noise_control.room_to_room

Room-to-room noise reduction: source room, partition, receiving room, criterion.

The other classic bookkeeping exercise of noise control, beside the duct-borne cascade of phonometry.noise_control.duct_path. A machine runs in one room, a partition separates that room from an occupied one, and the question is the octave-band spectrum in the occupied room and whether it meets a design criterion. The answer is a short chain, and every link of it already exists in the library: the reverberant level the machine builds up in the source room (phonometry.room.steady_state_spl over the room constant of phonometry.room.room_constant), the transmission loss of the partition (measured, or predicted by phonometry.building.prediction.panel_transmission), the absorption of the receiving room (phonometry.room.equivalent_absorption_area) and the rating of what arrives (phonometry.room.noise_criterion). What this module adds is the one step that lives nowhere else, and the result object that carries the whole path.

The step. Norton & Karczub, Fundamentals of Noise and Vibration Analysis for Engineers 2nd ed., 4.9, balance the steady-state power crossing the partition against the power the receiving room absorbs and the power that leaks back, and arrive at Equation (4.101):

NR = TL - 10 lo g_{10} [\frac{S _{w}}{S _{2} α _{2} + τ S _{w}}],

with $NR = L_{p 1} - L_{p 2}$ the noise reduction between the two reverberant fields, $TL = 10 lo g_{10} (1/ τ)$ the transmission loss of the partition, S_w the area of the partition and S_2 alpha_2 the equivalent absorption area of the receiving room. The tau S_w term is the power the partition itself passes back and is normally negligible beside the room absorption; it is switched on with include_partition_transmission.

The single most useful thing the equation says is that the noise reduction is not the transmission loss. A large partition into a hard room delivers less than its TL; a small partition into a well-absorbing room delivers more. Norton also warns that the measured noise reduction runs a few decibels below the prediction because of flanking transmission through mechanical connections and air leaks, which flanking_penalty applies as an explicit debit.

The source-room level. In a plant room the receiver of interest is the partition, not a point near the machine, so the level that drives the transmission is the reverberant field alone, $L_{p 1} = L_{W} + 10 lo g_{10} (4/ R_{1})$ . That is phonometry.room.steady_state_spl at distance=None, and this module delegates to it rather than repeating it. Norton’s Table 4.5 adds the choice of sound power model: a machine standing in the intersection of a floor and a wall radiates more power than its free-space L_W if it behaves as a constant-volume source, and source_model="constant_volume" is the conservative upper bound a design estimate uses.

The verdict. RoomToRoomResult.required_transmission_loss inverts the chain: given the source-room level and the criterion curve, the partition transmission loss that would just meet it, band by band. That is the number a partition is specified from.

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

DesignCriterion

DesignCriterion(
    family: str = 'NC',
    target: float | None = None,
    flanking_penalty: float = 0.0,
)

The design criterion the receiving room is held to.

The verdict half of the chain: which room-criterion family the received spectrum is rated against, the curve it has to stay under, and the allowance a design sheet keeps for the transmission the calculation does not model.

Parameters

Name	Description
`family`	Room-criterion family, `"NC"` (default) or `"RC"`.
`target`	The design criterion value (e.g. `45` for NC 45), or `None` for no target, which leaves the verdicts undefined.
`flanking_penalty`	Decibels debited from the predicted noise reduction for flanking transmission through mechanical connections and air leaks (Norton’s “a few dB”). Default `0`.

room_to_room_transmission

room_to_room_transmission(
    frequencies: ArrayLike,
    transmission_loss: ArrayLike,
    partition_area: float,
    receiving_absorption: ArrayLike,
    *,
    source: SourceRoom | None = None,
    include_partition_transmission: bool = False,
    criterion: DesignCriterion | None = None,
    label: str = 'Room to room',
) -> RoomToRoomResult

Sound transmission from one room to another (Norton 2e Equation (4.101)).

Computes the noise reduction the partition and the receiving room deliver together, and the reverberant spectrum in the receiving room. The source-room level is either given directly as source.level or built from a sound power level and the source room’s room constant, in which case the reverberant field alone is used (phonometry.room.steady_state_spl at distance=None), which is the level that drives the transmission across the partition.

Parameters

Name	Description
`frequencies`	Octave-band centre frequencies, Hz (1-D array).
`transmission_loss`	Transmission loss of the partition `TL`, dB; a scalar or one value per band. Measured, tabulated, or predicted by `phonometry.building.prediction.panel_transmission`.
`partition_area`	Area of the partition between the rooms `S_w`, m2.
`receiving_absorption`	Equivalent absorption area of the receiving room `S_2 alpha_2` per band, m2; e.g. from `phonometry.room.equivalent_absorption_area`.
`source`	The source room (`SourceRoom`): its level, or its sound power level with the room constant, directivity and sound power model that turn it into one. Exactly one of the two descriptions is required, so the empty default is rejected.
`include_partition_transmission`	When `True` the `tau S_w` term of Equation (4.101) is added to the receiving-room absorption, with $τ = 1 0^{- TL /10}$ . Default `False`, the form hand calculations use.
`criterion`	The design criterion (`DesignCriterion`): the family, the target curve and the flanking allowance. `None` is the default criterion, an `"NC"` family with no target.
`label`	A short human label of the chain.

Returns: A RoomToRoomResult.

Raises

Exception	When
ValueError	If the spectra do not share one value per band, if neither or both source descriptions are given, or if the criterion family or the sound power model is unknown.

