building.measurement.service_equipment

Sound from service equipment and activities in buildings, engineering method.

This is the engineering method of ISO 16032 for the sound pressure level a building’s own equipment makes in a room: taps, showers, baths and water closets, ventilation, heating and cooling, lifts, rubbish chutes, boilers and pumps, car-park doors and, new in the revision, activities, the sources whose operation nobody on site controls (a restaurant’s coolers, amplified music next door, a sports hall). It is the method a dispute over a noisy installation ends in; the survey method for the same quantities is ISO 10052 (survey_service_equipment_level).

Which text. The implementation follows the draft of the second edition, ISO/DIS 16032:2023, in its German publication E DIN EN ISO 16032:2023-05 (prEN ISO 16032:2023, German and English text). The draft is the only text of the revision the library holds, so everything implemented here follows it, and it is cited as a draft; clause numbers are those of the draft. What changed from the withdrawn ISO 16032:2004 is said where it matters below and in the guide.

Where it applies (Clause 1). The method is meant for rooms of about 300 m³ or less, in dwellings, hotels, schools, offices and hospitals. It is not meant for large auditoriums and concert halls, nor for a source so far from the building that the weather changes the propagation, although the operating conditions of Annex B can still be used there. Nothing in the library checks the volume.

The procedure. The linear band spectrum is recorded together with the A- or C-weighted level (a multispectral recording, Clause 6), in one of three quantities of Table 1: the maximum level with time weighting S or F, taken at the moment the weighted level peaks, or the equivalent continuous level over a stated integration time. It is measured at three microphone positions, one in the loudest corner of the room (7.2) and two in the reverberant field (7.3). When the three A-weighted readings differ by more than 3,0 dB the corner is measured again together with two new room positions, and a third time with two more if the spread of the six still reaches 6,0 dB (7.4.1, check_position_spread). Every reading enters the band average of Formula (1), rounded to one decimal (7.5):

A sound source in the room itself, a ventilation outlet for example, gets one additional position of its own (7.9, additional_microphone_position), whose reading is reported separately, never standardized and never averaged with the others.

Background (Clause 9). With the background band level $L_2$ and the measured one $L_1$, the difference $\Delta L=L_1-L_2$ decides: at 10 dB or more nothing is corrected; from 4 dB to 10 dB the band is corrected with Formulae (7) to (9),

and below 4 dB the correction is held at the 2,2 dB that a 4 dB difference gives, the band is reported as influenced by the background and the result is an upper limit of the equipment level (service_equipment_background_correction).

Standardization and normalization (4.8, 7.7). If a regulation asks for it, the corrected bands are referred to a reference reverberation time $T_0$ (0,5 s unless another value is specified) or to a reference absorption area $A_0=10$ m², Formulae (5) and (6):

Only the one-third-octave bands 50 Hz to 5 000 Hz (octave bands 63 Hz to 4 000 Hz) are standardized or normalized, because the reverberation time of the bands outside cannot be measured reliably; those bands still enter the weighted sums, unstandardized, and the report has to say so.

Weighted values (4.3, 4.4, 7.8). The A- and C-weighted single numbers are energy sums of the corrected bands plus the Annex A corrections, rounded to whole decibels. The A-weighted value uses either the restricted range 50 Hz to 5 000 Hz or the extended range 25 Hz to 10 000 Hz; the C-weighted value uses the extended range. The single numbers are named as Table 1 names them, for example $L_{\mathrm{A},\mathrm{Smax},\mathrm{n}T}$ or $L_{\mathrm{C},\mathrm{eq}}$. 4.2 also admits “a specific frequency range” for the bands, and the form of Annex C has a box for it, but 7.8 forms the weighted values over the two ranges above only, and so does the library: a band set that does not cover the chosen range is refused.

Precision (Clause 10). Table 2 gives the reproducibility standard deviation of a room-average band level, 1,9 dB at the lowest bands down to 1,0 dB from 800 Hz up, and 0,8 dB and 1,2 dB for the A- and C-weighted values of a steady, flat sound at least 10 dB above the background. A result carries them as ServiceEquipmentResult.reproducibility_db and ServiceEquipmentResult.weighted_reproducibility_db; the draft states no coverage factor, so no expanded uncertainty is formed.

