NASA's NESC says a messy hammer-tap pulse is no reason to throw out a vibration test

A NASA engineering bulletin published on November 2, 2026 says a hammer test on a spacecraft structure does not need a clean, single half-sine force pulse to be valid. If the full force and the response are measured accurately, ripples and double hits are simply known inputs, and a test should be rejected only for a specific measurement, excitation or coverage problem, the NASA Engineering and Safety Center wrote.

The one-page Technical Bulletin 26-09 comes from the NESC, NASA's independent engineering review group. Its conclusion is blunt: "Half-sine appearance neither establishes validity nor guarantees adequacy."

What a hammer test is for

Engineers tap structures with an instrumented hammer that records the force of each hit, while sensors record how the structure rings. The data feeds modal analysis, which maps the frequencies and shapes at which a structure likes to vibrate, and the tests can come before harsher pyroshock testing that mimics explosive separation events.

The textbook hit is a smooth single bump of force, the half-sine. Real taps are messier. The hammer can stay in contact and ripple, or bounce off and strike again, and the bulletin addresses records that show those shapes.

The ideal pulse has gaps of its own

The bulletin's most surprising point is about the ideal pulse itself. A finite half-sine excites a broad range of frequencies, but it is not flat: its spectrum drops to zero at three, five, seven and more times a base frequency set by how long the pulse lasts. A neat-looking hit can therefore leave some frequencies almost untested.

A messier hit can fill those notches. Rippling while the hammer stays in contact adds content, and the timing between separate contacts creates predictable interference patterns. The bulletin also warns that the usual way of plotting these spectra, normalized to their own zero-frequency value, can hide content, and recommends judging tests on the unnormalized measured spectrum and the signal-to-noise ratio of the response.

What NASA recommends instead

Between contacts the structure keeps vibrating, and a second hit acts on that motion. The bulletin says that is fine as long as the complete force history is measured and the analysis integrates through every contact. With an exact model, the same force and starting conditions give the same response; an approximate model can show different errors depending on which frequencies or nonlinear states a given hit excites.

The recommended checklist is short: measure the whole applied force with enough bandwidth, calibration and timing accuracy; make sure response signals are clean and not clipped; and confirm the excitation covers the frequencies, amplitudes and nonlinear behavior the test is meant to probe.

The bulletin rests on a September 2026 report by A. Majed, Joel Sills and colleagues on the validity of measured hammer tap inputs. It does not say how often tests have been rejected or repeated over pulse shape in the past.