hearing.earmuff_insertion_loss

The insertion loss of an earmuff on an acoustic test fixture (ISO 4869-3:2007).

ISO 4869-3 measures an earmuff without a listener. The acoustic test fixture (ATF) is a metal cylinder of 135 mm diameter with its end faces 145 mm apart, a pressure microphone flush with one of them and a spacer that carries the headband (5.1). In a random-incidence field or a plane progressive wave, the level at the microphone is measured without the earmuff and again with it seated on the fixture, and the difference in each one-third-octave band is the insertion loss (3.5, 5.4.2):

in bands from at least 63 Hz to 8 kHz (5.3), from at least three fittings unless the device’s repeatability is known (5.4.3), reported to the nearest 0,1 dB and drawn with increasing values downwards (Clause 6).

What it is for, and what it is not (Clause 1). The method checks production spreads for type approval or certification and the change of performance with age, and makes sure the samples sent for the subjective test of ISO 4869-1 are typical of their type. It is not the basic type test, and its data “are not intended to be quoted as representing the real-ear sound attenuation of an ear-muff, nor the protection provided by the ear-muff”; the Introduction adds that its results are not those of ISO 4869-1. So EarmuffInsertionLossResult is not an input of phonometry.hearing.assumed_protection_value, phonometry.hearing.hml_rating or phonometry.hearing.snr_rating: those take the real-ear attenuation of sixteen subjects, phonometry.hearing.real_ear_attenuation, and nothing here converts one into the other.

The test site (5.1.4, 5.2). check_random_incidence_field judges the random-incidence field of 5.2.2 with the diffuse-field judgement of ISO 8253-2:2009, 5.3, and this standard’s Table 1 (RANDOM_INCIDENCE_VARIATION_LIMITS); Annex A notes that the ATF itself may serve as the directional microphone where its front-to-random index, ATF_FRONT_TO_RANDOM_INDEX_DB, reaches 4 dB. check_plane_progressive_wave judges the plane wave of 5.2.3, and verify_fixture_isolation the acoustic isolation of the fixture with its isolation cup (5.1.4).

The uncertainty (Annex B). The insertion loss is modelled as the two levels plus three zero-mean inputs for the fixture, the sound field and the equipment (Formula (B.1)), all normal with a sensitivity coefficient of 1, so

(Formula (B.2)). The standard uncertainty of each level is the standard deviation of the mean of the repeated measurements; Table B.1 prints typical values, 0,5 dB open, 1,0 dB occluded, 0,3 dB, 0,5 dB and 0,2 dB, which give $u$ = 1,3 dB and $U$ = 2,6 dB (EARMUFF_INSERTION_LOSS_UNCERTAINTY).

Clause, table and formula numbers refer to ISO 4869-3:2007(E).

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

Constant (mapping).

ATF_FRONT_TO_RANDOM_INDEX_DB = {500.0: 1.7, 630.0: 2.2, 800.0: 2.8, 1000.0: 3.2, 1250.0: 4.6, 1600.0: 4.6, 2000.0: 6.3, 2500.0: 6.5, 3150.0: 5.9, 4000.0: 2.9, 5000.0: -0.6, 6300.0: 5.1, 8000.0: 5.9}

check_plane_progressive_wave(
    end_face_levels_db: ArrayLike,
    *,
    facing_levels_db: ArrayLike | None = None,
    facing_away_levels_db: ArrayLike | None = None,
    front_to_rear_index_db: ArrayLike | None = None,
    frequencies: ArrayLike | None = None,
) -> PlaneProgressiveWaveCheck

Is the plane progressive wave of the test site good enough? (5.2.3).

With the fixture removed:

• the levels at the two points the centres of the fixture’s end faces normally occupy, each read at 0° incidence, differ by 2 dB at most;
• from the 500 Hz band up, a directional microphone at the reference point reads at least 10 dB more facing the source than turned 180° away from it, the microphone having a front-to-rear sensitivity index greater than 15 dB (different microphones may serve different bands).

During the insertion loss measurement the fixture is turned so that the wave meets its end faces at grazing incidence; that is its orientation, not a condition judged here.

Parameters

Returns: PlaneProgressiveWaveCheck. Without the directional readings the directional test is not judged and the verdict does not pass.

Raises

check_random_incidence_field(
    position_levels_db: Mapping[str, ArrayLike],
    reference_levels_db: ArrayLike,
    *,
    directional_levels_db: ArrayLike | None = None,
    front_to_random_index_db: ArrayLike | None = None,
    frequencies: ArrayLike | None = None,
) -> DiffuseSoundFieldCheck

Is the random-incidence field of the test site good enough? (5.2.2).