On-site checks. The draft also sets numbers the operator meets while measuring, and each is a verdict here. The corner microphone keeps at least 0,2 m from any obstacle (7.2), judged from the distance measured on site by check_service_equipment_positions. A calibration that deviates from previous calibrations by more than 0,5 dB takes the equipment out of use (Clause 5, verify_calibration_deviation). The background is measured over approximately 30 s (7.6, check_background_duration, which reports each departure from 30 s and judges it against the tolerance the operator names, because the draft prints none). Where the background varies in time, its maximum watched for 10 min to 15 min at the corner and 10 dB or more below the equipment lets the result stand without correction (NOTE to Clause 9, check_varying_background). A maximum less than 5 dB above the equivalent level in the middle of the frequency range says a period was not disturbed by doors or footsteps (Clause 9, check_measurement_disturbance), and calculated single numbers more than 2 dB from the instrument’s reading send the calculation back for a check (NOTE to 7.8, check_instrument_agreement).

Operating conditions (Annex B). How each kind of equipment is to be run while it is measured, and for how long the equivalent level is integrated, is the data of SERVICE_EQUIPMENT_OPERATING_CONDITIONS, including the activity sources of B.10.

Printed defects. Table A.1 prints a C-weighting of 0 dB for every one-third-octave band from 1 600 Hz to 10 000 Hz and of -5 dB at 25 Hz. Its own octave column gives -0,2 dB at 2 000 Hz, -0,8 dB at 4 000 Hz and -3,0 dB at 8 000 Hz, and IEC 61672-1, which Clause 4 cites for the weightings, gives -0,1 dB to -4,4 dB from 1 600 Hz up and -4,4 dB at 25 Hz. The library uses the IEC 61672-1 values in those cells. The cells, and the other defects of the draft this module meets (a background correction sent to the reverberation clause, the corner procedure sent to Clause 6, and a position ladder with no rule at exactly 6,0 dB and 9,0 dB nor for three readings 6,0 dB or more apart), are registered in docs/ERRATA.md.

No numeric oracle. The draft prints no worked example. The conformance of this module rests on closed forms and on the numbers the draft prints: the 2,2 dB of a 4 dB background difference, Table A.1, Table 2, and the limits of the on-site checks met exactly and missed.

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

additional_microphone_position(
    room_dimensions_m: ArrayLike,
    source_position_m: ArrayLike,
    *,
    mounting: _Mounting,
) -> np.ndarray

The extra position for a sound source in the room (ISO/DIS 16032:2023 7.9).

A source in the room itself, a ventilation outlet in a wall or in the ceiling for example, gets one additional microphone position of its own: 1 m in front of a wall source at 1,5 m above the floor, or 1,5 m above the floor directly below a ceiling source. The reading there is reported separately, is neither standardized nor normalized, and is not averaged with positions 1, 2 and 3: pass it to service_equipment_level on its own, without a reverberation time.

Coordinates are those of check_service_equipment_positions: a rectangular room with one corner at the origin, x along the length, y along the width and z the height above the floor. A wall source belongs to the wall it is nearest to, and the position is 1 m from the source along that wall’s inward normal.

Parameters

Returns: The position (x, y, z), in m.

Raises

BackgroundDurationCheck(durations_s: np.ndarray, tolerance_s: float)

Whether each background was measured over approximately 30 s (7.6).

Attributes

property

How far each duration departs from 30 s, in s.

Returns: The duration less 30 s, one per background measurement.

property

The largest departure from 30 s, either way, in s.

Returns: The largest absolute value of departures_s.

property

The 30 s of 7.6, fixed by the draft.

Returns: The nominal background measurement time, in s.

property

Whether every background was measured over approximately 30 s.

“Approximately” is the operator’s tolerance_s, not a number of the draft.

Returns: True when no duration departs from 30 s by more than tolerance_s.

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

Plot each duration against the 30 s of 7.6 and the tolerance accepted.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

Per background, whether it departs from 30 s by no more than the tolerance.

A departure of exactly tolerance_s is within it.

Returns: One boolean per background measurement.

CalibrationDeviationResult(
    levels_db: np.ndarray,
    previous_levels_db: np.ndarray,
    deviations_db: np.ndarray,
)

Whether the instrumentation may be used, by its calibrations (Clause 5).

