electroacoustics.headphones

Headphones and earphones: the characteristics of IEC 60268-7:2010 that are computations.

IEC 60268-7:2010 (Edition 3.0) lists what the manufacturer of a headphone, headset, earphone or earset states, and how each characteristic is measured. Most of the clauses describe a set-up and a reading; this module holds the parts that are arithmetic on what was read, and the verdicts where the part sets a limit. The simulated programme signal several of them use is the one of IEC 60268-1:1985 Clause 7, in phonometry.electroacoustics.programme_signal.

A headphone is classified by the code 60268-7-IEC-XXXX-NNRN-N: four letters for the transducer principle, the type of earphone, the acoustic coupling to the ear canal and the radiation to the outside (TRANSDUCER_PRINCIPLES, EARPHONE_TYPES, ACOUSTIC_COUPLINGS, BACK_RADIATIONS), the impedance in ohms in “mantissa and exponent” form, and the number of channels. The clause prints three impedances, 8 ohm as 08R0, 32 ohm as 32R0 and 600 ohm as 06R2: a two-digit mantissa times ten to the digit after the R. The 600 ohm example settles the one choice the rule leaves open, between 06R2 and 60R1: the exponent takes every trailing zero, so 150 ohm is 15R1 and 300 ohm 03R2 (impedance_code). HeadphoneClassification writes the code and parse_classification_code reads it.

The rated impedance is a pure resistance the manufacturer states, “chosen so that the lowest value of the modulus of the actual impedance within the rated frequency range is not less than 80 % of the rated value”; a dip below that anywhere from 0 kHz to 20 kHz “should be stated in the specification”. The modulus is measured “at least over the frequency range 20 Hz to 20 kHz” (8.2.2.2). verify_rated_impedance judges both.

Every input of IEC 60268-7 is a source e.m.f. $E$ applied through the rated source impedance $R_{\mathrm{s}}$. The characteristic voltage (8.3.3) is the 500 Hz sinusoidal e.m.f. that produces 94 dB in the coupler or ear simulator; the headphone is linear there, so one reading $L$ at an e.m.f. $E$ gives it as $E10^{(94-L)/20}$ (characteristic_voltage). The simulated programme signal characteristic voltage (8.3.4) is the same with the programme signal, and 8.3.5 adds the A-weighting of IEC 61672-1 and the inverse of the free-field response of the head and torso simulator of IEC 60959 to the coupler’s output. The NOTE under Figure 3 says how to do that without filters: “Power summation of the 1/3-octave-analized data multiplied by filtering coefficients given by IEC 61672-1 and/or IEC 60969 gives the corrected voltage” (the second standard is IEC 60959, see docs/ERRATA.md). programme_signal_level is that power sum,

with $A_k$ the A-weighting at the exact centre of band $k$ and $F_k$ the free-field response, and programme_characteristic_voltage turns it into the e.m.f. for 94 dB, for each fitting of the headphone, and averages the e.m.f. of the fittings as 8.3.5 g) asks (“3 to 5 measurements”).

8.4 states each of these as a power instead, “derived from the corresponding voltages (8.3) and the rated impedance”: the power the e.m.f. dissipates in a pure resistance equal to the rated impedance $R$ connected in place of the headphone, $P=E^{2}R/(R+R_{\mathrm{s}}{)}^2$ (headphone_input_power, and headphone_source_emf the other way round). The working sound pressure level of 8.5.2 b) to d) is the level at the e.m.f. that dissipates 1 mW that way (working_sound_pressure_level). The method of 8.5.3 c) sets the e.m.f. differently, so that the voltage across the headphone’s own input connector is $\sqrt{1\mathrm{mW}\cdot R}$; the two agree only when the headphone’s impedance at 500 Hz is the rated one, and the function gives the second reading when that impedance is passed (see docs/ERRATA.md).

8.3.1 NOTE 3 recommends that the rated source e.m.f. “should, preferably, not exceed the characteristic voltage (see 8.3.3) by more than 10 dB to 15 dB”; the excess is $20\lg$ of their ratio, and a preference stated as a range is not a limit, so no verdict is offered on it.

