Measuring Sound Absorption in an Impedance Tube — and Why Your Number Disagrees With the Datasheet

Put a sample of material at one end of a rigid tube and a loudspeaker at the other. Drive the speaker with broadband noise. The sound travels down the tube, hits the sample, and part of it comes back. Incident wave going one way, reflected wave coming the other, superimposed in the same tube.

If you can separate those two waves, you know exactly how much energy the material absorbed. That is the whole measurement. Everything else is technique.

The old way (ISO 10534-1) used a single microphone on a sliding carriage: move it along, find the maxima and minima of the standing wave, and compute absorption from their ratio. It works, it is slow, and it does one frequency at a time.

The modern way (ISO 10534-2) uses two fixed microphones at known positions in the tube wall. Measure the complex transfer function between them, and because you know the spacing and the speed of sound, you can algebraically decompose the field into its forward and backward travelling components — across the whole band at once, in seconds. That is the two-microphone transfer-function method, and it is what a modern impedance tube does.

There are two hard limits, one at each end of the band, and both are geometry.

The upper limit is the tube diameter. The method assumes plane waves — the sound field is uniform across the tube's cross-section. Above a cutoff frequency set by the diameter, cross-modes start to propagate, the field stops being planar, and the two-microphone decomposition is simply invalid. A wider tube has a lower cutoff. This is why a tube that reaches 6 kHz is narrow, and why a narrow tube demands a small sample.

The lower limit is the microphone spacing. At low frequency the wavelength is long, and two mics a few centimetres apart see almost the same pressure with almost the same phase. The transfer function between them approaches unity, the decomposition becomes a difference of two nearly-identical numbers, and measurement noise swamps the result. Wider spacing helps the low end — but wider spacing also brings the upper limit down, because when the spacing approaches half a wavelength the maths becomes singular.

The consequence people do not expect: one tube cannot cover the full audio range. A serious absorption measurement uses two tube diameters and often two mic spacings, and the results are spliced. Anyone quoting 50 Hz to 6.3 kHz from a single tube and a single spacing is quoting a range they did not measure cleanly at both ends.

This is the most common support question in the whole subject, and it is not a fault in either measurement. They are measuring different quantities.

An impedance tube measures the normal-incidence absorption coefficient, written α₀. Sound arrives perpendicular to the sample, and only perpendicular.

A reverberation room (ISO 354) measures the random-incidence absorption coefficient, α_s — sound arriving from all directions at once, which is far closer to how a material behaves in a real room.

For almost every porous absorber, α_s is higher than α₀, sometimes substantially. Sound arriving at an angle travels a longer path through the material and loses more energy doing it. So the tube systematically reads lower than the room, and the datasheet — which usually quotes the room figure, because it is the one that sells — will not match your tube.

There is no conversion factor that fixes this honestly. Published correlations exist, they are approximate, and they fail for anything with structure — perforated panels, resonant absorbers, anything with a membrane. If you need the random-incidence number for a building-acoustics submission, you need a reverberation room. The tube cannot give it to you, and no amount of post-processing changes that.

What the tube is genuinely excellent at is comparison and development: ranking candidate materials, checking batch consistency, measuring the effect of a thickness change or an air gap, and validating a porous-absorber model. It is a fast, cheap, repeatable instrument for iterating — and then you take your final candidate to the room.

A sample that does not fit. A gap around the edge lets sound leak past the sample and into the cavity behind it, and the tube reports absorption that the material did not provide. Too tight, and you have compressed a porous material and changed its flow resistivity — which is the property doing the absorbing. Cut precisely, seat firmly, seal the rim with a thin bead of grease, and do not squeeze.

An unintended air gap. A backing air gap is a legitimate configuration and it dramatically shifts a porous absorber's low-frequency performance. What ruins a measurement is an accidental one — a sample sitting a few millimetres proud of the rigid termination because nobody checked. Always record the backing condition; a result without it is uninterpretable.

Skipping the microphone switching calibration. The two microphones will never have identical amplitude and phase response, and the method is a phase measurement — small phase mismatches produce large errors, especially at the low end. The standard calibration is to swap the mics between the two positions, measure again, and take the geometric mean, which cancels the mismatch. It takes two minutes. Skip it and your low-frequency data is decorative.

Ignoring temperature. The decomposition depends on the speed of sound, which depends on temperature — roughly 0.6 m/s per °C. Assume 20 °C in a 30 °C workshop and you have the wavenumber wrong by about 2%, which shifts everything. Measure the air temperature and enter it; do not assume it.

An impedance tube belongs in an R&D lab that is developing or selecting acoustic materials, in a QC role checking that incoming batches match the specification they were bought against, and in a teaching lab where students need to see absorption as a measured quantity rather than a table.

It does not belong in a workflow that needs certified random-incidence figures for a building submission, and a supplier who tells you otherwise is selling you the wrong instrument.

T-Sonus is built for the development loop — try a thickness, try an air gap, try a different density, and see the answer in seconds rather than booking a reverberation room.

We will tell you plainly which questions the tube answers and which it does not, because a mismatch there wastes far more money than the instrument costs.