Attributes

property

The largest deviation of a reading from the calibrations before it.

Returns: The largest of deviations_db, in dB.

property

The 0,5 dB of Clause 5, fixed by the draft.

Returns: The largest deviation that still lets the equipment be used, in dB.

property

Whether no reading deviates from an earlier one by more than 0,5 dB.

Returns: True when the equipment may be used; False takes it out of use until the reason is clarified and the sensitivity put right (Clause 5).

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

Plot each reading against the window the earlier calibrations leave it.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

check_background_duration(
    durations_s: ArrayLike,
    *,
    tolerance_s: float,
) -> BackgroundDurationCheck

Was the background measured over approximately 30 s (ISO/DIS 16032:2023 7.6)?

“The background sound pressure level shall be determined in frequency bands as the equivalent continuous sound pressure levels over a period of approximately 30 s just before or after each set of measurements. The same microphone positions as used for the service equipment sound pressure level measurements shall be used.”

The draft writes approximately and prints no tolerance, so the library sets none: the operator names the departure from 30 s, either way, that their report accepts as approximate, and each duration is held to it. A departure of exactly the tolerance is within it, and tolerance_s=0 holds every background to 30 s exactly. Each departure is reported (BackgroundDurationCheck.departures_s) whatever the verdict. Whether the background was taken just before or after each set, and at the same positions, is not a number and is left to the operator.

Parameters

Returns: BackgroundDurationCheck.

Raises

check_instrument_agreement(
    result: ServiceEquipmentResult,
    instrument_levels_db: Mapping[str, ArrayLike],
) -> InstrumentAgreementCheck

Does the calculation agree with the instrument (NOTE to ISO/DIS 16032:2023 7.8)?

“It can be useful to compare the corrected and calculated A- and C-weighted results with the values registered directly by the instrument. If the difference is more than 2 dB, the calculations should be checked for possible explanations.” The NOTE is new in the revision. The calculated value is the single number of the result as 7.8 rounds it, and a difference of exactly 2 dB is not “more than 2 dB”.

The instrument registers the level as measured, so a single number corrected for the background, or standardized, may differ from it for those reasons alone: the NOTE asks for an explanation, not a fault. Name the single number each instrument value is compared with; "LA,eq" against the A-weighted equivalent level the meter showed is the plain case.

Parameters

Returns: InstrumentAgreementCheck.

Raises

check_measurement_disturbance(
    maximum_levels_db: ArrayLike,
    equivalent_levels_db: ArrayLike,
) -> MeasurementDisturbanceCheck

Was a measurement period disturbed (ISO/DIS 16032:2023 Clause 9)?

“A simple check on-site is to compare the maximum level to the equivalent level in the middle of the frequency range during each measurement period, for many stable sources this difference should be less than 5 dB to indicate that the measurement has not been disturbed by closing doors, footfall noise etcetera.” The check is the paragraph’s, new in the revision: it holds for many stable sources, not for a source whose own level varies, and the draft does not name the band beyond “the middle of the frequency range”, which the operator chooses. A difference of exactly 5 dB is not “less than 5 dB” and marks the period.

Parameters

Returns: MeasurementDisturbanceCheck.

Raises

check_position_spread(levels_db: ArrayLike) -> PositionSpreadCheck

Are the positions measured so far enough (ISO/DIS 16032:2023 7.4.1)?

The A-weighted levels read directly from the instrument, without any correction, decide how many positions are measured. With three readings (corner, 2, 3) a spread up to and including 3,0 dB lets the corrections proceed; otherwise the corner is measured again with two new room positions (4 and 5), and the six must differ by less than 6,0 dB, then once more (corner, 6 and 7) and the nine by less than 9,0 dB. 7.4.2 applies the same steps to a maximum level, with the sound exposure level $L_{\mathrm{AE}}$ allowed in its place for short events.

The draft interrupts the session when the difference “is larger than 9,0 dB and is related to unpredictable time domain variations whereas the readings in the corner position confirm the sound source is stable”: the reasons are investigated before a new series, and no reading of the interrupted one is used. Levels alone cannot say what a difference is related to, so the check answers "interrupt" whenever no further stage can pass and leaves those two conditions to the operator: a spread still 9,0 dB or more after nine readings, and a spread that already reaches 9,0 dB after three or six, since more readings can only widen it. The third paragraph of 7.4.1 sends to positions 6 and 7 only a difference “less than 9,0 dB”.