The clipped programme signal of 8.3.2, which rates the voltages a headphone survives, has “a frequency distribution as specified in IEC 60268-1, and a peak-to-r.m.s ratio between 1,8 and 2,2” (check_limiting_test_signal). A protective device “causes a change of at least 1 dB in the sensitivity” at its protection voltage (8.3.6.2 b), which protection_voltage finds in a sweep of e.m.f. and level.

The coupler or ear simulator frequency response (8.6.2) is the level against frequency at the rated e.m.f., and its graph has “the same length representing 50 dB as represents one decade of frequency” (coupler_frequency_response). The crosstalk attenuation (8.12) is the difference of two such curves (crosstalk_attenuation).

The two subjective responses (8.6.3, 8.6.4) are the quotient of the reference field’s sound pressure by the e.m.f. that makes the headphone equally loud, “expressed in decibels referred to the value at the standard reference frequency”, averaged over at least eight test persons with the standard deviation in each band (field_comparison_response); a headphone calibrated on at least 16 persons may then serve as the reference of the substitution method. The ear-canal responses (8.6.5) read a probe microphone in the ear canal instead of a loudness judgement, by Formula (1),

after averaging the two fittings of the headphone and the two readings of the sound field, with the procedure repeated when the two fittings differ by more than 2,5 dB in any band (ear_canal_frequency_response). A headphone measured that way on at least 16 persons may replace the sound field of the indirect method (8.6.5.3, whose two references to 8.6.4.2 are read as 8.6.5.2, see docs/ERRATA.md). The microphone in the ear canal is specified by Annex B (verify_ear_canal_microphone).

The two reference frequencies are not the same. Formula (1) prints its reference band, 500 Hz, the standard measuring frequency of 7.2 b). The comparison responses are referred to “the standard reference frequency”, a term IEC 60268-7 uses without defining; IEC 60268-1:1985 Clause 3, which it cites, sets it at 1 000 Hz “in the absence of a clear reason to the contrary”, and 8.6.3.2 c) and 8.6.4.2 c) begin and end the test sequence on that band, so field_comparison_response refers to 1 000 Hz by default. The coupler response keeps 500 Hz, for the reason the NOTE of 8.3.3.1 gives: the coupler’s own resonances, leakage and standing waves at other frequencies. The frequency response itself has no tolerance: “It is not at present possible to set limits for the frequency range based on deviations from a flat, or defined, frequency response” (8.6.6 NOTE 2), so the rated frequency range is the manufacturer’s statement and enters only as the range a measurement has to cover.

The harmonic, modulation and difference-frequency distortions are those of IEC 60268-2, read with harmonic_distortion, modulation_distortion and difference_frequency_distortion, and expressed in decibels as $20\lg$ of the ratio. What 60268-7 adds are the test signals: 70 Hz and 600 Hz in the amplitude ratio 4:1 with the peak of the rated input voltage, “−1,9 dB at 70 Hz and −14,0 dB at 600 Hz” (NOTE 1 of 8.7.3.3; headphone_modulation_signal), and two tones 80 Hz apart, each at half the rated input voltage (headphone_difference_frequency_signal). The third-order modulation products are at 460 Hz and 740 Hz, as 8.7.3.3 b) says; Formula (3) prints $U_{470}$ (see docs/ERRATA.md).

The rated conditions (7.1, 8.6.6, 8.8, 8.13, 8.14) are stated, not computed. The external field (8.9) and the unwanted radiation (8.10) are readings. The sound attenuation of 8.11 is measured “as specified in ISO 4869-1”, which phonometry.hearing.real_ear_attenuation computes from the open and occluded thresholds of the subjects; for a headphone with active noise compensation, which the NOTE of 8.11.2 says “may require a modified procedure”, phonometry.hearing.anr_total_attenuation follows ISO 4869-6. Annex A is the geometry of a pinna simulator and Annexes C to E are informative practical conditions.

