Professional Documents
Culture Documents
Acoustics SANS Standards
Acoustics SANS Standards
z
ACOUSTICS m -1
BUILT ENVIRONMENT
ADVICE FOR THE DESIGN TEAM
SECOND EDITION
Edited by
DUNCAN TEMPLETON
[This document
contains 174 pages
Acoustics in the Built Environment
Page blank
in original
Acoustics in the Built Environment
Advice for the design team
Second edition
David Saunders
BSc(Hons), PhD
ArchitecturalPress
An imprintof Butterworth-Heinemann
Linacre House, Jordan Hill, Oxford 0X2 8DP
A division of Reed Educational and Professional Publishing Ltd
A member ofthe Reed Elsevierplc group
OXFORD BOSTON JOHANNESBURG
MELBOURNE NEW DELHI SINGAPORE
Figure 1.2 RAG Walsall — internal. Challengeofseparatingengine test beds from office areas
2 Acousticsin the Buift Environment
The scope for advice is more and more apparent: the The cost constraints are such that developers are very
environment is getting noisier, the standards demanded reluctant to have sealed buildings in speculative offices.
higher, ventilation andsound systems more sophisticated, Similarly, hotels sprout on busy interchanges to catch the
computer-aided instrumentation and prediction tech- passing trade, and commercial business parks crowd the
niques more reliable and accurate (Figures 1.1 and 1.2). airports. Leisure centres group innovative combinations
We want this book to appeal to a wide audience — clients, of noisy activity.
project managers, and students, as well as architects, Many developments are of such a scale now that new
mechanical, electrical and sound system engineers. infrastructure — transport, landscape and topography — is
Eachofthe headings generates a separate approachand entailed long before the building work starts. Many
different disciplines can refer to the relevant chapter, building complexes are of such a scale that the initial
although there is a greatdealofoverlap (Diagram 1.1). The contractis a mere shell for fit-out contracts by numerous
clean ideals of theory are inevitablycompromised on the tenants, so there is a shift in the approachto lookingafter
rack offast tracksite progress;judgementsandadvicehave the client's interests. Novel building types, like trading
often to be given based on half-truths and inadequate rooms, microchip production facilities, multiplex cinemas
information. Good acoustic study techniques are some- and theme parks, demand assessment in the absence of
times too cumbersome: on a recent auditorium physical published data. Relationships to other professions can get
modelling exercise, the project was tendered before complex (Diagram 1.5).
research results could be applied. If advice in this book Legislation is a key issue, not only because of the closer
nudges designers and engineers in the right direction once compatibility to European standards, but also because of
in a while,that is asmuchas we couldexpect. Timelyadvice far-reaching statutes, for example the Noise Act 1996, the
duringthe design, construction, and early use ofbuildings Noise at Work Regulations and the Town and Country
is the aim (Diagrams1.2 —1.4). Planning (Assessment of Environment Effects) Regula-
As practising acousticians we come across 'runs' of tions. The first gives local authorities powers to fine
design issues in design sectors. In offices it may be 'how peoplewho create excessive noise at night and confiscate
noisy can it possibly be before the building has to be noisy equipment,the second tightens the legal duties of
sealed rather than naturally ventilated to the perimeter?' employers, designers, manufacturers and suppliers, to
• Workmanship
ou
.9) •5
• Manufacturer
> selections
• Trouble shooting
• Remedialwork
Diagram 1.1 Checklist: stages ofdesign
Introduction
-
BUILDINGTYPE PLANNING BUILDING CONTROL ENVIRONMENTAL FACTORY OTHER
HEALTh INSTALLATIONS
Town & Country Control of Offices, Shops and Licensing
Planning Act, Pollution Railway justices
PUBLIC 1971 S.29 Act 1974 Public Health
premises Act,
1963 Acts, Amendment Act,
S.21 1980
Town& Country S.38, 62, 68 Control Private Places of
(Scotland) Acts, ofsources ofnoise Entertainment
1972, 1977 and vibration (Licensing) Act, 1967
COMMERCIAL Late Night
S60, 61 Limits on
site Construction Refreshment
noise Houses Act, 1969
Town& Country
Planning Environmental Health and Safety
(Assessment of Protection Act at Work(etc.) Act,
Environmental Effects) 1990 1974
Regulations 1988 BS 5228: 1997- EECdirect /188
INDUSTRIAL BS4142: 1990 NoiseControl on
Construction
and oDen sites
DOE circulars Noise and
10/73 Statutory Nuisance
2/76 Act 1993
1/85 TheNoise Act 1996
Building Regulations Noise at work
1992. SoundE1/213 HSE guidance/
SDD circular Building Standards Regs 1990
DWELLINGS 23/73 (Scotland) Regulations
1981 Building
Noise Insulation
Regulations (NI) 1971
Regulations 1975 as amended.
Hospital design
note 4 (Amendment
HEALTH HN 76/126)
DHSSDATA sheet
DH1.2
DESIGN/SUPERVISE I CONSTRUCT/INSTALL
________________________________________________ I
DESIGN
________________________________________________
I _______________________________________
I SUPPLIER/
TENDER SUBCONTRACTOR
I APPOINTED
___
_________________________________________________________
COMPONENT
I
I
_______________________
________________________________________________________
SUPPLY
__
SELECTION SAMPLES
___ CHECK
TOTAL
I
I
I
I LABORATORY TESTS,
I
SYSTEM VALIDATION
I
I
__
I _______________________ _______________________
I WORKMANSHIP
I
___________________________________________
STS
OR
WITNESSING
________________ I
_
MAINSTREAM PROFESSIONS SPECIALIST CONSULTANTS
I
I _________________________________________________
ARCHITECT I LIGHTING
QUANTITYSURVEYOR ACOUSTIC
TOWN PLANNER I ENERGY
\ CIVIL ENGINEER
STRUCTURAL ENGINEER
LANDSCAPE ARCHITECT
\ I _________
MANAGER
PROJECT
-
__________________________________
GENERAL PRACTICE
I
I
SPECIALIZATION
Diagram 1.5 Design disciplines: context ofacoustic consultancyto other skills in the buildingindustry
minimize hearingdamage; the third defines environmen- and reliable documentation; it becomes all too tempting
tal assessment for any major projects of more than local to 'leave it to the trade'. Sound systems are an intrinsic
importance or projects in sensitive areas. Newor amended part of any modern performancespace.
legislation is available relating to key aids, for example Professor Stephen Hawking, who popularized cosmol-
part E ofthe Building Regulations, BS 4142 and BS 6472. ogy and astrophysics with a bestseller, A BriefHistory of
Legal processes for noise abatement of course deserve Time, was advised by his publisher that 'each equation
proper legal advice, but ChristopherPenn's The Law and included would halve the sales'. Perhaps the publisher has
its Enforcementis a useful starting point. the last laugh, as the 1996 version is An Illustrated Brief
'Environmental acoustics' covers a topic largely over- History of Time. Our objective has been to produce
looked to date but is a growth areain consultancy because reference material of the greatest use in an attractive
of the real concern that Green issues raise. Increasingly, format— with minimal essentialformulaesupportingthe
road noise is universal and there are few truly quiet spots methodology.
left on the mainland UK. Thirty-two million people are In Chapter 5 we address the working knowledge of
exposed to 'high' (55—75 dBA) levels of noise. Protection acoustic terminology, relevant standards in the UK and
techniques to properties alongside roads vary from UK worldwide, and up-to-date information sources. Some of
rustic timberto Swiss curved glazing. Environmental noise the topics arise from the course notes for a university
has become an everydayissue: in 1978, there were 17980 degreein acoustics.Sucha courseis a general grounding
complaints to local authorities about environmental noise. for acoustics, as opposed to being specificallyrelated to
By 1982, the number ofcomplaints had grown to 33014, the built environment. The technology and analysis
rising each year to reach 111 151 in 1993, and nearly techniques are advancing quickly, so there will be in the
145000 in 1994, the last year that statisticsare available. near future more data available to analyse, define and
'Design acoustics' too has been well served in publica- accurately commission criteria set on projects. Acoustic
tions but the authors feel there is increasing need for measurement equipment has reduced in weight and size
more detailed advice specific to buildings' uses; one can dramatically, while its ability to collect, and hold in
no longergeneralize andsuggest a single set ofcriteria for memory, data has increased equally dramatically.Chapter
say studios or practice rooms. More and more, the 5 is a summary ofdefinitions overlapping with the topics
designeris setting performancecriteria only, for specialist covered in the chapters, intended as a quick reference
suppliers and installers to implement. source to ensure that terms quoted in performance
'Services noise and vibration' have been reasonably specificationdocuments are correctly ascribed, or alterna-
served by a number of publications to date, for example tively to interpret, in a dictionary style, terms comeacross
the Sound Research Laboratories' NoiseinBuildingServices in contractdocuments or technical reports.
and Beranek and Ver's Noise and Vibration Control Engineer- The most relevant standards have been selected and it
ing Chapter 3 complements existing advice rather than has been a difficult decision to decide how much to
competes. An increasing proportion of buildings are include on this; database keywords generate many hun-
mechanically ventilated, and economic and space pres- dreds of standards but the 'first port of call' reference
sures lead to a tendency to higher velocity duct systems should be given otherwise any oracle referred to will be
where good control is critical. too broad andmeaningless. A problemwith quotinglarge
'Sound systems' have been considered in systems numbers of standards is the constant updating; any
manuals and electronics guidesbut the applications here standardquoted herein should therefore be checked for
of interest relate to speech intelligibility (PA), audibility any amendments subsequent to publication.
(fire alarms), sound quality in particular spaces (sound The humorist Max Frisch defined technology as 'the
reinforcement) and electroacoustics (modifring the way knack of so arranging the world that we don't have to
auditoriasound),applications directly related to building experience it'. This book tries to make design acoustics
projects and to acoustics, rather than attemptingto cover less of a black art or science, by giving concise and
the rapidly-changing equipmentfield. There are a num- economically reasonable advice, topic by topic. This
ber ofprofessionswith a useful half-knowledge— electrical secondedition, promptedby the sell-outofthe first, allows
engineers, systemsspecialists, theatre consultants — butthe general updating of references and guidance with the
areafalls dangerously between professions as regards full introductionof some newcase studies.
Page blank
in original
This chapterlooks at the need for an acoustic appraisal,
Chapter 1 Environmental what needs to be considered for a site inspection or
survey, the types of environmental noise sources that
acoustics could be encountered, and identifies those factors that
need to be considered when investigatingthe impact of a
development on its surroundings, including environmen-
Peter Sacre tal impact assessments.
Needfor an acoustic appraisal
An appraisal may be required for one of the following
reasons:
• to assess possible site constraints
Environmental appraisals
• as part ofan Environmental Statement to accompany
Introdudion
a Planning Application.
The initial assessment that needs to be made for any In the latter case, an acoustics assessmentcouldbe oneof
development takes accountof its location in the environ- several issues to be covered or it may be required as an
ment. Thus environmental acoustics needs to be con- independentstudy.
sidered at the outset, whether it is the consideration of The route taken to identify the need for an appraisal is
planningissues which take accountof the possible effect shown on Diagram 1.1.
of a development on its surroundings or whetherit is the
effect of an external noisy climate on that proposed Specialist help
development. Once the need for an acoustic appraisal has been agreed,
The need to considerenvironmental acoustics has been specialist advice is available from a number of acoustics
given more emphasis now that environmental issues consultants. These can be found via professional bodies
generally are ofuniversal concern.The publication of the such as the Institute of Acoustics or the Association of
Government's White Paper on the Environment [1] Noise Consultants. Often the local authority will keep a
added weight to this consideration by requiring an register ofacoustics consultants ableto undertakeapprais-
environmental assessmentfor significant schemes. als within their area.
Effect on
Site inspection
Theappointmentofanacoustic consultant canbedirect The site survey is probably the most importantpart of an
to theclientoron asubcontract basisto thelead consultant acoustic appraisal, whetherit is only a site inspection or a
(often the architect),as part ofa design team. full measurement survey, since it will determine the
location of noise-sensitive areas and noise sources and
Method ofacoustic appraisal other local factors needed to make an accurate assess-
The basic requirements ofan acoustic appraisal are shown ment, e.g. local shielding. If a full survey is being
in Diagram 1.2. undertaken, the initial site inspection or pilot survey will
The site survey or inspection will enable importantsite- identify the preferred measurementlocations.
specificinformation to be obtained, suchas whetherthere The items that needto be considered in undertakingan
are any local noise and/or vibration sources which may inspection are identified in Diagram 1.3. Diagram 1.3 also
affect any new development, e.g. transportation routes or gives a checklist of the likely aims of an inspection. This
industry, or whether there are any nearby noise-sensitive includes reference to local topography, particularly
areas, for example housing. embankments or cuttings, which would provide significant
In setting design criteria for a development, reference acoustic shielding but the details ofwhich would not easily
will need to be made to such documentationas British be determinedfrom maps or plans. In determining noise
Standards, to establish acceptable intrusive noise or from transportation routes,data suchas type andgradient
vibration levels in a development, or possibly planning ofroad or type ofrailway trackproximity to airports— civil
conditions, which ensure that a development will not or military — need to be obtained.
affect a nearby noise-sensitive area. In some cases, Measurement locations need to be selected to be
researchstudies may need to be referred to in addition to, representative of the local noise or vibration climate and
or in the absence of, relevant standards. take account of site practicalities. This would include
A prediction exercisewould, in the majority ofcases,be whether noise measurements need to be made at heights
based on measured data taking account of site-specific greater than 1.5 m to obtain appropriate data. Short-
•• Development location
Location of nearestdwellings or otherbuildings and Contact local authority to discuss local factors i.e.
periods of operation —--. • Noise-sensitiveareas
•• Noise-producingactivities near the site •• Major noise/vibration sources in area
Vibration sources Complaints received
•• Local topography
routes
•• Transportation
Measurement locations
Indicative noise levels by short measurement
Organize site access and check that abnormal
activities such as site investigations will not be
taking place
*—Preferred
position
microphone
mechanical services plant for an office development will area. A small housing development affected only by a
affect the noise climate at different times of the day. The single industrial noise source known to maintain a
use of a development 24 h/day, 7 days/week, would also continuous noise level may only take 2—3h at one
identi1' the need to survey at weekends. location.
Typically, consideration needs to be given to assessment The durationofsamples will be dependent on the noise
at night and the lowestambient/backgroundnoise levels climate; 10—20 mm/h at different locations is normally
normally occur between 02:00and04:00 hours. However, sufficient.
if the background noise levels are very low between 02:00 Vibration monitoringwould typically be of short dura-
and 04:00 hours, it may be acceptable from a sleep tion since it is normally only the effectofvibration on the
disturbance pointofview to take 22:30to 00:00 hoursand proposeddevelopment that is of interest. Thus measure-
05:30 to 07:00 hours as the most sensitive periods. ments only need to take account of the maximum levels
In describing noise/vibration climates, the Depart- that would occur during, for example, train pass-bys or
ment of the Environment's Report of the Noise Working quarryblasting and the number ofoccurrences in a given
Party 1990 [21 defined the period 07:00 to 19:00 hours period.
as 'daytime', 19:00 to 23:00 hours as 'evening' and 23:00
to 07:00 hours as 'night-time'. The subsequent publica- Weather
tion of the Departmentof the Environment's Planning The preferred monitoring conditions are on a dry and
Policy Guidance (PPG 24) Planning and Noise [3] recom- clear day or night with a light wind blowing from the
mends the period 07:00 to 23:00 hours as daytime when source towards the measurement location, or when it is
considering the impact of general noise levels on calm. If the monitoringperiods are over a long duration
dwellings. then the effect of weather should not be important,
The durationof the survey will be dependent not only provided reasonably accurate information relatingto the
on the hours of operation but also on the site of a weather can be obtained, and it will only be necessary to
development and/or the noise-sensitiveareas andasurvey avoid long spells ofwindy and rainy weather.
carried out over a numberofdays would average out any High winds and heavy precipitation must be avoided
differences occurring due to weather conditions. For during surveys. High or even moderate winds result in
example, to establish the existing noise climate for a increased background noise levels dueto leavesrustlingin
development on the scale of the ChannelTunnel Project trees or hedges and wind noise in fences. Even with a
[41! required monitoring twice per year over a 2-week windshield, there can be wind 'roar' effects at the
period including weekends at approximately 15—20 loca- microphone itself. Therefore, conditions where wind
tions surrounding the proposed Terminal development speeds are greater than 5 rn/s should be avoided. Rain
Environmental acoustics 11
Parameters to be determined
Rail
Road
L
SEL (to determine L&eq)
LMO, T
Number and type of trains
Industrial
L
SEL (to determineLAeq) Number and types
L
LAeq,T
LA9O, T
could affect the measurement equipment and would range of units, it may be preferredto discount unwanted
create higher noise levels due to its impact on roofs or parameters at a later date.
trees or causing the surface of a road to become wet (in If vibration levels that are to be measured are steady
wet conditions, tyre noise increases). Temperature inver- then r.m.s. acceleration and/or r.m.s. velocity should be
sions could also affect monitoring where long distances determined. Where the vibration levels are caused by
are involved but it is likely that variations due to wind intermittentor impulsivesources then the peakaccelera-
would have more effect. tion and/or peak velocity should be measured. For
Reasonably reliable and up-to-date information can subsequent analysis, frequency spectra should also be
always be obtained from regional weather centres. obtained.
Weather information should always be recorded during
any environmental survey and include wind speed and Equipment
direction, temperature, humidity and cloud cover. The basic instrumentation for noise or vibration measure-
ments, together with a checklist of requirements for
Measurement units instrumentation, is given in Table 1.2. Measurement
The various units and parameters for measuring environ- equipment must be regularly calibrated, at least once
mental noise are definedin Chapter5. In undertakingan every 2 years, and this calibration must be traceable via a
environmental noise survey the values identified in Table laboratory accreditedfor testing by the National Measure-
1.1 shouldbe determinedfor differentnoise sources. The ment Accreditation Service (NAMAS).
table also suggests additional parameters that should be To determine environmental noise levels, a calibrated
obtained. sound level meter complying with the requirements of
Consideration may need to be given to obtaining preferably type 1 butatleasttype 2 as given in BS 6698 [5],
frequency spectra ofdistinctnoise sources, e.g. industrial or BS 5969 [6] should be used. The microphoneselected
plant, for subsequent design development purposes. should always be protected by a windshield and shielded
When the impact of a particular noise source on a from heavy rain.
development is being assessed in isolation it may be In additionto usingequipmentcalibrated to a National
possible to limit the range of parameters measured (see Standard, the equipment should always be calibrated on
Table 1.1) but since most equipment records the full site before and after any survey and at the beginning and
12 Acoustics in the BuiltEnvinmment
refrigerated vehicles, i.e. not diesel engine powered, traffic noise is currently described in terms of LAeq, but it
indicates that noise levels from equipmentwould typically has been converted from LAb levels.
be 65dBA at lOm. For most situations: LAeq,T LAb,T— dB. In 95% of
suchconversions the estimated LAeq,TLSlikelyto be within
Measurement unit ±2dB of the 'true' value.
The measurementunit that has historically been used to
described road traffic noise is LAb. LAb is the A-weighted Legislation and criteria
soundlevelwhich is exceeded for 10% ofthe time period. In the UK, the main legislation dealingwith road traffic
The period normally used is 18h (06:00—24:00 hours). noise is the Noise Insulation Regulations 1975 [9]. This is
The LAb,18 h noise level is the basis for determining issued under the Land Compensation Act 1973 [10].
eligibility under the Noise Insulation Regulations 1975 These regulations were brought into force to compensate
(see Legislation and criteria, below) [9]. residents subjected to additional noise due to the use of
LAeq, Tis the preferredunit formeasuring environmen- new roads. Road construction noise is also included. If
tal noise generally and is the A-weighted equivalent additional noise is at or above a specifiedlevel the affected
continuous sound level. However,in manyinstances, road residents receive a grant for acoustic double windows,
supplementary ventilation, and, where appropriate,vene-
tian blinds to control solar gain in south-facingwindows,
anddouble orinsulated doors. The specifiedlevel is 68dB
3.axl, commercial v,hicl.s LA10 b8h
3. axlecommercial vehidex These regulations do not apply to new housing. New
E 2- axle commercial vel,icles housing or development should be appraised by PPG 24
[3]. The guidance proposes the followingnoise exposure
N Busesand coaches
categories (NECs) based on a 15-years-ahead predicted
2- axle commercialv.hlc*es traffic flow:
> OOkg u.w.
Private cars andlitvse NEC D Planningpermission shouldnormally be refused
where externalfree-fieldnoise levels are in excess
0
C of 72dB LAeq (07:00—23:00hours),
a-
or 66dB LAeq (23:00—07:00hours)
NEC C Planning permission should not normally be
S,eed(km/hI grantedwhere external free-field noise levels are
in excess of 63dB LAeq (07:00—23:00hours),
Figure 1.2 Generalizedsound level/speedcharacteristicsfor or 57dB LAeq (23:00—07:00hours)
different vehiclecategories but less than those in NEC D.
14 Acoustics in the Built Environment
Prediction
An accurate procedurefor the prediction of noise due to
freely-flowing road traffic is given by Calculation of Road
90
85 - ......
80-
-C
.
.
.
.. .
j75
70
a)
->
a
65
.
.
.
I:: Iii
50 I I I
20
ILLUMINATED SHADOW
ZONE ZONE
15
0
Illuminated
Diffracting zone
C edge ——
0
0
10
Effectiv -- —
b
Shadow
zone
R
Reception
( point
8=a+b—c
5
0
1.0 0 1.0 2.0 3.0
Path difference, (m)
CORTN gives the basic noise level at 10m from the The propagation of traffic noise with distance is
nearside carriageway edge for traffic containing about predominantly based on distance to the source, angle of
10% heavygoods vehicles (those over 1.5 tonnes) at up to view of the road, interveningground cover, andwhether
60km/h. This also assumes that the road surface is any barriers exist between source and receiver. Typically,
bitumenandrelativelylevel (gradient less than 3%). Ifthe over ground covered with vegetation and a reception
traffic speed exceeds 60 km/h then the noise level will point not more than 4m above ground, the reduction in
increase at a rate of approximately 6dBA/doubling of noise level could be as much as 7 dBA/doubling of
speed. If the percentageof heavyvehicles is greater than distance. Over hard ground or an acoustically reflective
10% then an approximate factor of 2dBA/doubling of surface such as concrete or water, the reduction in noise
heavyvehicle content could be used. level will be 3 dBA/doubling of distance.
16 Acoustics in the Built Ezivironment
100
95-
90- Heavy vehicles
75
70
65
60
20 40 60 80 100 120
Speed (km/h)
Figure 1.5 Single-event road vehicle noise level at d!fferentspeeds at a distance of10m [12]
100
95
D
90
a)
>
85
a)
a) 80
U,
U)
a)
75
-D
C
70
0
U) 65
60
55
50
63 125 250 500 1000 2000 4000 8000
Octave band centre frequency (Hz)
Figure 1.6 Frequency spectrafor dfferent road traffic flow conditions equivalent to 76dB LAb, 7-
The effect of barriers depends on the path difference Figure 1.5 provides an estimate of the single-event
and it is importantto checkthat the line ofsight between exposure level at a distance of 10 m for light and heavy
source, 0.5m above road surface, and receiver, typically vehicles for different speeds. This can be used to
1.5 m or 4.0 m above ground (ground and first floor determinethe overall noise level in terms of LAeq at 10m
reception heights respectively) cuts the barrier (Figure for a number of vehicles. The overall level at a distance
1.4). The sourcefor light and mixed traffic is taken to be greater than lOm can then be estimated based on
0.5 m above the road surface, and for heavy vehiclesonly attenuation of 3—7dBA/doubling of distance depending
can be taken to be typically 1.5m above the road surface. on the ground cover. There are also occasions when a
The barrierneedsto extendasignificantway alongside the predictive exercise may become complicated anda meas-
road to provideeffective screening but if this is achieved urementis the only availablecourse ofaction, suchas at a
the reduction canbe determinedfrom Figure 1.4. traffic-lightedjunction or a roundabout.
Alternatively,ifvehicle flows are low and measurements The resultant noise levelsare normally given in terms of
difficult to make, it maybe preferable to use the following a level at a particular point. However, provision of
method based on the Noise Advisory Council guidance contours, particularlyon a site where the best location for
[12]. a building is being determined, can be helpful. The
Environmental acoustics 17
100
V
a)
>
a,
0
E
E
60
50 100 150 200
Speed (km/h)
predicted levels for road traffic noise will be overall speeds get higher, aerodynamic noise may become sig-
A-weighted single figures which can be converted to nificant, but that stage has not yet been reached in the
typical octave band levels using the graphs shown in UK
Figure 1.6.
Measurement unit
Railways Train noise is measured in terms of the A-weighted
Noise sources equivalent continuous noise level,LAeq,T Although in the
The predominantsources ofnoise dueto train movement past the period T has been taken as the full 24-h daily
are propulsion equipment and wheel/rail interaction. period for railway noise, The NoiseInsulation(Railways and
The propulsion equipment includes diesel locomotives Other Guided Transport Systems) Regulations 1995 [15]
anddiesel multiple units; noise from electric locomotives identif' the daytime (06:00—24:00hours) and night-time
andelectric multiple units is significantlylower than from (00:00—06:00hours) periods. PPG 24:1994 [3] identifies
diesel equivalents. In addition, auxiliary equipment,such the daytime (07:00—23:00 hours) and night-time
asventilation systems and othercarriage-mounted compo- (23:00—07:00hours) periods.
nents, can be sources of noise, and elevated structures, In order to determinethe LAeqovera given time period,
such as bridges, tend to increase noise levels but both are it is often preferred to undertake a calculation using
typically insignificant in the UK compared to diesel individual train pass-by levels. Thus the sound exposure
locomotive and wheel/rail noise. In the US there are level (SEL) is measured for different train types; typically
many steelelevated structures causing high noise levels. A this is at a distance of 25m. In addition, the maximum
similar situation also occurswith Docklands LightRailway noise level is often measuredin order to assess the effect
but it is predominantly at low frequencies [13]. The of train pass-bys on conversation and telephone use, for
maximum noise level at 25 m from diesel locomotives is example.
typically85—95dBA [14].
Wheel/railnoise is due to the vibration ofboth caused Legislation and criteria
by the action of one rolling over the other. The para- The Railway Noise Insulation Regulations [15] operate on
meters that can affect this noise are the type of track, i.e. a similar basis to the Noise Insulation Regulations [9] for
continuously welded rail (CWR) orjointed (+5 dBA), the road traffic noise. The specified levels are 68 dB LAeq 18h
type of braking system, i.e. disc- or tread-braked, and (day time) and 63dB LAeq6h (night time). If new or
maintenance of track/wheels, i.e. removal of corruga- additionalrailway noise togetherwith railway noise in the
tions. Noise due to tread-braked rolling stock can be vicinity is at or greater than these specified levels at the
10dBA higher than disc-braked, and badly corrugated facade ofa dwelling, the residents are entitled to a grant
track could cause increases of 10 dBA. Therefore rolling for noise insulation to habitable rooms. The noise
stock with discbrakes on CWRthat is regularly maintained insulation grant is for acoustic double windows, supple-
will result in the lowestnoise levels. Typical noise levels of mentary ventilation and, where appropriate, venetian
the different train types hauledby electric locomotivesare blinds to control solar gain in south-facingwindows, and
shown in Figure 1.7. Noise control measures to railways double or insulated doors.
are being brought in, in the form of 'Hush' rails, The Regulations are based on the findings of the
beneficial through their smaller cross section, and wheels DepartmentofTransport's Railway Noise and the Insulation
which are damped to reduce 'ringing'. In future, as train ofDwellings [16]. These Regulations do not apply to new
18 Acoustics in the Buift Environment
-6
Figure 1.8 Attenuation oftrain noise with distance over grassland [14]
housing; PPG 24 proposes the following noise exposure suggested in guidance given by the US Environmental
categories. Agency [17]. This relates to an external free-field noise
level of 65—70 dBA unless it is a sealed building.
NEC D Planningpermission should normally be refused
whereexternal free-fieldnoise levels are in excess Prediction
of 74dB LAeq (07:00—23:00 hours) Procedures for calculating noise from trains are given in
or 66dB LAeq (23:00—07:00hours) the Department of Transport's Calculation of Railway
NEC C Planning permission should not normally be Noise 1995 [18]. This calculation method requires the
granted where externalfree-field noise levels are number and types of trains in an 18-h daytime or 6-h
in excess of night-time period. Typical sound exposure levels, SEL,
66dB LAeq (07:00—23:00hours) for a variety of train types are given where measured data
or 59dB LAeq (23:00—07:00hours) is not available. In addition the type of track and any
butless than those in NEC D, or where individual bridges or viaducts crossed by the rail and the topog-
noise events regularly exceed 82dB (max- raphy need to be known. The railway is then broken
imum SPL on 'slow' meter setting) at night. down into segments, where necessary, and an overall
NEC B Noise should be taken into accountwhen deter- noise level is determined for a reception point. The
process is shown in Diagram 1.7.
mining planning applications and appropriate The overall LAeq noise level on a specific site can be
conditions requiring adequate noise protection
should be imposed where external free-field determined from the sound exposure levels (SEL), as
described in Chapter 5. In order to assess noise levels at
noise levels are in excess of other distances over grassland, the chart shown in Figure
55dB LAeq (07:00—23:00hours) 1.8 can be used. Typically this is 5dBA/doubling of
or 45dB LAeq (23:00—07:00hours) distance. Intervening properties such as semi-detached
but less than those in NEC C. and terraced housing could provide the following noise
NEC A Noise need not be considered as a determining reduction:
factor in granting planning permission where • single row of semi-detached houses 8 dBA
externalfree-field noise levels are below • subsequent rows, each 4dBA
55dB LAeq (07:00—23:00hours)
and 45dB LAeq (23:00—07:00hours) • terraced housing 13dBA
However, noise at the high end should not be Cuttings orinterveningground barrierscould be assessed
regardedas desirable. in a similar way to that described for roads. These
For dwellingswith windowsopen for ventilation procedures can also be used to determinethe 07:00—23:00
and 23:00—07:00h period noise levels for PPG 24 assess-
the noise levels in NECA indicate that 45 dB LAeq
is acceptable internally during the day and35 dBments. Although the barrier corrections for railways are
LAeq at night. slightlygreater, see Calculation ofRailway Noise, the barrier
attenuation in Figure 1.4 can be used. BS 8233 [11]
Fordevelopments with generalofficeswhere the impact contains estimated noise levels for a track carrying diesel-
on communication, either verbal or by telephone, must be hauledpassengers andfreighttrains at differentdistances
considered, a maximum internal level of 55—60 dBA is over open grassland andis reproducedhere asTable 1.4.