The draft states no rule at a spread of exactly 6,0 dB after six readings or 9,0 dB after nine, nor for three readings 6,0 dB or more apart. The ladder is read as a sequence: a stage that does not pass goes to the next while a later one can still pass, and a spread at a limit does not pass it (docs/ERRATA.md).

Parameters

Returns: PositionSpreadCheck.

Raises

check_service_equipment_positions(
    room_dimensions_m: ArrayLike,
    corner_position_m: ArrayLike,
    room_positions_m: ArrayLike,
    *,
    source_positions_m: ArrayLike | None = None,
    small_room: bool = False,
    corner_obstacle_distance_m: ArrayLike | None = None,
) -> ServiceEquipmentPositionCheck

Are the microphone positions far enough apart (ISO/DIS 16032:2023 7.2, 7.3)?

The corner position is preferably 0,5 m from the two walls and the floor, raised to 1,0 m or 1,5 m where furniture is in the way, and at least 0,2 m from any obstacle. The reverberant-field positions keep at least 1,0 m from each other and from the corner (1,5 m preferred), 1,5 m from any sound source in the room, 0,50 m from every room surface (0,30 m in a small room where 0,50 m cannot be met) and a height from 0,5 m to 2,0 m. Pass the positions of one stage of 7.4.1 at a time: positions 4 and 5 keep the distances of 7.3 like 2 and 3.

The verdict holds the corner position to its height of 0,5 m to 1,5 m and the reverberant-field positions to every distance and height of 7.3. The two preferences, 0,5 m from the walls of the corner and 1,5 m between positions, are reported beside it and do not decide it (preferred_corner_wall_distance, preferred_separation).

A room outline cannot show furniture, so the 0,2 m that 7.2 asks between the corner microphone and any obstacle is judged from the distance measured on site, when it is given: “The microphone position shall be at least 0,2 m away from any obstacle”, inclusive. The draft states it in the clause on the corner position and nowhere for the reverberant-field positions, whose 0,50 m from the room surfaces 7.3 sets instead.

Parameters

Returns: ServiceEquipmentPositionCheck.

Raises

check_varying_background(
    background_maximum_db: ArrayLike,
    equipment_levels_db: ArrayLike,
    *,
    observation_time_s: float,
    frequencies_hz: ArrayLike | None = None,
) -> VaryingBackgroundCheck

Can the result stand without correction (NOTE to ISO/DIS 16032:2023 Clause 9)?

Clause 9 corrects the bands for a background roughly constant in time, and its NOTE says that for a background varying in time, road traffic for example, “a reliable correction cannot be made. However, the maximum sound pressure levels of the background noise could be determined over a period of 10 min to 15 min in the corner microphone position. If the maximum level is 10 dB or more below the service equipment sound pressure level the result can be regarded valid without correction.” The NOTE also suggests checking “the validity in all relevant octave-bands”, words kept from the 2004 edition, which measured in octaves: the margin is judged in every band given, one-third octaves or octaves, and a single weighted value is one band.

Both ends of the 10 min to 15 min are included, and so is a margin of exactly 10 dB (“10 dB or more”). The 2004 edition took the maximum “in one of the microphone positions”; the draft names the corner.

Parameters

Returns: VaryingBackgroundCheck.

Raises

InstrumentAgreementCheck(
    calculated_db: Mapping[str, float],
    instrument_db: Mapping[str, float],
)

Whether the calculated single numbers agree with the instrument (7.8).

Attributes

property

The calculated value less the instrument’s, per single number, in dB.

Returns: Keyed like calculated_db.

property

The single numbers more than 2 dB from the instrument.

Returns: Their notations, in the order of calculated_db.

property

The 2 dB of the NOTE to 7.8, fixed by the draft.

Returns: The largest difference, either way, that agrees, in dB.

property

Whether every calculated value is within 2 dB of the instrument.

Returns: True when no single number calls for a second look at the calculation.

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

Plot each calculated-less-instrument difference against 2 dB.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

loudest_corner(
    c_weighted_levels_db: ArrayLike,
    *,
    excluded: Sequence[int] = (),
) -> int

The corner that becomes microphone position 1 (ISO/DIS 16032:2023 7.2).

