Building Acoustics – Terminology

For more guidance see a guide to office acoustics

www.acousticguide.org

A-weighted decibel, A modification of the measured sound pressure level that

dB(A) approximates more closely to the response of the ear over the
normal range of sound levels heard and thus correlates
reasonably well to the subjective reaction of loudness to sound.

Alpha w Weighted sound absorption coefficient – is also a method for

αw converting a wide frequency-based range of sound absorption
coefficient values into a single number. This is done using a
curve-fitting process. Although more complex to derive, Alpha w is
considered more representative of how the human ear interprets
sound. The method is fully described in BS EN ISO 11654 and at
the time of writing is the preferred European unit for making a
simple and rapid comparison of sound absorption performance.
The single-figure rating can have a modifier added to indicate if
the spectral shape is dominated by a particular frequency range:
L – Absorption is predominantly in the low frequency region.
M – Absorption is predominantly in the mid frequency region.
H – Absorption is predominantly in the high frequency region.
The absence of a letter following the rating indicates that the
absorber has no distinct area of sound absorption and has an
essentially flat spectral shape.

Apparent sound A measurement of sound reduction index but in the presence of

reduction index, R’ flanking sound transmission. Apparent sound reduction index can be
established for laboratory or field measurements in accordance with
BS EN ISO 140-3 or 140-4.

Ceiling attenuation class, A US-based single-number rating system expressed in decibels,

CAC of the laboratory-measured frequency-dependent room-to-room
sound attenuation of a suspended ceiling sharing a common
ceiling plenum above adjacent rooms.

Decibel, dB A unit of magnitude for sound pressure, sound intensity, sound

power and, in relation to sound insulation, the measurement of level
reduction. The dB is a logarithmic unit that compresses a wide
range of values into a smaller scale.
dBA See A-weighted decibel.
DnT,w See Weighted standardised level difference.

Dnc,w
See Weighted suspended ceiling normalised level difference.

Dn,e
See Element normalised level difference.

Dnf,w See Flanking normalised level difference

DnT,w See Weighted sound level difference and ISO 717 and 140.

Dw See Weighted level difference and ISO 717 and 140

Element normalised level From BSEN ISO 12354;2000 – the difference in the space and
difference time average sound pressure level produced in two rooms by a
source in one, where sound transmission is only due to a small
building element – for example, transfer air devices, electrical
cable ducts, transit sealing systems.
Dn,e is normalised to the reference equivalent sound absorption area.

Flanking sound Structure-borne transmission of sound between adjacent rooms or

spaces but which bypasses the obvious dividing barriers. Such
transmission paths may include side walls, floor slabs and ceiling
slabs. The effect of flanking sound is to lower the achieved sound
insulation between adjacent areas below that which would be
expected from the known performance of the identified dividing
barriers. Because flanking sound is always present (other than
 within the ‘ideal’ confines of an acoustics laboratory), practical site
performance between ‘non-isolated’ constructions will be limited
(typically to about R’w 55dB).

Flanking normalised Defines the sound insulation value from room to room, where a
level difference dividing partition abuts the underside of the ceiling with a plenum
(void) above. The laboratory test procedure involves use of a
massive partition wall, such that the derived performance is that of
the ceiling alone, with no flanking paths.

Floating floor A floating floor is part of a composite floor construction whereby

the upper surface membrane (possibly a concrete screed or
timber deck) is independently isolated (floated) from the lower
structural floor by the use of a resilient underlay or an array of
flexible pads or spring isolators. This separation results in better
air-borne and impact sound insulation than would be achieved by
an equivalent solid floor of the same overall mass, provided the
‘isolation’ is maintained throughout with no rigid connections
between the floating and structural floor, including around the
edges of the floating floor.

Frequency, f The rate at which sound or vibration is generated by a source.

The human audible frequency scale extends from about 20Hz to
20,000Hz. With such a potentially wide range of frequencies it is
necessary, for practical use, to break them down into manageable
groups or bands. In building acoustics, octave bands and one-
third octave bands are generally used.

Hertz, Hz The unit of frequency for a repetitive occurrence. One cycle per
second is termed one hertz.

Impact sound Produced when short duration sources such as footsteps,

wheeled trolleys or door slams, have a direct impact on a
structure. The sound will be heard as surface-radiating air-borne
sound within the area containing the source, but it may also be
transmitted as structure-borne sound to re-radiate as air-borne
sound in more remote locations.

Impact sound pressure This is the measurement of sound pressure levels in a receiving
level, Li room, when the floor/ceiling assembly is under test, including the
presence of a floor covering if applicable, is excited by a
standardised tapping machine in the room above.

Imperforate construction No air gaps through which noise can be transmitted.