Environmental acoustics 19
Distance correction
Air absorption correction
Ground correction
Ballastcorrection
A Screening correction For each segment determine
Angle of viewcorrection the correction factors for each
train type at the receiver posi-
tion for each track
STAGE 4 — Reflectioneffect
Facade correction
Reflectioncorrection
No
}
STAGE 6— Calculatethe total LAeq for the railway
Sum LAeq to obtain predicted
level from railway
For each time period combine the LA8qS for each segment
Table 1.4 Noise levelsfor a typical railway' The typical octave frequency band levels can be
determined either from on-site measurements or the
Distancefrom track over open grassland typical noise spectra given in Figure 1.9a and b for both
LAe9,18 h diesel electric locomotives and tread- or disc-braked
(m) (1B)
rolling stock hauled by electrically-poweredlocomotives.
25 67 Aircraft
50 64 Sourcenoise
100 59 The main concern relating to aircraft noise is associated
200 54 with take-offs and landings near an airport. In addition,
ground operation noise may also need to be considered.
aBased on BS 8233. Typical railway traffic is assumed to consist In termsofnoise dueto flying operations, the mainfactor
of a mixture ofa total of90 high-speed diesel-hauled is the type of aircraft. The maximum noise levels for
passenger and freight trains, per 18h day (06:00—24:00hours). various types of aircraft under different operating condi-
20 Acoustics in the Built Environment
100
Planning guidance andcriteria
95 PPG 24 [3] identifies the following noise exposure
90
85
categories for aircraftaffecting new dwellings.
80 NEC D Planningpermission should normally be refused
linE
where externalfree-field noise levels are in excess
of
72 dB LAeq (07:00—23:00hours)
or 66 dB LAeq (23:00—07:00hours)
NEC C Planning permission should not normally be
63 260 500 2000
1000 4000 8000 grantedwhereexternalfree-field noise levels are
Octave band centre frequency (Hz)
in excess of
66 dB LAeq (07:00—23:00hours)
H
100
itii
95 or 57dB LAeq (23:00—07:00hours)
f 90 butless than those in NEC D, orwhereindividual
85 noise events regularly exceed 82dB at
80
75
night.
70 NEC B Noise should be taken into account when deter-
65 mining planningapplications and, where appro-
60 priate, conditions requiringadequate noise pro-
56 tection should be imposed where external
50 free-field noise levels are in excess of
63 125 250 500 1000 2000 4000 8000 57dB LAeq (07:00—23:00hours)
Octavebandcentrefrequency (Hz)
Ib)
or 48 dB LAeq (23:00—07:00 hours) but less than
those in NEC C.
NEC A Noise need not be considered as a determining
::: factor in granting planning permission where
85- externalfree-field noise levels are below
80'- I
57dB LAeq (07:00—23:00hours)
and 48dB LAeq (23:00—07:00hours)
t5Li
c)
63
a
4000 8000
However, noise levels just below these limits
should not be regardedas desirable.
For dwellingswith windowsopen for ventilation
the noise levelsin NECA indicate that 45dB LAeq
is acceptable internally during the day and35dB
LAeq at night.
locomotive at 25 m; (b) tread-brakedpassenger train at 25m
with electric locomotive at 100km/h; (c) disc-braked passenger There are a number of airport grant schemes under
train at 25 m with electric locomotive at 100km/h which grantsare paid towardsthe cost ofsound insulation
in existing dwellings within defined areas around major
airports. The sound insulation package is similar to that
offered under the Noise Insulation Regulations [9] for
road traffic noise, plus increased roofinsulation.
dons are given in Table 1.5. Data are given for the UK
reference distance of 152m used to determine the
Reference Noise Level. Military airfields may also need to
be considered not only for noise near the airfield but Prediction
sometimes due to other operations suchas lowflying.Data Although a simple calculation can be carried out using
in this case needs to be obtained from the Ministry of the sound source data (SEL) from Table 1.5, the number
Defence. of aircrafttypes and a correctionfor distance, the error
in accuracy is likely to be great. In reducing the error,
Measurement units reliance will have to be placed on published contour
Historically,the Noise and NumberIndex (NNI) has been maps which are based on accurate flight profile data.
used as the noise unit for measuring aircraft noise. It takes However, consideration may need to be given to the
into accountthe maximum perceived noise level of each maximum noise level on a site from a design point of
aircraftfor the number of aircraft movements during a view. Measurement is obviously the easiest method of
12-h period (06:00—18:00 hours). However, in September determining maximum noise levels, but if this is not
1990 the Departmentof Transportchangedto the use of possible, then the maximum noise level at a particular
LAeq to describe aircraft noise over a 16-h period location can be obtained by calculating the slant distance
(07:00—23:00 hours). as shown in Figure 1.10 and applying the correction
EnnthI acoustics
Table 1.5 Typical noise levels ofaircraft at a distance of152 m
8 dB/doubling of distance. An indication of the fre- NNI contours are available it is possible to estimate the
quency content for typical jet aircraft types is shown in approximate equivalent LAeq 16 h value within ±2dB from
Table 1.6. the following table:
LAeq contour maps are available from the Civil Aviation
Authority or via the local authority or airportauthorities; NNI 35 40 45 50 55 60
examples are Gatwick, Heathrowand Manchester. If only LAeq, 16h 57 60 63 66 69 72
22 Acoustics in the Built Environment
I
N
>-
0
z
w
ci
w
U-
Figure 1.11 (a) Path dfference; (b) attenuation provided by noise barriers at dfferentfrequencies
Environmental acoustics 23
Table 1.7 Typical construction plant noise In addition to the above equipmentwhich is operating
normally out in the open, structureborne noise due to
drills or breakers maycause potentialnoise
Equipment Approx. 'eq (dB) hand-operated
problems buildings coupled to the construction under
in
consideration. The noise level due to structureborne
Tracked loader 109 noise varies significantly depending on local site condi-
Tracked excavator 109 tions and an estimate of the noise level likely to occur
Dozer 111 cannot easily be provided, although noise levels of
55—60 dBA in nearby areas during percussive drilling
Piling: could be anticipated.
Diesel hammer 130
Drophammer + wooden dolly 115
Augerbored 112
Measurement unit
Pneumatic breaker 116 The A-weighted equivalent continuous sound level,
Concrete pump 110 LAeq, is the preferred unit for describing construction
Truck mixer 110 noise. However, in addition, to take account of isolated
Concrete mixer 95 events and impulsive sources such as piling, it is recom-
105 mended that the maximum noise level, is also
Batching plant considered.In describing site noise, the particularperiod
Poker vibrators plus compressors 102 of the day should always be stated.
Compressors:
4m3/s 98
7m3/s 101 Legislation and criteria
Noise from construction sites is specificallyreferred to in
17m3/s 111
Sections 60 and 61 of Part III ofthe Control of Pollution
Generator 104 Act 1974 [19].
Pump 103
Crane 103 Section 60
Under Section 60 ofthe Act, a local authority may serve a
Notice on the contractor specifyingone or more of the
following:
the noise level on site from transport, is negated in •plant or machinery which is, or is not, to be used
determining the overall performance provided by a
•hours during which works may be carried out
barrier, unless it is on earth mounding with shallow-
•noise limits
sloped sides. The design of the barrier should ensure However, in specifyingany of the above, a local authority
that the length of the barrier is sufficient to protect the should have regard to:
whole development.
The barrier need only be relatively lightweight and •relevant Codes of Practice issued under this part of
the Act, viz. BS 5228: Parts 1—4 [201;
normally a close-boarded timber fence is quite adequate. •the need to ensure that the best practicable means
Other barrier types include metal sandwich construction
or precast concreteunit assemblies. The performanceof ('practicable' meaningreasonably practicable having
barriersalongside railways can be reduced by as much as regardamongstother things to local conditions, the
5 dBA where the side closest to the track is acoustically current state of technical knowledge, and the finan-
reflective.Consideration shouldbe given to a barrier type cial implications; 'means' includes design, main-
with an acoustically-absorbentsurface facing the track. tenance and manner and periods of operation of
There are barriers, of metal sandwich construction and plant and machinery and the design, construction
and maintenance of buildings and acoustic struc-
precast concrete faced with woodwool slabs, which will tures; this is provided safety and safe working
achieve this requirement.
conditions are met and regard paid to any provision
of BS 5228) are employed to minimize noise;
Construction noise •the interest of the recipient before specifying any
particular methods or plant or machinery, i.e. where
Sources alternative methods or plant more acceptable to the
construction operatorwould be substantiallyas effec-
Major noise sources involved with construction activities
include pilingrigs, earthmoving equipment suchas dozers tive in minimizing noise as those proposed by the
and excavators, and concrete pouring plant such as local authority;
concrete pumps and truck mixers. A range of construc- •the need to protect any personsfrom the effect of
tion equipment is given in Table 1.7 which includes the noise.
approximateaveragesoundpower level during the activity Any person served with such a notice may appeal to a
of each item. In most cases, diesel engine noise predom- magistrate's court within 21 days from The
inates but consideration needs to be given to piling and grounds for appeal and form of notices arereceipt.
outlined in
material handlingnoise. Departmentof the Environment's Circular 2/76 [21].
24 Acoustics in the Built Environment
Resultant
Operating
soundpower Distance Screening Mobileplant soundpower On-time
level, Distance attenuation attenuation correction level in 12 h
Plant and size (dB) (dB) (dB) (h)
'WAeq (dB) (m) (dB)
100 50 42 0 0 58 4
Compressor 7m3/min 0 69
Pneumatic breaker 116 50 42 5
Tracked excavator 109 30 38 0 0 71 8
Tracked loader 109 15 32 0 4 73 6
different approaches available to local authorities in Although there is no requirementto set a noise limitfor
dealing with construction noise: any ofthe three approaches, noise limits can be set based
on those aspectsdescribed earlierin the 'Code ofPractice'
• encourageapplication for prior consent section above.
• serve a Notice Noise nuisance caused byconstruction noise is normally
• allow work to be carried out andif any cases ofnoise ofshort duration.Often, there are no means availableto
nuisance arise dealing with them either under apply control, e.g. structurebornenoise due to hand-held
Section 58 or Section 60 of the Control of Pollution tools, and so it is very important that the contractorhas a
Act. Although construction noise nuisance should be good public relations policy. If a noisy activity has to take
dealt with under Sections 60 or 61, individuals may place which could affect neighbouring properties, the
complain to local authorities, who could then serve a neighbours should be warned, particularly with regard to
Notice under the Environmental Protection Act to the durationofthe activity. Itmay evenbe possibleto agree
abate the nuisance. to periodsofoperationthat are acceptable to both parties.
ofcomplaints. If by this method, noise from the proposed Appropriate conditions will need to be imposed to meet
development 'is likely to give rise to complaints' then these requirements and examples are given in Annex4 of
is
permission unlikely to be granted. In determining the PPG 24 [3]. Appropriate conditions have subsequently
predicted noise levelfrom a proposeddevelopment, it will been published in Appendix A of the Departmentof the
be necessary to take account of the plant operating at its Environment's Circular 1/85 [26] and are reproduced
maximum capability. here in Table 1.9. Although conditions relating to the
Planning guidance on noise from surface mineral physical characteristics of the development, the type and
extraction or landfillsites is given in MPG 11: The Controlof intensity of activity to be carried out, and hours of
Noise at Surface Mineral Workings [25]. These guidelines operation, are preferable, in some instances a condition
recommend a noise prediction model representing a laying down a maximum noise level at a particular
proposedmineral development anda methodfor settling locationor possiblydifferentlevels for different periodsof
noise limits. They also provide adviceon noise monitoring the day may be appropriate.
and noise control. If plant associated with the mineral If a proposed development were shown by a noise
extraction or landfill is fixed on site it seems appropriate assessmentto be acceptable during normal working hours
for this plant to be assessedon the basis ofBS 4142 [22]. but not at other times, it would be reasonable to apply a
If a local authority gives permission for the develop- condition restricting operation to certain specified hours
ment, theywill need to ensure that: rather than rejectthe application altogether.
Using this guidance, permission will be given for
• noisierprocesses than those proposedby the devel- developments against which the local authority is unlikely
oper are not allowed, and to find it necessary to serve a noise abatement notice
• all physical features of the submitted plans which under the Environmental Protection Act. However,it will
control noiselevels are in the finished
incorporated not necessarilyprovide protectionagainst legal action by
development. private citizens.
Table 1.9 Suggested models of acceptable conditions relatingto noisefor use in appropriate circumstances. Extractedfrom
Appendix A ofDepartment ofEnvironment Circular 1/85 [261
NB. Additions and amendments to these conditions are given in Annex4 ofPPG 24 [3].
28 Acoustics in the Built Environment
Under Section 80 of the Environmental Protection Act, computer model. This will require data relatingto sound
the local authority is empowered to deal with noise power levels and directivity and is likely to necessitate a
nuisanceby serving a Noise Abatement Notice. three-dimensional model taking into account different
BS 4142 Methodfor rating industrial noise affecting mixed heights above a ground level datum.
residential and industrial areas [22] describes methods for The sound power level data may need to be estimated
determiningnoise levels from factories, industrial prem- by converting,anticipated internallevels within a building
ises, or fixed installations and sources of an industrial to those radiating from an aperture or from wall elements
nature in commercial premises. The noise level deter- or obtained for external sources such as transportation
mined in terms of LAeq,1 is corrected for tonal and from references.
impulsive character to establish the rating level. This
rating level is then compared with the measured batk- Noise control
ground noise level. Even wherethe noise climate is always There are a numberofalternative methods ofcontrolling
affected by the industrial noise, it is possible to measure noise from an industrial development including:
the background noise level at another location whereit is
presumed to be equivalent. The process is shown in
• attenuators in ductwork of ventilation or extraction
systems
Diagram 1.9.
The difference between the rating level and the • enclosures around mechanical plant like fans and
background noise level indicates the likelihood of com- motors, selection of low-noise components
plaints. A difference of around 10 dB or more in LAeq T
• cladding on ductwork
indicates that complaints are likely. Adifference ofaround • silencers in pipework serving valves or engine
5dB is of marginal significance. At a difference below exhausts
5dB, the lower the value the less the likelihood that • building orientationduring design to avoid openings
A difference of —10 dB is a positive like doors facing sensitive areas
complaints will occur.
indication that complaints are not at all likely. In assessing • building construction
whethera particularprocess is causinga noise nuisance, • hours of operation to avoid night-time/evening
the local authority would normally use BS 4142: 1990. periods if possible
• methodsof operation to avoid high levels of impact
Prediction noise
As stated earlier, industrial noise sources can be ofvarious • lining bins collecting metallic waste material
forms and prediction of industrial installation noise Implementation of any noise control needs to take
requiresa clear understandingof the processes involved. account of other parameters, for example cooling needs
Ideally, noise levels should be measured during the and fire ratings, so requirements need to be checked for
operation of similar processes elsewhere. It may be overall practicality with the manufacturers and/or the
necessaryto takeaccountofan increase in size or capacity suppliers.
ofan operation.Onlyin the case ofmineralextraction are
there suitable data to enable reasonably accurate predic-
tions to take place and these are provided in BS 5228, Leisure noise
particularly Part4. Noise due to leisure activities can be considered as an
Data obtained in this way can make the prediction extension to industrial noise. Leisure activities include
exercise easier since it will normally only require a discotheques, night clubs, cinemas, ten-pin bowling and
distance correction to be applied. Where a large number clay pigeon shooting. In addition, certain apparently quiet
of sources are identified it may be necessary to preparea activities such as ice skating can be a potential noise
EnvhnmeiitaI acoustIcs 29
Control
If there is a likelihood of structural damage then,
obviously, an alternative form of the source of vibration
needs to be found or possibly the structure could be
reinforced. However, only rarely is it shown that structural
damage is likely to be caused and the method of control
normally available is the isolation of the recipient,
(H typically:
Figure 1.13 Criteriafor subjectivelimits and building • small sensitive equipment, for example optical
damage balances
• entire buildings, for example concert halls
• individual rooms, for example studios.
ference meetingrooms. Spaceswith windows to outside are Onlyoccasionallyis it possible to isolate the source, since
unlikely to be of concern since there will be a reasonably it is out of the control ofthe developer, although this has
high level of low frequency noise breakinginto the space been undertakenwith railway tracks and some industrial
viathe windowswhich will mask any structurebornenoise. sources such as presses. It may also be possible to control
Establishing acceptable levels of structurebornenoise in the operating time of the vibration source to avoid
sensitiveareas will need to be site specific. sensitive periods.
Iedidion
In predicting the likely vibration levels from a particular New
developments as a noise source
source, consideration needs to be given to: In assessing the new development as a noise source,
•
the type of source and its interaction with the consideration may need to be given to undertaking an
ground, Environmental Assessment or simply to meeting a plan-
•
transmission through the ground, ning condition.
Table 1.10 Transient vibration guide values for cosmetic damagefrom BS 7385: Part 2 [31].
Table 1.11 Multiplyingfactors used to specify satisfactory magnitudes of building vibration with respect to human response
(fromBS 6472) [33]. Extractsfrom BS 6472 are reproducedwith the permission ofBSI. Complete copies ofthe standard can be
obtained by postfrom BSI Publications, Linford Wood, Milton Keynes, MK14 6LE
Multiplyingfactors
Note 1. This table leads to magnitudes of vibration below which the probability of adverse commentis low (any acoustic noise
caused by structural vibration is not considered).
Note 2. Doubling of the suggested vibration magnitudes may result in adverse comment and this may increase significantlyif the
magnitudes are quadrupled(where available,dose/responsecurves may be consulted).
Note 3. Magnitudes of impulse shock in hospital operating theatres and critical working places pertain to periods of time when
operations are in progress or critical work is being performed.At other times magnitudes as high as those for residences are
satisfactoryprovided there is due agreement and warning.
Note 4. Within residential areas people exhibit wide variations of vibration tolerance. Specificvalues are dependentupon social
and culturalfactors, psychologicalattitudes and expected degreeofintrusion.
Note 5. The 'trade off' between number of events per day, their magnitudes and durationsis not well established. In the case of
blasting, and for more than three events per day, the followingprovisionalrelationship can be used to modify the factors for
residences in column 4 of this table. This involves further multiplying by the factor F= 1.7 N°5 T' where: Nis the numberof
events in a 16-h day; Tis the duration of the impulse and decay signal for an eventin seconds.
The duration of an eventcan be estimated from the 10% (—20dB) points of the motion time histories.
d = zero for Tiessthan 1 s.
Forshortdurationstimulithereis evidence that forwooden floors the human response d=0.32 and forconcrete floors d= 1.22.
This 'trade off' equation does not apply when values lower than those given by the factors for continuous vibration result.
Note 6. The root mean quad (r.m.q. = (l/Tf, a4(t) dt)"4) of the weighted acceleration signal a(t) may be used as an alternative
f
methodof assessmentfor impulsiveevents. The same relation between duration and acceleration maybe used to accumulate the
exposure to intermittentvibration occurring throughoutthe day (i.e. accumulated value = a4 (t)dt). The value obtainedby
this method, which shouldbe related to the boundaries for continuous vibration, allows greater magnitudes with shorter and/or
less frequent periods ofintermittentvibrations.
Note 7. The magnitudes for impulsive shock excitation in offices and workshop areas should not be increased without
considering the possibilityof significantdisruption ofworking activity.
Note 8. Vibration acting on operators ofcertain processes such as drop forges or crushers, which vibrate working places, maybe
in a separate category from theworkshop areas considered in this table. The vibration magnitudes specified in relevant
standards would then apply to the operators of the exciting processes.
Environmental assessments ment if they are 'likely to have significant effects on the
There is a formal requirement in the UK and the rest of environment byvirtue offactors such as their nature, size,
Europe for an Assessment of Environmental Effects and and location' and include developments such as mineral
the preparation of an Environmental Statement to be extraction, industrial complexes, food industry, infra-
undertakenfor certain projects. The projects that require structure projects, and holiday villages or hotel com-
an assessmentin every case are listed in Schedule 1 ofthe plexes. If the project under consideration is deemed by
Town and Country Planning (AssessmentofEnvironmen- the local planning authorities to be included in either
tal Effects) Regulations 1988 [34] andinclude oil refiner- Schedule 1 or 2 then it is likely that an acoustics appraisal
ies, power stations, waste disposal and chemical installa- would need to be included as part of the assessment. In
tions, andlarge transportation schemes. Schedule 2 ofthe some cases, the local authority may considerthat only an
Regulations identifies projects which require an assess- acoustics appraisal is needed. The method ofapproachis
32 Acoustics in the Built Euviroument
Establish baseline conditions during shown in Diagram 1.10; Schedules 1 and 2 are summa-
period representativeof development use rized in Table 1.12.
There is a proposal to amend the Regulations by
extendingSchedule 1 by 14 new categories andclarifying
Predict/estimate noise and/or vibration those projects requiring environmental assessment in
due to development use during both Schedule 2. It is likely that these amendments will come
construction and operation phases
into force at the end of 1997.
Included in Schedule 2 are wind farms and theme
Assessimpactof development by comparing parks. The assessment guide for noise from wind farms
existing and predicted levels was published in September1996 by ETSU [35].
] The assessment for an Environmental Statement will
need to consider the construction andoperationalphases
Discuss situation with local representative
] of the development, including likely transport move-
ments. It will also need to include those elements of
- acoustic appraisal identified earlier, particularly a site
Determine appropriate mitigation measuresif
I survey. Some site uses in sensitive areas will not be
necessary J acceptable even with noise control measures, and plan-
fling permission for a new development may be refused
Presentoutcome of environmental assessment on environmentalgrounds.
1
i.e. the best practicable means have been employed, there 3. Department of the Environment Planning Policy
couldbe a requirementfor a substantial degree of public Guidance PPG 24 Planning and Noise, HMSO, 1994.
relations activity to satisf' potentialcomplainants. 4. Wimpey Laboratories Ltd Residues and Emissions:
Sound and Vibration, Report No. 15, CTG Channel
Operationalphase Tunnel Project — Environmental Impact Assessment,
If the development is industrial then adequate guidance Hayes, Middlesex, September 1985
may be obtained from earlier guidance dealing with 5. BS 6698: 1986 (amd 1991) Specffication for integrating-
industrial noise. This is also likelyto be the case for those averaging sound level meters, British Standards Institu-
developments with mechanical plant for building services tion, Milton Keynes
as their only noise source — a topic covered in detail in 6. BS 5969: 1981 Specficationfor soundlevel meters, British
Chapter 3. Standards Institution, Milton Keynes
Suitable criteria have to be set at the most appropriate 7. Organisation for Economic Co-operation and Devel-
locations, normally the site boundary, and discussions opment, Fighting Noise in the 1990s, OECD Publica-
with the Environmental Health Departments of the local tions Service, Paris, 1991
authorities should lead to an agreement. 8. The Department of Transport Calculation of Road
In the case of steady noise from mechanical plant, a TrafficNoise, HMSO, London, 1988
noise level criterion matching the existing background 9. Statutory Instrument 1975 No 1763 Building and
noise level is normally appropriate. Consideration may Buildings NoiseInsulationRegulations, HMSO, London,
need to be given to a criterion 5dBA below the back- 1975
ground noise level in some cases to avoid any significant 10. Land Compensation Act 1973, HMSO, London,
increase in a particularly sensitive area, but 10dBAbelow 1973
the backgroundshould never be necessary. 11. BS 8233: 1987 Codeofpractice for sound insulation and
Noise from other activitiessuchas leisure, for example noise reduction for buildings, British Standards Institu-
night clubs or ten-pin bowling, will need to be carefully tion, Milton Keynes
considered since music or impulsive noise can be annoy-
12. The Noise Advisory Council A Guide to Measurement
ingeven at noise levels below the existing background. In and Prediction ofthe Equivalent Continuous Sound Level
some cases the only means of control may be a restriction Le:j, HMSO, London, 1978
on hours of use. 13. Shield, B. M. and Zhukov, A. N. A survey ofannoyance
If there are noise problems once a development is caused by noisefrom the DocklandsLightRailway. Institute
operational, the first line of complaint is frequently theofAcousticsProceedings, 13(5), pp. 15—23 (1991)
local authority Environmental Health Officer. He may 14. Nelson, P. (ed.) Transportation Noise Reference Book,
have been asked to commentby the planningauthority at Butterworth-Heinemann, Oxford, 1987
planningstage on whethera noise condition should apply,
15. Statutory Instrument1996No 428. The NoiseInsulation
and reacts to the later complaints by carryingouthis own (Railway and Other Guided Transport Systems)
checks. If he agrees that the complaint is reasonable and Regulations.
may constitute a Nuisance, he can issue an 'Abatement 16. DepartmentofTransportRailway Noise and the Insula-
Notice in respectofNoise Nuisance' under the provisions tion of Dwellings, Mitchell Report, London, HMSO,
of the Environmental ProtectionAct 1990which gives 28 1991.
days to remedy, restrict, or stop the noise. The action to
17. US Environmental Protection Agency Office ofNoise
improve matters mayvary from 'best practical means', i.e. Abatement and ControlInformation on Levels ofEnvi-
achieving as much as costs and practicalities allow to ronmental Noise Required to Protect Public Health and
provide some amelioration, to radical noise control Welfare with an Adequate Margin of Safety US Govern-
measures or proscribing activity or plant noise during ment Printing Office, March 1974
certain times or even altogether. Once a Notice is served,
18. Departmentof Transport Calculation ofRailway Noise
an Appeal can be made against the basis of the alleged 1995, London, HMSO, 1995.
19. Control of Pollution Act 1974
occurrences constituting the statutorynuisance. However,
the Notice provisions are not suspended until the appeal
20. BS 5228: 1992/1997 Noise control on construction and
court so decides. cpen sites, British Standards Institution, Milton
Alternatively, a Court Order to restrain a continuationKeynes
of a wrongful act or omission may be initiated. This is the
21. Department of the Environment Control of Pollution
most common legal action in private nuisance claims. Act 1974, Implementation ofPart 3: Noise, Circular 2/76,
Failure to comply could mean fines and even imprison- HMSO, London, 1976
ment. Unlike a Notice, the restraint requirement is 22. BS 4142: 1990 Method ofrating industrial noise, affecting
immediate. Further details can be found in Garner's mixed residential and industrial areas, British Standards
Environmental Law [36]. Institution, Milton Keynes [BS under review]
23. Environmental Protection Act 1990
24. Noise and Statutory Nuisance Act 1993.
References 25. Department of the Environment Mineral Planning
1. This CommonInheritance, Government White Paperon Guide MPG 11 The Control of Noise at Surface Mineral
the Environment, 1990 Workings, HMSO, 1993.
2. Department of the Environment Report of the Noise 26. Departmentof the Environment The Use ofConditions
Review Working Party 1990, Batho Report, HMSO, in Planning Permissions, Circular 1/85, HMSO, Lon-
London, 1990 don, 1985
34 Acoustics in the Buift Environment
27. Code of practice on Noise from Clay Pigeon Shooting, 32. Deutsches Institutfür Normunge.V. 4150: 1986: Part
Midland Joint Advisory Council for Environmental 3 Structural vibration in buildings: effects on structures,
Protection, 1989, revised 1991 Berlin
28. Noise AdvisoryCouncil Draft Code ofPractice on Sound 33. BS 6472: 1992 Guide to evaluation ofhuman exposureto
Levels in Discotheques, HMSO, 1986 vibration in buildings (1Hz to 80Hz), British Standards
29. British Water Ski Federation Code ofPracticefor Water Institution, Milton Keynes
Skiing and Noise, 1989/Consultation draft for revised 34. Town and Country Planning (Assessmentof Environ-
Code October 1996. mental Effects) Regulations 1988, SI 1988 No. 1199,
30. The Noise Council Code of Practice on Environmental HMSO, London, 1988
Noise Control at Concerts, The Noise Council, 1995. 35. ETSU The Assessment & Rating ofNoisefrom Windfarms,
31. BS 7385: Part 1: 1990 Evaluation and measurementfor The Working Group on Noise from Wind Turbines,
vibration in buildings: guidefor measurementofvibrations ETSU-R-97, Final Report September 1996.
and evaluations of their effrcts on buildings, British 36. Garner,J. E, Harris, D.J. (eds) Garner'sEnvironmental
Standards Institution, Milton Keynes. Part 2: 1993 Law, Butterworth-Heinemann, Oxford,first published
Guide to damage levelsfrom groundborne vibration. July 1942, updated three times p.a.