Auto-generated from the source docstrings by scripts/generate_api_docs.py (make api-docs). Do not edit by hand.

Constant (mapping).

ACOUSTIC_COUPLINGS = {'L': 'acoustically open (controlled leakage)', 'S': 'acoustically closed (minimum leakage)'}

Constant (mapping).

BACK_RADIATIONS = {'C': 'closed-back', 'O': 'open-back'}

characteristic_voltage(
    source_emf_v: float,
    sound_pressure_level_db: float,
) -> float

The source e.m.f. that produces 94 dB in the coupler (IEC 60268-7 8.3.3).

The characteristic voltage is “the sinusoidal source e.m.f. at 500 Hz which, when applied to the headphone through the rated source impedance, produces a sound pressure level in the coupler or ear simulator of 94 dB”. The headphone is linear at that level, so a reading of sound_pressure_level_db at source_emf_v scales to it: $E10^{(94-L)/20}$. The same scaling serves the programme signal of 8.3.4 with the level of the whole signal (programme_signal_level).

Parameters

Returns: The characteristic voltage, in volts (RMS).

Raises

check_limiting_test_signal(
    signal: Signal | NDArray[np.float64] | list[float],
    fs: int | None = None,
) -> LimitingTestSignalCheck

Is this the clipped programme signal IEC 60268-7 8.3.2.2 b) asks for?

“The clipped noise signal at the output of the amplifier shall have a frequency distribution as specified in IEC 60268-1, and a peak-to-r.m.s ratio between 1,8 and 2,2.” The record’s one-third-octave bands from 20 Hz up to the highest whose upper edge lies below the Nyquist frequency are judged against Table II of IEC 60268-1 (check_programme_signal), and its peak over its RMS against the two limits. simulated_programme_signal with peak_to_rms generates such a record.

Parameters

Returns: A LimitingTestSignalCheck.

Raises

coupler_frequency_response(
    frequencies_hz: ArrayLike,
    sound_pressure_level_db: ArrayLike,
    *,
    rated_frequency_range_hz: tuple[float, float] | None = None,
) -> CouplerFrequencyResponse

The coupler or ear simulator frequency response of IEC 60268-7 8.6.2.

“The variation of the sound pressure (level) in the coupler or ear simulator as a function of frequency”, at the rated source e.m.f. through the rated source impedance, over “at least the rated frequency range”.

Parameters

Returns: A CouplerFrequencyResponse.

Raises

CouplerFrequencyResponse(
    frequencies_hz: NDArray[np.float64],
    sound_pressure_level_db: NDArray[np.float64],
    rated_frequency_range_hz: tuple[float, float] | None,
)

The coupler or ear simulator frequency response (IEC 60268-7 8.6.2).

Attributes

property

Whether the frequencies span the rated range (8.6.2.2 b), None if unstated.

CouplerFrequencyResponse.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    iec_scale: bool = True,
    **kwargs: Any,
) -> Axes

Draw the level against frequency, 50 dB to the length of a decade.

8.6.2.2 c): “the preferred scale has the same length representing 50 dB as represents one decade of frequency (see IEC 60268-1 and IEC 60263)”. Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

CouplerFrequencyResponse.relative_response_db(
    *,
    reference_frequency_hz: float = 500.0,
) -> NDArray[np.float64]

The level relative to its value at a reference frequency, in dB.

Parameters

Returns: The relative response at each frequency, in dB.

Raises

crosstalk_attenuation(
    frequencies_hz: ArrayLike,
    driven_level_db: ArrayLike,
    other_level_db: ArrayLike,
) -> CrosstalkAttenuation

The crosstalk attenuation of IEC 60268-7 8.12.

“The ratio of the sound pressure produced in the coupler or ear simulator due to rated source e.m.f. applied to the channel under test to the sound pressure produced by rated source e.m.f. applied to another, stated channel”, expressed as the difference of the two levels against frequency.

Parameters

Returns: A CrosstalkAttenuation.