Position 1 is the corner with the highest C-weighted level, measured directly (a hand-held integrating meter will do: no band calculation, no reverberation or background correction) as the maximum level with time weighting S or F or as the equivalent level, under the chosen operating conditions or any steady condition that finds the loudest corner. A corner whose level is dominated by direct sound from a source in the room, a ventilation outlet for example, is left out.

Parameters

Returns: The index of the loudest corner left, counted from 0; the first of a tie.

Raises

MeasurementDisturbanceCheck(
    maximum_levels_db: np.ndarray,
    equivalent_levels_db: np.ndarray,
)

Whether each measurement period was free of disturbances (Clause 9).

Attributes

property

The maximum less the equivalent level of each period, in dB.

Returns: One difference per measurement period.

property

The periods whose difference reaches 5 dB, counted from 0.

Returns: Their indices, in the order given.

property

The 5 dB of Clause 9, fixed by the draft.

Returns: The difference a period must stay under, in dB.

property

Whether no period shows a disturbance.

Returns: True when every period’s maximum lies less than 5 dB above its equivalent level.

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

Plot each period’s maximum-to-equivalent difference against 5 dB.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

Per period, whether the difference is less than 5 dB.

A difference of exactly 5,0 dB is not “less than 5 dB”.

Returns: One boolean per measurement period.

OperatingCondition(
    clause: str,
    equipment: str,
    maximum_level: bool,
    equivalent_level: bool,
    conditions: str,
    maximum_cycle: str | None,
    equivalent_cycle: str | None,
    nominal_integration_time_s: float | None,
    parameters: Mapping[str, float] = ...,
)

How one kind of equipment is run while it is measured (Annex B).

Annex B is normative, but B.1.1 lets national requirements and regulations override it, and equipment it does not list is measured by the principles of the document with the chosen conditions reported in detail.

Attributes

PositionSpreadCheck(levels_db: np.ndarray)

Whether the readings so far are enough to average (7.4.1).

The readings are the only field. The stage, the spread, the limit of the stage and the action the draft prescribes are all read from them, so a check cannot be built, or rewritten with dataclasses.replace, with an action the readings do not lead to.

Attributes

property

What 7.4.1 prescribes for these readings.

Returns: "proceed" to the corrections of 7.5 to 7.7 when the spread is within limit_db, "add_positions" for the corner again and two new room positions while a later stage can still pass, or "interrupt" when none can: after nine readings, or sooner when the spread already reaches 9,0 dB, since more readings can only widen it.

property

Sample standard deviation of the corner readings, in dB.

7.4.1 takes it as representative of the room average when a statistical value (a 5 % value, the n-th highest) is wanted.

Returns: The deviation, or None with a single corner reading.

property

The spread this stage allows, fixed by the draft for the stage.

Returns: 3,0 dB after three readings (inclusive), 6,0 dB after six or 9,0 dB after nine (strict), in dB.

property

The room positions to add, (4, 5) or (6, 7), or ().

property

Whether the readings may be averaged as they are.

Returns: True when action is "proceed".

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

Plot the readings against the spread the stage allows.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

The difference between the highest and lowest reading, in dB.

property

How many sets of three readings there are: 1, 2 or 3.

service_equipment_background_correction(
    levels_db: ArrayLike,
    background_db: ArrayLike,
    *,
    frequencies_hz: ArrayLike | None = None,
) -> ServiceEquipmentBackgroundResult

Correct band levels for the background (ISO/DIS 16032:2023 Clause 9).

With $\Delta L=L_1-L_2$ the band is left as it is at 10 dB or more, corrected with $K=-10\lg (1-10^{-0,1\Delta L})$ from 4 dB to 10 dB (Formulae (7) to (9)), and below 4 dB the correction is held at 2,2 dB, the value a 4 dB difference gives. Such a band is influenced by the background, which the report has to state, and for a comparison with a limit the result is an upper limit of the equipment level. A background at or above the measured level is the same case: held at 2,2 dB rather than left undefined.

The method assumes a background roughly constant in time. Where it is not, the NOTE to Clause 9 suggests the maximum level of the background over 10 to 15 minutes at the corner position instead: if it is 10 dB or more below the equipment, the result stands without correction (check_varying_background).

Parameters

Returns: ServiceEquipmentBackgroundResult.