Sound waves in the air on the source side impinge on
Chapter 2 Design the partition and 'drive' it (i.e. cause it to vibrate), and in
turn the barrier radiates sound into the receiving space.
The main issues are the transmission path, the sound-
acoustics reducing capability of the separating structure, and the
components making up this structure. The practical
reduction ofairbornesound energy is not only dependent
Duncan Templeton on the direct path via the wall, but also on flanking
paths.
Sound insulation
Flankingpaths
Airbornesoundinsulation Flanking paths occur at the edges of the physical barrier,
its junction to floors, other walls, ceiling, or ductwork
Airborne sound insulation entails the separation by a common to source andreceiving rooms (Figure 2.1). The
physicalbarrierofthe space containing a noise source from indirecttransmission via a flanking path can be reduced
an adjacent space requiring protection. The physical
barriermaybeapartitionorwallbetween rooms orfloor for by increasing the massofthe flanking walls, increasing the
rooms above each other (Diagram 2.1). The term 'parti- partition mass and bonding it to flanking elements, or
tion' implies dry construction (modular panels or clad introducing discontinuity to side walls in the adjacent
studded frames) or lightweight,non-loadbearing masonry; rooms, for example by independentwallliningor floating
floors.
'wall' implies a masonry structuretied into other wallsand
supportingfloors or roofabove.There is, however, overlap
between the terms. The two systems have differing Sound level difference
characteristicsacoustically. Partitions may be more subject The soundlevel difference between two spacesis depend-
to edgecracking,flankingandresonance effects. Walls can ent on the sound-reducing capability of unit area of the
transmit structurebornesound more than studding-type partition,the area of the partition, the acoustic properties
partitions, or pass re-radiated energy,andwithoutspecial of the source and receiving spaces, and flanking effects
detailingare moredifficult to make discontinuous. (Figure 2.2 and Diagram 2.2).
• Area of
TRANSMISSION partition
PATH
•• Stiffness
SRI • Coincidence
Partition
• resonances
Composite
construction
•WaIIs
• Roofs
• Floors
COMPONENTS
•• Doors
• Ceilings
Partitions
0
0
LU
C-,
0
U)
A = Sourceroom
B = Receiving room
z0
I-
I-
4
1Via other rooms
2 Via flankingwalls
3 Via ceiling 0
4 Via floor 0
5 Direct path through wall
A, Absorption units
Partition Area, s = h x I m2
S
0= R—1O log —
A
where 0is sound level difference
A is sound reduction index
for unit area of partition.
Facade Area
(SI
separate calculation
for sound via other
facadesand openings
Inside-to-outsidesound level difference
OL1—L2
= R— lOlog S+ 14 + 20 log r
Figure 2.3 Inside-to-outside sound level dfference
Facade Area
(SI A, Absorption units
separate calculation
for sound via other
SRI facades and openings
(A)
Room-to-room sound level dfference. This is defined in receiving point, or in other situations as defined in
Chapter 5. It is often useful in a sequence of measure- Chapter5 (Figure 2.3).In practice, an assessmentofnoise
mentsfor spacesat differentstagesoffit-out to have a level break-out from say a factory building requiresa sequence
difference standardized to a reference half second rever- of calculations involving the contributions of sound via
beration time. The Weighted Standardized Level Differ- the roof, other facades, doors and windows. Little useful
ence (DflT,W) is defined in BS5821 [1] and referred to in informationexists aboutsoundradiation viaangled roofs.
the Building Regulations Part E [2]. In mostcases,itis the openings in industrial buildings that
Iftwo spacesare similar in acousticalcharacter, i.e. have determine the noise break-out to adjacentsites.
the same amount of absorption in both the source room
andreceiving room, the measured level difference will be Outside-to-insidesound level dfference. This is given by:
the same whichever is the source room, If one space is D R— 10 log S/A —6dB
much 'deader' than the other,or the spaces are ofsimilar This assumes that the measuring microphone is well away
reverberation timebutgreatly different involume, the level from the facade. As with break-out, break-in calculations
difference will be greater with the deader (or larger shouldanalyse components ofnoise from the other faces
volume) space as the receiving room. This is not an and roof (Figure 2.4). While separate checks follow for
anomalyalthougha partitionapparently more effectivefor break-in and break-out, both may be of concern on some
sound insulation in one directionthan in the other can
jobs; for example, a hospital may be considered a noise-
requiresome explanation to a client. In fairlydead spaces, sensitive building type but has significant 24-hour noise
forexample cinemaauditoria orstudios,a truereverberant from plant and activity, which may upset nearby housing
field in the receiving room may not be generatedby the
source in the room adjacentand there will be agradientof (Diagrams 2.3 and 2.4).
soundlevel awayfrom the separating partition. Sound reduction index
The sound reduction index (SRI) is the basic measureof
Inside-to-outsidesound level difference. This is given by: sound insulation andis the number ofdecibels that sound
D=R— 10 log S+ 14 + 20 log r power is reducedby transmission through the barrier. The
average sound reduction index is usually expressed over
for facade radiation to hemisphere, where Sis the outside 100—3150Hz one-third octavebands, andwill be similar to
wall area and ris the radius from the facade centre to the the single value at 500 Hz (see Chapter 5).
38 Acoustics in the Built Environment
• Compare
to
criterion
NOISE BREAK-OUT
• Neighbourhood
properties
or site boundary
Averagesound insulation index rating, R theoretically increases by 6dB per octave. It may be seen
The average sound insulation index rating, R, is the from Figure 2.5 that the empirical mass law curve based
weighted single-figuredescriptordefined in Chapter5. Ason results is below the theoretical curve due to coinci-
with the A-weighting, greatersignificance to midandhighdence and resonance effects, and approximates to 5 dB
frequencies is given than for, say, the direct arithmetic
per mass doubling. The theoretical soundinsulation over
average of the SRIs. This more accurately reflects the the frequency range 100—3150 Hz is given by:
subjective effect of insulation due to the ear's reduced R = 7.6 + 20 log MdB
acuity at lower frequencies.
where M is in kg/rn2; alternatively, to determine the
Mass law theoretical performance at a particularfrequency:
The mass law indicates that in theory sound insulation
increases by 6dB for every doublingofweight ofdividing
R= 20 log (JM) — 47dB
f
element per unit thickness. The sound insulation also where is the frequency of the incidentsound.
Design acoustics 39
OBCF (Hz)
Masonry/blockwoi*
102-mm single-leaffairfaced 36 37 40 46 54 56 45
Single-leafplastered both sides 240 34 37 41 51 58 60 47
Cavitybrickwork with ties 480 34 34 40 56 73 76 52
Double leaf brickwork plastered both sides 480 41 45 48 56 58 60 51
100-mm lightweight blockwork fairfaced 125 32 32 33 41 49 57 41
100-mm blockwork plastered both sides 32 34 37 45 52 57 43
100-mm blockwork with plasterboard on dabs both sides 28 34 45 53 55 52 45
200-mm fairfaced lightweight blockwork 250 35 38 43 49 54 58 46
200-mm blockwork plastered both sides 37 39 46 53 57 61 49
200-mm blockwork plasterboard on dabs both sides 33 39 50 55 56 60 49
Three-leaf brickwork plasteredboth sides 720 44 43 49 57 66 70 55
Two leaves of 100-mm dense concrete blocks, 50-mm
cavity, 13-mm plaster both sides, cavity ties 35 41 49 58 67 75 52
Stud partitions
9-mm plasterboard on 50 X 100mm studs at 400mm
centres 15 31 35 37 45 46 35
13-mm plasterboard on 50 X 100mm studs at 400mm
centres 25 32 34 47 39 50 38
13-mm plasterboard on 50 X 100mm studsat 400mm
centres, 25mm mineralwool between studs 25 37 42 49 46 59 43
6-mm ply on 50 X 50mm studs at 600mmcentres 10 14 22 28 42 42 26
Double 13-mm plasterboard on 146-mm steel studs at
600 mm centres 32 41 47 49 53 58 47
Sheetmaterials/boards
9-mm ply on frame 5 7 13 19 25 19 22 18
25-mm T&G timber boards 14 21 17 22 24 30 36 25
5-mm ply/1.5-mm lead/S-mm ply composite sheets 25 26 30 34 38 42 44 36
Two layers of 13-mm plasterboard 22 24 29 31 32 30 35 30
1.2-mm steel sheet, 18 g 10 13 20 24 29 33 39 26
6-mm steel plate 50 27 35 41 39 39 46 38
40 Acoustics in the Buift Environment
OB CF (Hz)
An example of the effect of increasing weight may be tion is typically at lower to mid frequencies, 200—800Hz.
seen bythe performance ofa brickwall.A single-leafbrick However, there may be a high frequency benefit in
wall may be rated 45 dB average, a 225-mmwall 50dB, but practice also, because the quilt has an attenuatingeffect
it takes a thickness of 450mmto achieve 55dB, and this on sound via weaknesses at partition leafjunctions.The
performance may well be compromised by edge flanking schematic sound insulation related to frequency is
effects. The performances of typical constructions are increased from single-panel6dB per octave slope to 12 dB
scheduled in Table 2.1. A schedule ofdensities (kg/m3) of by double leaves plus absorption, so the overall effect is
common building materials may be found in BS 648 greater improvement at higher frequencies.
[3]. The sound reduction index will never rise to the
arithmetic sum ofthe Sills ofthe individual leavesbecause
Double leaves the two leaves canneverbecome totally isolated. However,
Double leaves with a gap between allow greater sound it is usual to obtain a higher SRI from a double-skin
insulation than a single layer ofequivalent weight. There construction than from the equivalent-weightsingle skin.
are two main means of transmission: The principle applies for horizontal or vertical dividing
•
radiation from the first panel into the air space elements, of i.e. roofs and floors as well as partitions. For a
partition two leaves of like mass (inkg/m2) with
excites the second panel, which radiates energy into
the receiving room; separation leaves d (in metres),
of
•
structureborne transmission between the two leaves 85
by mechanical links, the second leaf radiating the
transmitted vibrational energy.
I rnd
70
60
0
(/)
a)
x0)
a)
C)
co 40
a)
> 0
U
-D
30 0)
20 0
U)
10
ii;ft/
100 1000
Surfacemass (kg/rn2)
-D 50
55
,ii / ()
125 200 315 500 800 1250 2000 3150 5000
100 160 250 400 630 1000 1600 2500 4000
rd octave bandcentrefrequency(Hz)
U) (2) /
a)
C)
Ct
45
40
57
.-.
10
3
35
-.-1
125 4
30
10 100 1000 5
construction:effect on SRI
Figure 2.6 Double-leaf
1 0.7 mm steelskin
The effect is shown graphically in Figure 2.6. The 2 60 mm mineral wool (90 kg/rn3)
minimum useful gap is 50mm and wide cavities improve 3 2 x 13 mm plasterboard
the low frequency performance. Different characteristics 4 Metalstuds at 600 mm Cs
5 25 mm glassfibre
(weight, thickness) to one leaf offers a further improve- 6 Single-leaf common brickwork, plasteredboth sides (1920 kg/rn3)
ment, as the many tests on double glazing combinations
demonstrate. Cavity walls in brickwork with ties offer Figure 2.7 Dry liningversus masonry
negligible benefit to solid 225-mm construction. A good
example of the performancepossible is the use of wide-
cavityseparating wallsin multiplex cinemas where double
15-mm plasterboard either side of a 250-mm gap with fraction of the weight, to achieve 50dB average SRI, but
100-mm quilt inside consistently achieves 65—70 dB more care is required at the edge and at penetrations. In
DflTW. dry construction, if substantial acoustic doors are used, a
structural steel 'H' subframe bolted to the floor and the
Dry construction versus masonry underside of the floor above shouldbe used to hold the
Dry construction versus masonry is a frequent design doorset firm; standard partition metal studs allow too
comparison (Figure 2.7). A plastered block wall can be much flexing. This has to be borne in mind in any cost
replacedby double-layer plasterboard with quilt inlay at a comparisons between systems. The acoustic integrity of
42 Acoustics in the Built Environment
dry partitioningis more easily compromised by services Discontinuity implies the separation of structural ele-
penetrations, sockets and fixtures. ments so thatvibrations are not easily transmitted around
the mainstructure to cause intrusive noise in other areas
Stffness by re-radiation. The ingredients for 'box-in-box' rooms
The stiffness of thin panels is important because of the employing a consistent standard of discontinuity are
susceptibilityof leaves to be more easily driven by a noise double or even treble walls, a floating floor, and a
source on one side at certain frequencies. The effectcan substantial ceiling and slab above.
be seen in duct systemswhere thin duct walls may easily
transmit low frequencies of in-duct sound. Composite construction
Composite construction is that consisting of surface areas
Coincidenceeffect ofdifferentsound reduction indices, for example a brick
The coincidence effecthappenswhen soundwaves falling wall containing a door and a window. The total sound
on a panelexcite bendingwaves in it, the velocityofwhich power through a composite structure is the sum of the
depends on frequency. Sound transmission is greater at components of sound power transmitted by each compo-
the frequencies where the coincidence effect is greatest nent separately (see Chapter 5).
and the theoretical R is reduced by as much as 10 dB
below the level derived from mass law calculation (see Sound leaks
Chapter 5 for full description). Sound leaks can have a serious deleterious effect on the
performanceofa partition, wall, floor or roof. The effect
Partitionresonances is more marked at high frequencies. Figure 2.8 shows that
Partition resonances happen when standing waves are a hole of 0.001 m2 makes the composite SRI 40dB for a
formedwithin the partition.At the frequencies at which 45-dB-rated SRI wall of 16m2. Gaps at door edges are a
this occurs the resonances will reduce performance.In a typical example of sound leakage.
single-leaf partition or wall, the fundamental resonant
frequency is determined by its stiffness. At higher fre- Buildingenvelope:rooft
quencies, there are other performance 'dips' at har- Roofs are typically oflighter construction than outer walls
monics ofthe resonantfrequency. For many partitions the and of relatively large area compared with walls and
resonances occur at low frequencies outside the range of openings. The exposure to road, rail or industrial sources
usual interest and the effect can often be ignored. is less, but a building can be vulnerable to aircraftnoise or
However, it may be of interest when, for example,
checking a specific curtain wall glazing arrangementfor
f=--
where
tr\;
low frequency components of traffic noise. The most
importantis the fundamental resonantfrequency, calcu-
lated from:
11 1
I—+—
[a2b2
a and b are the partition dimensions (m), t is its
1:1 000000
0
CD
1:500000
1:250 000
1:125000
1:64000
1:32 000
1:16000
thickness (m), E its Young's Modulus (Pa), and p density
1:8 000
(kg/rn3) (Table 2.2). a)
-c 1:4000 1H1N-n-I
0)
1:2 000
Discontinuity
The discontinuity of rooms within a building can get C
0 1:1 000 II I IIILltIIi
complicated at junctions and is most practically imple- CD 1:500
mentedon smaller studiospaces than on majorauditoria. (a
C
a)
1:250
1:125 'II1
0 1:64 0
Table 2.2 Values of Young'smodulus and density 1:32
I
Cd,
CD
a) _______.'.1'-H-I-rtuInhIlC
1:16 ________ Cc?
Density, p ftHNII1'WIIi'JtI
I II
Material 0
(Pa) 1:4
(kg/rn3) 0
CD
cc
1:2 ##f'IJ
H _________
-
roof-mounted items ofplant. Traditional slated roofs have be cased and absorption is reduced compared to the
a reasonable surface mass, butgaps up the lapping slates proprietarytiled grid.
and the need to ventilate under the lapping slates make
such roofs poor insulators (27 dB average) on their own. Walls
The usual use of a roofvoid with its thermal insulation Blockwork
and plastered ceiling increases this to 38 dB average. Blockwork performs reliably if well constructed and of
adequate mass. Lightweight thermal blockwork
Flat roofs (350—700kg/m3) frequently usedin the absence ofadvice
Flat roofs of built-up felt on thermal insulation on metal otherwise, is poor. Unplastered blockwork loses sound
decking or composite construction of profiled metal, insulation by its fissures and movement cracks: plastering
insulation and liner tray metal sheeting, achieve only can improve this. The best blockwork is 2000kg/m3 solid
30—35 dB. Roofs of similar surface mass to floors — no-voids dense concrete masonry (dcm), a thickness of
screeded topping to precast concrete, for example, with 190 mm achieving 50dB SRI.
asphalt and insulation above — manage 45—50 dB. For Target densityvalueswill have to be set outas there is no
performance exceeding 50 dB, the roof will have to be set definition of 'dense' except in terms of blockwork
supplemented by a barrier ceiling below. A conventional strength.Some strong blockwork (7 N/rn3) is not neces-
lay-in grid mineral tile ceiling will be of little additional sarily very dense at 1400kg/rn3.
value to the roof,particularly if open grilles in the ceiling
allow its use as a supply or extract air plenum. Brickwork
Brickwork is usually better than blockwork; the smaller
Lightweightroofs units can be built around partitions more easily and
Lightweight roofs with a profiled metal outer face are movement cracking is less. The heaviest (2300kg/rn3)
subject to rain and hail drumming, and can also 'click' construction is obtained by using solid engineering
and bang during thermalmovement. Damping the outer bricks; an acceptable everyday use is commons laid with
skin by having quilt directly behindit muffles the sound to frogs up (1700—2000kg/rn3). Mortar density is typically
a degree. Composite metal roofswith a soffit ofperforated 1800kg/rn3.
metal are often used in sports halls — gymnasia,ice rinks
and swimming pools — to absorb soundwithin the space, Partitions
butthe position of the vapour barrier above the perfora- With care, lightweight construction can outperform
tions needs considering carefully.Too thick a membrane masonry, certainly mass-for-mass, andsometimes even for
will blank offthe absorption capabilityof the quilt above; similar thicknesses. In plasterboarded partitions, metal
some systems have the membrane embedded in the studding has largely replaced timber studding and gives
thermal/absorptionquilt, but special fixings through this better SRI performance because the leaves are coupled
arrangementare needed. across the studs more resiliently. Plasterboard itselfis used
less often as the range of metal-skinned modular panel
Cethng partitioning diversifies and becomes more competitive.
To uprate the sound reduction capability of a roof, the The panels take the form of50-or 100-mm-thick elements
suspension of a barrier ceiling can be included. The with absorption material in a core, for offices.For studios,
performance of a timber floor can be altered from 42dB noise havens or music practice rooms, panels can be
to 58 dB by the addition of a British Gypsum M/F ceiling perforated on the inner face for absorption and fixed to
as shown in Figure 2.9. The system uses straightforward isolated floor and ceiling panels.
metal straps; some other specialist systems use resilient A strong combination is masonry plus independent
hangers, and any design will have to address problems of lining, with quilt in the cavity Some examples are shown
suspending ductwork or further decorative ceilings below in Figure 2.10. Care must be taken not to make the cavity
the barrier ceiling. too small, otherwise low frequency resonances can render
the dry lining disadvantageous rather than advantageous
Ceiling voids to insulation.
Ceiling voids are familiar transmission routes for sound
between rooms, where partitions are not carriedthrough Folding partitions. These shouldbe avoided if possible.
to the roof or floor above. Carrying the partitions up not Theyare frequently installed and then complained of in
only breaks the ceiling but inhibits moving them and places where low background noise levels and need for
affects ventilation arrangements — ducted supplies need confidentiality exist, e.g. solicitors' offices and board-
the cross-talk attenuationdiscussed in Chapter 3. Ceiling rooms. Because of gaps around the suspension gear, the
manufacturers should be able to quote room-to-room level difference either side is similar to a door's: 15—20 dB
transmission characteristics as measured in BS 2750: Part average. Folding panel rather than concertina types are
9: Laboratory testing [411. marginally better, but beware of the claims of suppliers
who quote high sound insulation values, even supported
Suspended ceilings by tests, that relate only to the body ofthe panels and not
Suspended ceilings tend to be selected for their light to the total assembly. The best types have some closure
weight (hence economy) andfor absorption, rather than seal — pneumatic or mechanical — which can lock the
sound insulation: the level difference through a ceiling is panels in place. A robust ceiling at head and a division of
limited to 10—15dB. For greater performance a closed ceiling void are necessary. By care,separation in the order
plasterboard ceiling can be used but recessed lights must of around 35dB can be obtained: this can be related to
44 Acoustics in the Built Environment
100
9C — — — — — — —
8C —
—
— —
/ — N ——
— —
V \
1
71 ———————————————
0
U)
.
\
-.-.. N
0.
E
a)
N
3 E
z0
20
125 200 315 500 800 1250 2000 3150
100 160 250 400 630 1000 1600 2500
rd octave band centre frequency (Hz)
1 Timber floor
3 (b) 2 Timber floor and suspended ceiling
Impact sound
4
6 2
0
III 200mm
18 mm chipboard
1
195 x 45 mm timberjoists 600 cs
2
3 13 mm gyproc V
4 150 mm mm. spacing/ceiling void x
50
5 80 mm gypglas 1000 C
6 13 mm gyproc on M/F suspension system C
(a) 0
C)
V
a,
V
Figure 2.9 Sound insulatingsuspended ceiling. (Courtesy of C
0
British Gypsum) U)
"Average sound reduction index, 100—3150 Hz. speech privacyneeds as shownin Table2.3. An alternative
method is to add the background noise level and the
partitionSRI and workto a total exceeding 65 (Table 2.4).
Sound insulation performance is particularly important
for speech privacy in the frequency range 500—2000Hz.
Doors
The typical domestic door, hollow-coredwith a loose fit,
achieves 15—20dB average. A solid-core door with fire
ratingrebatesto the frameimproves this marginally. The
fittingofintegralblade or compression seals to edgesand
the threshold help the value up to about 30dB. Purpose-
'± 2 made timber or metal doorsets can be selected in the
range 35—45 dB average. The weight of acoustic doorsets
\
2
—X- 3
xa,
\I N
'4
C
C
0
C,
a'
÷4 C
0
U)
___
111111111 125 200 315 500 800 1250 2000 3150
200 ri,rr 100 160 250 400 630 1000 1600 2500 4000
1 single-leaf common brickwork rd octave band centre frequency (Hz)
48 mm glass fibre (24 kg/rn3)
2
3 13 mm plasterboard (wallboard) on 48 mm. -•-1 Rw=44dB ±2 Rw=58dB
'I' section metalstuds at 600 mm cs —*3 Rw=55dB +4 Rw=59dB
4 2 x 19 mm Gyproc' plank, adhesive between layers
5 2 x 13 mm plasterboard masonry/dry-lining combinations
III 11111
0 200 mm
V
x
Va)
C
C
0
U
V
VC
I 0
U)
125 200 315 500 800 1250 2000 3150 125 200 315 500 800 1250 2000 3150 5000
100 160 250 400 630 1000 1600 2500 100 160 250 400 630 1000 1600 2500 4000
rd octave band centre frequency (Hz) rd octave band centre frequency (I-lz)
Source: British Gypsum Source: British Gypsum
Impactsound (c) Airborne sound insulation
Absorptionand insulation
17oom Absorption and insulation are not to be confused. The
Timber floors as used in dwellings perform as shown in application ofsound-absorbing finish to a separatingwall
Figure 2.14. Between flats, Building Regulation Part E will not have any discernible effecton its sound insulation
recommended detailsare a reference source [2]. The 1992 properties at all. All surfaces absorb sound to a greater or
version increased newparty floorstandards, e.g. abase floor leser extent: bare concrete or marble have a low sound
slab in a composite system is increased from 220kg/rn2 to absorption coefficient, and hence absorb little sound and
300 kg/rn2, and brought in onerous provisions for con- reflect back almost all incident energy.
50 Acoustics in the Built Environment
70
-J
0
Ct
C-)
a-
E
-a
a)
z0
' 4 10
125 200
100 160
315
250 400
500 800 1250 2000 3150
630 1000 1600 2500
rd octave band centre frequency (Hz)
Durabella Westbourne flooringwith 22-mm battens on
a 200-kg/rn2 concrete floor
(b) Impact sound
x6
/
Figure 2. 5 Floatingtimberfloors. ((a) Courtesy of
Contiwood(Durabella) Ltd; (b) courtesy ofPheonix Floors Ltd)
1 19mm chipboard
2 8mm isolating semi-rigid foam
fused to base of battens
3 Concrete subfloor
4 Cover moulding on mastic
5 19mm chipboard on hardwood/foam
isolation battens •0
(a) 6 Existing floor construction a)
C-,
a
a)
>
a)
-J
Absorption coefficients
Absorption coefficients are not considered dependent on
the angle of incidence of sound striking the medium:
randomincidence is assumed. Absorption coefficientsare
normally given for the frequency range 125—4000Hz
(Table 2.5). Third octave band values of absorption
coefficients will not differ much from octave band values
(octave bands average the third octave values, as opposed
to NR values where one-third values are additive in 10
producing octave bands); NRC and dBA are correspond- 125 200 315 500 800 1250 2000 3150
ingly different in derivation. 100 160 250 400 630 1000 1600 2500 4000
rd octave band centre frequency (Hz)
Types of absorber
Types of absorber include fibrous absorbers, fibrous (1, structural floor; 2, structural and floating floor)
absorbers with impervious membrane facing, and Airborne sound insulation
fibrous absorbers covered with perforated panelling. An Figure 2.16 Increaseofairborne sound insulation by use of
example of the first type is quilt batts mounted directly concretefloatingfloors. (Courtesy ofSoundAttenuators Ltd)
Design acoustics 51
OBCF (Hz)
'Hard'finishes
Water or ice 0.01 0.01 0.01 0.01 0.02 0.02
Smooth concrete, unpainted 0.01 0.01 0.02 0.02 0.02 0.05
Smooth concrete, sealed or painted 0.01 0.01 0.01 0.02 0.02 0.02
Concrete blocks, fairfaced 0.05 0.05 0.05 0.08 0.14 0.20
Rough concrete 0.02 0.03 0.03 0.03 0.04 0.07
Brickwork, flush-pointed 0.02 0.03 0.03 0.04 0.05 0.07
Brickwork, 10-mm-deeppointing 0.08 0.09 0.12 0.16 0.22 0.24
Plastered walls 0.02 0.02 0.03 0.04 0.05 0.05
Painted plaster 0.02 0.02 0.02 0.02 0.02 0.02
Ceramic tiles 0.01 0.01 0.01 0.02 0.02 0.02
Marble, terrazzo 0.01 0.01 0.01 0.01 0.02 0.02
Glazing (4mm) 0.30 0.20 0.10 0.07 0.05 0.02
Double glazing 0.15 0.05 0.03 0.03 0.02 0.02
Glazing (6mm) 0.10 0.06 0.04 0.03 0.02 0.02
Ceilings
13-mm mineraltile, direct to floor slab 0.10 0.25 0.70 0.85 0.70 0.60
13-mm mineraltile, suspended 500mm below ceiling 0.75 0.70 0.65 0.85 0.85 0.80
Metal planks, slots 14% free area, mineralwool overlay and
void 0.50 0.70 0.80 1.0 1.0 1.0
Metal tiles 5% perforated, 20-mm quilt overlay andvoid 0.13 0.27 0.55 0.79 0.90 1.0
Woodwoolslabs 0.40 0.40 0.70 0.70 0.70 0.80
Panels
Solid timberdoor 0.14 0.10 0.06 0.08 0.10 0.10
9-mm plasterboard on battens, 18-mm air space with glass
fibre 0.30 0.20 0.15 0.05 0.05 0.05
5-mm ply on battens, 50-mm air space with glass fibre 0.40 0.35 0.20 0.15 0.05 0.05
Suspended plasterboard ceiling 0.20 0.15 0.10 0.05 0.05 0.05
Steel decking 0.13 0.09 0.08 0.09 0.11 0.11
Ventilation grille (perm2) 0.60 0.60 0.60 0.60 0.60 0.60
13-mm plasterboard on frame, 100-mm air space with glass
fibre 0.30 0.12 0.08 0.06 0.06 0.05
13-mm plasterboard on frame, 100-mm air space 0.08 0.11 0.05 0.03 0.02 0.03
2 X 13-mm plasterboard on frame, 50-mm air space with
mineralwool 0.15 0.10 0.06 0.04 0.04 0.05
22-mm chipboardon frame, 50-mm air space with mineral
wool 0.12 0.04 0.06 0.05 0.05 0.05
16-mm T&G on frame, 50-mm air space with mineral wool 0.25 0.15 0.10 0.09 0.08 0.07
22-mm timber boards 100-mm-wide, 10-mm gaps 500-mm
air space with mineralwool 0.05 0.25 0.60 0.15 0.05 0.10
Treatments
Curtains in folds against wall 0.05 0.15 0.35 0.40 0.50 0.50
25-mm glass fibre, 16 kg/m3 0.12 0.28 0.55 0.71 0.74 0.83
50-mm glass fibre, 16 kg/m3 0.17 0.45 0.80 0.89 0.97 0.94
75-mm glass fibre, 16 kg/m3 0.30 0.69 0.94 1.0 1.0 1.0
100-mm glass fibre, 16 kg/m3 0.43 0.86 1.0 1.0 1.0 1.0
25-mm glass fibre, 24 kg/m3 0.11 0.32 0.56 0.77 0.89 0.91
50-mm glass fibre, 24 kg/m3 0.27 0.54 0.94 1.0 0.96 0.96
75-mm glass fibre, 24 kg/m3 0.28 0.79 1.0 1.0 1.0 1.0
100-mm glass fibre, 24 kg/m3 0.46 1.0 1.0 1.0 1.0 1.0
50-mm glass fibre, 33 kg/m3 0.20 0.55 1.0 1.0 1.0 1.0
75-mm glass fibre, 33 kg/m3 0.37 0.85 1.0 1.0 1.0 1.0
100-mm glass fibre, 33 kg/m3 0.53 0.92 1.0 1.0 1.0 1.0
50-mm glass fibre, 48 kg/rn3 0.30 0.80 1.0 1.0 1.0 1.0
52 Acoustics in the Buift Environment
OBGF (Hz)
75-mm glass fibre, 48 kg/rn3 0.43 0.97 1.0 1.0 1.0 1.0
100-mm glass fibre, 48 kg/rn3 0.65 1.0 1.0 1.0 1.0 1.0
25-mm acoustic plaster to solid backing 0.03 0.15 0.50 0.80 0.85 0.80
9-mm acoustic plastic to solid backing 0.02 0.08 0.30 0.60 0.80 0.90
9-mm acoustic plasteron plasterboard, 75-mm air space 0.30 0.30 0.60 0.80 0.75 0.75
50-mm mineralwool, 33 kg/rn3 0.15 0.60 0.90 0.90 0.90 0.85
75-mm mineralwool, 33 kg/rn3 0.30 0.85 0.95 0.85 0.90 0.85
100-mm mineral wool, 33 kg/m3 0.35 0.95 1.0 0.92 0.90 0.85
50-mm mineral wool, 60 kg/rn3 0.11 0.60 0.96 0.94 0.92 0.82
75-mm mineral wool, 60 kg/rn3 0.34 0.95 1.0 0.82 0.87 0.86
25-mm mineral wool, 25-mm air space 0.10 0.40 0.70 1.0 1.0 1.0
50-mm mineral wool, 50-mm air space 0.50 0.70 0.90 0.90 0.90 0.80
50-mm mineral wool (96 kg/rn3) behind 25% open area
perforated steel 0.20 0.35 0.65 0.85 0.90 0.80
Floorfinishes
Cord carpet 0.05 0.05 0.10 0.20 0.45 0.65
Thin (6-mm) carpet on underlay 0.03 0.09 0.20 0.54 0.70 0.72
Thick (9-mm) carpet on underlay 0.08 0.08 0.30 0.60 0.75 0.80
Wooden floor boards onjoists 0.15 0.11 0.10 0.07 0.06 0.07
Parquetfloor on timberjoists and deck 0.20 0.15 0.10 0.10 0.05 0.10
Parquetlaid concrete 0.04 0.04 0.07 0.06 0.06 0.07
Vinyl or linoleum on concrete 0.02 0.02 0.03 0.04 0.04 0.05
Vinyl and resilientbacking on concrete 0.02 0.02 0.04 0.05 0.05 0.10
Miscellaneous
Audience on timber seats (1/rn2) 0.16 0.24 0.56 0.69 0.81 0.78
Audience on timber seats (2/rn2) 0.24 0.40 0.78 0.98 0.96 0.87
Audience per person, seated 0.33 0.40 0.44 0.45 0.45 0.45
Audience per person, standing 0.15 0.38 0.42 0.43 0.45 0.45
Seats, leather covers (per m2) 0.40 0.50 0.58 0.61 0.58 0.50
Upholstered seats (per m2) 0.44 0.60 0.77 0.89 0.82 0.70
Floor and upholstered seats (perm2) 0.49 0.66 0.80 0.88 0.82 0.70
Areas with audience, orchestra, or seats, including narrow
aisles 0.60 0.74 0.88 0.96 0.93 0.85
Orchestra with instruments on podium, 1.5 m2/person 0.27 0.53 0.67 0.93 0.87 0.80
Shading factor (apply to finishes under seats, x coefficient) 0.80 0.70 0.60 0.50 0.40 0.20
Air 30% RH (per m3 at 20°C) 0.005 0.01 0.04
Air 50% RH (per m3 at 20°C) 0.005 0.009 0.03
Air 70% RH (per m3 at 20°C) 0.005 0.009 0.02
Office furniture (perdesk) 0.50 0.40 0.45 0.45 0.60 0.70
on a wall surface, or carpet on a floor. An example of is 5% or less, the panels are reflecting except for
the third type is metal perforated suspended ceiling myriad Helmholz resonators formed by the holes. 'Bass
tiles with quilt inlay above; provided the open area of traps' are used in studios to provide broadband absorp-
the perforations exceeds say 20%, the quilt and air tion right down to very low frequencies. They consist of
cavity behind the metal tiles is almost as efficient at a lined labyrinth air space within which negligible
soaking up sound as if the tiles were not present. In reflection results.