With the fixture removed, the level read by an omnidirectional microphone kept in one orientation at six positions 150 mm from the reference point, on the front-back, right-left and up-down axes, stays within ±2,5 dB of the level at the reference point, and the right and left positions within 3 dB of each other; from the 500 Hz band up, the levels a directional microphone reads at the reference point in any two directions, the directions of the extremes among them, stay within the variation Table 1 allows for its front-to-random sensitivity index (RANDOM_INCIDENCE_VARIATION_LIMITS).

This is the diffuse-field judgement of ISO 8253-2:2009, 5.3, with this standard’s Table 1, and it returns the same DiffuseSoundFieldCheck. The fixture itself may be the directional microphone in the bands where its index (ATF_FRONT_TO_RANDOM_INDEX_DB) reaches 4 dB; pass the index band by band, NaN where it has none, and the bands it cannot judge are left unjudged.

Parameters

Returns: DiffuseSoundFieldCheck.

Raises

earmuff_insertion_loss(
    open_levels_db: ArrayLike,
    occluded_levels_db: ArrayLike,
    *,
    frequencies: ArrayLike | None = None,
    isolation_cup_levels_db: ArrayLike | None = None,
    fixture_uncertainty_db: float = 0.3,
    sound_field_uncertainty_db: float = 0.5,
    equipment_uncertainty_db: float = 0.2,
) -> EarmuffInsertionLossResult

The insertion loss of an earmuff on the acoustic test fixture (5.4).

The level at the fixture’s microphone is measured without the earmuff, the earmuff is seated on the fixture (5.4.1), and after about 30 s the level is measured again (5.4.2); the difference in each one-third-octave band is the insertion loss. With repeated fittings (5.4.3) it is the mean open level less the mean occluded level:

Annex B’s uncertainty is evaluated band by band: the standard uncertainty of each mean level is the standard deviation of the mean of its repeated measurements (B.3), or Table B.1’s typical 0,5 dB open and 1,0 dB occluded when fewer than two were given, and the three $\delta$ terms default to Table B.1’s 0,3 dB, 0,5 dB and 0,2 dB.

The result is an insertion loss on a fixture. It screens production and ageing and is not the real-ear attenuation ISO 4869-2 needs (Clause 1); use phonometry.hearing.real_ear_attenuation for that.

Parameters

Returns: EarmuffInsertionLossResult.

Raises

Constant (phonometry.hearing.earmuff_insertion_loss.InsertionLossUncertaintyBudget).

Constant (tuple).

EARMUFF_TEST_BANDS_HZ = (63.0, 80.0, 100.0, 125.0, 160.0, 200.0, 250.0, 315.0, 400.0, 500.0, 630.0, 800.0, 1000.0, 1250.0, 1600.0, 2000.0, 2500.0, 3150.0, 4000.0, 5000.0, 6300.0, 8000.0)

EarmuffInsertionLossResult(
    insertion_loss_db: np.ndarray,
    repetition_insertion_loss_db: np.ndarray,
    open_levels_db: np.ndarray,
    occluded_levels_db: np.ndarray,
    open_uncertainty_db: np.ndarray,
    occluded_uncertainty_db: np.ndarray,
    fixture_uncertainty_db: float,
    sound_field_uncertainty_db: float,
    equipment_uncertainty_db: float,
    frequencies: np.ndarray,
    isolation_cup_levels_db: np.ndarray | None = None,
)

The insertion loss of an earmuff on the test fixture (5.4, Annex B).

A screening quantity, not an attenuation at the ear: Clause 1 says the data are not to be quoted as the real-ear attenuation of the earmuff nor as the protection it provides, and nothing in the library feeds them to the ISO 4869-2 methods.

Attributes

EarmuffInsertionLossResult.budget(
    frequency: float,
) -> InsertionLossUncertaintyBudget

The budget of Table B.1 for one band of this measurement.

Parameters

Returns: InsertionLossUncertaintyBudget.

Raises

property

The expanded uncertainty per band, $U=2u$ (B.4), in dB.

Returns: One value per band.

property

Per band, whether the isolation cup reads 10 dB lower at least (5.3).

Returns: One boolean per band, or None without cup levels.

property

The mean occluded level less the isolation cup’s, per band, in dB.

Returns: The margin, or None when no cup levels were given.

property

How far the open level strayed from its first measurement, in dB.

5.3 keeps the test signal within ±1 dB of the level set before the measurement; with one open measurement nothing can be seen and the drift reads 0.

Returns: The largest absolute departure per band.

property

Per band, whether the open level stayed within ±1 dB (5.3).

Returns: One boolean per band.

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

Draw the insertion loss as Clause 6 asks, increasing downwards.