L’nT,w See weighted standardised impact sound pressure level

Mass law A relationship used to predict the approximate sound reduction

index (SRI) of a uniform single-skin dividing barrier based on its
mass.
Noise
Unwanted sound resulting in distraction and disturbance,
 interference with speech, and stress or damage to hearing.
However, noise can be subjective – a noise that upsets one person
may go unnoticed by another.

Noise criteria curves, NC A US method of rating broadband sound against a set of

standardised curves that broadly equate to curves of equal
loudness.

Noise rating curves, NR A European method of rating broadband sound against a set of
standardised curves that broadly equate to curves of equal
loudness.

Used to describe sound from mechanical ventilation systems in

buildings.

Noise reduction A US single-number rating for random-incidence sound absorption

coefficient, NRC coefficients. The term is defined in ASTM 423 as the arithmetical
average of the measured sound absorption coefficients for the
four one-third octave band centre frequencies of 250Hz, 500Hz,
1,000Hz and 2,000Hz, which is then rounded to the nearest 0.05.

Normalised impact sound The corrected measurement of impact sound pressure level, to
pressure level, Ln take into account the sound absorption of the receiving room.

Normalised level The corrected value of field-measured level difference between two
difference, Dn rooms, to take into account the sound absorption of the receiving
room.

Octave band A group of adjoining frequencies where the value of the upper
limiting frequency is twice that of the lower limiting value.

One-third octave band A group of adjoining frequencies (formed by dividing each octave
band into three) that are used when more detailed information is
required than is available when octave bands are used.

Pink noise Broadband (electronically generated) noise, whose energy

content is equal per frequency bandwidth, which is often used as
a sound source for acoustic measurements.

Pugging A loose-fill material inserted into the joist cavity of a lightweight

timber floor in order to improve its sound reduction properties. Dry
sand, gravel, ash or dense granular man-made mineral fibre can be
considered and the material is either laid onto the rear of the ceiling
or supported on boards fixed between the joists. Performance
increases with mass.

Random incidence sound A measure of sound absorption derived from tests undertaken in
absorption coefficient, αs a reverberation chamber of an acoustics laboratory, over the one-
third octave frequency bands 100-5,000Hz, in accordance with
BS EN 20354 or ASTM C 423.

Rating of sound Most measurements of air-borne or impact sound insulation,

insulation whether field or laboratory, are conducted over a range of
frequencies so as to obtain a detailed picture of performance. If
these are made in accordance with national or international
standards, at least 18 individual one-third octave band
 measurements over the 100Hz to 5,000Hz range should be
obtained.
For specifiers, who are trying to compare the performance of
alternative building products, or for suppliers of product data, such
a large range of values can be too cumbersome and make
comparisons awkward. A single-number rating (that distills the wide
range of measurement results into a single value) is therefore more
appealing and easily handled.
A simple arithmetic average (adding up all the values in the range
and dividing by the number in the range) has been used – see
Average sound reduction Index – but the final result can be
misleading in that it takes no account of the shape of the frequency
spectrum. Using this method, three entirely different-shaped
spectra could all have the same average value.
Consequently, specific rating methods have been devised and
introduced that do take into account the spread of measurements,
against frequency, by comparing them with a standard curve.
Further, as the standard curves adopted are chosen so as to
simulate the human response to sound, the values obtained
generally correlate well with the subjective impression of common
internal noise sources.
BS EN ISO 717-1 and 717-2 and ASTM E413 and E989 give the
rating methods and procedures.

Reverberation The persistence of sound in an enclosure, due to its continued

reflection or scattering from surfaces or objects, after the sound
source has ceased. Reverberation is significant in determining the
quality and level of sound within internal spaces.

Reverberation time, T or This is the time, in seconds, required for reverberant sound in an
RT enclosure to decay to one-millionth (equivalent to a drop of 60dB)
of its original energy level after the cessation of the sound source.
It is the most common (and easily obtained) measurement or
predictor of a room’s potential sound quality. The reverberation
time for speech needs to be reasonably short or successive
speech sounds will overlap, leading to loss of intelligibility. Music,
however – particularly classical or with unamplified instruments –
benefits from a longer reverberation time a blending of successive
notes resulting in a fuller tone.

See Weighted sound reduction index, Rw – also

Rw
ISO 717 and 140

Sound absorption The loss of sound energy when striking or transmitting into a
boundary surface material or obstacle or when causing a volume
of air to resonate. The reduction of energy is generally due to
dissipation into heat by friction but it may also be lost when sound
passes into adjacent areas, or to the outside through an opening,
 but does not return.

Most materials absorb sound to a greater or lesser extent and the

more common forms of sound absorbers can be conveniently
classified into three main types – dissipative (porous), membrane
(panel) and resonant (cavity).