studios, deep boxes with thin membranes can be pur- An extremecase of absorptive materials installation is
pose-designed or selected to even out the reverberation the semi-anechoic and anechoic chambers in acoustic
characteristics at different frequencies. Resonance laboratories (Figure 2.17): deep wedges of foam above,
absorption can be produced by selecting appropriate below andto all sides reduces the reverberation time to a
perforation and air space depth. If the perforation rate very low value at all audible frequencies.
53
Parallel pattern
(1/rn2
II —1
Largerspaces
Larger spacesproducesound decay characteristicsin poor 0.2
agreement with Sabine or In
Norris-Eyring. 'amorphous
space' areas, such as shopping malls or industrial halls, 0.1
thereisnot atrue reverberant fieldacross thespace and the
sound characterwill vary in different parts ofthe space. On 0
theother hand,suchspacesare not 'free-field' and the SPL 125 250 500 1000 2000 4000
will decay at something less than 6dBforeach doubling of Octave band centre frequency(Hz)
distance from the source. Empirical data suggest 4dB/ 1 parallel pattern
doubling of distance across typical industrial halls. Alter- 2 cross pattern
native forms of calculation have been proposed for
industrial halls. Complex spaces can sometimes be sub- Figure 2.18 Overheadsound absorbers. (Courtesy of
divided into individual coupled volumes: if alcoves off a Rockwool)
54 Acoustics in the Built Environment
(b)
(d)
(f)
Figure 2.19 Sound decay analysis: ODEON
ampiffied music events: these included stalls examples of Initial investigation, by BDP Acoustics, consisted of
+3dB on direct sound after 300 ms time delay. extensive baseline measurements at 14 receiver positions
A fabric velarium was the original means of acoustic andtwo source positions, for unoccupiedhall andduring
control, removedin 1949. The replacement suspensions events. Results were validated to an ODEON model
were 102 grp 'mushrooms' at 25 m height and a cumber- (Figure 2.19e). Balcony seating with tiering — existing and
some orchestral canopy at 11 m height, installed in proposed— werelaboratorytestedprior to installation and
1969. remeasurement in the hall.
56 Acoustics in the Built Environment
90
14
80
-O
w
i 1Z
1H
12 C.)
70
+
a,
0
I: I 1 1
60
C
Co
50 -:
U,
5 5
40
30
125 200 315 500 800 1250 2000 3150
100 160 250 400 630 1000 1600 2500
PROJECTION
ROOM
3
4
dB
65—74
(variance
Figure 2.22 Odense concert hall, CarlNielsen Hall
over 10 projects>
are not even axial to the platform: Aalto's concert halls Seating
put more seating on the 'keyboard' side of the piano Seating is a key issue as there are so many constraints —
soloist; Segestrom Hall, Orange Free State, California, sightlines, travel distances, aisle steps, seats per row and
interlocks two narrow halls as a means of avoiding the balance of seats at different tiers. 'Vineyard' seating as
disadvantages of fan-shaped auditoria. In the UK there pioneeredat the Berlin Philharmonie workswellin milder
are impressivenew semi-surround major concert halls in form: 'seating trays' of several hundred can be optimally
Manchester (Bridgewater Hall) and Belfast (Waterfront set to face the platform and have local side-reflecting
Hall). Figure 2.23 shows the 2400-seat Bridgewater Hall surfaces (Figures 2.25a and b).
during design and at completion. Figure 2.24 gives
similar coverage of the 2250-seat Waterfront Hall.
Reference can be made to a number of case study
collections,for example AcousticalSurvey ofElevenEuropean Table 2.6 Optimum volumesfor performancespaces
Concert Halls [61 and Halls for Music Performance:
1962—1982 [7]. Excellent recent works are Beranek's
Optimum volume (m3/occupant)
Concert and Opera Halls: How They Sound [8], Barron's
Auditorium Acoustics andArchitecturalD n
[91, andAndo
and Noson's iVlusic and Concert Hall Aoustics [10]. Minimum Recommended Maximum
Theatres 2.5 3 4
Volume Rooms for speech — 3 5
The volume should be adequate for a full-bodied sound: Opera houses 4 5 6
the old rule of 'RT X 4 equals volume per person' is if Concerthalls 8 10 12
anythingon the low side and 10 m3 per person even in a Churches 6 10 14
full-size hall is advisable (Table 2.6).
Design acoustics 59
(f)
(a)
(b)
(c) (d)
Building services and lighting noise were designed to zones, there are massive twin wall and roof structures to
PNC 15, with main plant in an isolated tower (Figure exclude noise (Figures 2.23fand 2.23g).
2.23e). The entire 25000-ton concert hall is supportedon
280 steel springs to prevent disturbance from adjacent
Metrolink tranis, External noise break-in is minimized by The WaterfrontHall, B4fast
wrapping the ancillary accommodation around the audi- The 2250-seat elongated hexagon concert hall results
torium. Where the auditorium rises above these buffer from a collaboration of acousticians Sandy Brown Asso-
Design acoustics 61
(e) (f)
AUDITORIUMSECTION
0 IOU
(g)
Figure 2.24(e)—(g) The WaterfrontHall, Belfast
ciates, with architects Robinson and Mcllwaine.A large- roof comprises two concrete slabs, the outer domed to
volume reverberant space has evolved from St David's form a significant void between the slabs.
Hall, Cardiff, in arrangingseating trays for good diffusion A displacement ventilation system includes very low air
and local soundreflections. Overhead reflectors, galleries, velocity supply air via special terminals under the seats.
and platforms provide local reflections for performers, to Special flexible bellowsintroduce air to stage elevators to
assistensemble andbalance. accommodate the range of elevator movement.
Early use of a 1:50 physical scale model led to design Early objective measurements and subjective tests indi-
modifications and further testing. Flexibility of use is cate an excellent acoustic for orchestraluse, a warm well-
accommodated by removal of stalls seating to allow a balanced yet intimate sound.
central arena format, and acoustic adjustment by the
movement ofhigh level areas offabric. The fabric reduces Further acousticparameters
reverberation timesfor events where speech intelligibility
or amplified music are the main consideration. Early decaytime
Ancillary accommodation — foyers, dressing rooms, The early decay time (EDT) is the most important
offices— cocoon the auditorium from external noise. The criterion in evaluating a hall's acoustics, followed by
62 Acoustics in the Built Environment
Figure 2.25(b) Seating trays: Berlin Philharmünie (morerecently built Recital Hall)
Design acoustics 63
QUADRATIC
RESIDUE SEQUENCE
I ! — j I I I I I I
THICK
GRG
BANNER
DROP
ZONE
I'
-
III
(a) (b)
II 4
Figure 2.27 Concert hail detail.: (a) high-backseats at rear; (b) ceiling reflectors; (c) side-wallabsorberbanners
66 Acoustics in the Built Environment
• Public meetings:
political, religious
• Pop festivals
D
• Spectator sports:
cricket, rugby,football
• Participation sports:
running
• Trade shows
Exhibitions
E • Dances
a, I Banquets
a) I
C
C5) Ca
Wa,
E
I • Cinema
2 c
• Conferences
Pop concerts
•
o I
• Theatre
I Staged musical events
I j Ballet
Opera
I TV studiotheatre
I • Classical concerts
I Choral
I Chamber
I I
— __I
• BBC
ILR (IBA)
Recording studios
0Z
Diagram 2.5 Publicperformance
spaces
Activity RT Building te
Broadcast 0.2—0.25 Sound dubbing, announcerbooths
0.3 Small speech studios
L0—2.0 Large classical music studios
Speech 0.6—1.2 Council chambers, law courts, lecture theatres, meeting rooms,
conference halls
Drama 0.9—1.4 Theatres, functionrooms
Amplified sound 0.5—1.2 Multiplex cinemas, pop concert venues, discotheques, videowall
settings
Multiuse 1.0—1.7 School assemblyhalls, community halls, sports/arts halls
Opera 1.0—1.6 Opera houses, theatres with orchestrapits
Soloists, ensembles 1.2—1.7 Recital halls, orchestra rehearsal halls, chamber music salons
Orchestral music 1.7—2.2 Concert halls
Organ and choir music 2.0—5.0 Ceremonial halls, organ concert halls, churches, cathedrals
68 Acoustics in the Built Environment
Discotheques
The attraction ofdiscotheques is precisely their noise and
freneticactivity. The problems associated are noise break-
out to neighbouring properties, hearing damage to
employees,andstructurebornere-radiated sound to other
facilities in the same building. Average sound pressure
Table 2.8 Specific needs ofcourts and surroundingrooms levels on the dancefloor climb to the range 90—llOdBAas
the night goes on, with a high component of low
Good include frequency sound. Levels tend to be 8—10dBA higher by
sound sound the endofthe disco session. Where amplified live music is
insulation absorption performed,levels can be even higher. Such sound power
levels exceedclassical forms ofmusic — orchestras in loud
Main courtrooms •• •• passages maybe a more modest80 dBAorso. The Code of
Practice on sound levels in discotheques [11] identifies
Juvenile courts
Sound lobbies
• •• the hearing risk for staff in particular. The Code recom-
mends an LAeq not to exceed 100dB at the nearestpoint
Magistrates' retiring rooms
Court hail
• • in the premises to operating loudspeakers, the value
referred to as the Maximum Permissible Exposure Level.
Duty solicitor's room
Social workers' offices •• It assumes 25% of the total public area is given to rest
areas, otherwise a MPEL of 95 dB would apply. It is not
Holding/waiting rooms
Internal circulation
Clerk to the justice's office •• • desirable to have direct-to-earsound, diffuse andreflected
sound being better to even out exposure. Previous GLC
guidance recommended an LAeq,8h not exceeding 93dB,
Deputy clerk's office
Secretary's office •• or for external audience protection 93dB at 50m.
Discotheques are an obvious candidate for assessment
Legal/administration/accounts areas
Public zone — counter
Interview rooms •• •• against the Noise at WorkHSE statutoryguidance [12]. In
considering an Entertainments Licence, any authority will
Consulting rooms •• consider the following:
• objection at any public hearing
Refreshments area
• duration and timing of concert
Design acoustics 69
• frequency of concertsat the same premises busy roadsor airports, or in city centres. The main issues
• noise complainants at previous concerts will be noise break-in from outside, privacybetween rooms
• location (in relationto noise-sensitive buildings) and to public rooms, and ventilation noise.
Noise-limiting devices on sound systems can be Windows
included although commercially it is not realistic to set Windowsin hotels have openinglights even in the noisier
values below 90dBA. Premises should have full mechan- situations; good weatherstripping and double glazing are
ical ventilation (direct-to-atmosphere extracts with no essential. Some protection to road noise can be given by
attenuationwill in most cases be unacceptable because of inset balconies.
the impact on the environmental noise climate) and
lobbied doorset accesses at the entrances. Re-radiated Privacy
structurebornesound demands great care in design, one Privacybetween rooms will be of a reasonable standard if
approach being to isolate the disco walls by— using a separating walls and floors are selected with an average
drylining 'shell' and effectivelya triple floor a dance SRI of 50dB (for example, plastered 200-mm dcm,
floor on a concreteslab on isolators on a secondconcrete blockwork and solid precast concrete floor units with
slab — to alleviate the problem on the floor below. structural topping). Creation of a lobby outside the en
suitebathroomwill give isolation to corridor noise. Cross-
Education buildings talk attenuation to bathroom extracts will prevent this
Draft revised Building Bulletin 51 Acoustics in Education being a route for plumbingsounds. If single doors from
corridors are used, these should be 35-dR rated, i.e. solid
Buildings [13] and Design Note 17 [14] have yet to core plus seals, well rebated. Bathroom — corridor walls
materialize in final form. Design Note 25 Lighting and
Acoustic Criteria for the Visually Handicapped and Hearing should have an average SRI of 45dB. Partitions must
extend full height, structural floor-to-floor, and weak-
Impaired in Schools [15] can be referred to for those with nesses like back-to-backelectrical sockets mustbe avoided.
special needs. BS 8233 [161 classifies four groups with Room televisionandradio sets shouldnotbe fixed directly
sound insulation requirements varying from 25 dB to
45 dB. Classroomconditions should be controlled to 0.75 s to the room separating walls.
at middle frequencies and 40dB average separation Ventilation noise
between reading areas. More importantly, overhead ceil- Ventilation noise should be kept within NR 35 in any hotel
ing surfaces can usefully be sound reflective if side and and down to NR 25 in good-standard bedrooms. It is
rear walls are panelledin saypin boardingto dampsound- arguedthat unless a systemis audible, guests will thinkthat
blurringcross reflections. General mechanical ventilation it is inoperable. Atmosphere connections andchiller plant
shouldbe designed to within NR35 in teaching rooms. In shouldbe remoteto hotelbedrooms, or wellscreenedand
primary schools in particular, openingdoors and windows attenuated.Time clockson, say,kitchen extracts could help
from classrooms are expected, so a consideration of
facade aspectwill avoid distracting external noise levels. byproviding a cut-offtime so plantis not noisyin the early
hours. Plumbing noise, particularly 'water hammer',
School theatres are no longer exclusively an adaptation should be avoided by a 'head' to water pipework
of the assembly hall but mimic public theatres. A (307720mmUS practice) or a balloon-typereliefvalve.
reduction in scale, including platform height and size
allows for the lesser projection ofchildvoicescomparedto
Housing
adult. A BRE Report, Building Regulations and Health[17],
Health buildings mentions a 1980 survey where 18% of residents of new
Noise control rather than room acoustics is important.
Hospitals are highly serviced and reference to Building Table 2.9 Recommended noise ratingsfor health care
Notes, Technical Memoranda, Health Circulars and Hos-
facilities
pital Data Sheets should be made. Recommended design
criteria are shown in Table 2.9, as a more detailed
interpretation to rating values included in Chapter 3. Facilities NR
Separation between rooms has to be carefullyconsidered
given the requirement to stop most partitions offatceiling Quiet wards, overnight stay rooms, chapel, 25—30
level.
New major hospital developments have considerable resuscitation
impact on the local community and the services centre Children'swards, treatment and recovery 35
with its standby diesels, boilers and chillers needs partic- rooms, staff rest rooms, staffbases, offices
ular attention.Attenuation in the ventilation systemsis by 40
theatres, circulation, utility rooms,
absorptive material protected by plastic membrane and Operating day rooms, pharmacy, reception areas
perforated sheet to avoid the risk of fibrous particles
release. Sound-absorbing ceilings (cleanable) and cush- Kitchens, laundry, changing rooms, OT 45
ioned vinyl floor finishes will contribute some noise exercise areas, X-ray process areas, clean
control within wards. rooms
Hotels Utilityrooms, stores, cleaners' rooms 50
Hotels vary in their standards, most new-built projects
being 2- or 4-star. To serve guests, they will usuallybe near dBA levels approximate to NR + 6.
H
70 Acoustics in the Buift Environment
o o"o
determine issues like partywalls and the surface mass of
walls that can be taken through party floors (in flats). The
minimum sound insulation for party walls is 52 dB
(DflTW). The corresponding figure for party floors is :
.
51 dB, and for impact noises the maximum value is 62 dB
Lr, Some sample Approved Document Constructions
are illustrated in Figures 2.30 and 2.31.
p
Mean performance
54
dB
4 tl I
Plastered
brickwork 1. 18 mm T&G Floor boarding or 22 mm flooring grade chipboard
>415 kg/rn2 53 on 50 x 50 mm battens
2. 13 mm/36 kg/rn3mineralfibre quilt resilient layer
3. 75 mm reinforced concrete screed
4. Cored precast concrete units, at least220 kg/m2*
5. In-situ concreteon permanent forrnwork
6. Plastered soffit
7 liii! II Ill100
0 200 mm
amended to 300 kg/rn2
in June 1992
Plastered BuildingRegulations
lightweight
blockwork Figure 2.31 Selectedfloatingfloor constructkms. (Source:
>250 kg/m2' 52
ApprovedDocument ofBuildingRegulations [2]; [18]
Noise Act [22] deals with noisy neighbours but has no established at definedfrequencies.
direct design guidance. In setting noise hazard limits the following should be
considered:
Industrialbuildings
1. continuous noise sources (to avoid hearing loss),
The main issues are noise break-out (particularly for 24-h
2. impulsive noise sources (to avoid temporarythreshold
operating buildings like printingworks, bakeries and flour shift),
mills) and good conditions for workers within. The 3. perception of danger warnings, signals (fire alarms,
planningauthority may be expected to set a limiting value
at the boundaryof the industrial premises, particularly if machinery start-up),
4. adequacy of speech communication.
housing is involved. Noise break-out may occur via the
body of the building, if airborne noise levels are high For only modest local control (up to 9—1OdBA reduc-
inside and the building is constructed of lightweight tion) a noisy process, for example band resaw machines,
cladding. Break-out will also occur via atmosphere pro- can be screenedoff using flexible loaded PVC hung full-
cesses and ventilation plant connections, for example height, of around 5 kg/m2 surface density lined with
flues, smoke andprocess extracts, and via goods doorways 25-mm polyurethane. Solid panelling of 13-mm fibre-
(particularly roller shutters) and personnel doors. board with 25-mm bagged rock fibre lining instead can
The traffic flow to industrial buildings may itselfbe a lower the level at operatorposition around by 12 dBA. A
noise source problem, for example vehicles parked modular metal panel system can achieve a 25dBA
outside a dairy-produce factory with refrigerator plant reduction even with small openings for conveyorsto pass
running continuously. through, whilst full enclosure test cells can obtain a
Ancillary sources like signal klaxons, public address 30 dBAreduction. Modular GRG equivalents are not quite
sound leakage or externalPA,and occasional soundfrom as effective,with a reduction potentialof 20 dBA. Applica-
tests of emergency procedures — standby generators, tions are illustrated in the HMSO/Health and Safety
smoke shutters, valve releases — can add to the process and Executive publication, 100 Practical Applications of Noise
ventilation sources as regards noise break-out. Reduction Methods[231. Another useful reference is SRL's
Within industrial premises, the protection ofemployees Noise Control in Industry[24].
is afforded by legislation, in particularthe 'Noise at Work Often noise in industrial premises is a mixture of high
Regulations' of the Health and Safety Executive. The constantnoise (engines, compressors and conveyors)and
effectofthe updatingof this statuteon iJanuary 1990was high maximum noise events (hammering, grinding and
that three times as many employees became involved. stamping). Pneumatic power processes can have intense
Failure to comply can mean prosecution or even closure high-frequency noise components. Localized activity in a
for an employer. There is a general duty to reduce risk to large industrial interior forms noise 'hot spots'. If noisy
employees' hearing, on employers and designers, by processes are spread out, general treatments will help
reducingexposure to the lowest level reasonably practica- because workers will each experience noise as a mix of
ble. In known noisy places of work, noise assessments direct sound from other nearby activity and reverberant
should be made by a Competent Person, and records of sound, as well as near-fieldnoise from theirown efforts. If,
assessmentskept until new ones are made. Ear protection on the other hand, direct sound from very noisyprocesses
zones are designated areas where noise levels will trigger nearby dominate, the reduction of the reverberant com-
the 'Second Action Level' as defined by the regulations. ponent ofnoise will be of littlebenefit. Onlyboth a noise
The unit which applies is the Daily Personal Noise survey and an understanding of activitiescan throw light
Exposure Levelwhich relates the potentialfor damage to on this.
hearing to both level and duration. A room treatment of ceiling absorbers can be ofbenefit
as follows:
1 'A(t)l
1T,,
dt • sound decay away from industrial noise sources will
LEPd=lOlo_J be greater, perhaps becoming —5dB/doubling of
distance rather than —3 dB/doublingof distance;
where 7 = duration of exposure, 7
= 8h, PA(t) =
instantaneoussound pressure(Pa) varying with time, P0 = • reverberant levels will be lower and there will be a
20 X 10_6Pa. slight reduction of continuous noise, for example
The trade-off has an additional 3 dBA on noise level from extract fans, due to the added room
offset by a halving of duration, 85 dBA LEP d is the First absorption;
Action Leveland90dBALEP d is the Second Action Level. • reflected sound will be reduced;
Claims ofindustrial deafness hinge on causation, showing • the 'ringing' character of sound impacts will be
loss of hearing on the balance of probability is due to lessened.
noise at the place of work. The provision of protection An 'applied' treatment like hung absorbers can be
and regard for levels and duration plus a plaintiff's efficient (Figure 2.18) but a built-in inclusion of absorp-
exposure to noise prior to employment by the defendant tion is cost-effectively dual purpose.The roofsoffit can be
may show that there has not been a Breach of Duty. rendered absorptive rather than reflective by using a
Contributory Negligence by the plaintiffmay occur if he perforated profiled metal deck rather than a plain
has not worn ear protection when it was offered or has profiled metal deck, or a lining treatment that is inher-
chosen to ignore directiveson durationofnoise exposure. entlyabsorptive. The roof offers greater scope than walls
The standard method for testing hearing is Pure Tone because in a large factory, wallsurfaces will be ofrelatively
Audiometry, where the employee's hearing threshold is less surface area and perimeters may be remote to
72 Acoustics in the Built Environment
working areas. One problem with perforated soffit roof Thelecture theatre,unlikesaythe council chamber, has
decks is the tendency for a vapour barrier directly behind set locations for speakerand listener, so finishes can be
the perforations to some degree blank off the higher tailored to flatter the speaker. The front two-thirds of the
frequency absorption of the quilt behind the vapour ceiling should be sound reflective rather than absorptive,
barrier. This can be reducedby placing the vapourbarrier the rear absorbing; better still, reflecting ceiling panels
as an interlayer between absorption and thermal quilt can be optimally tilted to give strong early reflections via
layers. the ceiling, reinforcing the direct sound. Goodsightlines
are essential for adequate sound reception: a dais at the
Lecture and conference rooms front and tiered seating rows enable this.
Lecture rooms Reflectingsurfaces at the speaker endcan be 'hard' but
Lecture rooms, purpose-built for 50 up to 500 persons, are best modelled to avoid local cross reflections and resultant
a feature of many education and business centres. Up-to- flutter echoes which will be off-putting to the speakerand
date techniques have revolutionized forms of presenta- will lessen speech intelligibilityfor the audience. The rear
tion: video conferencing, simultaneous translation, back of the lecture room should be sound-absorbing to damp
projection, video recording (for traininguse) and CCTV, long sound path reflections from the back (Figure 2.32).
BARCO projection, satellite TV links, and computer- In considering whether overhead reflecting surfaces
generated images. usefully reinforce the direct sound, one can consider a
Section
Sound-absorptivefinishes to
rear part of ceiling, rear Half plan
wall and most of side walls 6. Projectionroom (202-seat room)
limiting ellipse within which surfaces can reflect back reflecting ceilings but edge absorption strips and absorp-
without echo, but outside of which surfaces should be tive wall linings to at least half the wall surfaces (Figure
either absorptive or reflect incident sound so it remains 2.33). PSA and CIBSE recommendations suggest for
outside the ellipse. A discrete second image of sound is intrusive/ventilation noise criteria, NR 25 for 'large'
discerned if the reflected sound energy arrives more than conference rooms (>50 persons), NR 30 for rooms
40 ms after the direct sound, i.e. if the distance travelled holdingmore than 20 persons, and NR35 for the smallest
approaches or exceeds 14m. The notional limiting ellipse rooms. Soundlobbies shouldbe plannedinto both larger
is thereforedefined byAB +AC< 14 BC+ 14 where Cis the lecture rooms and conference rooms.
source point, B the receiver location, andA the surface of
reflection. Librariesand museums
In a space with good room acoustics for speech (RT Library activity varies from busy popular fiction and
0.5—0.75s) the following guide applies: cassette loan areas to quiet reference areas, so the
• subdivision of the facility by bookstacks and exhibition
up to 15m relaxed listening
• 15—20m good intelligibility screening can allow this. Sound-absorbingcarpet,acoustic
ceilings and soft furnishings can help keep reverberant
• 21—25m satisfactory soundlevelslow. Ventilation noise shouldbe controlled to
• 30m limit of acceptability NR 30, intrusive noise from traffic to 45dB (LAeq,TL
Reading lip movements helps intelligibility,which is of Museums should be lively centres of activity and
assistance up to 15m. When the distance between the 'interactive'/participatory exhibits may have to compete
speaker and farthest listener exceeds 10—15m, a speech noisilywith repetitive video presentations. Careful zoning,
reinforcementsystem may be considered. sound absorption materials, and good-qualitydirectional
sound systems can help. 'Theme tours' are a new
Video conferencing derivative. The close arrangementof different 'sets' in a
Video conferencing is a newdevelopment entailing voice- tour can allow the effects to be spoiled if noise from one
activated cameras and microphone systems to connect area is distracting and intelligible in other areas.
specially-adaptedmeetingrooms. Ideally the rooms could
be designed to talks studio standards butmore frequently Musicpracticerooms
standardconference rooms are adapted. The wall behind The standardoffacilitiesvaries widelybetween the rooms
the seated participants should be fully treated with provided in state school music departments — little
absorptive facing. Intrusive outside and ventilation noise different to normal classrooms— and isolated,controlled
sources should be kept within NR 25. environmentsprovided for professional, and trainee
professional, musicians.The smallestpractice rooms hold
Conference rooms only two (instructorand soloist); slightly largerones take
Conference rooms vary from meeting rooms to large a small group. Large rehearsal spaces can hold sections of
auditoria where massed delegates can attend a conven- the orchestra. Surfaceswithin can usefully be sparse (RTs
tion. Ideally, office conference roomsshould have sound- typically 1 s at500 Hz for practice rooms, 1.5s for rehearsal
rooms), and non-parallel: offsetting alternate walls in a
row of rooms by 7° or more is adequate to prevent
distracting cross reflections. Low-frequencyabsorbers may
be useful to balance reverberation characteristics, and
velour curtains can allow some user-choice of playing
conditions. Ventilation and steady intrusive noise should
be controlled to within NR25 and in all but the quietest
settings, rooms should be fully mechanically ventilated.
