Tonality: Why a Steady Tone Disturbs More Than Equally Loud Hiss

Two noises with the same decibel figure can be experienced completely differently. One goes unnoticed; the other keeps you awake. The main difference is usually tonality.

What tonal noise is

Broadband noise — wind, rain, traffic background — spreads its energy fairly evenly across the frequency range. Human hearing registers it as "background" and ignores it relatively easily.

Tonal noise concentrates a substantial part of its energy in a narrow frequency band. The result is an audible note — a hum, a whine or a whistle.

Industrial installations produce tonal noise almost by definition, because they contain rotating parts. Fan blades turning at a given speed produce a tone at a calculable frequency: rotational speed times blade count. That note stays constant for as long as the machine runs.

Why hearing singles out a tone

The auditory system is evolutionarily tuned to find structured sounds in a background — it is how we pick a voice out of a crowded room. A tone is a structured sound, so the brain foregrounds it automatically, including when the listener would rather it did not.

The practical consequence: tonal noise cannot be un-heard. It is also why people so often describe it as intrusive rather than loud.

The tonality correction

This is not a detail. A +5 dB correction can turn "compliant" into "non-compliant" without a single new measurement being taken.

How to check whether it was applied correctly

Look for two things in the report and compare them against each other:

1. The K_t value. If it is 0.0, no tonality was found.
2. The spectrum analysis — third-octave band data or a description of it.

If the spectrum section describes a markedly dominant band — one band, say, 15–20 decibels above its neighbours — while K_t = 0, that raises a fair question:

Was tonality assessed at the receiver point using the procedure the applicable standard prescribes, and is that assessment documented?

The wording matters. It is not an assertion that the report is wrong. It is a question about procedure, and only documents can answer it.

Why tonality may legitimately not be found

There are several entirely proper reasons why K_t can be 0 even where the source is tonal:

• tonality is assessed at the receiver, not at the source — at distance a tone can merge into the background;
• the procedure requires the tone to be audible against the background at the moment of measurement;
• if measured during the day, traffic background can mask the tone entirely.

The last point is the practically important one, and it connects to everything else in this subject: a daytime measurement usually will not register tonal industrial noise. Which is exactly why evening and night measurements are the ones that show anything.

What to do with this

If your situation involves a persistent hum, it is worth asking the competent authority to assess three things: whether tonality was evaluated in the evening and night periods, whether that evaluation is documented, and whether the spectrum data confirms or rules out a dominant band.