More recently developed forms of engineered sound absorbers

have become available, which use thin micro-perforated plates
and films and thin non-woven porous tissues. Their performance
can be optimised by selecting specific values for their flow
resistance and spacing them in front of reflective surfaces.

Sound can also be absorbed within air, although this is only

significant at high frequencies and depends on temperature and
humidity.

Increasing the sound absorption within an enclosure will result in

a reduction of the (frequency-dependent) reverberant sound
pressure level and reverberation time. Conversely reducing the
sound absorption will result in an increase in the reverberant
sound pressure level and reverberation time.

Sound absorption For a given material, this is the fraction of incident sound energy
coefficient,  absorbed at its surface. It is expressed as a value between 1.0
(perfect absorption or no reflection) and 0 (total reflection or no
absorption). The value varies with frequency and the angle of
incidence.

Sound insulation (also 1. A term used in relation to the room-to-room transition of

known as sound sound via a common ceiling plenum.
attenuation)
2. A general term relating to the reduction in transmission of
sound pressure levels between one internal area and
another.

3. The reduction of noise levels associated with airflow

within ventilation equipment and ductwork systems.

Sound(noise) leakage Air-borne sound transmission via gaps or cracks around or

through building elements and services that allows sound to
escape from one area to an adjacent area and thus lower the
element’s potential sound reduction.
Typical, although not the only, sound leak paths include:
 Keyholes or air vents in doors or around their edges, resulting
from poor frames or threshold acoustic seals
 Openings in walls or floors through which pipes, ducts, wiring
or similar distribution services pass
 Junctions at the perimeter, and joints between the modular
elements, of partitions, ceilings and raised floors
 Recessed air-handling luminaires within the suspended ceiling
to a common ceiling plenum above adjacent areas
 Recessed continuous lighting troughs or linear air diffusers in
 the suspended ceiling, which span across a common partition
 Ventilation grilles or diffusers that share common ductwork
systems or are connected to each other via a common ceiling
or raised floor air-moving plenum (cross-talk)
 Continuous perimeter piped heating or power/data cable
conduit systems running behind partitions
 Continuous perimeter curtain walling and internal lining
systems.

Sound Internal multi-occupancy spaces such as open-plan offices or

masking/conditioning cellular offices with low sound insulation between them often
suffer from poor acoustic conditions. This results in disturbance
due to the persistence of fluctuating activity and speech noise
between working zones. Such problems may be mitigated by the
use of sound masking.

This term describes electronically generated sound, of a specified

level and frequency content, which is introduced into such
environments (normally by an array of loudspeakers concealed in
the ceiling void) to provide masking of the fluctuating noises and
to enhance speech privacy.

Sound reduction index, The property of a dividing barrier that characterises its ability to
SRI or R reduce the level of sound transmitting through it.

It is determined from acoustic laboratory measurements made in

accordance with BS EN ISO 140-3, when the barrier is mounted
between two reverberant chambers. Since SRI is a function of
frequency, it is desirable that the measurements are made over
the one-third octave frequency bands 100-5,000Hz.

Sound transmission A US single-number rating, expressed in decibels, of the

class, STC laboratory-measured frequency-dependent air-borne sound
insulation properties of a dividing barrier. The test can distort
figures when compared with Rw as a different range of
frequencies is used when calculationg a single-figure result.

Sound wave A pressure disturbance in air proceeding at a finite velocity that

results from the activity of a sound source.

Descriptor used to define intelligibility in a room or a space.

Speech transmission
index STI

Spectrum adaptation The single-number rating method defined in BS EN ISO 717-1

terms, C and Ctr uses a standard reference curve to determine the weighted value
of air-borne sound insulation. The spectrum adaptation terms C
and Ctr can also be evaluated to take into account different
source spectra as indicated in annex A of BS EN ISO 717-1. C is
an A-weighted pink noise spectrum, while Ctr is an A-weighted
urban traffic noise spectrum. Ctr can also be added to Dnt,w or
Rw to take into account low-frequency noise.

Suspended ceiling The laboratory-measured frequency-dependent level difference of

normalised level a suspended ceiling sharing a common ceiling plenum above
difference, Dnc adjacent rooms that is corrected to take into account the sound
absorption of the receiving room.

Tapping machine A portable electro-mechanical device used as a standardised

impact sound source to rate the normalised sound pressure levels
or impact noise insulation of floors.

UKAS United Kingdom Accreditation Service. Acoustic test laboratories

are accredited under this scheme, which was formally the
National Measurement Accreditation Service (NMAS).

Vibration The oscillatory (backwards and forwards) movement of a solid

body about a reference position. If the movements are of
sufficient amplitude and frequency, sound will be generated.

Vibration isolation The use of devices such as steel springs, rubber mounts or
similar resilient materials used to support a vibrating source or
structure and thus minimise the transmission of its vibration
energy into the surrounding structure.