Adequate cross-talk attenuationis essential.
The critical issue is the isolation afforded to the rooms.
School rooms tend to be a lower specification both
because ofcosts andthe desire of teachers to know pupils
are practising in individual rooms. Single acoustic doors
rather than using sound lobbies may have to suffice.
Figure 2.34 shows the commissioned results for a school,
Reflected The isolation from separated wall leaves is not fully
ceiling realized because of the flanking effects of roof and floor
plan
continuity. The more costly but effective isolating ceiling
and floor shown in Figures 2.35—2.37 show a worthwhile
1. Sound reflectivecentral ceiling to 'carry' gain in performance, again by carewith acoustic doorsin
speech across the conferencetable soundlobbies. Even with this degreeofseparation, music
2. Sound absorptive edges to ceiling
practice will be discernedin the adjacentpractice room.
3. Sound absorptivewall panelling/piriboard Care in
to at least 50% of wall surfaces workmanship and supervision is needed, as
4. Carpetfloor finish contractors find it hard to resist tying structures together
for stability during building.
Figure 2.33 Reflectingand absorbingsurfaces in a small An alternative approach, comparable with studio tech-
meetingroom nical facilities, is to use modulardry-construction 'boxes'
74 Acoustics in the Built Environment
a)
1)
a)
a)
>
a)
-J
C
0
riD
I I I
0 5m
Figure 2.34 Music practice rooms: !vlanchesterHigh Schoolfor Girls
Design acoustics 75
1. Subfloor: rc slab
2. Floating floor:
100-mm rc slab on
neoprene bearings/
50-mm air space
3. double ceiling:
metal lath + plaster
4. floor above: 200-mm
rc slab
5. wall: plastered 110-mm
brick leaves, 50-mm
quilt, 150-mm cavity
(total 540 kg/m2)
6. Wallhead mastic seal
111111 I
0 im
Offices
V Complaints from office workers arise from intrusive
0)
U outside noise, high noise levels within offices, and poor
C
0) insulation between cellular offices. BS 8233 recommends
LAeq,T values of 40—45 dB for private offices and office
0)
N
3
/N
frequency absorbers but are otherwise strongly sound
reflective. Combined with hard floor finishes and wall
claddings,aclattery,reverberantcharacterwill result unless
a proportion, say 25%, of the wall surface is clad in
a)
C-)
C
a) -/ — absorptive panelling. Internal modelling serves to diffuse
the sound and features like trees, banners, umbrellas,
a) kiosksand other furniture all serve to soak up sound and
0 reduce reverberant sound pressure level. Water features
a)
> canprovide usefulmasking sound (70dBAatclose range).
a)
-J There is little information on atrium acoustics.Apaper
C by de Ruiter[25] compares shopping centre, office and
0
(I)
hospital examples. In the UK, Gaughan of the Institute of
Environmental Engineering, South Bank Polytechnic,has
taken extensive measurements in London atria at the
Broadgate Centre (two, each 20m X 11 m X 18m) and
the SedgewickCentre (35m X 20m X 15m).
Broadgate'smetal, marble and glass finished interior
court, four floors high, has reverberation times of 3s at
125 Hz increasing to 9s at 500 Hz. The average absorption
coefficient is only 0.05.
20 Sedgewick'sseven-floors heightby contrast has amean a
63 125 250 500 1000 2000 4000 of 0.2 and diffuse, almost Sabine, characterof RT 3s at
rd Octave Band Frequency (Hz) 125 Hz, 500Hz and 1 kHz.Both centres haveambientnoise
levels around NC 50 due to ventilation plant and
sound level difference between continuous escalators operation. Three-dimensional
music practice rooms
sound propagation contours show greatly varying and
Figure 2.36 Music practice rooms: Birmingham School of uneven decay rates of sound from source position to
Music. (Courtesy ofBAP) position, due to multiple andcomplex reflection patterns.
Design acoustics 77
225 350
8 225!
7
9 9
T
5 5
2 Brass .3 3 Percussion
I
0
liii I
im
4. 100-mm r.c. slab on
Neoprene mounts (jack-up type)
8.
9.
Ground floor slab
Services void
11LI
a)
U
laser printers are typically 64 dBA at 1 m compared to
fly'
C
a) 83dBA for mechanical printers. It may be tempting to
a) consider the use of a sound conditioningsystem. This
0
a)
>
a,
-J
---
/ --
consists of concealed loudspeakers emitting masking
white noise. Care is required as the working sound level
for efficient use is narrow: too noisy and the sound is
objectionable, or at least draws undue attention to the
/
-o
C 7C-
0 — sources; too quiet and the system is ineffective.
/
C,)
60- Privacy
Privacy between work places is only in the order of
17—20dBA between open-plan screen-based work-stations
50 at 12m2/work-station, and this may be compared with
speech privacy needs as set out in Tables 2.3 and 2.4.
These reflect the subjective reactions of office workers,
125 200 315 500 800 1250 recordedin Table 2.10. Screens can be tested for sound
100 160 250 400 630 1000 absorption (BS 3638) [261 — a NRC of 0.6 to 0.8 being
desirable — and speech privacy noise isolation class (NIC).
rd Octave BandCentre Frequency (Hz) There is a hierarchy of privacy in offices as shown in
sound level difference between
music practice rooms Diagram 2.6. Surveys have indicated that the thought
interruptiondueto office noise can amountto significant
Figure 2.37 Music practice rooms: CentralLondon Music work 'downtime' for employees, and good privacy
College. (Courtesy ofBAP) arrangements in open-plan officescan increase productiv-
78 Acoustics in the Built Environment
Staff in an adjacent
Activity+ ventilationnoise
(dBA)
work-stationannoyed
normal speech (%)
l 10
35 65
8
40 40
45 25 C
0
47 16
55 4 Co
C 6
C
Co
Co
0
0
0 2 4 6 8 10
gap width (mm)
effectof small gap above an
office partition on Sound insulation
Computerrooms
Computer rooms have inherently high noise levels, the
acceptability of which will depend on the occupants. If
Diagram 2.6 Offices: zoning forprivacy and quiet staff work-stationsare within, a background level of NR
Design acoustics 79
45—50should be aimedfor; ifonly intermittentoccupancy values rather than In some live studios now under
is involved, for example in machine room areas, NR 60 construction as fast news centres, background activity is
may be acceptable. Measures to reducenoise in the room put on show so a standard more exacting than NR 20 is
airhandling units include lower air velocities, ducted not called for. In control rooms, NR 20 for sound dubbing
supply and return with silencers, and double-skin and control facilities will be adequate; the background
casings. levels in such rooms may be determined by equipment
cooling fans typically NR25+ rather than ventilation noise
Sports and leisure or intrusive noise.
Swimming pools are inherently noisy: hard surfaces for Largerstudios may double as theatre audience venues,
wear and hygiene and the reflective qualities of water for example for game shows. Ventilation noise control to
exacerbate the shouting and splashing. Ice halls have NR 25 will be adequate, provided that for lower lighting
similar problems. Good sound systems andbuilt-in sound and occupancy loads a lower ventilation noise value (NR
absorption by ceiling or banners will give more comfort- 15 or 20) could be provided for, say, recordingdrama.
able conditions to achieve Sports Council recommenda-
tions (1.8—3s mid-frequency values for empty halls). High-quality radio, recording, broadcasting, or live television
Swimmingpools and ice rinkshave 24—h operational plant facilities
of significant duty, so care is needed re environmental Local radio stations met IBA standards by complying with
noise break-out to the local community. Ten-pin bowling pass/fail tests. The standards are set out in the Engineer-
has made something of a come-back;high impulsivenoise ing Code of Practice for Independent Local Radio,
levels occur close to the racking machines during 'strikes' uprated as the IBA Specificationfor Studio Centres [27].
(typically peaking at 1102dBAclose to) andduring racking The document itself should be studied in detail but
by the machines beyond. Absorbent surfaces and enclo- covers:
sureofthe machine room will containthis, andabsorptive
ceiling finishes improve conditions near the lanes. Slab
• background noise
isolation under the racking machines and under the lanes • reverberation times
(to stop ball-rolling thunder) is required to prevent
• performance characteristics for line path, micro-
structureborne sound transmission. Indoor or outdoor phone path, and storage media
shooting facilities demand specialist advice. Acoustic criteria for transmitting equipment are also
specified. Typically double- or triple-leaf walls and room-
Televisionand radio facilities dedicated floors and 'lids' were coupled with quiet (NR
Video production 15) ventilation systems and 'dead' room acoustics (RT
Video production demands that the mainstudio shouldbe 0.16—0.3s). The European Broadcasting Union's Report
reasonably isolated to outside and to other facilities, for R22 [28] is an important reference for standards includ-
example by double-leafwalls, and a roofofreasonable SRI ing preferred volumes of rooms, proportions of spaces
(35+ dB average). Acoustically-rateddoorsets should be (avoid single integral ratios between length, width and
usedto the control room suite but not necessarilyto other height), monitoringpositions, reverberation times (over
rooms. Room acoustics aspects will not be exacting. The one-third octave bands 200—2500Hz, the average RT
additional cost to standard accommodation of basic should be 0.3 ± 0.1 s). At low frequency one-third octave
specification is in the order of 25%. Facilities like this are band 50Hz, the RT should not exceed 0.45s. The RT
sometimes built as a fit-out of modern industrial estate/ should not vary by more than 0.04s between adjacentone-
business park units or conversion of older buildings to third octave bands in the range 200—10000Hz, and
multi-tenancyunits. In the case of the latter, transmission ventilation noise should be less than NR 15 with no cyclic
throughthe floormay needto be checked. Thestudios are variation or pronouncedtones.
suitable for promotional orvideo films,centringon sets. Recording studios demandsimilar high standards. The
quality criteria here are dynamic range, distortion
Commercial television (attenuation, phase or non-linear: harmonic, intermodu-
Commercial television management inevitably took a lation or amplitude), noise (ambient, system), wow and
short term view on capital expenditurereturns because flutter (short-term speed fluctuations) and electronic
franchises were, until 1991, renewed on a 5-year cycle. cross-talk. Detailed advice is given in Borwick's Sound
Small firms producingprogrammes on the fringes of the RecordingPractice[29]. Recommended reverberation times
main networks may have standards no better than for vary from 0.3 to 0.4 s for small speech studios to 1—2s for
video production. Regional franchise holderswill have a classical music recording. Monitor rooms, separated by at
range of accommodation from property stores to ware- least 8/200/12 glazed windows, should be designed to
houses with larger sets inside, applying standards no 0.2—0.35s. The introduction of stereo recording and
better than for outside broadcast, to control rooms and broadcast imposed furtherdiscipline to studiotechniques.
controlled environment studios. Standards in commercial Conditions should not be 'dead'/semi-anechoic, as this
television production improved as the value of good- would be unpleasantto work in. The ear is better able to
quality music scores in drama became recognized. locate an image in a stereo fieldwith some reflections and
The construction standard will be higher than for there is the 'single pass' concept of monitoring: part of
smaller facilities. Nevertheless, because the programmes the room is left 'hard' behind the loudspeakers, with the
in production are recorded and not 'live', the very area behind the engineers treated with absorption lining.
occasional intrusionofa loud noise can be accepted. This Control room ambient noise should match the main
implies that design is determined by, say, LAb or LAeq studio conditions.
80 Acoustics in the Built Environment
The need to have reliably acceptable conditions at all rooms also works well for fitting out 'shell' interiorsand
times follows from 'live' broadcast, where a retake cannot consistsofmodularstudios madeofstandardwall,roofand
be relied upon if some intrusion occurs. British Broad- floor panels. These are made of double metal skins with
casting Corporation advicemaybe takenfromRose's Guide mineralquiltfilling to the cavity between skins. Theroom-
to AcousticPractice [30]. The technique of mix ofoffice and side metal skin can be perforated: this reduces the sound
technical facilities by installing factory-built rooms-in- insulation propertiesbut contributes most of the general
absorption within rooms. Low frequency absorber boxes
may also be required. Small announcer booths can be
installed on a subfloor, each with a standard dedicated
100 room air conditioning unit, either recirculation or con-
____________________________________100I—t-i—I-100 nectedto a headersupply and extractducts via attenuators
IJ
+ (Figure 2.39). The system is one-third the weight of the
+ H equivalent masonry structure, is quickto erect, and can be
reconfigured relativelyeasily.
BBC checks of background noise are against three
reference curves, which very approximately equate to
NR5, NR1O, and NR15:
2
octave band values/octave
63 125 250 500 1 k 2 k 4 k (Hz)
1. Radio light
entertainment
-r
— 1
170
-r —
i— studios
Other radio
41 31 24 18 13 9 6
2.
studios, control
1 100-mm modular floor units on isolating rails with cable voids rooms.
2 100-mm modular wall units: metal skins, inner face perforated Television
absorbentquilt fill studios 36 26 19 13 8 4 1
3 Dedicated room ventilation unit,
recirculatedair with some fresh air 3. Radio drama
intake studios 31 21 14 8 3 —1 —4
a)
a,
a,
80
70
60
-_—
—-- —
7 — -_
—
/ the curve at upper frequencies. Flow noise at secondary
attenuators or diffusers can be a means of selectively
adding back higher frequency noise.
Theafres
The term 'theatre' covers a wide range of auditoriafrom
community halls (naturally lit and ventilated, mukiuse
and small) to large national theatres for resident com-
>
-J
0
C 7 panies. Traditional theatres and opera houses have a
'hammerhead' plan shape, a proscenium opening
z
0 between the stage and audience, an orchestra pit, and
acoustically comprise a high, bare space (alternatively
U,
I
4
Figure 2.40 Overhead reflector panels in a 1000-seat theatre. (Courtesy ofBucks CC/BDP)
There have been several excellent new guides on theatre 2.41 and Table 2.11. The basis of the system is the
design including Buildings for the Performing Arts (Apple- sampling ofthe direct sound fieldat the stage, by an array
ton) [33] and Making Spacefor Theatre: British Architecture of microphones. Signalsare relayed to a single processing
and Theatre Since 1958 (Mulryne andShewring) [34]. The rack which, by equalization and signal delays, simulates
most stunningimages of theatres, if not technical detail, early reflections and modified reverberation. Broad-band
are in Steele's Theatre Builders [35]. Lottery-funded
schemes are much in evidence, with community arts and
multi-media uses as well as traditional drama. Table 2.11 AGS electroacoustics system as installed in the
Overhead ceiling panels can be sound-reflecting and Gordon Craig Theatre, Stevenage
angled to give useful early reflections to 'carry' sound to
rear seats (Figure 2.40). Finishes should be sound-
Setting'3 RT1 (s) Use
absorbing except near the stage, and perimeter surfaces
should be non-parallel.
Balconyfrontsmay need to be modelled or have sound 0 1.0 System off: natural hall acoustics
absorption applied, in order to avoid distracting back for speech and theatrical
reflection effects for performers on stage. Angled pro- performances
scenium wallscan be modelled to reflect sound to seating 1 1.3 Piano,jazz, ballet, musicals
by facetting, which can improve audibility.As with concert 2 1.5 Chambermusic, recitals, opera
halls, acoustic modelling can be useful. In providing a 3 1.8 Chamberand Baroque music
setting for other events, like opera or music concerts, a 4 2.0 Symphonies
means of adapting the acoustics to these uses can be 5 2.1 Symphonies
incorporated. Banners or similar devices do not help, as 6 2.6 Choral music
the space is already 'dead' acoustically. There are a 7 3.0 Organ (dedicated setting)
number of electroacoustic systems which are able to
increase the RT artificiallyand compensate on stage for "Commissioneduse of different settings.
the lack of local reflecting surfaces. An example that is bMeasured reverberation time at 500Hz; plotted on Figure
installed in an existing UK theatre is illustrated in Figure 2.41.
82 Acoustics in the Built Environment
OrganChambersfar
//
ChristieOrgan Id Soot—yore
Speakers for Early Reflections AUOITORIUM RESERRERATION —AIJDITDRILJM
mocntedoertically
II ssr-So2e
. bOLDOACE
REOETRERATION TOSTAGE
N J.
"1\ t
Microphones fined to
55cr -\ Lighting Bridge
scit Prolectian Room
0 MOOr—MOte
MICROPHONES PLYRAR ra, STAGE
9
acditoricm ceiling.
O Mctg—Mo2a
ORGANMICROPHONES IN CEILING
s,i
ns'c
scm cnn
H RemoteControl
I Location
Mi
0 . 7 cern
Microphoneslocated
ecYc. 55
/
rnt
along flybar ri--f —- -i scar /
ccoi
Mains ELCB
and breaker. ¶oc: siccc3, ¶sa ccsc ssos /
AcocstinControl Systems SpeakersforEarly Reflections
Processing gack mocstnd oerrically
(a)
2.8
z
2.6
C)
0)
2.4 >
C)
C) C)
E U,
I- C
C
C
C -C
C
CO C
C)
C
-o 1.2
C)
S0)
> C)
to
C)
0.8
0.6
0.4
0.2 .2 .5 1 2 5 10 20
I I I
Communication distance )m)
125 250 500 1000 2000 4000
(b) Frequency (Hz) Figure 2.42 Effect ofbackground noise on speech
intelligibility
oACS1 +2 03 a4 es 06
Figure 2.41(a) and (b) AGS: Adaptable reverberationby shell, and ability to balance early andlate sound as well as
electroacoustics. (Source:Shuttlesound)
increasing lateral sound. There may be some overlap with
the house sound system.
Such systemsare in a fast stateof development andcare
loudspeakers around the hall transmit the modified in commissioning is required so that any artificial acous-
signals to supplement the natural sound decay in the tics are not unrealistically 'special effect' and hence
auditorium. The systemclaims to 'reshape' ahall as well as unconvincing to professional musicians.
just increase reverberation, giving early reflections to an For good speech intelligibility, the seating should be
orchestra to make up for the absence of an orchestral grouped as closely as possible to the stage and good
DesigN acoustics 83
25. de Ruiter, E. Ph. J., Atria in Shopping Centres, Office 30. Rose, K. Guide to Acoustic Practice, 2nd edn, BBC
Buildings and HospitaLs, IOA proceedings, 10(8), Engineering, Oxford, 1990
1988 31. Talaske, R. H. (ed.) Theatres for Drama Performance,
26. BS 3638: 1987 Method for measurement of sound Acoustical Societyof America, London, 1985
absorption in a reverberantroom, British Standards 32. Forsyth, M. Auditoria: Designing for the Perfanning Arts,
Institution, Milton Keynes American Institute of Physics, NewYork, 1987
27. Specification for Studio Centres, Engineering Code of 33. Appleton, I. Buildings for the Performing Arts: A Design
practicefor Independent Local Radio- Issue 2, Independ- and Derelapment Guide, Butterworth Architecture,
ent Broadcasting Authority, London, February 1996
1988 34. Mulryne, R. andShewring,M. MakingSpacefor Theatre:
28. Acoustical Properties of Control Rooms and Listening British Architecture and Theatre Since 1958, Mulryne &
Rooms far the Assessment of Broadcast Programmes, Shewring Ltd, 1995
ReportR22, European Broadcasting Union, London, 35. Steele, J. Theatre Builders: A CollaborativeArt, Academy
1985 Editions, September1996
29. Borwick,J. SoundRecordingPractice,OxfordUniversity 36. Binney, M. and Runciman, R. Glyndebourne— Building
Press, Oxford, 1987 a Vision, Thames and Hudson, 1994
Settingdesign objectives
Chapter 3 Services noise The areasofdesign forwhich noise criteria need to be set
are as follows:
and vibration • Central plant. 'Plant rooms' are often split (air-
handling units in roof-level housing; boilers and
pumps in ground level or basement rooms). Chiller
Peter Sacre and Duncan Templeton condenser units may be in a louvre-screened com-
pound open to atmosphere.
• Local plant. Decentralized ventilation systems can
have advantagesoflower cost (no long duct runs) and
Background flexibility (zoned units).However,plantnoise sources
The control ofnoise from mechanical and electrical plant are taken out of the sound-insulating plant room and
can be a vital area of design, as failure to meet criteria is into userspacesandso careis neededin their siting.
more readily perceived than, say, room acoustics criteria. There are already a number of guides on the noise
The designerneeds to be aware ofthe need to limitnoise control of building services. For a number of years, the
inside and outside the building: noise control products trade has concentrated on noise
• in occupied internal areas, where noise can be from fans in central ventilation or air-conditioning plant
transmitted through air distribution ductwork. Specialist
irritating or distracting, or can affect working acoustic suppliers via technical sales personnel can select
• efficiency;
in industrial premises, where processes rather than and supply the appropriatepackage ductwork and asso-
ciated attenuators. As a result, there should be relatively
rooms are serviced;
• occupied
in the areas immediately surroundingthe building, few noise problems in buildings due to inadequate fan
noise silencing through ductwork. However, many prob-
which may be used for circulation or leisure, where
excessive noise can be intrusive and may present an lems exist due to poor ductwork layout or high velocities
environmentally unsatisfactory character; causing regeneratednoise.
• beyond the site boundary — excessive noise from For the typical commercial or public sector building,
the following potential noise sources and transmission
plant may cause nuisance, leadingto complaints and
legal proceedings, especiallyin residential areas. paths may need to be considered:
The importance of noise depends mainly on two factors: • Internal noise
Central
—
air-handling plant
• the typeofbuilding whetherits use dependson low
—
Fan noise to ducts
• noise levels, and
the location particularly the proximity of other
—
Airflow-generated noise in ductwork, at duct
fittings or dampers
noise-sensitiveareas beyond the boundary. Noise break-out through duct walls
Noise control should be an integral part of the design Noise generated at grilles and diffusers
procedure. Too often noise aspects are introducedinto a
— Local
air-conditioningplant and room units
Fan-coil units
design too late, and in an ad hoc way. The role of any Volume-controlterminalunits
advice from an acoustics consultant or silencing specialist
should be proactive rather than reactive. The cost is many Heat pumps
times greater for a retrofit compared to the original Local extract fans
inclusion of adequate noise control measures. Fan convectors
The key participants who can influence services noise Warm air curtains
are:
—
Piped services
Pump noise or flow noise radiated from pipe-
• the client: including noise criteria in the brief, from work or from building surfaces to which pipe-
site to room data sheets work is fixed
• the layout
architect: provision of adequate structure, sen- Water hammer
sible location and area allocation of plant rooms, Flow noise from drains, particularly WC, soil
planningin distribution routes pipes and rainwater pipes from roofs
• the mechanical engineer: total design of HVAC — Electrical
equipment
systems, setting criteria Emergency generators
• the electrical engineer: design of substations, emer- Uninterruptiblepower supplies, generator sets
gency generators, lifts Transformers
• the mechanical and electrical engineeringsubcon- Thyristor speed controllers and light dimmers,
tractors: detailed selections of components and fluorescent lamp ballasts
installation — Airborne or structurebornenoise transfer
through
• the acoustics consultant or engineer: design advice plant room envelope to adjacentareas
from briefing to commissioning — Other sources
Figure 3.1 Noise havens: printing hail aSeealsoTable 2.9 with regard to health care facilities criteria.
88 Acoustics in the Built Environment
OBCF (Hz)
15 NR 66 47 35 26 19 15 12 9 7
NC — 47 36 29 22 17 14 12 11
PNC 58 43 35 28 21 15 10 8 8
RC — — 35 30 25 20 15 10 —
20 NR 69 51 39 31 24 20 17 14 13
NC — 51 40 33 26 22 19 17 16
PNC 59 46 39 32 26 20 15 13 13
RC — — 40 35 30 25 20 15 —
25 NR 72 55 44 35 29 25 22 20 18
NC — 54 44 37 31 27 24 22 21
PNC 60 49 43 37 31 25 20 18 18
RC — — 45 40 35 30 25 20 —
30 NR 76 59 48 40 34 30 27 25 23
NC — 57 48 41 35 31 29 28 27
PNC 61 52 46 41 35 30 25 23 23
RC — 55 50 45 40 35 30 25 —
35 NR 79 63 52 45 39 35 32 30 28
NC — 60 52 45 40 36 34 33 32
PNC 62 55 50 45 40 35 30 28 28
RC — 60 55 50 45 40 35 30 —
40 NR 83 67 57 49 44 40 37 35 33
NC — 64 56 50 45 41 39 38 37
PNC 64 59 54 50 45 40 36 33 33
RC — 65 60 55 50 45 40 35 —
45 NR 86 71 61 54 49 45 42 40 38
NC — 67 60 54 49 46 44 43 42
PNC 67 63 58 54 50 45 41 38 38
RC — 70 65 60 55 50 45 40 —
system because ambientnoise will tend to dominate, and madeto Guide toAcousticPractice [1], EBUReportNo. R22,
the client will not be gettingvalue for money in an over- AcousticalProperties of Control Roomsand Listening Roomsfor
silenced installation. If the levels are too high, the system the Assessment of Broadcast Programmes[21, and the IBA's
itself is obtrusive. Around or slightlybelow (0 to —5dBA) Specificationfor Studio Centres[31.
averaged activitynoise is usually found to be acceptable;
ventilation noise can have a useful masking effectin room-
Externalnoise
Environmental noise control is generally covered in
to-room speech privacy (as indicated in Table 2.3 in
Chapter 2). In order not to add noise from different
Chapter 1, under Industrial Noise. Emission limits from
fixed plant andprocesses may be written in as a planning
services sources, 'local' ventilation noise from room
condition, or should be established early in the design
supply and extract grilles should exceed by at least5 dBA
process to avoid the local environmental health officer
noise break-out, or re-radiated noise, from primary plant.
agreeing with complainants thatthe plantnoise is a public
In considering a criterion, the 'steady state' ventilation
noise to be introduced has to be considered relative to
nuisance to neighbouring properties and is therefore
both background noise from traffic noise break-in and
actionable under statutes.
equipment noise like computer fans, andvarying ambientDifferent local authorities take different attitudes.
noise from occupants' activity. Some, not wanting prevailing neighbourhood noise levels
to 'creep' up due to the addition to newexisting sound,
Broadcasting authorities will ask for existing levels not to be raised. This is very
The BBC, IBA and EBU (European BroadcastingUnion) onerous, as to reliably enable this, the new sources will
all have recommended criteria. Reference should be have to be 10dBA less than existing levels. It is normally
Services noise and vibration 89
1.
0)
E
C
0
a)
a)
0
0
0.01
0.00W
Figure 3.2 Building vibration x- and y-axis curvesfor (a) peak accelerationand (b) peak velocity. Extractsfrom BS 6472 are
reproducedwith the permission ofBSI. Complete copies ofthe standard can be obtained ly post from BSI Publications, Linford
Wood, Milton Keynes, MK14 6LE
0
a)
m
a)
>
a)
a)
0
a
0
C
0
C),
Typically, turbulent flow is random and causes broad- has a useful initial guide (Table3.4).Axial and centrifugal
band noise across the audible frequency range. There are fans produce similar sound power, with axial fans having
pure tones at the blade rotationfrequency and its higher higher high-frequencyvalues (Figures 3.4 and 3.5).
harmonics, as an overlay to this broadband noise.
Fan noise data of concern are the octave band sound
Air-handling plant
power levels ofnoise via the intake, exhaust terminations, Air-handling plant of modern design consists of the fan
and as radiated via the fan casing and external motor. unit itself (which may be centrifugal or axial), flexible
Reference bodieswhich may be quoted in a performance
connections, casing and chassis, filter, coils, mixing boxes,
specification include CIBSE [8] in the UK and ASHRAE and possiblyintegralattenuators and dampers.
in the US. Pending selection or as a cross check of
manufacturers' claims, an estimate can be made from
empirical formulae andtypical spectra (Figure 3.3) based Cooling tower/condeizser units
on the duty of the fan. Noise arises from the fan, fan motor assembly, and water
An empirical formula for sound power level at inlet or turbulencedown to sumps. Regenerated noise may occur
outlet is: via water circulation pipes. An indicative sound power
level is given by:
I=40+10logV+2Ologh
L.N = 11.5 + 10 log P
where Vis delivered volume (m3/s), his fan static pressure
(N/rn2). SRL's book, Noise Control in BuildingServices[9], where P is the total rated fan power output (in watts).
