Tonality Analysis

Tonality quantifies the degree to which discrete tonal components are perceptible within a sound. Prominent tones in broadband noise significantly increase annoyance, which is why many noise regulations impose tonal penalties (Tonzuschlag) on products that emit clearly audible tones. Deepsona implements two standardized tonality assessment methods.

Supported Standards

Standard	Method	Scope
DIN 45681:2005 (+ Corrigendum 2006)	Determination of tonal components and the tonal adjustment (Tonzuschlag) for noise immission assessment	Stationary and quasi-stationary sounds
ISO/TS 20065:2022	Objective method for assessing the audibility of tones in noise — Engineering method	Broadband and tonal noise from industrial and commercial sources

Theory

Tone Detection

Both methods begin with a high-resolution FFT of the input signal. Spectral peaks are identified as tone candidates if they protrude above the local broadband noise floor by a frequency-dependent threshold. The DIN 45681 method uses critical-band masking levels to determine audibility, while ISO/TS 20065 applies a prominence-ratio approach.

Prominence Ratio

The prominence ratio (PR) quantifies how much a tonal component exceeds the surrounding masking noise within its critical band:

PR = L_tone − L_noise [dB]

A tone is considered prominent when the prominence ratio exceeds a frequency-dependent threshold, typically between 6 dB (low frequencies) and 8 dB (high frequencies).

Tonal Adjustment (Tonzuschlag)

DIN 45681 defines the Tonzuschlag K_T as an additive penalty in dB that is applied to the overall A-weighted sound pressure level to account for the increased annoyance caused by tonal content:

K_T = max(0, ΔL_t − A_v) [dB]

Here ΔL_t is the level difference between the tonal component and the narrow-band masking noise, and A_v is a frequency-dependent threshold value. The Tonzuschlag can range from 0 dB (no tonal penalty) to a maximum of 6 dB.

Parameters

Parameter	Options	Description
Standard	DIN 45681 / ISO/TS 20065	Assessment method to apply
FFT Size	4096 – 65536 samples	Frequency resolution; larger FFT provides finer resolution for separating closely spaced tones
Averaging	Linear / Exponential	Spectral averaging mode for time-varying analysis
Number of Averages	1 – 100	Number of spectral blocks to average (stationary analysis)
Frequency Range	50 Hz – 10000 Hz	Analysis bandwidth; tonal detection outside this range is excluded
Signal Type	Stationary / Time-varying	Determines single-block or running analysis

Output Quantities

Quantity	Unit	Description
K_T	dB	Tonal adjustment (Tonzuschlag) per DIN 45681
Tone Frequency	Hz	Center frequency of each detected tonal component
Tone Level	dB(A) or dB	Level of each detected tonal component
Prominence Ratio	dB	Tone-to-noise prominence for each detected tone
Tonality(t)	dB	Tonzuschlag vs Time trace for time-varying signals

Tonality vs Time

In time-varying mode, tone detection and Tonzuschlag computation are performed on a block-by-block basis. The resulting trace reveals:

How tonal prominence changes during engine run-up/coast-down as orders sweep through resonant frequency bands
Intermittent tonal events such as belt squeal or pump cavitation
The temporal stability of tones, which affects their perceived prominence

Tonality vs RPM

Tonality is also available as a vs RPM variant. The result is a colour map in which the time axis of the vs Time analysis is replaced by the speed of rotation, so that the prominence ratio is shown over frequency and operating speed at once. A tonal order appears as a diagonal ridge, because its frequency is proportional to speed; a fixed-frequency tone appears as a horizontal ridge.

Standards: DIN 45681 (default) or ISO/TS 20065, the same choice as for the vs Time variant.
Applicable to run-up and coast-down measurements; the recording must contain speed information.
Speed step: 50 rpm by default; the slope (run-up or coast-down) is detected automatically unless it is set explicitly.
The spectral parameters — FFT size, overlap and averaging time — are inherited from the analysis settings; the default averaging time is 3 s.
Result: prominence ratio in dB over frequency and speed, with the tonal level difference and its frequency, the averaged level difference and the uncertainty reported as single values.

For the parameters shared with the other speed-resolved psychoacoustic analyses and for advice on reading the maps, see Psychoacoustics vs RPM.

Interpretation Guidelines

Regulatory significance: In many jurisdictions (particularly in Germany under TA Lärm), the Tonzuschlag K_T is added to the rated sound level before comparing against limit values. A K_T of 3 – 6 dB can determine whether a product passes or fails a noise regulation.

K_T = 0 dB — No audible tonal content; no penalty applied.
K_T = 1 – 2 dB — Marginal tonality; tone is near the audibility threshold.
K_T = 3 – 4 dB — Clearly audible tone; significant annoyance potential.
K_T = 5 – 6 dB — Dominant tonal component; strong penalty applied.

Practical Application Notes

FFT size is critical: a resolution of at least 1 Hz (or finer) is recommended to separate closely spaced tonal components from broadband noise. For a 48 kHz sample rate, use FFT size of 65536 for 0.73 Hz resolution.
DIN 45681 requires a minimum analysis duration of 10 seconds for stationary assessment. Shorter recordings may not provide stable spectral estimates.
When multiple tones are present, each is assessed individually, and the highest Tonzuschlag determines the overall penalty.
ISO/TS 20065 and DIN 45681 may yield different results for the same signal because they use different tone detection criteria and threshold functions. Always specify which standard was used when reporting results.

Standard References

DIN 45681:2005 + Berichtigung 2006 — Acoustics — Determination of tonal components of noise and determination of a tone adjustment for the assessment of noise immissions
ISO/TS 20065:2022 — Acoustics — Objective method for assessing the audibility of tones in noise — Engineering method
TA Lärm (Technische Anleitung zum Schutz gegen Lärm) — Technical Instructions on Noise Abatement, Germany
E. Zwicker & H. Fastl, Psychoacoustics: Facts and Models, 3rd ed., Springer, 2007