Raises

CrosstalkAttenuation(
    frequencies_hz: NDArray[np.float64],
    driven_level_db: NDArray[np.float64],
    other_level_db: NDArray[np.float64],
)

The crosstalk attenuation of a multi-channel headphone (IEC 60268-7 8.12).

Attributes

property

The difference of the two levels at each frequency, in dB.

property

The smallest attenuation over the frequencies, in dB.

CrosstalkAttenuation.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    **kwargs: Any,
) -> Axes

Draw the attenuation against frequency.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

ear_canal_frequency_response(
    frequencies_hz: ArrayLike,
    earphone_levels_db: ArrayLike,
    sound_field_levels_db: ArrayLike,
    *,
    reference_frequency_hz: float = 500.0,
) -> EarCanalFrequencyResponse

The ear canal sound pressure level frequency response of IEC 60268-7 8.6.5.

For each test person a probe microphone in the ear canal reads each one-third-octave band of pink noise twice in the sound field (items b and e) and twice with the headphone, refitted between the two (c and d). The two pairs are averaged and the response is Formula (1), $L_{\mathrm{f}}=L_{\mathrm{e}}-L_{\mathrm{s}}-(L_{\mathrm{e}}-L_{\mathrm{s}}{)}_{500}$, averaged over at least eight persons (h). The result also carries the two checks of the procedure: the fittings within 2,5 dB in every band (f), and the 500 Hz band of the first fitting within 3 dB of the sound field’s (c).

Parameters

Returns: An EarCanalFrequencyResponse.

Raises

EarCanalFrequencyResponse(
    frequencies_hz: NDArray[np.float64],
    earphone_levels_db: NDArray[np.float64],
    sound_field_levels_db: NDArray[np.float64],
    reference_frequency_hz: float,
)

The ear canal sound pressure level frequency response (IEC 60268-7 8.6.5).

Attributes

property

The largest difference between the two fittings of each person, in dB.

property

For each person, whether the fittings agree within 2,5 dB in every band.

8.6.5.2 f): “if there is a difference exceeding 2,5 dB […] in any 1/3 octave band, the whole procedure is repeated.”

property

The first fitting’s level in the reference band minus the sound field’s, in dB.

8.6.5.2 c): the level is adjusted “using the 1/3 octave band of noise centred on 500 Hz, so that the filtered microphone output signal level is within 3 dB of that due to the sound field in the same frequency band”.

property

For each person, whether level_match_db is within 3 dB.

property

Formula (1) averaged arithmetically over the persons, in dB (8.6.5.2 h).

property

Whether at least eight persons were measured (8.6.5.2 h).

property

The number of test persons.

EarCanalFrequencyResponse.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    iec_scale: bool = True,
    **kwargs: Any,
) -> Axes

Draw the persons’ responses, their mean and its standard deviation.

8.6.5.2 h): the results “may be tabulated or presented graphically, the scales being preferably chosen so that 50 dB and one decade of frequency are represented by the same length”. Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

Whether the panel is large enough to calibrate a reference (16 persons).

8.6.5.3: the indirect method replaces the sound field by “a headphone, previously calibrated by the method of 8.6.4.2, using at least 16 test persons”; the method meant is the direct ear-canal measurement of 8.6.5.2 (see docs/ERRATA.md).

property

Formula (1) for each person, one row per person, in dB.

property

The standard deviation of the persons’ responses, in dB (NOTE 5).

EarCanalMicrophoneVerification(
    entrance_area_mm2: float,
    canal_section_area_mm2: float,
    ear_canal_area_mm2: float,
    volume_mm3: float,
    neighbour_difference_db: float,
    sealed_attenuation_db: float,
)

A probe microphone judged by IEC 60268-7:2010 Annex B a) to e).

Attributes

property

The microphone’s area over the ear canal’s, in the rest of the canal.

property

Whether all five requirements hold.

EarCanalMicrophoneVerification.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    **kwargs: Any,
) -> Axes

Draw each requirement’s value as a share of its limit.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

Each of a) to e), keyed by its item letter, and whether it holds.