OBCF' (Hz)
Refrigeration units
Fridge plant compressors may be annoying by virtue of
intermittency of operation.Typical spectra are shown in
Figure 3.7.
Airflow
Boilers
Sound pressure levels are fairly similar for different types
offuel and representative octave band SPLs are given in
Figure 3.8. The principal noise sources are the fuel burner
units and combustion air fans. Noise will also be dis-
charged up the flues and is predominantly of a low
Figure 3.4 Axialfan frequency character. Prediction formulae are given in
CIBSE/ASHRAEbut caution is necessary in view of the
large number of variables (flue height, directivity, cross-
sectional area, linings, etc.) inherent in empirical
formulae.
Generators
In many public buildings, emergency powerfor light and
Airflow
safety procedures is provided by battery sets. Some
11 industrial premises, hospitals, broadcasting centres, and
officeswith vital constantpower needsforcomputers, etc.,
will include emergency power generators, frequently in
the form of diesel engines. Noise comes from:
• engine itself,
• the
Airflow
'1 • exhaust,
• air intake,
cooling fan,
• ventilation openings t&engine enclosure.
It is misleading to measure engine sets when run-up in
routine tests because full load cannot be applied. Noise
control measures shouldbe applied as a kit:
• enclosure, with controlled ventilation supply and
Figure 3.5 Centrifugal fan extract openings by attenuators,
-o
>
a
0
0
Figure 3.6 a
Typicalfrequencyspectrum ofair-cooled condensersat distance of3m based on 300kW model
Services noise and vibration 93
a,
>
a,
a,
J
a,
C
0
U)
Octaveband centrefrequency. Hz
Figure 3.7 a
Typicalfrequencyspectrumofreciprocatingchillersat distance of 1 m based on 600kW model
95 -
90-
a,
>
85-
4'
4,
a,
C
0
C',
8. Design criterion NR 35 52
9 Attenuation required 13
To less noise-sensitiveareas
Vaned bends 1.5 m3/s
L 12m
OBGF (Hz)
500 5 7 10 15 23 17 13 11
1000 8 11 19 31 48 37 28 21
1500 10 16 27 45 50 50 39 31
"The attenuatorperformancedependsnot only on its length but also on the ratio of airway-to-splitter width (refer to Figure
3.12). These insertion loss figures are for an attenuatorunit with an approximate ratio of 1:2.5 (airway:splitter).
Regeneratednoise
The basic layout of the ductwork and the air velocity Table 3.8 Cross-talk attenuation
within it influences noise levels most. The optimal
placement of attenuators and other in-line duct items is
critical. Recommended maximum duct velocities are as Requirement in Attenuator length Noise reduction
shown in Table 3.7, for low velocity systems. receiver room (mm) at 500 Hz (dB)
The flow rate of air in a duct can be checked by a
calibrated inlet device or by static suction in the early part NR 40 750 25
of the system. NR35 1000 30
Regeneratedduct noise can be created by: NR3O 1250 35
•• bends
transition pieces, NR25 1500 40
(turningvanes alleviate noise),
Servicesnoiseand vibration 97
Determine Lw at grille
in source room taking into
account source room
losses
A = area of grille
INo
The disadvantagesare:
casing. A layer of 12kg/rn2 lead on mineral wool will
increase the sound reduction of the ductwork by 5 dB/ • mixed responsibilities of main contractor and duct-
octave. work supplier,
Duct shape influences in-duct noise and duct noise
break-out characteristics. Circular ducts are more rigid
• difficultiesend
in avoiding pressure drop-off from a take
off at one to one at the other,
andofminimum perimeterfora particularcross-sectional
area thus reducingnoise transmitted into rooms or ceiling
• good workmanship is required to ensure an airtight
chamber.
void. Hence circular ductwork is often preferred in
The builder'swork details at ductwork penetrationsof
exposed system installations within spaces. Rectangular wall need careful attenuation (Figure 3.10).
ducts have less rigid wallsandthe flat metal is more easily
excited, and although it may provide low-frequencyin- Riser ducts are a feature of the distribution from plant
duct attenuation, it allows more noise break-out at low rooms in multi-storey buildings. They can be either
frequencies. This can lead to 'drumming' heard
within masonry or dry construction. Access doors to high velocity
the room through which the duct passes. riser ductwork should be acousticallyrated.
Builder's work ducts can be used for low velocity systems Transfergrillesare frequently usedto save ductwork runs
in, for example, auditoria. These are long plenum in ventilating adjoining small rooms, but negate acoustic
chambers formed in airtight masonry or plasterboard separation. They may be used only for acoustically non-
critical partitions or doors.
(Figure 3.9). The advantages are:
• lower cost than equivalent very large-scale metal Large public spaces
Large public spacespresent a conflict between large-scale
• ducts,
easier installation, air distribution and noise generation, particularly in large
noise-sensitive volumes like auditoria. High occupancy
• efficientuse of building's space.
98 Acoustics in the Built Environment
Plant rooms
1. Lintels over
Plant in older buildings was placed on the basis of 2. Mineral wool packing to
'boilers in the basement, tanks on the roof. Modern keep cavity clear
plant rooms house not only boilers and pumps but also 3. Dense mineral wool
slabs to all sides of
air-handling units, lift motors, compressors, and open-to- duct
atmosphere chiller plant. Plant-room noise levels are 4. Mortar pointing
typically in the range NR 70—85 for boiler rooms, NR between slabs and
60—75 for air-handling plant rooms. Masonry structures brickwork reveals
— concrete floors, concrete, brick or blockwork walls — 5. Metal flanges bedded
in non-setting mastic
are essential, with metal acoustically-rated access doors.
An SRI of 50dB (100—3150Hz) is a minimum require- Figure 3.10(a) Builder's work penetrations: duct through
ment for walls and floors. Additional airborne attenua- wall
tion through the floor can be achieved by the introduc-
1. Oversize stubmetaltube
tion of a floating floor. 2. Preformed dense mineral
The roof structuresound insulation will need consider- wool
ing in the case of roof-mounted freestanding air-handling 3. Pipe throughwall, not
units often used on commercial, retail or multiplex mechanically fixed at
cinemas projects. the wall
4. Hole filled with mortar
Absorptionin plant rooms may reducereverberant sound
pressure levels by about 5 dB but it is usually more cost
effective to have noise control at source or increase the
sound insulation of the plant-room structure. If possible,
expensive shrouds to units should be avoided as after
initial maintenance there is a tendency to leave enclosure
panels loose or detached altogether.
Plantroom structure
Once the location of plant or plant rooms is fixed,
consideration needs to be given to providing an adequate
plant room structure. In addition to determining the
appropriatemain construction which is typically masonry,
any openings and penetrations by ductwork or pipework
have to be carefully designed.
Acousticdoorsets may have to be specified. Metal doorsets
are capable of achievinga highersoundinsulation than a
timber type. In specif'ingacoustic doorsets, care must be
taken in selecting the appropriate performance from
manufacturers' data. Figure 3.10(b) Builder's work penetrations: pipe through
Ventilationopeningswill normally need to be acoustically wall
controlled by attenuatorunits or acoustic louvres.
Service penetrations will need to be effectively sealed.
Suitable details are given in Figure 3.10. or openings into fully enclosed plant rooms — air inletsor
exhausts, and plant room naturalventilation.
Externalplant Condensers are often roof mounted and therefore do
Noise to the outside can be from plantwhich by its nature not benefit from ground attenuation or natural screen-
needs to be open to the atmosphere, screened areas or ings by proximity to walls. Suchunits often have to run at
freestanding plant rooms holdinggenerators or chillers, night and so have to be considered relative to low
100 Acoustics in the Built Environment
Advantages Disadvantages
• Lower cost • fights natural convection
• underseast noise break-in
weakness
•• difficult to avoid draughts
throw (and hence noise pro-
duced) too great for concert
halls
Cladding
• equivalent;
splitters should ideally be constructedwith builnose
fairings to both entry and exit, thus reducing the
resistance to airflow;
Acousticlouvres • duct connections need to match the duct flanges or
spigots.
Screens/partial enclosures In some instances there are requirements to use plastic
ductwork or provide special lining materials.
Diagram 3.2 Requirements
— nouecontrol
from building
services
Acoustic louvres
Acoustic louvres are slatted blades angled to keep out
Acoustic performance a = airway width rain and block line of sight through, but have an
depends on length, I of unit b = splitter width
and ratio a/b additional feature over conventional louvres: the under-
sides of the blades are lined with absorptive material
behind perforated mesh. The arrangement allows
attenuation from one side of the louvre bank to the
other, whilstmaintaining a relatively high free area value.
The deeper the louvre bank, the more effective it is, but
it would provide a high pressure loss as a penalty.
Compared to in-duct silencers, the insertion loss per-
formance is poor at about 12 dB at 500 Hz for a 300-mm-
casing to ductwork deep louvre (Figure 3.13).
specification Besides controlling air intakes or exhausts direct into
Splitters: acoustic plant rooms, acoustic louvres can also be found as
infill retained by metal channel atmosphere terminations to plenum chambers and as
and with bull-nosed ends or
fairings to improve air-flow
open screens around chiller plant on roofs or around
characteristics Optional acoustic
transformer plant in the open. As screens, the limited
pod provides through-the-vanes performance is not as important, as it
Lengthof unit relative greater acoustic only has to prevent more sound goingthrough the screen
to diameter of performancebut
connecting greaterpressure
than over the top by diffraction.
ductwork drop
Cladding
Cladding, or lagging, is a measure to be considered to
either damppipes or ducts which are excitedby air or gas
turbulence within, or to uprate the thin duct or pipe wall
to prevent noise break-out from high airbornenoise levels
Circular ductwork internally within. The former can re-radiate noise to user areas that
lined with acoustic infill
they run through either direct via the duct or pipe walls,
orviavibration effects arising from transmitted excitation
Figure 3.12 Acousticattenuatorconstruction to casings or suspension to local structure. A light
102 Acoustics in the Built Environment
600 mm
300mm Acoustic curtains
Acoustic curtains, in the form of limp blankets hung
from and draped over a framework, can restrict noise
break-out from noisy machine areas within an industrial
Screens/partial enclosures
To control noise from externally located plant, it is
sometimes necessary to shield the plant. This is achieved
by a solid barrier ofrelativelylightweight material, it only
having to preventmore sound going through the barrier
screen than over the top by diffraction, or in some instances it
may be an acoustic louvre as described earlier.
Provision ofa screen should allow adequateventilation
of mechanical plant.
metal
DOUBLEBANK SINGLE BANK
Section
Screen designed
to maintain shielding
Plan
Screen
Plant
Section
Plant
Screen design
to maintain shielding
Plan
Table 3.9 Staticdeflection: guidancefor resilience to provide vibration isolation and avoid resonances. (Source: A. T Fry)
'The floor span refers to the largest dimension between supporting columns.
The equipmentis assumed to be at mid-span.
In selecting vibration isolation for plant, account must There are normally three options, simplyshown in Figure
be taken not only of the lowest forcingfrequency of the 3.15:
machine (typically its running speed) but also the
resonant frequency of the loaded supporting structure, • provide antivibration mounts directly between the
for example floor slab. equipment and the supporting structure. This
An example of the typical selection charts used to includes isolators as an integral part of a packaged
determinethe preferredstatic deflection is given in Table unit,e.g. isolation ofthe fan from the casing ofan air-
3.9 for different types of plant and supporting floor handlingunit, (a);
spans.
Once the static deflection has been determined, the
• support the equipment on a steel base frame and
then provide antivibration mounts between the
method of mounting the equipment has to be assessed. frame and the main supportingstructure, (b);
Services noise and vibration 105
Base frame
improves stability
Rangeofstatic
Type Material deflection (mm)
Pumps, compressors —
rubber/neoprenein shear
damped metal springs
—
• mount the equipment on a concrete inertia block — inertia slabs+ mounts
which is isolated from the main supporting struc-
tures, (c). Air-handling units —
(direct fix if integral AVs)
—
rubber/neoprenein shear
Unless the equipment is rigid and not liable to
—
undamped metal springs
distortion if supportedby its existing feet individuallyon Extract fans — resilient
hangers + flexible
isolators, then either a steel base frame or concrete connection
inertia block should be used. Both the frame and the — solid
block addrigidity to the equipment andcan be designed Standby diesels layerpads
to a size, greater than the base size of the equipment,to
—
rubber/neoprenein shear
damped metal springs
—
improve the stability and lower the centre of gravity of a
system. The provision of the concrete inertia block will Transformers —
rubber/neoprenecomposites
increase the mass of the system, giving a more even in compression pads
weight distribution (where equipment alone could be
106 Acoustics in the Built Environment
significantly heavier at one end than the other), will In ensuring that the most efficient form of vibration
minimize the vibration effects of changing equipment isolation is selected, the mechanical engineerin formulat-
speeds or loads, will reduce problems likely to occur due ing his specification should include the following:
to coupled vibration modes, and can act as a local • advice from an acoustics engineer;
acoustic barrier.
Once the preferred mounting method has been se-
• an appropriateschedule detailing the equipment to
be isolated and its method of isolation;
lected, the type and location of vibration isolators or • all isolators to provide the required minimum static
antivibration mounts can be determined. The types of
vibration isolation commonly available are as shown in deflection given in the scheduleunder the imposed
Tables 3.10 and 3.11. load of the equipment, the selection of isolators
In areas where high isolation, i.e. high static deflection, taking account of eccentric load distribution.
is required, such as in the semiconductor manufacturing To enable the correct selection of vibration isolation,
industry, air mounts are likely to be required. These the following details are required:
consist of an air-filled bag which can be coupled to a • type of equipment,
levelling control device enabling equipment to be main- • weight of equipment,
tained at the correct level regardless of load.
The selection of isolators should be undertakenby an •• centre of gravity location,
number and position of mountingpoints,
acoustics engineeror the vibration isolator supplier. They •
will need to determine the preferred location of the operating speed and nature of the operating
mechanism,
isolators,taking into account the loads imposed at support • details of the supportingstructural floor, particularly
positions.
In describing vibration isolation, a simple system is spans.
normally used taking into account the vertical direction
only. To minimize the risk of the selected isolation being Isolator types
reduced by the effect of other modes of vibration, it All isolators offered to the contract must meet the
should be ensured that the horizontal stiffness of the specifications and must be suitable for the loads, and
isolation is similar to the vertical stiffness, and the centre operating and environmental conditions which prevail.
of gravity of the system is as low as possible. Mats and padsshould be manufactured from synthetic
In additionto careful selection ofthevibration isolation rubber, neoprene, or glass-fibre coated with neoprene.
for equipment, itwillbe necessary to ensure that there will The materials anddesign should render them impervious
be no flanking/shortcircuiting/bridging of the isolation. to contamination from oils and attacking chemicals and
Flanking could occur due to service connections, poor be rot andvermin proof.
installation, or the impact of ancillary operations, for Turret compression mounts should be fabricated from
example infill 'builder'swork'. synthetic rubber or neoprene between two steel plates.
Flexibility for pipe, duct and other connections is The materials must be oil and corrosion resistant with the
essential to avoid unacceptable stresses in stiff pipe and metal protected from corrosion by painting. Friction
other connections. This can be provided by the use of surfaces must be provided to the bottom and the top. Bolt
flexible connectors in the pipe or duct systemclose to the holes must be provided to allow fixing.
equipment. Spring compression mounts should comprise a high-
In high-pressurepipe systems, allowancesmust be made strength, low-stress, laterally-stable, open-steel spring
for the force tending to stretch or compress the flexible locatedby a steel pressure plate on the top and bottom.
hose, and restrainedconnectors would normally be used. The bottom plate should include a bonded ribbed
In this case, the connection will be stiffened and suitable neoprene pad to the underside of minimum thickness
pipe layouts must be used to improve the situation. In 6mm, and be pre-drilled for bolting down. Each isolator
addition, reliance is placedon the naturalflexibilityofthe should be identifiable by a colour-code mark and pro-
pipework itself and this will be increased by changes of vided with a levelling facility with final lock nuts. The
direction. However,until the vibration has been reduced spring element should have an overload capacity of 50%
effectively, the pipework needs to be isolated from the with an outside diameterof at least80% of the operating
supportingstructure, typically for a distance of 50 pipe height. The horizontal stiffnessshouldbe not less than the
diameters from the equipment in normal circumstances. vertical spring stiffness.
This isolation is achieved by hangers or clamps, having a Captive spring mounts are as above but the steel spring
similar performance to the isolators under the equip- shouldbe encased in a neoprene-covered bodyto achieve
ment. Any other rigid connections should be flexibly horizontal andvertical snubbing.
joined to the isolated system, e.g. electrical conduits. Care Vertically-restrained spring mounts are as for the spring
will also need to be taken during installation to ensure compression mounts but the steel spring should be
that bridging of the isolator does not occur, e.g. via mounted within a hanger box constructedfrom steel of
hangers or by fixing bolts of isolator touching the minimum thickness 1.6mm. The box shouldbe vertically
equipmentbase. restrained by noise-isolated bolts. In addition, any hor-
On site,other trades may influence the performance of izontal buffers or snubbers shouldbe manufactured from
an apparently isolated system. Examples of problems are synthetic rubber or neoprene.Duringnormal operation,
concrete being spread such that it covers isolators or the snubbers should be out of contact.
debris building up under isolated equipment bases, Hangers are turretor spring compression mounts (or a
particularly inertia blocks. combination of the two) to the specifications above
Servicesnoise and vibration 107
incorporatedwithin a hangerbox. The hangerbox should Airflow testing facilities need a silenced air supply to
be constructedfrom steel of 1.6 mm thickness complete ensure that tests reflect the device performance — the
with a hole for the suspension rod andan enlarged lower device may be an attenuator, diffuser, grille, louvre
hole for the drop-rod to equipment. The lower hanger terminal unit, damper or ventilator.
rod should be allowed to move laterally at least 15° Reverberation roomshave bare walls, floor and ceiling so
without any contactwith the hanger box. that there is a diffuse and even distribution of sound
Steel baseframes should be purpose-built using a welded energy around the room.
steelframedwith attachmentpointsfor suitable vibration Anechoic chambers (Figure 2.17) have deep wedges of
isolation. The frame shouldprovide adequate support of absorption on all surfaces to create a 'dead'acoustic.They
the equipment without flexing or significant deflection. give 'free-field' conditions so the direction of sound
Concrete inertia bases should be purpose-built, using energy from a test item can be investigated.
welded steel frame formwork containing reinforced con- Transmissionsuitesare a pair ofreverberant rooms to test
crete 35 mm above the bottom of the base. The base the sound reduction index to BS 2750 Part 1 [13], of
shouldbe designed with sufficient strengthandrigidity to doors, windowsor equipment like cross-talkattenuators or
support the equipment and compensate for dynamic acoustic louvres. An area of 10 m2 is a minimum surface
reactions due to operation of the equipment.The size of area of aperture. Floor and ceiling transmission can be
the base should be sufficient to give support for all tested in a vertical suite of rooms.
integral parts of the equipment including inlet and Test methods need to be standardized to reference
discharge manifolds. The design of the base should bodies' test codes, either:
provide a minimum clearance of 18mm between the
undersideofthe base and the structural floor below, with • British Standards Institute, BSI
the installed mountings operating at their design static • InternationalOrganisation for Standardisation, ISO
deflection and under full plant rating. • European Heating and Ventilation Society,
Pipeworkflexible couplingsprovide flexible hosecouplings Eurovent
for connecting pipework, comprising nylon fabric or a •American Society of Heating, Refrigeration and Air
steel mesh carcasswith a waterproof cover. The design of Conditioning Engineers, ASHRAE
the couplingshould takeinto accountthe conveyed fluid
temperatures andpressures imposed. A schedule of relevant codes is given in Chapter 5. A
Ductworkflexible connectionsprovide flexible coupling of summary for services noise components is as follows:
fans/air-handling units and connecting ductwork
designed to provide a minimum operating length of •attenuators —
BS4718[14]
100 mm. •impact sound — BS 2750,
In addition to vibration isolation of the main items of BS 5821 [15]
plant by these methods, there are likely to be situations •sound absorption coefficient —
BS3638[16]
where small plant items need to be isolated from the •airborne sound reduction index — BS 2750
structure to prevent structureborne noise. This can •power-generated SWL
normally be achieved or
by neoprene synthetic rubber measurements —
BS4718,
inserts. BS 4773 [17]
BS 4856 [18]
BS 4857 [19]
BS 4979 [20]
Installation •in-duct plates or transformations — BS 1042 [21]
•fan testing — BS 848 [22]
Laboratory tests
An awareness of laboratory tests in services noise control
is importantin order to understandthe limitations ofthe
Installation: waittests
data provided by manufacturers. The concerns are:
Reasonsfor tests
• adequacy of the facilities and the reputation of the On a large project, there may be many room units ofone
testing house, type, so it is important to know the characteristics of a
• relevance of the data quoted to design application, sample unit. Theunit may bean air-conditioning package,
• traceability to British, US or International fan coil unit,generator, or otheritem ofplant.Frequently,
standards. the manufacturer's data is inadequate, or not specific to
the model ordered.
The 'laboratory' should be an approved testing house
rather than a corner of the works, because the facilities Timing
must provide for testing between 40Hz and 16kHz with The timingof tests are critical — prototype testing and the
effective isolation to background noise and vibration. obligation for the supplierto cooperate must be builtinto
There shouldbe an airflow testing rig and testroomsfor the programme andcontractdocumentation. The respon-
both reflection and absorption available, by means of sibility and cost of any retests needs to be set out, as a
reverberation room andanechoic chamber. Linked rooms disincentive for too early 'time wasting' testing or too
enable the testing of sound insulation in a transmission hasty requests for retestwhen the supplierknows full well
suite. that the unit under retestwill still not comply.
108 Acousticsin the Built Environment
head height, for a variety of positions. Any 'hot spots' 10. ESDU 82002 Reduction of Sound in Ventilation and
directly under supply or extract grilles, or discernible SimilarAir Distribution Systems, 'Noise in Air Condition-
rattles (it is amazing what people leave in ducts) need to ing Systems', Fluid Mechanics in Internal Flow, 9,
be noted. Engineering Services Data Unit, London,July 1982
There maybe a requirement to includemeasured levels 11. ESDU 81043 Sound in Low-Velocity Ventilation Ducts,
in engineering manuals at handover. Care should be 'Noise in Air Conditioning Systems', Fluid Mechanics
takento take measurements for equipmentin normal use, in Internal Flow, 9, Engineering Services,Data Unit,
or at its various settings to provide complete information. London, December1981
This may be difficult for automatically controlled chiller 12. ESDU 82003 Example toIllustratethe UseofData Items on
plant, measured in winter. Diagrams 3.2 and 3.3 show Noise from Ducted Ventilation and Air Conditioning
requirements anda basic problem-solvingpath. Systems, 'Noise in Air Conditioning Systems', Fluid
References considering commissioning and testing of Mechanics in Internal Flow, 9, Engineering Services
systems are as follows: CIBSE Commissioning Codes: Data Unit, London, December 1982
Series A Air Distribution, Series B Boiler Plant, Series R 13. BS 2750:Part11980Measurement ofsoundinsulationin
Refrigerating Systems [23]; BSRIA Applications Guide: buildings and of building elements, British Standards
Operating and Maintenance Manuals for Building Ser- Institution, Milton Keynes
vices Installations [24]; RlBAArchitect'sJob Book [25]. In 14. BS 4718: 1971 Methods of test for silencers for air
the absence ofInternationalStandards in general use for distribution systems, British Standards Institution, Mil-
measuring services noise, the Association of Noise Con- ton Keynes
sultants have issued guidelines for instrumentation, meas- 15. BS 5821: 1984 Methodsforrating the sound insulation in
urement units, and techniques for 'steady','non-steady', buildings and of building elements, British Standards
and 'impulsive' service noise sources. Institution, Milton Keynes
16. BS 3638: 1987 Method for measurement ofsound absorp-
tion in a reverberantroom, British Standards Institution,
Milton Keynes
References 17. BS 4773: 1989 Methodsfor testing and rating air terminal
1. Rose, K Guide to Acoustic Practice, 2nd edn, BBC devices for air distribution systems, British Standards
Engineering, London, 1990 Institution, Milton Keynes
2. European Broadcasting Union, Acoustical Properties of 18. BS 4856: 1978 Methodsfor testing and rating fan coil
Control Rooms and Listening Roomsfor the Assessment of units, unit heaters, and unit coolers, British Standards
Broadcast Programmes,Report No. R22, EBU, 2nd edn, Institution, Milton Keynes
1985 19. BS 4857: 1983 Acoustic testing and rating, British
3. Specificationfor Studio Centres, Engineering Code of Standards Institution, Milton Keynes
Practice for Independent Local Radio — 1984, Issue 2, 20. BS 4979: 1986 Aerodynamic testing of constant and
Independent Broadcasting Authority, London, Feb- variable dual or single duct boxes, single duct units and
ruary 1988 induction boxesfor air distribution systems, British Stan-
4. BS 4142: 1990 Method of rating industrial noise affecting dards Institution, Milton Keynes
mixed residential and industrial areas, British Standards 21. BS 1042:various dates 1981 to 1993Pressure differential
Institution, Milton Keynes (under review) devices, British Standards Institution, Milton Keynes
5. BS 6472: 1992 Guide to evaluation of human exposure to 22. BS 848: Part2: 1985 Fansforgeneral purposes: methodsof
vibration in buildings (1Hz to 80Hz), British Standards noise testing, British Standards Institution, Milton
Institution, Milton Keynes Keynes
6. BS 7385 Evaluation and measurement for vibration in 23. CIBSE Commissioning Code Series A: Air distribution,
buildings CIBSE, London, 1996
Part 1: 1990 Guide for measurement of vibrations and CIBSE Commissioning Code Series B: Boiler plant,
evaluation oftheir effects on buildings CIBSE, London, 1975
Part 2: 1993 Guide to damage levels from groundborne CIBSE Commissioning Code Series R: Refrigeration
vibration plant, CIBSE, London, 1991
7. Steffens, R. J. Structural Vibration and Damage, BRE 24. BSRIA Operating and Maintenance Manualsfor Building
Report, HMSO, London, 1974 ServicesInstallations, Building Services Research and
8. CIBSE, Sound Control, Section B12, Guide Volume B: Information Association Applications Guide, Brack-
Installation and equipment data, Chartered Institute of nell, 2nd edn, 1990
Building Services Engineers, London, 1986 25. RIBA CommissioningServicesInstallations, Royal Insti-
9. Fry, Alan (ed.) Noise Control in BuildingServices, Sound tute of British Architects, Architect's Job Book, 6th
Research Laboratories Ltd, Pergamon, Oxford, 1980 edn, 1995
Local acoustic and environmental conditions must also
Chapter 4 Sound systems be fully taken into account at the system planning and
design stage, together with zoning and emergency over-
ride and priority requirements.
Peter Mapp Table 4.1, taken from BS 6259 [1], gives a brief
summary of typical building types and sound installation
requirements. In many buildings such as hotels, con-
ference centres and theatres, more than one type of
Introduction system may be required, e.g. stage music, speech rein-
Sound and communications systems are becoming forcement, and perhaps even electroacoustics,systemsfor
increasinglyintegrated into building design, formingpart the main auditorium, public paging/ announcementfor
of an emergency warning/fire alarm and evacuation the front-of-house, circulation and ancillary areas, and
system, a general announcement(paging) system, or as a staff or technical paging/ announcements behind the
special design feature enabling background music to be scenes. These self-contained systemsmay, however, need
played, or live entertainmentto be played or relayed. An to be linked or interfacewith othersystems or subsystems
even greaterneed for early consideration andappropriate for emergency announcements, evacuation or fire alarm
integrationofthe sound systemoccurs in public buildings purposes.
housing auditoria, lecture theatres and conference facili- The design and installation of a sound system may
ties, where high-quality sound reinforcementsystemswill involve a number of different disciplines including:
be required. The system chosen must complement the
natural acoustics of such halls.
• the architect, who will have interest in the loud-
In buildings with hostile acoustic environments, suchas speaker types, appearance and placement, and the
overall integration of the system into the project
noisy and/or reverberant leisure centres, ice rinks and design;
swimming pools, special care needs to be taken with the • the M & E consultant, who will be involved with the
loudspeaker type and locations and the overall system provision of electrical power supplies and cable
design, in order to ensure that adequate speech intelligi- routes, etc., and if a voice alarm system is included,
bility and clarity will be achieved. possible interfacing to the fire alarm detection and
In sports stadia and similar venues where crowd noise control system;
levels can be high, the system must be capable of
comfortably overcoming the anticipated noise level and
•
specialist consultants, including theatre consultants,
audiovisualconsultants and acoustic consultants, who
providing a clear and distortion-free signal. Similar will all be interested in the way the sound system
considerations also apply to noisy industrial process areas interfaces with their particular areas and
and plant rooms. installations;
In addition to the moreusual speech and music signals,
sound systems may be used to transmit warning tones and
•
sound systemsdesign consultants: althoughmany of
the smaller systemsare designed either by the M & E
signals, or information tones/signals (for example, fac- consultant often with the aid of a manufacturer,
tory processes, swimming pool sessions). Sound systems specialist independent sound-systems design con-
are frequently required to interface with or form an sultants are available,who are able to design a system
integral part of other systems such as fire alarm and without a vested interest in a particular range of
emergency evacuation control systems, or systemsfor the equipment. Equally,a specialist consultant will have a
hard ofhearing, audiovisual(A/V) systems, video andfilm wider view of the market, equipment performance,
projection systems, languages translation/interpretation current techniques, and experience of systems
systemsandbroadcast and relay systems. design.