Raises

service_equipment_level(
    levels_db: ArrayLike,
    frequencies_hz: ArrayLike,
    *,
    quantity: _Quantity,
    band: _Band = 'third',
    background_db: ArrayLike | None = None,
    reverberation_time_s: ArrayLike | None = None,
    volume_m3: float | None = None,
    reference_reverberation_time_s: float = 0.5,
    a_weighting_range: _Range = 'restricted',
) -> ServiceEquipmentResult

Sound pressure level of service equipment in a room (ISO/DIS 16032:2023).

The chain of Clauses 6 and 7: the band levels of every reading are energy-averaged by Formula (1) and rounded to one decimal (7.5), corrected for the background by Clause 9, standardized by Formula (5) or normalized by Formula (6) where a reverberation time is given (7.7, 50 Hz to 5 000 Hz only), and summed into the A- and C-weighted single numbers of Table 1 with the Annex A corrections, rounded to whole decibels (7.8).

The readings are the linear band spectra of the measurements of 7.4: positions 1, 2 and 3 at least, with the corner reading repeated in each further stage that check_position_spread asks for. Take them at the instant the A- or C-weighted level peaks for a maximum level (Clause 6), or over the integration time of Annex B for an equivalent level.

Parameters

Returns: ServiceEquipmentResult.

Raises

Constant (mapping).

Constant (mapping).

SERVICE_EQUIPMENT_REPRODUCIBILITY = {'third': {25.0: 1.9, 31.5: 1.9, 40.0: 1.9, 50.0: 1.9, 63.0: 1.9, 80.0: 1.9, 100.0: 1.9, 125.0: 1.9, 160.0: 1.9, 200.0: 1.5, 250.0: 1.5, 315.0: 1.5, 400.0: 1.2, 500.0: 1.2, 630.0: 1.2, 800.0: 1.0, 1000.0: 1.0, 1250.0: 1.0, 1600.0: 1.0, 2000.0: 1.0, 2500.0: 1.0, 3150.0: 1.0, 4000.0: 1.0, 5000.0: 1.0, 6300.0: 1.0, 8000.0: 1.0, 10000.0: 1.0}, 'octave': {31.5: 1.9, 63.0: 1.9, 125.0: 1.9, 250.0: 1.5, 500.0: 1.2, 1000.0: 1.0, 2000.0: 1.0, 4000.0: 1.0, 8000.0: 1.0}}

Constant (mapping).

SERVICE_EQUIPMENT_WEIGHTED_REPRODUCIBILITY = {'A': 0.8, 'C': 1.2}

Constant (mapping).

SERVICE_EQUIPMENT_WEIGHTING = {'third': {'A': {25.0: -44.7, 31.5: -39.4, 40.0: -34.6, 50.0: -30.2, 63.0: -26.2, 80.0: -22.5, 100.0: -19.1, 125.0: -16.1, 160.0: -13.4, 200.0: -10.9, 250.0: -8.6, 315.0: -6.6, 400.0: -4.8, 500.0: -3.2, 630.0: -1.9, 800.0: -0.8, 1000.0: 0.0, 1250.0: 0.6, 1600.0: 1.0, 2000.0: 1.2, 2500.0: 1.3, 3150.0: 1.2, 4000.0: 1.0, 5000.0: 0.5, 6300.0: -0.1, 8000.0: -1.1, 10000.0: -2.5}, 'C': {25.0: -4.4, 31.5: -3.0, 40.0: -2.0, 50.0: -1.3, 63.0: -0.8, 80.0: -0.5, 100.0: -0.3, 125.0: -0.2, 160.0: -0.1, 200.0: 0.0, 250.0: 0.0, 315.0: 0.0, 400.0: 0.0, 500.0: 0.0, 630.0: 0.0, 800.0: 0.0, 1000.0: 0.0, 1250.0: 0.0, 1600.0: -0.1, 2000.0: -0.2, 2500.0: -0.3, 3150.0: -0.5, 4000.0: -0.8, 5000.0: -1.3, 6300.0: -2.0, 8000.0: -3.0, 10000.0: -4.4}}, 'octave': {'A': {31.5: -39.4, 63.0: -26.2, 125.0: -16.1, 250.0: -8.6, 500.0: -3.2, 1000.0: 0.0, 2000.0: 1.2, 4000.0: 1.0, 8000.0: -1.1}, 'C': {31.5: -3.0, 63.0: -0.8, 125.0: -0.2, 250.0: 0.0, 500.0: 0.0, 1000.0: 0.0, 2000.0: -0.2, 4000.0: -0.8, 8000.0: -3.0}}}

ServiceEquipmentBackgroundResult(
    measured_db: np.ndarray,
    background_db: np.ndarray,
    frequencies_hz: np.ndarray | None = None,
)

Band levels corrected for the background by Clause 9.

