2025 M4 MacBook Air A3240 Backlight Repair Without Schematics
Liquid-damaged 820-03597 logic board reverse-engineered with an A3113 schematic reference and an A3240 donor board.
Repair overview
Item | Confirmed detail |
Device | 13-inch MacBook Air with M4, introduced in 2025 |
Model | A3240 |
Logic board | 820-03597 |
Customer area | Melbourne CBD, postcode 3000 |
How the customer found us | ChatGPT |
Reported fault | Water spill; startup chime present; internal screen black |
Initial USB-C reading | 20 V; 0.10 A -> 0.25 A -> 0.45 A -> 0.23 A |
Final repair | Restored the PPBUS_AON feed to R2 and UP800 pins 9/10 |
Outcome | Backlight restored and Mac fully functional |
Fault description
A customer from Melbourne CBD brought in a liquid-damaged 2025 M4 MacBook Air. The Mac produced its startup chime, but the internal display remained black. Because it still chimed, the fault did not behave like a conventional no-power failure.
The initial USB-C meter sequence was 20 V at 0.10 A, rising through approximately 0.25 A and 0.45 A before settling near 0.23 A. That pattern was consistent with normal startup activity. A current profile alone cannot prove that every rail is healthy, but together with the startup chime it showed that the computer was progressing through boot and focused the investigation on the display and backlight path.
Why this repair took a full day
With an accurate schematic and boardview, this fault should normally have taken about two hours to diagnose and repair. The A3240 is a newer model and an A3240 schematic was not available, so the backlight circuit could not simply be followed component by component.
The available A3113 schematic was useful because the earlier 13-inch MacBook Air uses a related backlight architecture. However, Apple altered the A3240 implementation. The A3113 circuit therefore had to be treated as a functional reference rather than an exact wiring map. Copying the older circuit blindly would have risked connecting the wrong nodes.
Working without a current schematic A working or donor board becomes the physical schematic. Components may need to be removed so hidden pad connections can be checked directly with continuity and diode-mode measurements. |
Why the 820-03285 schematic can still help
The 820-03285 schematic cannot be copied connection for connection onto logic board 820-03597. It is nevertheless a strong functional reference because the two boards use the same identifiable backlight-controller and eFuse families. The component markings provide a bridge between the documented A3113 circuit and the undocumented A3240 implementation; donor-board continuity testing is then used to confirm what Apple actually changed.
1. The same UP800 backlight-controller family
UP800 on both the A3113 and A3240 boards carries a top marking beginning with 48B1. In the 820-03285 schematic, UP800 is identified as an LP8548B1-family backlight boost controller, including Apple-specific programmed variants. The shared 48B1 base marking, package and circuit position therefore make the older schematic valuable for identifying likely pin roles such as VSENSE_N, VSENSE_P, SD, SW and FB.
The characters following 48B1 should still be recorded. They may distinguish a programmed, revision or production variant, and the publicly available TI marking lookup does not provide a complete decode for these Apple-specific LP8548B1 versions. Matching only the first four characters is useful evidence of the controller family, but it is not sufficient proof that two chips are fully interchangeable.
2. The same UP950 eFuse
UP950 on both boards is marked 2L7H. Texas Instruments’ official package-marking information maps 2L7H to TPS259461LRPWR: a TPS259461L latch-off eFuse in the 10-pin RPW VQFN-HR package. This confirms the UP950 pin-function reference used during tracing, including pin 2 OVLO, pin 5 IN, pin 6 OUT and pin 9 ILM.
Observed marking | Official identification | Why it matters |
2L7H | TPS259461LRPWR | Confirms the TPS259461L latch-off eFuse variant |
RPW | 10-pin VQFN-HR package | Matches the physical package and published pinout |
2L6H | TPS259461ARPWR | Related auto-retry version; not the same fault-response variant |
How to trace 2L7H to TPS259461L
Step | Action |
1 | Open Texas Instruments’ Packaging Part Marking Lookup. |
2 | Select ‘Marking on the part’ and enter 2L7H exactly. |
3 | The result identifies TPS259461LRPWR in the RPW 10-pin package. |
4 | Open the TPS25946xx datasheet and check the Package Option Addendum: it lists 2L7H beside TPS259461LRPWR. |
5 | Treat any additional line on the IC as lot, assembly or trace information unless the manufacturer identifies it as part of the device marking. |
Manufacturer references: TI Part Marking Lookup and TPS25946xx datasheet.
Reverse-engineering the A3240 donor board
To establish the actual A3240 topology, we spent a day studying an A3240 donor board. UP800, LP800, RP800, UP950 and two unidentified 0-ohm links were removed so their pads and surrounding traces could be mapped. For clarity in this case study, those two links are called R1 and R2; these are investigation labels, not confirmed Apple reference designators.
Three unpopulated paired footprints were also labelled E1, E2 and E3 during the investigation. Each footprint connected between PPVOUT_LCDBKLT_SW and ground, indicating optional backlight-output filter capacitor positions.