In order to achieve an appropriatesound quality, clarity
and intelligibility, a sound system must meet certain A specialist consultant will take responsibility for his
performancecriteria, which will be dependent upon the design andbecause he is not trying to sell anything can be
type and application of the system concerned. The basic acceptably integrated into the design team. He is there-
parameters to be considered include: fore in a much better position to argue the case for critical
•• sound level, loudspeaker locations or the requirement for specific
acoustic treatments, and realistic budgets, than can a
overall frequency response, manufacturer's representative, or a non-specialist con-
• signal-to-noiseratio, sultant.
•• direct-to-reverberant sound level ratio, Most ofthe larger manufacturers offer a design service,
freedomfrom echoes and distortion. butschemes will be basedon their own equipment ranges,
Other secondary requirements include system stability which may not necessarilybe suitable for the project in
and reliability, freedom from interference both to and hand. Furthermore,few manufactuers' designers have a
from other systems operating within the vicinity. Even background in acoustics and loudspeaker technology— an
humidity and climatic conditions influence how a system area which currently is one of the less well understood
performs on individual occasions. areas ofsystem design.
Systems used for fire alarm and emergency announce- A numberof computer-aided design programs are now
ment purposes must be appropriately fire protectedand available to assist with sound systems design, and are
monitored, and incorporate secondary power supply primarily aimed at improving the loudspeaker coverage/
back-up and switch-over facilities. room acoustics problem. However, the majority of the
Table 4.11 Ba.sic recommendations for sound system applications
Thne
signals
excluding High
Live Paging and Emergency fire Back- quality
reinforce- general announce- alarm Radio/ ground recorded Assisted
Building type ment announcement merit CCTV (see 5.18) TV music programme hearing Interface Notes
* * * .. * *
Airports FA/E Multiple inputs and zones
required. Noise levels vary
Auditoria General SM XS Rehearsal only * * * * FA/E Paging only during
E/S A/V rehearsals. Relay to other
areas
Canteens (SM) * * R (see R ' R (see note) R (see note) FA/E May be used for live
note) entertainment: use a
separate system (see also
auditoria)
Clubs SM XS As auditora * R — R * * * FA/E Zone areas (see theatre)
Theatres A/V E/S
Concert halls (see SM XS — — R * *
Rehearsal only, see FA/E Relay to other areas. Zone
theatres) theatres E/S A/V peripheral areas
Council chambers S — * R — — — — — FA/E Relay to other areas. Zone
A/V separately
Other areas of . * — —
May act as overflow area.
congregation Zone accordingly (see
offices)
Court buildings — R 5' — — — —
FA/E Announcement to public
area
Court rooms s — * — — — R — FA/E High quality recording
R A/V and playback usually
required
Conference rooms/ S (XS) (M) 5' R R R Areas of 5' *
(See note) FA/E Paging generally restricted
centres 5' R R R
(SI) circulation ES A/V to areas outside
only SI conference rooms. Zone
systems
— R * — — R — —
Department stores R FA/E Zone different floors/
areas. (Microphone ports
for spot announcements)
Discos S (XS) 5' * 5' 5' —
Only in FA/E Sound level noise limiters.
SM areas A/V M/S FA/E overrides. Zone
isolated areas w
=
from dance
floors
Key: () = Possibly necessary; * = Usually or often necessary; R = May be necessary/should consider; XS = External system (brought in). May operate and need to interface
with permanent systems, for example a 'rock' band PA system or broadcast system; SI = Simultaneous Interpretation Systems; M = Music; S = Speech; FA = Fire Alarm
System; E = Emergency Evacuation System;A/V = Audio Visual.
Extract from BS 6259 is reproduced with the permission of BSL Complete copies of the Standard can be obtained by post from BSI Publications, Linford Wood, Milton I-
I-
Keynes MK14 6LE I-
1)1
I..'
Table 4.1 Continued I-
Tone
signals
excluding High
Live Pagingand Emergency fire Back- quality
reinforce- general announce- alarms Radio/ ground recorded Assisted
ment announcement ment CCTV (see 5.18) TV music programme hearing Interface Notes
Building type
Key: 0 = Possible necessary; * = Usually or often necessary; R = May be necessary/should consider; XS External system (brought in). May operate and need to interface
with permanentsystems for example a 'rock' band PA system or broadcast system; SI = Simultaneous InterpretationSystems; M = Music; S = Speech; FA = Fire Alarm System; :
E = Emergency Evacuation System; A/V = Audio Visual.
114 Acoustics in the Buift Environment
In terms of the required amplifier power output, a In highly reverberant areas, it is often desirable to limit
10dB margin is a significant factor, being equivalent to a the low frequency response of the system in order to
multiplication factor of 10. However, techniques such as reduce the sound masking effect of the low frequency
compression and limiting of speech input signals are reverberation and hence improve the potential intelligi-
available which enable smaller power margins to be bility of the reproducedspeech. Although the basic tone
employed. The amountof'headroom' requiredwill vary a control filters found on most Public Address (PA) and
great deal upon the type of system beingconsidered. For soundreinforcementequipmentcan be extremely useful,
speech paging/announcement systems, a signal head- their control range is limited. The use of a multi-band
room of6dBmay be adequate, particularly ifcompression filter or graphic equalizer to contour and adjust the
or limiting is incorporated. overall frequency response of a sound system is an
Where ambient noise levels vary considerably, the extremely usefuland powerful aid to system intelligibility
signal-to-noise ratio of the sound system can be main- and should be considered for all but the smallest
tainedby incorporating either local groupor zone volume system.
controls, which allow the level of the sound system signal
to be adjusted as required. Alternatively, automatic noise
sensing and level adjustment circuitry may be incorpo- Echoes and reverberation
rated. This automatically and continuously monitors the Secondarysources ofsound, for example reflections from
noise level at a point andadjusts the output of the sound room surfaces, or the output from supplementary or
system accordingly in order to maintain an adequate repeater loudspeakers, which arrive at a listening position
signal-to-noise(S/N) ratio. within a period ofup to 35—50 ms afterthe original sound,
will integrate or merge with the original direct sound
(whether it is a person speaking or a loudspeaker) and
Frequency response combine to produce one overall louder sound (i.e.
When selecting equipment for a sound system, due intelligibilityis enhanced).
consideration should be given to the frequency response Secondary sounds arriving after 50ms do not fully
characteristicsofthe equipmentandthe nature ofthe task integrate and, depending on their level relative to the
it is required to perform. initial sound, may be heard as echoes which will have the
The minimum frequency range over which a sound effect of reducingthe overall intelligibility of the speech
system must operate for intelligible speech is 400 Hz to signal.
4kHz. However, the quality of such a limited range is Sound systemsshould be designed to ensure that the
extremely poor. For reasonable quality reproduction a generation of long-delayed secondary sound signals,
sound systemshouldbe capable ofreproducing the range caused either by structural reflections or by secondary
of at least 150Hz to 6 kHz. For high quality sound loudspeakers, do not occur or are well controlled.
reinforcement, the range should extend to 8 or 10 kHz. Loudspeakers should be placed no more than
Forhigh quality reproductionofmusic, a rangeextending 12—15m apart when facing in the same direction or
from at least 100Hz to 10kHz is required. 25m apart when directed to cover an area as shown in
The coverage (dispersion) angles of the loudspeakers Figure 4.1. The loudspeaker should be angled down-
must be taken fully into account when designing the wards to cover the appropriate area and to limit the
layout and potentialcoverage/soundlevel variations. The overspill to adjacent areas. A row of loudspeakers even
coverage angles are defined as the angles at which the though individually spaced at 12—15m apart can still
output from the loudspeakers is 6 dB less than that give rise to an appreciable echo effect due to the
produced on the main axis of the loudspeaker. Usually, difference in distance between the nearest and sub-
the total included angle is quoted.This anglewill usually sequent sets of loudspeakers whose output, due to the
change significantly with frequency, decreasing as the effects of the Inverse Square Law, may still be
frequency increases with most loudspeaker types. When significant.
designing sound systems, the 4kHz coverageangleshould
be used.
When laying out and positioning loudspeakers, con-
sideration should be given to the difference in Sound
Pressure Level (SPL) which will be produced between a
position on the main axis ofthe loudspeaker, and an 'off-
axis' position; at the coverageanglefor example where, by
definition, the SPL will be 6dB lower than on the main
axis. It may in fact be slightly more than 6dB lower when
account is also taken of the additional angular path
distance involved. A 6 dB difference is quite noticeable,
and is sufficient to affect significantly the potential
intelligibility of the system in a noisy or reverberant
environment by decreasing the apparentS/N (D/R) ratio
as calculated for the on-axis case. In such circumstances,
loudspeakers should be positioned so that their coverage
angles either meet or preferably overlap so there are no
gapsin the coverage. Figure 4.1 Typical loudspeakerarrangement to reduce echoes
118 Acoustics in the Built Environment
measurably different than at 50m. Figure 4.2(a) Basic sound system components
Ambient noise-sensing
microphone
r— Compressor/
p
I limiter
ATIC
,
Microphone I Pre-amplifier Graphic
Equalizer
_____ Loudspeaker A'
Pre-amplifier I LEVEL
Mixer
—I_ -u:1
Microphone I Compressor/
r—j udspeaker'B'
I limiter CONTROLL_J
i-ca
Local or
-—--1 i-.— —i
L__.J group
volume
Tape L Graphic I
I control
Equalizer L_
Delay
line
Figure 4.2(b) Simplfied sound system schematicdiagram ofsignalprocessingequipment
Sound systems
considered as the normally permissible ratio (1.2—2 is the power amplifier(s) feeding the particulararea or zone
optimal). Compressor/limiters may either be incorpo- in question. The signal from the microphone is time-
ratedinto eachpre-amplifier inputor alternativelya single averaged so that occasional transientevents do not affect
unit may be connected between the output of the pre- the overall level. When an announcement is made, the
amplifier and the input to the power amplifier. level is automatically set so that the announcementdoes
Each microphone or system input should be capable of not fluctuate, in order to maintaina nominally consistent
being individuallycontrolled. Microphone inputs should S/N ratio.
ideally be provided with tone controls or equalization Correct equalization of a sound system is becoming
facilitieswhich are separate from music or other auxiliary increasingly recognized as an essential aid to optimizing
inputs. In live sound reinforcement applications, it is both speech intelligibility and sound quality/clarity, as
generally not possible to pre-set the input level controls aswell as helping to ensure that maximum gain before
the volume requiredwill depend on the voice level of the feedback is maintained. An equalizer is essentiallya multi-
person speaking and the additional acoustic damping frequency tone control, typically consisting of either
provided by the audience. In reverberantrooms with little 10—12 octave band filters or preferably 27—30 one-third
natural sound absorption, e.g. by carpet or acoustically- octave band filters.
absorbent finishes, the audience can significantly affect The one-third octave band filter provides extensive
the degree ofgain availablebefore feedback. Outputsfor control over the entire audio band with usually ±12 dB of
recordingor feeds to other areas' services,e.g. simultane- cut or boost available at each of the one-third octave
ous interpretationsystemsor audio induction loops, may frequencies. Although the unit is particularly suitable for
also be required. Wherever possible, a sound reinforce- sound reinforcementapplications, it is becomingwidely
ment system should be controlled by a trained operator. used in more general PA systems. The more restricted
Paging and announcement systems where the micro- control of the octave band equalizer tends to limit it to
phone is not in the same acoustic space as the loud- only basic PA applications. Other forms of tunable
speakers can be pre-set, though some compensation for equalizer are also available.
differentvoice levels may be required— the useof either The equalizer is used to help compensate both for
an automatic gain control (AGC) or compressor is deficiencies or irregularities in the response of the system
preferable. loudspeakers and/or the acoustic characteristics of the
space that they are serving. The response of most horn
Signalprocessing equipment loudspeakers for example begins to roll offabove 3—4kHz
Apart from compression and limiting, a number of other whilst the response of many cone loudspeakers begins to
additional items of electronic signal processing equip- fall off above 5 or 6 kHz — the equalizer may be used to
ment are commonly incorporatedinto the sound system extend and flatten the response of the loudspeakers.
design. Two processors that are particularly important for In difficult acoustic environments, e.g. those with long
optimizing speech intellibility are automatic noise sens- reverberation times and/or high ambient noise levels,the
ing/level control, and equalization. Other processors equalizer is used to shape the response of the broadcast
include phase/frequencyshifters for improving the gain sound to either help reduce reverberant excitation or
of a sound system before feedback, and audio delay lines accentuate the frequencies most importantto intelligibil-
which enablethe signal arrivals from distant loudspeakers ity and clarity, i.e. 1—5kHz.
to be synchronized in order to overcome potential echo Where differenttypes ofloudspeakers are employed, or
problems or to enhance apparent localization of a sound where a number ofacoustically different areas are served
source. by the PA system, a numberof equalizer channelsmaybe
Automatic noise-sensing and level controllers consist required. In addition, care should be taken with the
essentially of two basic elements, the ambient noise- routingof such signals, as the equalization filter curve set
sensing microphone, and the PA signal and control for one type of loudspeaker may not be suitable for
circuit. another. Furthermore, the curve set for a reverberant
Where the ambient noise level in an area regularly hangar or noisy workshop is unlikely to be suitable for
fluctuates considerably, say by more than 6—8dBA, an office areas or other 'deader', i.e. less reverberant, spaces.
ambient noise sensing and automatic level adjustment Figure 4.2b illustrates how the above signal processing
system can help to maintain an appropriate S/N ratio. connects into the signal chain. In large PA systems, several
Examples of such areas are industrial or process areas, equalizers and level controllers may be required, as each
workshops or locations, and spaces affected by traffic or difficult area should ideally be individuallytreated.
occupational noise,including airports, rail termini,sports
stadia and swimming pools. A sensing microphone is Poweramplifiers
locatedat a suitable point where it is able to monitor the The signal presentedto the power amplifier is normally at
ambient noise level within the designated area. The the standardaudio signal line level of 0dBm. The input
sensing microphone must be carefully positioned to should normally be balancedwhere a substantial distance
ensure that it is not affected by very local events. It is separates the pre-amplifier (or control unit/mixer) from
therefore usually located at a fairly high position, away its power amplifier. Unbalanced inputs maybe usedwhere
from local noise sources suchas machinery, air-condition- the distance/cable run is short.
ing equipment,or isolated groups of occupants. Amplifier outputs may be either of low impedance, e.g.
The signal from the microphoneis used to drive, after to suit 4—8 loudspeakers, or high impedance for 100 V
suitable processing, a voltage-controlled amplifier/ line working. The majority of PA systems employ 100 V
attenuatorwhich adjusts the level of the signal routed to line working to reduce line losses and for ease of
Sound systems 121
thereby limiting the inrush current requirements. The teens, office areas,etc. (typically 3—5m)
heat load generatedby the amplifiers must also be taken For variable coverage,e.g. cor- — Spacing dshould be 2 to 3h or
into account. ridors, general areas 3 to 4h in corridors (typically
5—9m)
Loudspeakers
The loudspeakerconvertsthe electrical power from the Figure 4.3 Loudspeaker
coverage
amplifier into an acoustic signal via the vibration ofa cone
or diaphragmwhich sets up sound pressure waves. The
importantparameters to note are: imately 25 m2 for a ceiling loudspeaker at a mounting
• rated or maximum power, capacity, height ofbetween 2.8 and 4.8m, as opposed to 30—50 m2
•• for a wall cabinetpaging/PA system, based on a 60—90°
sensitivity(SPL at 1 W, 1 m), cone. When calculating the area covered by a ceiling
frequency response,
• directional information, i.e. dispersion or coverage loudspeaker, the height of the listeners' ears above floor
level must be taken into account, as shown in Figure 4.3.
angles. A high-density ceiling loudspeaker installation can form
The choice of loudspeaker depends on the intended the basis of a very high quality sound reinforcement
use, power/SPL output capability, directional character- system with very uniform sound distribution and cov-
istics, quality required, location (e.g. indoor or outdoor erage. When designing a high quality sound reinforce-
use) and local acoustic environment. The aesthetic ment system, the coverageangle at 4kHzshould be used;
requirements of the unit may also need to be considered. typically, this will be 60° or less for a 200-mm cone
The most commonforms of loudspeakers are: loudspeaker.
• the cabinetloudspeaker, Bidirectionaland wedge loudspeakersmay either be directly
mountedon a wall or ceiling, or suspended below a high
• recessed ceiling loudspeaker,
ceiling by a chain or wire. The loudspeaker produces two
• the bidirectional and wedge cabinetloudspeaker, conesof sound (60—90°).Directlyunder the loudspeaker,
• the column loudspeaker, a high-frequencies dead spot can occur.
• the re-entranthorn loudspeaker,
• the ConstantDirectivityhorn loudspeaker, Columnorlinesourceloudspeakers provide wide coverage in
the horizontal plane but deliberately restrict the sound
• the full-range, integrated, high power loudspeaker.
output in the vertical plane (e.g. 90—120° horizontal,
Cabinet loudspeakers are suitable for music, speech and 15—20° vertical). This characteristic enables the column
paging systemsin areas where the floor-to-ceilingheight is loudspeakers to have the potential of providing greater
typically 4m or less. Good reproduction units typically intelligibilityin reverberantareas orgreater coveragethan
provide a coverage of a 60—90° cone from wall a conventional cone loudspeaker in less hostile environ-
mountings. ments. The column loudspeakers may be usedfor both PA
Recessed ceiling loudspeakers may be used in areas with and sound reinforcement applications. It is essential
ceiling heights up to 5—6m, as well as lowerceiling areas. however to correctly aim the column loudspeaker so that
The coverage from a ceiling-mounted loudspeaker is the beamof sound it produces is directedat the listeners.
generally less than from a wall-mounted cabinet; approx- Normally, column loudspeakers should not be mounted
122 Acoustics in the Built Environment
/
,-ft5-2O°
Sound dispersion • by use of a centralized loudspeaker systemcluster,
If • by use of a low-level grid of ceiling-mounted loud-
speakers or distributed system of wall or column
mounted loudspeakers.
1 1
{\ Typically
Centralizedloudspeakersystems typically comprise column,
CD horn, or full-range loudspeaker cabinets. The units
1.5m to 2.5m
may either be mounted on either side of the stage or
Figure 4.4 Column/linesourceloudspeakers rostrum, or preferably they may be mounted centrally
over it in the form of a cluster. Use is made of the
loudspeaker's directional properties to direct sound into
flat against a wall but should be provided with suitable the audience and away from reflecting wall surfaces
thereby reducing the degree of reverberant excitation
angle brackets allowing correct alignment. By aiming the producedby the system and aiding intelligibilityandgain
column loudspeaker towards the centre of the area to be before feedback.
covered, a fairlyuniformcoverageisobtainedas the nearer Constant directivity (CD) horns are becomingincreasingly
listeners will be located out of the main beam of sound
which can therefore be increased to reach the rear of the popular in high quality sound systems applications, or
area.When usingrepeatercolumns, e.g. in long churches, where high soundpressurelevelsare required.Theyhave
the repeaterunits shouldbe locatedwithin 12—15m ofthe many advantages over re-entrant or other horn types
which may be described as follows:
primary or any other supplementary loudspeakers, unless
signal delaylines are employed (see Figure 4.4). (a) They are extremely efficient, typically exhibiting a
Re-entrant horn loudspeakers have a restricted frequency 1W/i m sensitivity of around 113dB for a 60 X 40°
response and are used mainly for speech. They are more horn.
efficient than either the cabinet or column loudspeaker (b) They can provide extremely uniform coverage as they
and are therefore suitable for use in areas with high exhibit a reasonably constant directivity with fre-
background noise levels, large enclosed areas, and out- quency, e.g. within ±100 over their operational range
doors. The coverage angles of horn loudspeakers vary of 500 Hz to 16kHz plus, for a well designed large-
considerably but typically are between 40 and 80°. format device.
Loudspeakers should be selected and locatedto provide (c) They exhibit considerably lower distortion character-
as uniform a coverage of an area as practicable. The istics than a re-entrant horn.
inversesquarelawandcoverageangledata should be used (d) They provide a very much smoother and uniform
initially to predict the coverage and plan an installation. frequency response of high fidelity (hi-fl) rather than
ForPA/pagingsystems, the maximum variation should be PA quality. CD horns may therefore be used for high
less than 10dB at 2 kHz. In areas of high noise or quality sound systemsin theatres and concert halls as
reverberation, a smaller variation will be required if well as large stadia or reverberant exhibition halls. A
intelligibilityis not to be downgraded. In sound reinforce- CD horn usually operates over the range 500 Hz to
ment systems, the variation should be within 6 dB and over 15 kHz (or from 800 Hz for the smaller types).
preferably within 4dBat 2 and 4kHz. An associated low-frequencyloudspeaker is therefore
The majority of loudspeakers used for PA and SR required to form a complete system operating from
systemsare operated on a 100-V line distribution system, well below 100Hz (e.g. 50Hz) to over 15kHz. A
with each loudspeaker fitted with an individual tapped dedicated cross-over unit is employed to block low
transformer. Tappings provide a useful range of output frequencies being fed to the horn and high fre-
adjustment and enable the loudspeakers to be set to quencies to the bass driver. The cross-over may be
compensate for different local conditions such as noise either active or passive, dependingon the particular
level, room volume, and length of throw, as described application and system configuration desired.
earlier in this chapter. (e) The controlled dispersion of a CD horn ensures that
Group or zone volume controls can be employed to high frequency beaming and 'hot spots' do not occur,
adjust the overall sound level in a given area. butinstead a uniform distribution of sound is created
Loudspeakers are affected by the acoustic environment at all frequencies within the working range of the
and peaks in their response coupled with the room horn. The controlled radiation of the CD horn also
Sound systems 123
High impedance
Resistance 5011 10011 20011 1 00011 5 00011 10 00011
Equivalent ohms/ Low impedance
conductor 1000 feet 200W/100V 100 W/100V 50W/100V 1OW/100V 2W/100V 1W/100V
size area AWG (300m) 411 811 1611 100W/70V 50W/70V 25W/70V 5W/70V 1W/70V W/70V
5.2mm2 10 1.00 120 240 480 1500 3000 6000 30000 150000 300000
3.3mm2 12 1.59 75 150 300 940 1800 3800 18000 94000 180000
2.08mm2 14 2.50 48 96 190 600 1200 2400 12000 60000 120000
1.3mm2 16 4.02 30 60 90 370 740 1500 7400 37000 74000
0.87mm2 18 6.39 19 38 76 230 460 920 4600 23000 46000
0.52mm2 20 10.1 12 24 48 150 300 600 3000 15000 30000
should be given to the use of satellite amplifiers where clearly intelligible and to whom? The answers to these
long cable runs are required to deliver high power loads questions are less obvious than at first sight. For example,
(see Table 4.2). it is quite possible to designa perfectly intelligible sound
system but end up hearing almost totally unintelligible
'Deafaid loop' systems announcements. This may be due to either defective
Systems for the hard ofhearingmayeither be basedon an hearing on the partofthe listeneror poor articulation on
audio frequency induction loop system (AFILS), which the part ofthe announcer. Equallyit is possible that some
enables any standardhearing aid with a 'tele pick-up coil' environmental factors not takeninto account could affect
(usually identified by the aid having an additional switch the audibility of the system, e.g. intrusive noise from
position marked 'T') or via an infra-red (IR) transmission outside or high ventilation noise levels.
system. The latter system requires a special receiver but The problem is that apart from dealing with the
has the advantage of a potentially greater bandwidth and electronic andelectroacousticsaspects ofthe systemwhich
ability to simultaneously transmit up to 12 separate are reasonably controllable, a PA system has to interface
channels, making it very suitable for simultaneous inter- with people, both at the source and at the receiving end,
pretation systemsor 'museum tour' systems. In theatres, and people are highly variable. For this reason, when
infra-red systemsare normallyemployed, primarily to help testing a systemsubjectively it is very importantto:
overcome potential interference problems with other
audio or AV systems, but also because of their generally
• use a large enough and representative enough
sample of listeners, and
greateraudio quality.
AFILS and JR systems may also be used to transmit
• use several announcers.
speech or warning signals to personnel working in very The effect an announcer can have on the perceived
noisy conditions where hearing protectors are required. intelligibility from a sound system is profound. Systems
Special protectors incorporating a receiver and in-built with 'good' intelligibility can be transformed into total
miniature loudspeaker/earphoneare available, enabling unintelligibility by different users of the same micro-
'wireless' communication to be maintained. phone. One way round the announcerproblemwould be
Guidance is given in BS 7594 [4]. to use standard recorded messages of known good
articulation. However, such messages would need to have
an informationcontentandstyle ofdeliverysimilar to that
Speech intelligibility of normal system announcements.
The practicality of setting up statistically valid tests
Assessment complete with an appropriatelistening panel to test out a
With the ever-increasing use of public address (PA) sound system so that an absolute intelligibility score can
systemsfor fire alarm andemergency warning announce- be obtained, is generally an impossible goal. Instead, in
ments, more emphasis than ever before has been placed practice a small group of people, maybe as few as 2 or 3
on the intelligibilityandclarity ofsuchsystems. Whilst it is typically, wanderround the designated area andlisten to a
obvious that an emergency, PA or sound reinforcement few test messages. In most cases a consensus view can be
system must be intelligible, many specifications currently reached, but often listeners may well be confusing sound
include such phrases as 'the PA system must be clearly quality with intelligibility, and an impasse over the
audible' or 'the PA system must be capable of producing intelligibilityrating poses difficulties.
clear and intelligible speech'; but what do 'audible' and Ideally, what is requiredis some totallyobjective method
'intelligible' mean, for they are far from being the same which can be implemented and produces an easy-to-
thing, and under what circumstances has the systemto be understand rating. Such methods do now exist and a
126 Acoustics in the Built Environment
significant number of sound system specifications are • the message language and familiarityof the listener
beginning to require specific levels of intelligibility to be to that language.
met.
It is useful to consider also and at greater length the
main sound system parameters which have an effect on
Techniques ofmeasurement
At present four generic techniques are available for speech intelligibility.There are a number of parameters,
as follows:
assessing the potential intelligibility of a sound system.
The methods and implementation vary considerably,but •
frequency response,
only onetruly takes both background noise andreverbera- •
noise or more precisely signal-to-noiseratio,
tion into account. •
reverberation or more preciselydirect-to-reverberant
A number ofsubjective aspects must also be taken into ratio,
account, particularly when deciding on an appropriate •
freedom from echo.
target value. These include: Distortion isalso an influencing factorbut it is generallyof
• the level of difficultyofinformation that needs to be secondary importance unless gross overload or similar
understood, e.g. the requirements for a supermarket distortion occurs.
staff announcementsystem are somewhat less oner-
ousthan a system requiredto reproduceShakespeare Frequency response
or perhapsrelay specificinstructions inside a nuclear Frequency response was discussed earlier. Figure 4.8 shows
power station; the intelligibilityrange of concern.Although the primary
• the general hearing ability of the listeners and their speech information is contained at higher consonant
environment; frequencies, the main power of the voice comes from the
• the articulation or intelligibility of the original low and mid frequency vowel sounds with most sound
signal; energy centredat around200—600Hz. Typicallythe higher
• the rate at which the speech signal is to be frequency information is 15—20dBbelow these levels and
delivered; can be easily masked or lost if the soundsystem places too
40
35
a'
ii
£30
S LI
C
C
C
S
S
0a
C
20 -
0
a
C
0 15
U
S
0
S
I
C
S
S
0
S I.
S. -
H 4.
-I-- - - - 4- -J:i
5.
'FEft
ft5 T1'FI
0
125 250 500 1k 2k 4k 8k
Octav. Band C.ntrs Frsqusnoy (Hz)
70
65 S S S
60• 2U _____
155
V
45
1 i:
a.
C
40'
'1'
' —. —
-S
IH1t
Hr.
a,
:
-*-
':: -.-
—S.—-
-—-
30 - S.
ltS.l'
25 -
20 -
125
____
250
tii:
500 1k 2k
Octave Band Centre Frequency (Hz)
ii4k 81
315 0.001 0
400 0.001 4 COOL
4
500
630
0.001
0.002 0
0.002 0
" .
,
/
800
1K 0.002 4 ou
1.25 K 0.003 0
1.6 K 0.003 7 VE8VL0L
7
2K 0.003 7
2.5 K 0.003 4 NIL
3.15 K 0.003 4
4K 0.0024 02 04 LEO 08
5K 0.002 0 ARIl(ULAIII)N
Having now established the basic factors affecting ity. Although a generalized subjective responsescale can
speech intelligibility, measurement techniques can be be formulated for example:
studied.