Constant (mapping).

EARPHONE_TYPES = {'C': 'circumaural', 'E': 'intra-concha', 'H': 'earshell', 'I': 'insert', 'M': 'supra-concha', 'S': 'supra-aural', 'T': 'stethoscopic'}

field_comparison_response(
    frequencies_hz: ArrayLike,
    field_level_db: ArrayLike,
    source_emf_v: ArrayLike,
    *,
    field: Literal['free', 'diffuse'],
    reference_frequency_hz: float = 1000.0,
) -> FieldComparisonResponse

The free-field or diffuse-field comparison response of IEC 60268-7.

“The quotient, as a function of frequency, of the sound pressure of the reference [free or diffuse] sound field by the source e.m.f. to the headphone which is required to produce a sound subjectively equal in loudness to the […] sound field. It is normally expressed in decibels referred to the value at the standard reference frequency” (8.6.3.1, 8.6.4.1). In each band $k$ and for each person,

and the persons’ responses are averaged in each band, with their standard deviation (8.6.3.2 e). At least eight persons are heard (d); a headphone calibrated on at least 16 may serve as the reference of the substitution method (8.6.3.3).

Parameters

Returns: A FieldComparisonResponse.

Raises

FieldComparisonResponse(
    field: Literal['free', 'diffuse'],
    frequencies_hz: NDArray[np.float64],
    response_db: NDArray[np.float64],
    reference_frequency_hz: float,
)

A free-field or diffuse-field comparison frequency response (IEC 60268-7 8.6.3, 8.6.4).

Attributes

property

The response averaged over the persons in each band, in dB.

property

Whether at least eight persons were heard (8.6.3.2 d, 8.6.4.2 d).

property

The number of test persons.

FieldComparisonResponse.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    iec_scale: bool = True,
    **kwargs: Any,
) -> Axes

Draw the mean response as bars with the standard deviation of each band.

8.6.3.2 e): “the resulting bar graph is presented as the free-field comparison frequency response of the headphone. The standard deviation of the results in each band should be indicated on the graph. The scales for the graph preferably shall be such that 50 dB is represented by the same length as one decade of frequency.” Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

Whether the panel is large enough for a substitution reference (16 persons).

8.6.3.3 and 8.6.4.3: a headphone measured “using a panel of at least 16 persons, may be used as a loudness comparison reference”.

property

The standard deviation of the persons’ responses in each band, in dB.

The sample standard deviation; zero with a single person.

headphone_difference_frequency_signal(
    fs: int,
    seconds: float,
    *,
    upper_frequency_hz: float,
    rated_source_emf_v: float,
) -> NDArray[np.float64]

The difference-frequency distortion test signal of IEC 60268-7 8.7.4.

“Two sinusoidal signals, separated in frequency by 80 Hz, each giving half the rated input voltage”: tones at $f_2-80$ Hz and $f_2$, each of RMS value $U/2$, in cosine phase so that the peak is that of the rated voltage. Read the products with difference_frequency_distortion.

Parameters

Returns: The signal, in volts.

Raises

headphone_input_power(
    source_emf_v: float,
    *,
    rated_impedance_ohm: float,
    rated_source_impedance_ohm: float,
) -> float

The power corresponding to a source e.m.f. (IEC 60268-7 8.4).

“Specifications in terms of power can be derived from the corresponding voltages (8.3) and the rated impedance”: the power the e.m.f. $E$ dissipates, through the rated source impedance $R_{\mathrm{s}}$, in a pure resistance equal to the rated impedance $R$ connected in place of the headphone, the arrangement 8.5.2 b) describes for 1 mW,

Parameters

Returns: The input power, in watts.

Raises

headphone_modulation_signal(
    fs: int,
    seconds: float,
    *,
    rated_source_emf_v: float,
) -> NDArray[np.float64]

The modulation distortion test signal of IEC 60268-7 8.7.3.

