A network router mainboard is dead, doing nothing at its main SoC. A technician's first move is to order a replacement SoC and plan a reball. What does the section teach?
Select one answer.
Only the flux — the SoC is fine but its joints need reflowing, so a quick reflow of the existing chip, rather than checking its inputs, is what a dead SoC calls for first. Nothing — the central chip is the most likely part to fail, so replacing the SoC is the correct first move on a dead board, and reballing a new one is the fastest route to a repair. The big chip is the weak point, so a dead board almost always means a dead processor, making an immediate replacement the efficient path and any checking of the surrounding circuitry a waste of time better spent on the rework that the fault clearly requires at the heart of the board. Prove the SoC's inputs before condemning it: the processor is among the most robust and the most expensive, hardest-to-replace parts on the board, so a dead one is far more often stopped by a missing prerequisite — power, clock, or reset — than actually failed, and its power, clock, reset, and boot inputs are proven before any reball is attempted. Only the firmware — the dead board means corrupt firmware, so reflashing the SoC, rather than checking whether it is even running, is the first thing to try.