Attributes

The difference, the regime, the correction and the corrected level are read from the two measured levels and the 10 dB and 4 dB of Clause 9, so they are read-only properties and not fields.

property

$L=L_1-K$ per band, in dB (Formula (7)).

property

The correction $K$ per band, in dB (Formulae (7) to (9)).

property

$\Delta L=L_1-L_2$ per band, in dB (Formula (9)).

property

Whether any band is influenced by the background (Clause 9).

Returns: True when at least one band was held at 2,2 dB, which the report has to state.

property

Per band, whether the background held the correction at 2,2 dB.

Returns: One boolean per band; True marks an upper limit.

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

Plot measured, background and corrected band levels.

Bands held at 2,2 dB are marked as upper limits. Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

Per band, "none", "corrected" or "limited" (Clause 9).

ServiceEquipmentPositionCheck(
    room_dimensions_m: np.ndarray,
    corner_position_m: np.ndarray,
    room_positions_m: np.ndarray,
    source_positions_m: np.ndarray,
    small_room: bool,
    corner_obstacle_distance_m: float | None = None,
)

Whether the positions keep the corner height of 7.2 and the distances of 7.3.

Positions are coordinates in metres in a rectangular room with one corner at the origin: x along the length, y along the width and z the height above the floor.

The distances, the heights and the verdicts are read from the positions and the distances the clauses print, so they are not fields: a check cannot be built to pass positions the clauses fail.

Attributes

property

The height of the corner position, in m.

property

The corner position from 0,5 m to 1,5 m high.

property

At least 0,2 m from any obstacle (7.2).

None when the distance was not given, and then not judged.

property

The corner position’s distances to the two walls nearest to it, in m.

7.2 prefers 0,5 m (preferred_corner_wall_distance, advisory).

property

Every room position from 0,5 m to 2,0 m high.

property

The heights of the room positions, in m.

property

Whether the distances and heights of 7.3 and the corner of 7.2 hold.

The corner position enters through its height and, when its distance to the nearest obstacle was given, through the 0,2 m that 7.2 asks of it. The 0,5 m from its walls is a preference (preferred_corner_wall_distance), like the 1,5 m between positions (preferred_separation).

Returns: True when the five requirements hold, and the sixth, the obstacle distance, holds or was not given.

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

Plot the room in plan with the positions and their clearances.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

Whether position 1 is 0,5 m from both walls of its corner, as 7.2 prefers.

7.2 places the corner microphone “preferably” 0,5 m from the walls and the floor, and where furniture is in the way raises only its height, which corner_height_ok judges. The draft gives the wall distance no tolerance and no lower or upper bound, so a corner position elsewhere is a departure the operator reports rather than a failure.

Returns: True when both corner_wall_distances_m are 0,5 m; advisory, not part of passes.

property

Whether the positions are 1,5 m apart, the distance 7.3 prefers.

Returns: True at 1,5 m or more; advisory, not part of passes.

property

The shortest distance between any two positions, corner included, in m.

property

At least 1,0 m between positions.

property

The shortest distance from a room position to a source, in m, or None without sources.

property

The distance 7.3 asks between a room position and a source, in m.

Returns: 1,5 m.

property

At least 1,5 m from every source; True without sources.

property

The shortest distance from a room position to a wall, the floor or the ceiling, in m.

property

The surface distance 7.3 asks of this room, in m.

Returns: 0,30 m in a small room, 0,50 m otherwise.

property

At least 0,50 m (0,30 m in a small room) from every surface.