“The sum of two sinusoidal signals, at 70 Hz and 600 Hz, with amplitude ratio 4:1. The peak voltage of the signal shall be equal to that of the rated input voltage”: amplitudes $0,8\sqrt{2}U$ and $0,2\sqrt{2}U$ for a rated voltage $U$, which NOTE 1 states as “−1,9 dB at 70 Hz and −14,0 dB at 600 Hz”. Both start at their crest (cosine phase), so the record reaches that peak at its first sample. Read the products with modulation_distortion at f_low=70, f_high=600: the second order at 530 Hz and 670 Hz, the third at 460 Hz and 740 Hz.

Parameters

Returns: The signal, in volts.

Raises

headphone_source_emf(
    input_power_w: float,
    *,
    rated_impedance_ohm: float,
    rated_source_impedance_ohm: float,
) -> float

The source e.m.f. that corresponds to a power (IEC 60268-7 8.4, 8.5.2).

The inverse of headphone_input_power, $E=\sqrt{PR}(R+R_{\mathrm{s}})/R$. With 1 mW it is the e.m.f. of the working sound pressure level of 8.5.2 b) to d).

Parameters

Returns: The source e.m.f., in volts (RMS).

Raises

HeadphoneClassification(
    principle: str,
    earphone_type: str,
    coupling: str,
    radiation: str,
    impedance_ohm: float,
    channels: int,
)

The classification of a headphone by IEC 60268-7:2010 Clause 4.

Attributes

property

The code, for instance "60268-7-IEC-DCSC-32R0-2".

property

The four letters in words, in the clause’s terms.

impedance_code(impedance_ohm: float) -> str

The NNRN form of an impedance in the code of IEC 60268-7 Clause 4.

A two-digit mantissa, R, and a one-digit exponent of ten: “8 Ω as “08R0”, 32 Ω as “32R0” and 600 Ω as “06R2"". The exponent takes every trailing zero of the whole number of ohms, which is the reading the 600 ohm example fixes (06R2, not 60R1).

Parameters

Returns: The four characters, for instance "32R0".

Raises

LimitingTestSignalCheck(peak_to_rms: float, spectrum: ProgrammeSignalCheck)

The test signal of the limiting voltages judged by IEC 60268-7 8.3.2.2 b).

Attributes

property

Whether both the ratio and the spectrum conform.

property

Whether the peak-to-RMS ratio is between 1,8 and 2,2.

LimitingTestSignalCheck.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    **kwargs: Any,
) -> Axes

Draw the spectrum against Table II, with the ratio in the title.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

parse_classification_code(code: str) -> HeadphoneClassification

Read a classification code of IEC 60268-7:2010 Clause 4.

Spaces around the hyphens are ignored, so the clause’s own layout, 60268-7 - IEC - XXXX - NNRN - N, reads as well as the compact one.

Parameters

Returns: A HeadphoneClassification.

Raises

programme_characteristic_voltage(
    source_emf_v: ArrayLike,
    frequencies_hz: ArrayLike,
    band_levels_db: ArrayLike,
    *,
    a_weighted: bool = False,
    free_field_response_db: ArrayLike | None = None,
) -> ProgrammeCharacteristicVoltage

The simulated programme signal characteristic voltage of IEC 60268-7.

The source e.m.f. of the simulated programme signal of IEC 60268-1 that produces 94 dB in the coupler or ear simulator (8.3.4), or 94 dB after the A-weighting and the free-field compensation of 8.3.5. Each fitting’s one-third-octave band levels, read at a known e.m.f., are power-summed as the NOTE under Figure 3 describes (programme_signal_level) and scaled to 94 dB; 8.3.5 g) removes and refits the headphone between readings and states “the average value of source e.m.f. of 3 to 5 measurements”.

Parameters

Returns: A ProgrammeCharacteristicVoltage.

Raises

programme_signal_level(
    frequencies_hz: ArrayLike,
    band_levels_db: ArrayLike,
    *,
    a_weighted: bool = False,
    free_field_response_db: ArrayLike | None = None,
) -> float

The level of the simulated programme signal from its one-third-octave bands.

