Macbook Air 15.3 A2941 won't turn on and power cycles

[mgollob7]

Hi everyone,

I’m new to board level repair, so apologies if I’m missing something obvious.
I have a MacBook Air 15.3” (A2941) that won’t turn on. No keyboard, no trackpad. It does show up in DFU mode using a DFU Blaster Pro. On all ports it power cycles between 5V and 18V.
I’ve attached a video of the cycling and a screenshot of the DFU Message
Any idea where to start looking, or what I should measure first?

Thanks!

[mon2]

What is the voltage to ground of PPBUS_AON?

[mgollob7]

PPBUS_AON is not stable: with the charger connected it ramps up to ~12.3 V, collapses to 0 V and repeats. The light on the magsafe charger flashes orange in sync.

[mon2]

Flux and remove the fuse(s) linked to the PPBUS_AON. This is to isolate the regulator from the downstream consumer side.

Then power up again. Is PPBUS_AON on the regulator side stable or still power cycling?

[mgollob7]

On the 15” board there is no fuse on PPBUS_AON — the charger output cap bank sits directly on the rail, nothing in between. So I can’t isolate the regulator by pulling a fuse. What should I lift instead Or did i miss something?

[mgollob7]

Update — a lot of measurements since my last post. No short found, and I’m now stuck at interpreting the behavior.

What I measured:

• Diode mode on PPBUS_AON: 0.5 to ground at the tantalum bank and at U5700 VIN, same everywhere. No short found on the rail.

• On a 5V USB 2.0 supply, U5200 boosts PPBUS_AON to a stable ~12.3V and holds it.

• PP3V8_AON stable at 3.8V the whole time.

• DFU mode works: device enumerates, ECID reads, both Apple Configurator and DFU Blaster Pro identify it as Mac14,15.

• Every restore attempt fails at the same point though. Configurator: “The device is not connected”. DFU Blaster: “error restoring the device with status 1” right after _AMDFUModeDeviceFinalize, then it retries.


Rail behavior during restore (measured live):


The NAND rails (PP2V58_AWAKE_NAND_VCC from BUCK6/L7760, PP1V25_AWAKE_NAND_VCCQ, PP0V88_AWAKE_NAND_VDD, PP1V2_NAND0_S5E0_PCI_AVDD_H — 1.26V measured) come up during the early restore steps and hold stable for a few seconds. When the restore reaches the later step, they go down and the cycle repeats. PPBUS and AON stay up through all of it.

Also noticed: without DFU, the board power cycles on every power connect (the 5V–18V cycling from my first post). In DFU it sits quietly until a restore runs.

Board is the 256GB config, single NAND (UN000 only, UN100 not stuffed).

Where I’m at: Supply side looks healthy to me, and the failure seems tied to the point where the NAND gets involved — but I’m not sure how to read this. Is this pointing at the NAND itself / BGA joints, or is there something else that gets enabled at that stage that I should check first? What would you measure next before considering a reflow on UN000?

[ed-service]

I had an A2991 that wouldn't turn on and would flash orange when the charger was connected; the problem was the battery.

[ed-service]

I fixed the battery, and the MacBook turned on.

[mgollob7]

Okey thank you I will try that
Did it Turn on without the Batterie connected ?

[ed-service]

No