Dead element tolerance

Elements fail. They fail as open circuits, as short circuits, and as elements that still fire but at reduced amplitude, and a probe accumulates them over its life. The element check is therefore part of periodic equipment verification, not something to be done only when a problem is suspected.

The generally applied acceptance is that no more than a quarter of the elements in the active aperture may be inactive, and that no two inactive elements may be adjacent. The first limit protects sensitivity, which falls roughly as 20·log10((n − d)/n). The second protects the beam shape: a single missing element is a small perturbation, but a gap of two or more punches a hole in the aperture and raises side lobes, which puts spurious energy at angles the focal law does not account for.

Losing elements at the ends of the aperture is worse than losing them in the middle, because the ends contribute the largest delays and therefore the most steering and focusing authority. A probe that passes the count but has failures clustered at one end will show asymmetric sensitivity across a sectorial sweep.

Whatever the general rule, the number that governs is the one in the written procedure and the equipment verification standard being worked to. Record the element map, not just a pass or fail.

maximum inactive = floor(n × limit% / 100)
inactive fraction = d / n
sensitivity loss ≈ 20 · log10((n − d) / n)