RoomToRoomResult

RoomToRoomResult(
    frequencies: np.ndarray,
    source_level: np.ndarray,
    transmission_loss: np.ndarray,
    partition_area: float,
    receiving_absorption: np.ndarray,
    noise_reduction: np.ndarray,
    received_level: np.ndarray,
    flanking_penalty: float,
    source_power_level: np.ndarray | None,
    criterion: str,
    target: float | None,
    label: str,
)

The room-to-room chain of one partition (Norton 2e, 4.9).

Built by room_to_room_transmission. The four spectra are the four rows a hand calculation writes down: the level in the source room, the transmission loss of the partition, the noise reduction the partition and the receiving room deliver together, and what arrives.

Attributes

Name	Description
`frequencies`	Octave-band centre frequencies, Hz.
`source_level`	Reverberant sound pressure level in the source room `L_p1`, dB.
`transmission_loss`	Transmission loss of the partition `TL`, dB.
`partition_area`	Area of the partition `S_w`, m2.
`receiving_absorption`	Equivalent absorption area of the receiving room `S_2 alpha_2` per band, m2.
`noise_reduction`	The delivered noise reduction `NR` per band, dB: Equation (4.101) less `flanking_penalty`.
`received_level`	Reverberant sound pressure level in the receiving room $L_{p 2} = L_{p 1} - NR$ , dB.
`flanking_penalty`	The debit applied to the predicted noise reduction for flanking transmission and air leaks, dB.
`source_power_level`	The source sound power level `L_W` the source level was built from, dB re 1 pW, or `None` when `source.level` was given directly.
`criterion`	`"NC"` or `"RC"`, the room-criterion family.
`target`	The design criterion value (e.g. `45` for NC 45), or `None`.
`label`	A short human label of the chain.

RoomToRoomResult.criterion_curve

property

The design curve of target at the analysis bands, dB.

None when the chain declared no target. Comparable band by band with received_level.

RoomToRoomResult.exceedance

property

How far the receiving room is over the criterion, band by band, dB.

Positive where the curve is exceeded, None without a target. It is also the transmission-loss deficit of the partition, band for band.

RoomToRoomResult.meets_target

property

True when the receiving room stays under the curve everywhere.

None without a target. The band-by-band test of a design sheet, not the two-step designation procedure behind rating.

RoomToRoomResult.plot()

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

Plot the source-room level, the noise reduction and what arrives.

The source-room and receiving-room spectra are drawn against the design criterion curve, with the band-by-band noise reduction on a twin axis. Requires matplotlib (pip install phonometry[plot]).

Parameters

Name	Description
`ax`	Existing axes, or `None` to create a figure.
`language`	Label language, `"en"` (default) or `"es"`.
`kwargs`	Forwarded to the received-level `Axes.plot`.

Returns: The axes.

RoomToRoomResult.rating

property

The rating of the spectrum arriving in the receiving room.

An NCResult for the "NC" family, an RCResult for "RC".

RoomToRoomResult.required_transmission_loss

property

Partition TL that would just meet the criterion, per band, dB.

The chain of Equation (4.101) solved for the transmission loss,

TL_{r e q} = L_{p 1} - L_{p 2, target} + 10 lo g_{10} (S_{w} / S_{2} α_{2}) + penalty,

with $L_{p 2, target}$ the design criterion curve. The tau S_w term is left out of the inverse (it depends on the answer), which is how the equation is used to specify a partition. None when no target was declared.

RoomToRoomResult.table()

RoomToRoomResult.table() -> list[dict[str, Any]]

The chain as a list of printed rows, source room to verdict.

Each entry has label, kind (one of "source_power", "source", "transmission_loss", "absorption", "noise_reduction", "flanking", "received", "criterion", "required") and values (a per-band array). The rows are the ones a hand calculation writes down, in that order.

Returns: The list of row dictionaries, in printing order.

SourceRoom

SourceRoom(
    level: ArrayLike | None = None,
    power_level: ArrayLike | None = None,
    room_constant: ArrayLike | None = None,
    directivity: float = 1.0,
    model: str = 'constant_power',
)

The source room of the chain: what drives the partition.

Two ways to say the same thing, and exactly one of them is given. Either the reverberant level in the source room is known (level), or the machine’s sound power level is (power_level), in which case the room constant of the source room is needed to build the reverberant field $L_{p 1} = L_{W} + 10 lo g_{10} (4/ R_{1})$ (Norton 2e, 4.7), and the sound power model of Table 4.5 decides whether the position of the source in the room raises or lowers the power it radiates.

Parameters

Name	Description
`level`	Reverberant sound pressure level in the source room `L_p1`, dB (scalar or per band). Mutually exclusive with `power_level`.
`power_level`	Sound power level of the source `L_W`, dB re 1 pW (scalar or per band). Requires `room_constant`.
`room_constant`	Room constant of the source room `R_1`, m2 (scalar or per band); from `phonometry.room.room_constant`.
`directivity`	Directivity factor `Q` of the source in the source room (`1` in free space, `2` on one plane, `4` in an edge, `8` in a corner). Only affects the level through `model`, because the reverberant field itself is position-independent.
`model`	Sound power model of Norton Table 4.5: `"constant_power"` (default, the radiated power does not depend on the source position), `"constant_volume"` (the conservative upper bound, the power rises by $10 lo g_{10} Q$ ) or `"constant_pressure"` (the lower bound, it falls by $10 lo g_{10} Q$ ).

Created and maintained by José Manuel Requena Plens ORCID All projects