liquid can move around while the bag is carried. It can reach the logic board, USB-C area, keyboard, trackpad, screen layers and connectors.

A power-on test before cleaning can make the situation worse. If corrosion or liquid residue is bridging a rail, applying USB-C power can burn pads, damage small ICs, damage NAND-related rails or make data recovery harder.

The correct first step was:

Open the MacBook
Remove the logic board
Ultrasonic clean
Dry properly
Inspect under microscope
Replace visibly corroded components
Then start controlled electrical testing

Initial Repair Around UB400 / CD3215

After cleaning, we found corrosion around the UB400 / CD3215 USB-C controller area. We replaced:

9 corroded resistors and capacitors around UB400 / CD3215

After this, the USB-C meter still did not show normal behaviour. On all four ports, the USB-C meter display only flashed briefly for about one second, then went blank.

That meant the board was trying to start but immediately collapsing.

First Measurements

A normal multimeter measurement showed:

PPBUS_G3H = 0V

But this did not tell the full story. Using an oscilloscope, we saw:

PPBUS_G3H rises to about 8.9V for 2ms, then drops to 1.2V
PPDCIN_G3H rises to 5V, then drops to 0V

This is different from a rail that is simply missing. The board was attempting the early power sequence, but something forced the system to shut down almost immediately.

That is why the oscilloscope was important. A multimeter only showed “0V”, but the scope showed that the rail briefly appeared and then collapsed.

Why We Followed the Power-On Sequence

We used the 820-00850 power-on sequence diagram to guide the diagnosis.

The logic was:

USB-C input appears briefly
CD3215 / USB-C stage starts briefly
PPDCIN_G3H rises briefly
PPBUS_G3H rises briefly
Then everything collapses

When PPBUS_G3H and PPDCIN_G3H pulse briefly and then collapse, the next step is to check the early always-on support rails. According to the troubleshooting sequence, PP3V3_G3H_RTC is one of the important early rails.

So the next logical check was:

L6900 / PP3V3_G3H_RTC

PP3V3_G3H_RTC Short Found

At L6900, we measured:

PP3V3_G3H_RTC = 0V

Then diode-mode reading was:

0.03

That is a heavy short to ground.

This explained the symptom. If PP3V3_G3H_RTC is shorted, the early power logic cannot stay alive. The board may briefly try to start, but the shorted always-on / RTC-related 3.3V rail causes the sequence to collapse.

Voltage Injection and Thermal Camera

Because PP3V3_G3H_RTC is a 3.3V rail, we did not inject high voltage.

The safe method was:

Start at 1V
Slowly raise voltage
Do not exceed the normal rail voltage
Monitor current and heat
Use thermal camera to locate the hot component

At about:

2.1V

the thermal camera showed:

CB300 hot

We removed CB300, and the short was gone. After replacing CB300 with a good component, the MacBook worked again.

Circuit Logic

The repair logic was:

Bottle leak in bag
↓
Too risky to power on
↓
Ultrasonic clean first
↓
Replace 9 corroded resistors/capacitors around UB400 / CD3215
↓
USB-C meter flashes briefly on all 4 ports
↓
Multimeter: PPBUS_G3H = 0V
↓
Oscilloscope: PPBUS_G3H rises to 8.9V for 2ms, then drops to 1.2V
↓
PPDCIN_G3H rises to 5V, then drops to 0V
↓
Board is trying to start, then shutting down
↓
Follow power-on sequence
↓
Check PP3V3_G3H_RTC at L6900
↓
PP3V3_G3H_RTC = 0V, diode mode = 0.03
↓
Inject voltage carefully, no more than rail voltage
↓
CB300 hot at 2.1V
↓
Remove CB300, short gone
↓
Replace CB300
↓
MacBook works again

Repair Timeline

Step	Result
Customer reported bottle leak in bag, no power	Confirmed
Initial USB-C test	Skipped, too risky before cleaning
Ultrasonic cleaning	Completed
Corroded parts around UB400 / CD3215	9 resistors/capacitors replaced
USB-C meter after cleaning	Brief flash, then no display on all 4 ports
Multimeter PPBUS_G3H	0V
Scope PPBUS_G3H	8.9V for 2ms, then 1.2V
Scope PPDCIN_G3H	5V briefly, then 0V
Followed power-on sequence	Next check: PP3V3_G3H_RTC
L6900 / PP3V3_G3H_RTC	0V
Diode mode	0.03, shorted
Voltage injection	1V slowly raised to 2.1V
Thermal camera	CB300 hot
Removed CB300	Short gone
Replaced CB300	MacBook working
Final result	Data recovery successful

Key Lesson

A multimeter alone can be misleading in a fast power-cycling fault.

In this case:

Multimeter: PPBUS_G3H = 0V
Oscilloscope: PPBUS_G3H rises briefly, then collapses

That scope reading changed the diagnosis. The board was not completely dead; it was trying to start and then shutting down.

Following the power-on sequence helped us avoid random parts replacement. Instead of guessing around the USB-C controllers or charger IC, we moved to the next required early rail and found the shorted PP3V3_G3H_RTC rail.

Final Fix

The final repair included:

ultrasonic cleaning before power-on;
replacement of 9 corroded resistors and capacitors around UB400 / CD3215;
oscilloscope testing of PPBUS_G3H and PPDCIN_G3H;
power-on sequence tracing;
finding PP3V3_G3H_RTC shorted at L6900;
controlled voltage injection;
thermal-camera location of hot CB300;
CB300 replacement;
MacBook powered on and data recovery completed.

Final result:

MacBook working again
Customer data recovered

If you are interested in the deeper diagnostic side of Mac repair, we have more real logic board fault cases documented here:
Mac Logic Board Repair Case Studies

Need Board-Level Mac Repair?

MacBook USB-C and CD3215 Repair

We specialise in MacBook logic board repair, USB-C charging faults, CD3215 controller replacement, water damage repair and component-level diagnostics.