ServiceEquipmentResult(
    frequencies_hz: np.ndarray,
    band: str,
    quantity: str,
    a_weighting_range: str,
    readings_db: np.ndarray,
    average_db: np.ndarray,
    background: ServiceEquipmentBackgroundResult | None,
    corrected_db: np.ndarray,
    standardized_db: np.ndarray | None,
    normalized_db: np.ndarray | None,
    ratings: Mapping[str, int],
    unrounded_ratings: Mapping[str, float],
    reproducibility_db: np.ndarray,
)

Service-equipment sound pressure level, engineering method.

Produced by service_equipment_level. Every band array runs over frequencies_hz.

Attributes

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

Plot the band spectrum through each step of the method.

The average, the background, the corrected level and, when formed, the standardized one, with the bands the background limits marked and the A-weighted single numbers in the title. Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

property

The number of readings averaged by Formula (1).

Returns: The number of rows of readings_db.

property

Per band, whether 7.7 lets it be standardized or normalized.

Returns: One boolean per band, True inside the range 7.7 gives for band.

property

The bands that enter the weighted values without standardization.

7.7 keeps 25, 31,5, 40, 6 300, 8 000 and 10 000 Hz (octaves 31,5 Hz and 8 000 Hz) out of the standardization and normalization, and asks the report to mention them whenever they contribute to a weighted value.

Returns: The centres, in Hz, of the bands inside a weighted range that were left unstandardized; empty when nothing was standardized.

ServiceEquipmentResult.upper_limit(weighting: str) -> bool

Whether a weighted value is an upper limit because of the background.

A band the background held at 2,2 dB (Clause 9) that lies in the range of the weighted sum makes that sum an upper limit, and Clause 9 asks the report to say whether the A- and the C-weighted value are influenced.

Parameters

Returns: True when a band of the range was held at 2,2 dB.

Raises

property

The Table 2 standard deviation of each single number, in dB.

0,8 dB for every A-weighted value of ratings and 1,2 dB for every C-weighted one. Footnote a of Table 2 limits both to a constant sound with a relatively flat spectrum from 25 Hz to 10 000 Hz, at least 10 dB above the background; a fluctuating source, a maximum level most of all, is more uncertain, and the draft states no coverage factor.

Returns: The reproducibility standard deviation keyed like ratings ("LA,eq", "LC,Fmax,nT"…).

VaryingBackgroundCheck(
    background_maximum_db: np.ndarray,
    equipment_levels_db: np.ndarray,
    observation_time_s: float,
    frequencies_hz: np.ndarray | None = None,
)

Whether a result stands without correction for a varying background.

The route the NOTE to Clause 9 offers when the background varies in time.

Attributes

property

The 10 dB of the NOTE to Clause 9, fixed by the draft.

Returns: How far below the equipment the background maximum must lie, at least, in dB.

property

How far the background maximum lies below the equipment, per band.

Returns: The equipment level less the background maximum, in dB.

property

Per band, whether the background maximum is 10 dB or more below.

Returns: One boolean per band.

property

Whether the background maximum was watched for 10 min to 15 min.

Returns: True from 600 s to 900 s, both included.

property

Whether the result can be regarded valid without correction.

Returns: True when the background maximum, watched for 10 min to 15 min, lies 10 dB or more below the equipment in every band.

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

Plot the equipment level against the background maximum, per band.

Requires matplotlib (pip install phonometry[plot]).

Parameters

Returns: The axes.

verify_calibration_deviation(
    calibration_levels_db: ArrayLike,
    *,
    previous_levels_db: ArrayLike | None = None,
) -> CalibrationDeviationResult

Has the sensitivity moved more than 0,5 dB (ISO/DIS 16032:2023 Clause 5)?

“At the beginning and at the end of the measurements, verify the sensitivity of the instrumentation with a sound calibrator class 1 according to IEC 60942. If the calibration measurement deviates from previous calibrations by more than 0,5 dB, do not use this equipment until the reason for this deviation has been clarified and appropriate actions have been taken.” Each reading of this measurement is held to every calibration before it: the earlier calibrations given, and for the reading at the end, the one at the beginning as well. A deviation of exactly 0,5 dB is not “more than” it and passes. The 2004 edition asked for the two calibrations without a limit; the 0,5 dB is new in the revision.

The readings are compared as levels, so they must come from the same calibrator at the same stated level; readings taken at different stated levels are passed instead as each one’s departure from its stated level.

Parameters

Returns: CalibrationDeviationResult.

Raises