IEC 60268-7:2010 Figure 3, NOTE: “The output signal correction can be done numerically without use of any filtering devices. Power summation of the 1/3-octave-analized data multiplied by filtering coefficients given by IEC 61672-1 and/or IEC 60969 gives the corrected voltage.” The coefficients are the A-weighting of IEC 61672-1, taken at the exact base-ten centre of each band (where its Table 3 is computed), and the inverse of the free-field response of the head and torso simulator of IEC 60959 (printed “IEC 60969”, see docs/ERRATA.md):

Parameters

Returns: The level of the whole signal, in dB re 20 µPa.

Raises

ProgrammeCharacteristicVoltage(
    frequencies_hz: NDArray[np.float64],
    band_levels_db: NDArray[np.float64],
    source_emf_v: NDArray[np.float64],
    corrections_db: NDArray[np.float64],
    a_weighted: bool,
    free_field_compensated: bool,
)

The simulated programme signal characteristic voltage (IEC 60268-7 8.3.4, 8.3.5).

Attributes

property

The corrected band levels at the characteristic voltage, in dB re 20 µPa.

The mean of the fittings’ corrected bands, each scaled to the characteristic voltage; their power sum is 94 dB when the fittings agree.

property

The arithmetic mean of voltages_v, in volts: 8.3.5 g).

property

Whether 3 to 5 fittings were averaged, as 8.3.5 g) asks.

property

The corrected level of the whole signal in each fitting, in dB re 20 µPa.

ProgrammeCharacteristicVoltage.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    **kwargs: Any,
) -> Axes

Draw the band levels at the characteristic voltage and their 94 dB sum.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

The e.m.f. that gives 94 dB in each fitting, in volts (RMS).

protection_voltage(
    source_emf_v: ArrayLike,
    sound_pressure_level_db: ArrayLike,
) -> ProtectionVoltage

The protection voltage of a headphone’s protective device (IEC 60268-7 8.3.6).

8.3.6.2 b): “The source e.m.f., at the standard reference frequency, is increased until operation of the protective device causes a change of at least 1 dB in the sensitivity of the headphone.” The sensitivity at each step is $L-20\lg E$; the first step, the lowest e.m.f., is taken as the linear one, and the protection voltage is where the change from it first reaches 1 dB either way, interpolated linearly in $20\lg E$ between the two steps that straddle it.

Parameters

Returns: A ProtectionVoltage.

Raises

ProtectionVoltage(
    source_emf_v: NDArray[np.float64],
    sound_pressure_level_db: NDArray[np.float64],
    protection_voltage_v: float | None,
)

A protective device’s operating point from a level sweep (IEC 60268-7 8.3.6).

Attributes

property

The e.m.f.s 1 dB below and 1 dB above the protection voltage, in volts.

8.3.6.2 b): “measurements are then made of the impedance and sound pressure level at voltages 1 dB lower and 1 dB higher than the noted voltage”.

ProtectionVoltage.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    **kwargs: Any,
) -> Axes

Draw the sensitivity change against the e.m.f., with the 1 dB line.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

The sensitivity at each step relative to the first, in dB.

RatedImpedanceVerification(
    frequencies_hz: NDArray[np.float64],
    impedance_ohm: NDArray[np.float64],
    rated_impedance_ohm: float,
    rated_frequency_range_hz: tuple[float, float],
)

A rated impedance judged against the measured modulus (IEC 60268-7 8.2).

Attributes

property

Whether the modulus was measured at least from 20 Hz to 20 kHz (8.2.2.2 c).

property

The frequencies from 0 kHz to 20 kHz where the modulus is below 80 %.

8.2.1 b): “If the impedance at any frequency between 0 kHz and 20 kHz is less than this value, this should be stated in the specification.”

property

80 % of the rated impedance, in ohms (8.2.1 b).

property

The frequency of minimum_ohm, in Hz.

property

The lowest modulus within the rated frequency range, in ohms.

property

minimum_ohm over the rated impedance.

property

Whether the rated impedance is at most 1,25 times the lowest modulus in range.