Wavelength,  The distance a sound wave travels in one cycle.

Laboratory measurement of sound reduction performance of a
Weighted sound
reduction index, Rw building element – there is no flanking (indirect) loss, so only the
element under test needs to be considered. Rw is used by
specifiers to compare one building element with another.

Weighted apparent sound A single-number rating, expressed in decibels. This is a field

reduction index, R’w measurement of air-borne sound insulation between rooms – but
unlike DnT,w it will give you the performance of the building
element under test and not the performance of the whole room.

This test is used when the area of the separating wall (and
volumes of the rooms, cinemas, leisure centres and so on) are
large, as this means the flanking transmission conditions do not
come into play.

Weighted sound level This term is not defined in BS EN ISO 717-1: 1997 (so no
difference, Dw recognised layman’s definition exists). It is a site measurement
for noise reduction from one space to an adjacent space. In
schools, Dw is felt by acousticians to be a more realistic
parameter than DnTw because a single-reference reverberation
time cannot be applied to all rooms in schools, and is more
onerous to allow for the needs of hearing-impaired children in
mainstream education. Dw is used by acoustic consultants in
school specifications (BB93).

Weighted standardised This is the level difference achieved from one room to another,
level difference, D nT,w normalised to a standard reverberation time of 0.5 seconds and
weighted to provide a single-figure value. This is a field
measurement of air-borne sound reduction between one room
and another – there will be flanking transmission, so the whole
room needs to be considered because the element under test,
such as the separating wall, may not be the weakest path. DnT,w
 is used in the Building Regulations and will provide an overall
performance of the building element being tested in a ‘real’
environment.

Weighted standardised This field test value measures the reduction between one room
level difference, L nT,w and another (floor/floor) – there will be flanking transmission loss,
so the whole room needs to be considered because the element
under test, such as the separating floor, may not be the weakest
path. LnT,w is used in the Building Regulations to provide the
overall performance of the building element (floor) being tested in
a ‘real’ environment.

Weighted normalised Laboratory measurement of sound performance of a building

impact sound pressure element (floor) – there is no flanking (indirect) transmission loss,
level, Lnw so only the element under test needs to be considered.
A European single-number rating for random-incidence sound
Weighted sound
absorption coefficients αs calculated by reference to BS EN ISO
absorption coefficient, αw
11654. With this method, one-third octave band values of sound
absorption, measured in accordance with EN 20354 or ISO 354,
are arithmetically converted into octave bands at 250Hz, 500Hz,
1,000Hz, 2,000Hz and 4,000Hz (termed practical sound
absorption coefficients αp), and these are plotted onto a graph. A
standard reference curve is then shifted towards the measured
values in steps of 0.05 until the sum of the unfavourable
deviations (the values of αp below the reference curve) are less
than or equal to 0.10. At this point the practical sound absorption
coefficient value of the reference curve that intersects a vertical
line at 500Hz is deemed the ‘weighted sound absorption
coefficient, αw’ (see BS EN ISO 11654 for full details). The derived
value of αw will vary between 0.00 and 1.00 but is only expressed
w can equal 0.65 or 0.70 but
not 0.68.
If a practical sound absorption coefficient exceeds the reference
curve value by 0.25 or more, at one or more frequencies, a ‘shape
indicator’ is applied to the weighted sound absorption coefficient.
If the excess occurs at 250Hz, the notation (L) is used. For
excesses at 500Hz or 1,000Hz (M) is used and for 2,000Hz or
4,000Hz (H) is used. So, for example, a weighted sound
absorption with an excess at 500 w =
0.65(M).
It should be noted that the weighted sound absorption coefficient
is only intended for routine applications in normal offices,
corridors, classrooms, hospitals and so on. It is not appropriate for
specialist environments requiring expert acoustic design nor for
non-plane items such as chairs or screens. Also as the lower limit
for the reference curve is 250Hz, the rating method is not suitable
for products that principally exhibit low-frequency sound
absorption below this frequency, such as panel or resonant
absorbers.
In addition, and depending on the calculated value of the weighted
 sound absorption coefficient, a ‘sound absorption class’ (A to E)
may also be assigned to describe the performance. Such
classification can be useful for easy comparison between
materials.
Weighted suspended
A single-number rating, expressed in decibels, of the laboratory-
ceiling normalised
level difference, Dnc,w measured frequency-dependent room-to-room sound attenuation
of a suspended ceiling sharing a common ceiling plenum above
adjacent rooms. The calculation is normalised to a reference
value for the absorption of the room rather than reverberation
time, as is the case with DnT,w.
Dnc,w is used by specifiers to compare one suspended ceiling
with another.
NB: this is now superseded by Dnfw – see ISO 12354

White noise Broadband (often electronically generated) noise, but can also be
attributed to running water, air movement and other random
sounds.