Confirmed donor-board connections
Observed point | Confirmed connection | Meaning |
R1 (0 ohm) | UP800 pin 11 to UP950 pin 2 (OVLO); the node also has 37.5 kOhm to ground | Links the controller node to the overvoltage-lockout network |
R2 (0 ohm) | PPBUS_AON to the common UP800 pins 9 and 10 | Supplies the UP800 input stage |
RP800 pins 3/4 | No connection on the inspected A3240 donor board | The A3113-style Kelvin current-sense route is not populated here |
E1/E2/E3 | Each paired footprint runs from PPVOUT_LCDBKLT_SW to ground | Unpopulated optional output-filter capacitor locations |
A3113 schematic versus the actual A3240 board
Area | A3113 reference | A3240 donor-board finding |
Documentation | Schematic provides the older functional topology | No A3240 schematic available; continuity mapping required |
UP800 supply | Documented through the older input network | Pins 9/10 join and receive PPBUS_AON through R2 (0 ohm) |
OVLO node | Useful for identifying the protection function | UP800 pin 11 links through R1 to UP950 pin 2, with 37.5 kOhm to ground |
Current-sense path | Older design shows a populated sensing/switching arrangement | RP800 pins 3/4 are not connected on the inspected board |
Conclusion | A guide to circuit purpose | A simplified or altered implementation that must be verified physically |
The physical evidence suggests that the A3240 omits part of the A3113 switching or current-monitoring implementation. It is reasonable to describe this as a simplified backlight input stage. The donor board alone cannot establish Apple’s engineering or cost motivation, and it does not prove that the Mac lacks all current protection elsewhere on the logic board.
What OVLO does
OVLO means overvoltage lockout. UP950 pin 2 uses this node to supervise an input-voltage threshold and prevent operation when the monitored voltage is too high. OVLO is not a current-sense input. However, the TPS259461L eFuse itself still provides internal current-limit and short-circuit protection through its power path and ILM configuration. In this A3240 layout, RP800 pins 3 and 4 being unconnected confirms that the older UP800 Kelvin-sensing arrangement is not being used in the same way; it does not mean that all current protection has been removed.
The actual fault and final repair
The liquid-damaged board had lost the supply-side pad or trace connection serving R2. The donor-board map showed that this missing side of R2 had to connect to PPBUS_AON. A fine insulated jumper wire was installed from the R2 supply point to a confirmed PPBUS_AON point.
This repair did not force the backlight enable signal and did not connect the high-voltage output directly. R2 is a 0-ohm supply link, so the jumper reconstructed the intended factory connection: PPBUS_AON through R2 to UP800 pins 9 and 10. As soon as that input supply path was restored, the backlight operated normally.
Post-repair validation
Test | Result |
Backlight output | Bright and even |
Brightness control | Full adjustment range working |
Sleep and wake | Passed |
Lid close/open | Passed |
Charger operation | Passed |
Battery operation | Passed |
Maximum-brightness soak test | Passed after 20-30 minutes |
Overall operation | Mac fully functional |
Repair timeline
1. Confirm normal boot activity
Recorded the 20 V USB-C current sequence and confirmed the startup chime.
2. Isolate the display-lighting fault
Used the normal boot behaviour to prioritise the backlight supply path rather than the main power sequence.
3. Compare the previous model
Used the A3113 schematic to understand the likely functional blocks.
4. Map the new board physically
Removed donor-board components and traced the real A3240 connections.
5. Rebuild the missing supply
Connected the R2 supply side to PPBUS_AON with a jumper.
6. Validate every control state
Tested brightness, sleep/wake, lid operation, charger, battery and sustained maximum brightness.
Key lessons
A startup chime and a normal USB-C current sequence can prevent a black-screen Mac from being misdiagnosed as a dead logic board.
A schematic from the previous model is valuable for understanding circuit purpose, but every connection on a revised board must be verified before power is applied.
For the newest Macs, the absence of schematics makes diagnosis slower, not impossible. A donor board, careful pad tracing and a clear understanding of power-stage behaviour can still lead to a reliable, original-board repair.
Case summary
Confirmed final fix PPBUS_AON -> R2 (0 ohm) -> UP800 pins 9 and 10. Restoring this missing input-supply path returned normal backlight operation while preserving brightness and sleep control. |
This 2025 M4 MacBook Air was returned fully functional after a board-level repair that would not have been possible by parts swapping alone. The case demonstrates why donor-board analysis remains essential for current Mac models when manufacturer schematics are unavailable.
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 help with a liquid-damaged or black-screen MacBook?
IT-Tech Online Mac Repair Specialist performs chip-level MacBook logic-board repair in Melbourne, including new models with limited technical documentation. We also support regional Victoria and Australia-wide mail-in repairs. If your Mac chimes, charges or appears to start but the screen remains black, the display assembly may not be the only possible cause.
Contact IT-Tech Online for a board-level assessment. Opening hours: Monday-Friday, 10:00 am-6:00 pm.