The scale is IEC 60263’s 50 dB per decade of frequency, which Clause 6 asks of every graph, so given axes take that aspect too.

Parameters

Returns: The axes.

property

The number of fittings measured.

Returns: The count of occluded measurements.

property

Whether at least three fittings were measured (5.4.3).

Fewer are allowed when the device’s repeatability is known, which is the tester’s judgement.

Returns: True from three repetitions.

property

The insertion loss as Clause 6 reports it, to the nearest 0,1 dB.

Returns: One value per band, halves rounded upwards.

property

Whether adjacent bands of the open level differ by 5 dB at most (5.3).

Returns: True when the system’s one-third-octave response has no step larger than 5 dB between neighbouring bands.

property

The combined standard uncertainty per band (Formula (B.2)), in dB.

Returns: One value per band.

FixtureIsolationCheck(frequencies: np.ndarray, isolation_db: np.ndarray)

Whether the test fixture isolates its microphone well enough (5.1.4).

The requirement is the clause’s, read from the bands as required_db, so a check cannot be built against another one.

Attributes

property

The isolation less the requirement per band, in dB.

Returns: Negative where the fixture falls short, NaN where nothing is required.

property

Whether the fixture’s isolation meets 5.1.4 in every band.

Returns: True when no band falls short.

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

Draw the isolation against the requirement, band by band.

Parameters

Returns: The axes.

property

The least isolation 5.1.4 asks per band, in dB.

Returns: 50 dB from 63 Hz to 250 Hz, 65 dB from 315 Hz to 4 kHz and 55 dB above; NaN below 63 Hz, where it asks nothing.

property

Per band, whether the isolation reaches the requirement.

Returns: One boolean per band; True where nothing is required.

InsertionLossUncertaintyBudget(
    open_level_db: float,
    occluded_level_db: float,
    fixture_db: float,
    sound_field_db: float,
    equipment_db: float,
)

The uncertainty budget of an insertion loss (Annex B, Table B.1).

Five normal inputs with a sensitivity coefficient of 1 (B.3), so the combined standard uncertainty is their root sum of squares (Formula (B.2)) and the expanded one twice that (B.4).

Attributes

property

$u=\sqrt{\sum (c_i{u}_i{)}^2}$ (Formula (B.2)), in dB.

Returns: The combined standard uncertainty, unrounded.

property

The five standard uncertainties in the order of Table B.1, in dB.

Returns: The components.

property

$U=2u$ (B.4), in dB.

Returns: The expanded uncertainty, unrounded.

property

The five input quantities of Table B.1, in its order.

Returns: Their symbols.

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

Draw the five contributions as bars, with the combined value.

Parameters

Returns: The axes.

PlaneProgressiveWaveCheck(
    frequencies: np.ndarray,
    end_face_difference_db: np.ndarray,
    front_to_back_db: np.ndarray,
    front_to_rear_index_db: np.ndarray,
)

Whether the plane progressive wave of the test site is good enough (5.2.3).

Attributes

property

Per band, whether the directional test applies (500 Hz and up).

Returns: One boolean per band.

property

Whether every band from 500 Hz up was read with a suitable microphone.

Returns: True when none is left unjudged.

property

Per band, whether the two end-face positions are within 2 dB.

Returns: One boolean per band.

property

Per band, whether the microphone’s front-to-rear index exceeds 15 dB.

Returns: One boolean per band; False where no index was given.

property

Whether the plane wave qualifies, every requirement judged.

Returns: True when every band meets both conditions.

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

Draw both conditions against their limits, band by band.

Parameters

Returns: The axes.

property

Per band, whether facing the source reads at least 10 dB more.

Bands below 500 Hz count as meeting it; a band from 500 Hz up without a suitable microphone’s reading reads False.

Returns: One boolean per band.

Constant (tuple).

RANDOM_INCIDENCE_VARIATION_LIMITS = ((5.0, 5.0), (4.0, 4.0))

verify_fixture_isolation(
    open_levels_db: ArrayLike,
    cup_levels_db: ArrayLike,
    *,
    frequencies: ArrayLike | None = None,
) -> FixtureIsolationCheck

Does the test fixture isolate its microphone well enough? (5.1.4).

With the test signal of 5.3 at the actual test site, the microphone is covered by an acoustic isolation test cup sealed to the fixture, and the acoustic isolation (3.7) is the level without the cup less the level with it. 5.1.4 asks for at least 50 dB in the bands centred from 63 Hz to 250 Hz, 65 dB from 315 Hz to 4 kHz, and 55 dB above. Airborne and structure-borne paths both count, which is why the fixture sits on a resilient mounting (5.1.2).

Parameters

Returns: FixtureIsolationCheck.

Raises