RatedImpedanceVerification.plot(
    ax: Axes | None = None,
    *,
    language: str = 'en',
    **kwargs: Any,
) -> Axes

Draw the modulus against frequency with the rated value and its 80 %.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

Constant (mapping).

TRANSDUCER_PRINCIPLES = {'D': 'electrodynamic (moving coil)', 'E': 'electret (self-polarizing)', 'F': 'piezo-electric (polymer)', 'M': 'electromagnetic (moving armature or diaphragm)', 'P': 'piezo-electric (ceramic)', 'S': 'electrostatic (externally polarized)'}

verify_ear_canal_microphone(
    *,
    entrance_area_mm2: float,
    canal_section_area_mm2: float,
    volume_mm3: float,
    pink_noise_band_levels_db: ArrayLike,
    open_levels_db: ArrayLike,
    sealed_levels_db: ArrayLike,
    ear_canal_area_mm2: float = 45.0,
) -> EarCanalMicrophoneVerification

May this microphone measure inside the ear canal by IEC 60268-7 Annex B?

The five measurable conditions of the normative Annex B: a) a cross-sectional area of 5 mm² or less within the concha and the first 4 mm of the canal; b) in the rest of the canal, an area less than 0,6 of the canal’s (“The average ear canal area of an adult is 45 mm²”); c) a volume, mounting parts included, less than 130 mm³; d) responses to neighbouring one-third-octave bands of pink noise that “do not differ by more than 3 dB”; e) an output level with the sound entrance sealed “at least 15 dB below that with the entrance open, at all measuring frequencies”. Items f) and g), the suspension and the physician’s certificate, are not numbers.

Parameters

Returns: An EarCanalMicrophoneVerification.

Raises

verify_rated_impedance(
    frequencies_hz: ArrayLike,
    impedance_ohm: ArrayLike,
    *,
    rated_impedance_ohm: float,
    rated_frequency_range_hz: tuple[float, float],
) -> RatedImpedanceVerification

Is the rated impedance of a headphone chosen as IEC 60268-7 8.2.1 asks?

The rated impedance passes when the lowest modulus of the measured impedance within the rated frequency range is at least 80 % of it. The result also lists the frequencies up to 20 kHz where the modulus falls below that value, which the specification should state, and whether the measurement covered 20 Hz to 20 kHz as 8.2.2.2 c) requires.

Parameters

Returns: A RatedImpedanceVerification.

Raises

working_sound_pressure_level(
    sound_pressure_level_db: float,
    source_emf_v: float,
    *,
    rated_impedance_ohm: float,
    rated_source_impedance_ohm: float,
    headphone_impedance_ohm: complex | None = None,
) -> float

The working sound pressure level of IEC 60268-7 8.5.2 b) to d).

The level produced by the e.m.f. “of such value that 1 mW would be dissipated in a pure resistance equal to the rated impedance of the headphone, connected in place of it”, scaled from a reading at another e.m.f. With a 500 Hz sine it is 8.5.2 b); with the level of the simulated programme signal (programme_signal_level) it is c), and d) when that level is A-weighted and free-field compensated.

The method of 8.5.3 c) sets that e.m.f. another way: “so that the voltage across the input connector of the headphone is such that it would cause 1 mW to be dissipated in a pure resistance equal to the rated impedance”, $\sqrt{PR}$ across the headphone itself, whose impedance $Z$ at 500 Hz takes its share of the e.m.f., $E=\sqrt{PR}|Z+R_{\mathrm{s}}|/|Z|$. The two readings agree when $Z$ is the rated impedance and part when it is not: a 32 ohm headphone of 40 ohm at 500 Hz on a 120 ohm source reads 1,49 dB lower by 8.5.3 c). Pass headphone_impedance_ohm for the reading of 8.5.3 c) (see docs/ERRATA.md); it applies to the 500 Hz sine of 8.5.2 b), since a noise signal has no single impedance to divide by.

Parameters

Returns: The working sound pressure level, in dB re 20 µPa.

Raises