0.2 or less unacceptable
Arliculalion index 0.2—0.3 marginal
The articulation index is a longstanding methodfor rating
the effect of background noise on speech intelligibility
with its procedure formally adopted in an American SPEECHLEVEL MINUS NOISE LEVELdBA
Standard (ANSI 53.5). " LESt XII(XIII'lALIX—12 —9 0
—6 +3
—3 +6 *9 +12
'
'II,
octave bands is measured. The 20 S/N ratios are then
yf N(ES
individually weighted according to the speech informa-
t
I,lSTENERS PRESENTATION
—
referred to as the articulationindex (Al). The method in ,'
WHITE
_
T7 I1OWN ASWERS
fact operates using the rms levels of speech peaks, a /N / NONSENSE —
+12 dB correction factor being adopted to convert the
B
r 1000DIFFERENTSYILABLESI
that these RT corrections are not particularly effective. Figure 4.10 Intelligibilitycurves. Relationship between
The Al uses a scale from 0 to 1.0 where 0 is total articulation index and various measuresofspeech intelligibility
unintelligibility and 1.0 is equivalent to 100% intelligibil- and signal to noise ratio ('A' weighting)
Sound systems
0.3—0.4 acceptable/fair ant (reflected energy) ratio. From these two parameters
0.4—0.5 good the equivalent % Alcons can be computed.
0.5—0.6 very good Oneofthe most useful features ofthe % Aicons method
>0.7 excellent is that one can correlate measurement with prediction,
and the % Alcons is still effectivelythe only method we
As previously described, the subjective rating heavily have of predicting the potentialintelligibility of a sound
depends upon the type of speech information being system before it is installed.
imparted. Figure 4.10 compares the Al to various types of Using the established simple formula:
speech signal, for example from a known limited vocabu-
laryofwords to completesentences or single words which 200 D2 T2 (n + 1) + K
% Alcons =
do not have any contextual clues to help identif' them. QVm
Figure 4.10 providesan extremely useful set ofintelligi-
bility curves, clearly showing that when the subject is Alternatively:
requiredonly to understand a limited numberofwords or % Alcons = 100 (10-2[(A+BC) -ABC]) + 0.015
familiar sentences, a much lowerAl can be tolerated than
when listening for the first time. Equally, the effect of where
removing the context from around a given word, e.g.
when it is used in a sentence, can be seen clearly with / LR+LN
much higher Al scores (and hence S/N ratios) being A=—0.321og( forA>
— 1 A= 1
\1OLD+LR+LN)
required.
A simplified version of the Al procedure usingjust 5 / LN \
octave bandsis also available, but its accuracy is not up to B=—0.321og{ )forB1,B=1
the full 20-band test. In the absence of measured speech \1OLR+ LN/
level data, the Al method provides an idealized speech
spectrum (male speech) which has an overall equivalent C = —0.5 log
level of65dBC at 1 m (dBC is used rather than dBAas this (Rmo)
latter weighting too severely attenuates the lower
frequencies). where:
While very good correlations can be obtained between reverberation components. The reduction in the modula-
the D/R ratio at 2 kHz and %Alcons/intelligibility,using tion transfer function is measured over a range of
this technique only tests the system over one frequency 14 modulating frequencies at the 7 normal octave band
band. The method effectively assumes that the system is 'carrier' frequencies of 125Hz to 12.5kHz. The result is
well behaved and controlled at other frequencies. It is that a 98-point measurement matrixis produced, i.e. 7 X
therefore essential to carry out supplementary measure- 14. Each set of 14 MTFs is reduced down to a single
ments such as overall frequency response and impulse transmission index value. The 7 individual transmission
response (D/R ratios) at other frequencies, for example index values are then weighted and further combinedto
500 Hz, 1 kHz and 4kHz. produce a single-figure value, the speech transmission
index (STI). A direct and strong correlation between STI
Speech transmission tests (STJand RAST1) and perceived intelligibilityhas been shown to occurand
The speech transmissionindex (STI) and its shorter a simple rating scale has been devised. The scale operates
derivative, RASTI, are the newest and most complex from 0 to 1.0 and is divided into 5 categorized bands from
intelligibilitymeasurement techniques currently available. 'bad' to 'excellent' as shown below:
It is only the adventof modern microprocessor and desk Bad Poor Fair Good Excellent
topcomputer technology that have enabledthe technique 0—0.3 0.3—0.45 0.45—0.6 0.6—0.75 0.75—1.0
to be implemented on a practical basis.
The STI is not as such based on an impulse/direct The values which correspondto 5, 10 and 15% Alcons
reverberantratio type of technique, but instead measures are 0.65, 0.52 and 0.45, respectively. In use the scales have
the modulation transfer function (MTF) between source been found to be extremely sensitive; for example, the
and receiver. The test signal can be thought of as a author's living room sound system only measuring 0.85.
complex modulated carrier whereby the modulation The STI intelligibility measurement technique automati-
depth of the received signal is compared to the perfect cally takes both backgroundnoise and reverberation into
(100% modulation) originally transmitted signal. The account. Furthermore, because the measurement is under
signal path, both electronic and acoustic, between the automatic computer control there are no operator deci-
origination source andthe receiver modifies and degrades sions to make, such as where to place integration and
the modulation of the signal by adding noise and divisional cursors. However, setting up a measurement
Sound systems 131
system to give valid results does require some experience Measurements at only 500 Hz and 2 kHz cannot possibly
and skill, particularly setting up relative levels through a define the complete performanceof a sound system.
sound system. Setting the correct operating level is A major advantage of STI and RASTI is the ability to
obviously essential if the equivalent speech level to readily carry out 'what if' speculations and predictions by
background noise is to be accurately accountedfor. post-processingthe noise componentdata; for example, if
As can be imagined from the 98-point measurement measurements were taken during a quiet period, it is
matrix, the measurement and calculation ofthe STI takes possible to manually input new noise data and get a
considerable computing power. In order to reduce the recalculated STI. In some implementations of STI and
complexity of the measurement and the measurement RASTI it is possible to completely isolate out the noise
time and hence the required computer power, a simpler component and so see the effect due to reverberation
derivativeofthe full STI method has been developed. The alone. An increment of 3 dB in the signal-to-noise ratio
rapid speech transmission index (RASTI) carries out increases the STI by 0.1 whereas a doublingofearly decay
systemmeasurements at only 500 Hz and2 kHzand atjust time (see Chapter 2 for definition) decreases the STI by
9 (shared) modulationfrequencies. This has enabled a 0.15. RASTI/STI assumes that S/N ratios >15dB do not
dedicated hard-wired system to be developed. The advan- influence the potentialintelligibilityanyfurther (cf. 25 dB
tage of the Bruel & Kjaer RASTI system is that the % Alcons and 30 dB for At).
transmitter andreceiver are totallyindependentunits and In a similar manner to % Alcons, STI can be predicted
so require no loop-back interconnection cabling. The butit takes a relativelycomplex computerprogramto do
B&K RASTI equipment can carry out the foreshortened this. Although RASTI and STI measurements are reason-
STI in just 16 to 32s. ably straightforward to make, setting up the system
Themajordisadvantage ofthe RASTI method is that no particularly with regard to signal-to-noiseratio is highly
information is gatheredover the whole operatingrange of critical. Under certain conditions it is quite feasible to
a sound system, e.g. at 125 and 250 Hz at low to mid obtain false readings.
frequencies and at 4 and 8kHz at the upper end. The
systeminherently assumes that both the system and space
it is in are reasonably linear. This, however,is frequently References
far from being the case and an overly optimistic (or 1. BS 6259: 1997 Code of Practice for design planning,
sometimes pessimistic) view of the situation is presented. installation and maintenance ofsound systems
At present RASTI is the only method that has been 2. BS 5839 Fire detection and alarm systems for buildings,
formalized and recognized within the International Stan- British Standards Institution, Milton Keynes
dards (lEG 268) and so is rapidly gaining authority as a 3. BS 7443: 1991 Specficationfor soundsystemsfor emergency
reference. It is subject to misinterpretation as it is often purposes, British Standards Institution, Milton Keynes
takenas a measure of the overall systemquality. This is not 4. BS 7594: 1993 Code of Practice for Audio-Frequency
what it is intended for, nor how it should be used. InductionLoop Systems (AFILS)
begin to sense sound painful when the sound pressureis
Chapter 5 Technical around 20 Pa.
Atmospheric pressurehas a value of i05 Pa.
information I pascal = 106 micropascal (pPa)
= 1 newton/metre2 (N/rn2)
= 10microbar (pbar)
David Saunders
Range ofhuman hearing: Most healthy young people will
be able to hear sounds with frequencies that range from
around 20Hz to 18 000 kHz (k is used to denote a factor
Definitions
of 1000). As peopleget older the high frequency acuityof
the ear deteriorates and this deterioration will be accen-
Basic concepts tuated byprolongedexposure to high noise levels (85dBA
and above).
Analogue and digital signals: Analogue signals are signals
that vary continuously with time, for example, the sound Prevention of hearing damage: The Noise at Work
pressure from a noise source. A digital signal is a signal Regulations 1989 definethree actionlevels for employees
that only hasvalues at discrete time intervals. Itis obtained at work
by sampling an analogue signal at fixed time periods. First Action Level: 85 dB LAeq,8h
Much present-day analysis equipment works with digital Second Action Level: 90 dB LAeq,8h
signals. Peak Action Level: 140dB
Angular frequency: The angularfrequency,w, is related to Protection must be
f
the frequency, by provided to employeesexposed to the
second or peak action levels. An employee may request
Ct) = 2 X rr X frad/s. hearing protection if they are exposed to the first action
Anharmonic: A frequency of a system which bears no level.
simple relationship to the fundamentalfrequency. Sound: Sound is one form of energy: it is the energy of
mechanical vibration of the molecules of a medium
Frequency: If vibrations are given to a medium in a throughwhich the sound is passing. Unlike lightandheat,
regular manner then a regular progression of pressure sound must have a medium to propagate through. The
maxima and minima will be observed at any point in the most commonmedium is
medium. When two adjacentmaxima or minima, or for well air; however, sound will travel
through most media.
that matter any two adjacentpaths of equalpressure, pass
the points, we say that onecomplete cyclehas passed. The Sound pressure/acousticpressure: When a sound wave
numberofsuchcycleswhich pass per second is referredto propagates in air there are local variations in the air
as the frequency of the wave. pressure. At anypoint the pressure will oscillate about the
In practice, sources do not generally produce single ambient pressure. These oscillations are known as the
frequencies although, for example, fans do tend to sound pressure or acoustic pressure and itis these changes
produce a characteristic tone which is associated with a in pressure which the ear detects and which, to a large
particular frequency. However, most sources will tend to extent, determine the loudness of a sound. For normal
produce sound that is composed of many different sounds the soundpressures are extremely small compared
frequencies covering a wide range. with the ambient atmospheric pressure. The smallest
The frequency ofa soundis a very importantcharacter- changes that the human ear can detect are about ten
istic as many acoustic phenomena are frequency thousand million times less than atmospheric pressure
dependent. while sounds which cause pain are about one thousand
Fundamental frequency: The lowest frequency of oscilla- times less than atmospheric pressure (Table 5.1).
tion of a system. Soundsources: Anysolid objectthat vibratesmechanically
Harmonic: A frequency of a system which is integrally can communicate the vibrations with the surrounding
related to the fundamentalfrequency. medium and generate sound. Sound is also commonly
generated by objects moving through a medium (e.g.
Hertz (Hz): Herizis the unit offrequency; it is the same as blades ofa fan moving through air) and by turbulence in
cycles per second. gases and liquids (e.g.jeteffluxes andturbulentflowofair
Noise: Noise is the term often used to describe unwanted and liquids in ducts and pipes).
sound, i.e. sound that annoys, interferes with activities or Sound wave: When sound energy travels through a
damages hearing. It is also usedto describe a combination medium, a sound wave is said to be propagating. The
ofsounds which vary randomly with time and which cover speed of the wave or the speed at which energy travels
a wide frequency range such as the sound producedby a from one point to another depends on the medium: in
jet engine or many industrial processes. general, sound travels faster in solids than in liquids and
Pascal (Pa): Pascal is the unit of pressure and so sound faster in liquids than in gases. The speed of soundin air is
around 340 rn/s although it will depend on the air
pressures are measured in pascals.
The smallest sound pressure that the average human temperature.
ear can detect is around 2 X l0- pascal (Pa). Humans Velocity (m/s) = 331 + 0.6 X temperature (°C)
Technical infonnation 133
Music I= (L.i+L2+Is+...+L)/ndB
Orchestral music in concert However, if a number of sound levels are taken in a
hall,excludingquietpassages 78—95 reverberant field then the logarithmic average is the
Discotheque music 101 appropriateaverage to take as this represents the mean
Live pop music venue 107 energy value of the field
= 10 X log10((10'P1i'0 + 10Lp2/10
Outside l0'p''°)/n) dB
Suburban streetat night 40 or
Citycentrestreetonpavement 75 = 10 X log1o(10'P1"10+ +. .
—
10Lpn/10) —10 logn dB
134 Acousticsin the Built Environment
20
0
a
4)
ci
a
C..
0
rI 15
J 4)
U) 'U
a C
aa
4)
0 4)
.4) 4 10
C.
0a
£
13
a
11
I
o
r1 E
.4) 0
ci C.
a LI-
C.
C.
0
o 0
Combination of sound from several sources: When a Decibel (dB): The decibel is the unit that is used as a
number of noise sources operate simultaneously the total measure ofa number ofacoustical quantities. The decibel
sound pressure level is is equal to 10 bels andthe bel is a logarithmic unit,strictly
the logarithm to the base ten of the ratio of two power
L = 10 X logio(10PU'lO + 10Lp2/10+ 10LPn/1O) dB related quantities. In the field of acoustics the decibel is
commonly used as the unit for sound pressure level,
L,,, L2, L,, are the sound pressure levels of the sound intensity level and sound power level. Other
individual sources. assessment units which inherently depend on these
Alternatively, the total level can be found using the quantities will also have the decibel as theirunit.
chart shown in Figure 5.2. Using this chart, the individual
noise levels must be added two at a time. For example, to Directivity factor (0: The directivityfactoris the ratio of
add 80, 82, 84 and 86dB, the procedureis: the intensity in a given direction from a source to the
(i) Difference between 82 and 80 is 2, hence the intensity in the same direction had the source radiated
correction is 2.1 dB and the total level 84.1 dB. uniformly. For a particular source Q would normally be
(ii) Difference between 86 and84dB is 2, the correction determinedby measurement in an anechoic room.
is 2.1 dB and the total level 88.1 dB.
(iii) Difference between 88.1 and 84.1 is 4dB, the Directivity index (DI): The directivityindex is defined as
correction is 1.5dB and the total level 89.6dB. DI = 10 x log10 QdB
An approximate way to add sources is where Q is the directivity factor of the source in a given
direction.
Difference (dB) Correction (dB)
0 3 Intensity level: Intensity level is often used instead of
1 3 sound intensity level.
2 2 Relation between sound power level and sound pressure
3 2 level: For a source that radiates uniformly, the sound
4
5
2
1
pressure level, L., at a distance rmetres from the sourceis
given by:
6 1
7 1 L= Lp—2OXlog10rlldB
8 1 or
9 0 L.=10Xlog10P—20Xlog40r+109dB
Technical infonnation 135
DD
4-.-
2
WI-I
nqj
>
OW
-I-il
CL
OW
.r1C
4J
Ui-I :1.
WI
L
LO
Q4J
U
0
0 5 jo
Difference Between Levels (MB)
Figure 5.2 Adding noise sources
will be the A-weighted sound level if L or P is as, for example, when measuring sound intensity directly
with an intensity probe.
A-weighted.
If the sourceradiates non-uniformly then the equations The sound intensity level should always be quotedwith
must be modified to a reference quantity, e.g. 120dB re 10_12 W/m2.
= — 20 X log10 r— 11 + DI dB Sound pressure level: The sound pressure level, L,, is
defined as
or
L,1
= 20 X log10 (p/p0) dB
= 10 X log10 P— 20 X log10 r+ 109 + DI dB
L where p is the sound pressure and p° is a reference
where DI is the directivity index of the source in the pressure which for propagationin air has the value 2 X
direction of i iO- Pa.
Hence,as 20 X log (2 X 10-/2 X 10) = 0, a sound
Sound intensity level: The sound intensity level, IL, is
defined as of2 X i-
pressure level ofzero dB is equivalent to a sound pressure
Pa.
Also, as 20 X log10 (20/2 X 10-s) = 120, a sound
L1 = 10 X log10 (I/Is) dB pressure level of 120 dB is equivalent to a sound pressure
of20 Pa. The normal range ofhumanhearingthus covers
where I is the acoustic intensity which is defined as the the range 0 to 120dB.
power passing through a unit area perpendicular to the When a sound pressure level is given it should always
direction of travel of the power. Intensity, therefore, has have an associated reference quantity, e.g. 120dB re 2 X
units of watts per square metre (W/m2). iO-5 Pa.
I
is a reference intensitywhich forpropagation in air is
chosen as 10_12 W/m2.
An increase of 3dB in the sound pressure level of a
noise is thought to be the smallest change that is
In many cases the sound pressure level and the sound subjectively definitely noticeable under normal testing
intensity level have the same numerical value for a given conditions. An increase of 10 dB on average represents a
sound and they can be used synonymously. However, doublingin loudness of the noise.
circumstances do exist where this equivalence does not
hold andthus itis better if the sound intensity level is used Sound level/noiselevel: Sound level and noise level are
exclusively fordescribing the ratiooftwo sound intensities often used instead of sound pressure level.
136 Acoustics in the Built Environment
0
113
0 B
A. C
C
0
.1.1
U
C-
C-
0
C-)
0 0 00 0 00 0 00 00 00 00
(U 113 0 0
(U 113 0 0 0 00 00
(U 113
1 (U
Frequency (Hz)
Figure 5.3 A—D weightings
Sound power level: The sound power level, LD is defined instrumentbetween the microphone and the display. The
as A-weightingattenuateslow and high frequencies relative
to 1000Hz. The standardA-weighting curve is shown in
L
= 10 X log10 (P/P0) dB
Figure 5.3 and detailed in Table 5.2. The A-weightingis
where Pis the acoustic power of the source in watts (W), the most frequently-usedweighting network, especiallyfor
and P0 is a reference sound power chosen in air to be rating environmental noise.
10-12 W
As 10 X log10 (1/10_12) = 120, 1 acoustic watt is
B-weighting:The B-weighting is similar to the A-weighting
equivalent to a sound power level of 120dB re 10_12 W except that there is less attenuationat low frequencies as
L. = 10 X +
log10 (P) 120 dB shownin Figure 5.3 andTable 5.2. The B-weightingis little
used.
Source on a reflective plane:When a source, assumed to
radiate uniformly is placed on a reflective plane, e.g. on a
concretesurface, the energy radiatedabove the plane is C-weighting: The C-weighting is essentially flat except
effectivelydoubled. The source directivity factoris 2 and below 50 Hz andabove 5000Hz as shown in Figure 5.3 and
the directivitv index is 3dB. Hence, Table 5.2. It is not often used although it has been
suggested that it should be usedto describe low-frequency
L.=L—20 X log10 r—8dB short-duration events.
or
Constant bandwidth filters: These are filters that have a
L = 10 X log10 P— 20 X log10 r+ 112dB bandwidth which is constant independent of the band
Fora source at thejunction between two reflecting planes, centre frequency.
e.g. a door in a wall on a hard ground, DI is 6dB.
Digital filters: Digital filters are computer algorithms
which filter digital signals in the same way that electrical
Weightingnetworks andfrequency bands networks filter analogue signals.
A-weighting: Human hearing is not equally sensitiveat all
frequencies. In addition, the variation with frequency is a Fast Fourier transform: The Fourier transform is a
functionof the sound pressure level. To try and account method for transposing information in the time domain
for this variation when measuring sound, electronic to information in the frequency domain. The fast Four-
weighting networks are incorporated in the measuring ier transform (FF1) is a computational algorithm which
Technical information 137
Standardfrequencies (Hz)
Octave Third-octave
Preferred centre
Band No. frequency (Hz) Band limits Centre Centre Band limits
22.39 22.39
14 25 25.12
28.18
15 31.5 31.62 31.62
35.48
16 40 39.81
44.67 44.67
17 50 50.12
56.23
18 63 63.10 63.10
70.79
19 80 79.43
89.13 89.13
20 100 100.00
112.20
21 125 125.89 125.89
141.25
22 160 158.49
177.83 177.83
23 200 199.53
223.87
24 250 251.19 251.19
281.84
25 315 316.23
354.81 354.81
26 400 398.11
446.68
27 500 501.19 501.19
562.34
28 630 630.96
707.95 707.95
29 800 794.33
891.25
30 1 000 1 000.00 1 000.00
1122.02
31 1 250 1 258.93
1 412.54 1 412.54
32 1 600 1 584.89
1 778.28
33 2 000 1 995.26 1 995.26
2 238.72
34 2500 2511.89
2 818.38 2 818.38
35 3 150 3 162.28
3 548.13
36 4000 3981.07 3981.07
4 466.84
37 5000 5011.87
5 623.41 5 623.41
38 6 300 6 309.57
7 079.46
39 8 000 7 943.28 7 943.28
8 912.51
40 10 000 10 000.00
11 220.18 11 220.18
Technical information 139
Average sound level, L, This is the same as the For example, if a source of noise level 100dB is on for
equivalent continuous sound level Leq. 0.5 h in 8 h its effective 8-h Leq will be
Correctednoise level (CNL):This is the A-weighted sound =
Leq,81i 100 + 10 log 0.5/8
pressure level which has been modified to take accountof =
any distinguishable characteristics of the noise. It is Leq,8h 88dB
defined in BS 4142. Remember tand Tmusthave the same units, whetherit is
If the noise has a noticeable tonal component, e.g. a seconds, hours or days.
whine or hiss, is impulsive or in any way is such that it The corrections to obtain the value can be obtained
draws attention to itself, then 5 dBA is added to the from the chart shown in FigureLeq 5.4.
measured sound level. If one or many distinguishing
features exist, 5 dBA is only added once.
Percentageon-time
Day/night equivalent sound level, DNL (LON): This is a 0.1 0.5 5 10 50 100
rating, based on the equivalent continuous sound level 0
Leqwhich has its origins in the USA. The acoustic energy
is averaged over a 24-h period but the noise level during
the night-time period (22:00—07:00 hours) is penalized CC
by the addition of 10dBA. 0a)
1 22 00
LDN = 10 log [— I 10LA/1O dt C
0
[24 J7 a)
r7 1
+ + '°"° dt dBA
0
J22 J
where LA is the instantaneous A-weighted sound pressure
level. dB
For effectively constant noise levels an estimation of Figure 5.4 Correctionto measured
noiselevel forpercentage
LDN can be made in the same way as the normal Leq 5 on-time
estimated (see above). LDN has found widespread accep-
tance in the USA for environmental noise assessment.
Ifthere are a numberofsources the corrections can be
Equivalent continuous sound level, L T (dB): The applied to each source in turn and then the total
continuous equivalent sound level, Leq,1' is a notional Leq i'
soundlevel. It is the soundlevel which ifmaintainedfor a obtained by decibel addition, e.g.
given length of time would produce the same acoustic Source Level On time Correction Corrected
as a noise over the same time (dBA) (h) (dBA) level
energy fluctuating period.
It is defined mathematically as 1 80 4 —3 77
L
Iii p2 (t) dtl1dB
r=lOXlogI—l
2
3
85
77
2
8
—6
0
79
77
[Ti j 8-hour Leq = 10 log (10 + iO'° + 10) = 82.5dB
wherep(t) is the acoustic pressure which varies with time; Alternatively, one may use
Tis the total time over which the Leq,TiS calculated; p° is Leq,8h = l0141,'10 x t1 + .. . + 10'-P"° — 10 log T
2 X iO Pa. L1 is soundpressurelevel which is on for time t1, etc.
If p(t) is A-weighted before the Leq,T15 calculated then So, for the above example, the 8-hour Leq
the Leq.Twill have units of dBA.
The above formula is implemented electronicallyin all
= 10 log (108 X 4 + 1085 X 2 + iO X 8) — 10 log 8
= 82.5dB.
good sound level meters and it is customary to measure
rather than calculate the equivalent continuous sound The individual noises can be all on together, on
level. It should be remembered that any value of Leq,r separately,or overlap.Itwill makeno difference to the Leq
should be accompanied by the time over which it was level.
measured. Leq, T is widely used to measure any environ-
mental noise which varies considerably with time. Maximum sound pressure level, This is the max-
imum value of the sound pressure level that occurs during
Estimation of the equivalent continuous sound level: It is
any given period.Itsvalue will depend upon the frequency
possibleto estimate the equivalent continuous sound level weighting and meter time characteristic. The maximum
for a source or numberof sources if theyhave effectively slowA-weightedsound level during an aircraft flyover is
constantnoise levels over known periods of time. used in the assessment of aircraft noise (see 'perceived
Ifa sourceofnoise level, L,, is on for a period of time noise level').
then the Leq value, over a period T where T is greater
than or equal to t is Minimumsoundpressurelevel, l.:This is the minimum
soundlevel that occurs duringany given period.It is little
Leq, r L1, + 10 log t/TdB used in assessment procedures, comparedto L10.
=
140 Acoustics in the Built Environment
Noise andnumber index, NNI: This is an indexwhich was where p(t) is the sound pressure which varies with time
used in the UK for rating aircraft noise until 1990. It is over the period T The soundpressureis oftenA-weighted
defined as before the SEL is calculated.
The sound exposure level is related to continuous
NNI = averagepeak perceived noise level +
equivalent level by
15 logN-8O
1-Seq. r
= SEL — 10 log T dB
where Nis the number of aircraft having a PNL greater where T is the time in seconds over which the
than 80 PNdB in the specifiedperiod.The average PNL is Leq is
required.
given by If identical SEL values occur in the time 7 the Leq is
N 1 given by
10 log I—
IloL/N I dB 4q,T = SEL — 10 log T+ 10 log NdB
If the events are not identical then the 4q equivalent of
each event must be found and then the total
whereL is the peak value of PNL during the passage of calculated
the ith aircraft. by adding the individual Leq values as decibels
are added.
NNI has been replaced by 'tq, P
An approximate relationship between NNI and the 16-h Statistical level, L1: This is the sound pressure level that is
A-weighted Leq is exceeded for 1% of the measurement time. It gives an
indication of the maximum sound levels that occur.
NNI 16-h Leq (dBA)
35 57 Statistical level, L10: This is the sound pressure level that is
45 63 exceeded for 10% of the measurement time. Conse-
55 69 quently it isindicative ofthe higher levels that occurin the
Peak sound pressurelevel: This is the sound pressurelevel measurementperiod.
In the UK the A-weightedL10 value is used to measure
corresponding to the peak sound pressure that occurs in and assess traffic noise. For fairly noisy traffic
anygiven period.Its valuewill dependuponthe frequency
weighting and the time characteristics of the measuring
system.
L Leq + 3dB
It is often usedto quantif' short durationimpulses, e.g. Statistical level, L90: This is the sound pressure level that is
gunfire, explosions and high level impact noise. To exceeded for 90% of the measurement time. Conse-
measure the true peak sound pressurelevel of an event, quently it is indicative of the general ambient noise level
great care must be taken in the selection and use of in the absence of any higher level short-duration events
that occur during the period.
microphone and measuring system.
Perceived noise level, PNIL: The perceived noise level is a
The A-weighted L value is often used as a measure-
ment of background noise in environmental assessment.
rating for single aircraftflyovers based originally on jury
judgements of perceived noisiness, but it is now com- Sound insulation
monly derived by an extensive calculation procedure. sound insulation refers to the
Sound insulation: Airborne
Formost generalpurposes the perceived noise level can
be obtained from measurements of the maximum slow process separating, by physical barrier, a space to be
of a
protected from a space containing a noise source. With
A-weighted sound level that occurs during a flyover. noise insulation the sound is effectively prevented from
travelling in a specific direction by an impervious barrier.
PNL = dBA + 13 PNdB. The greater the surface mass of the barrier, the greater
the insulation will generally be. Unlike soundabsorption,
Single event noise exposure level, SENIELor L_: This is sound insulation does not remove energy from the sound
equivalent to the sound exposure level. In the original field; it merely redirects it.
definition it was assumed that the sound pressure was
A-weighted andthat the integration was over the time that Sound reduction index, Sm: The sound reduction index
the signal was within 10 dB of the maximum value. is the measurement generally used to express the insula-
tion propertiesof a partitionin decibels. It is definedas
Sound exposure level, SF1.. or L: The sound exposure
level is a notional level. It is the sound level that if SRI = 10 log10 (1/transmission coefficient) dB
maintained constant for 1 s contains the same acoustic
The soundreduction indexis frequency dependentandis
energy as a varying noise level. usually measured in octave or one-third octave bands.
It is normally used to quantify short duration noise If the transmission coefficient T = 0.01, i.e. 1% of the
events such as aircraft flyover, single vehicle bypasses, incidentsoundenergy is transmitted by the partition, then
impact or impulsive noise. the sound reduction index is 20dB, while if'r = 0.001 the
It is defined mathematically as SRI is 30dB, etc. Hence for 50 dB insulation, which, for
p2(t) dtdB example, may represent a reasonable reduction between
SEL=lOXlogf 2 flats, the incidentintensity must be reduced by a factor of
Jo Po 0.000 01.
Technical infonnatio.i 141
The octave band soundreductionindex,R is obtained Sound level difference between two spaces: The sound
from the one-thirdoctave band indices R1, R2 and R3 as level difference between two spaces separated by a
follows (see addition of attenuations, part (b)) partitiondepends upon the value ofthe sound reduction
R = 10 log 3 — 10 log [10_Riiio + 10o + 10/1oI dB index, the area of the partition and the acoustic proper-
ties of the two spaces.
R = 4.77 — 10 log[l0h/10 + 10?/o + l0-''°I dB
(a) room-to-room
R = 4.77 + C
= L1 — R + 10 loglo (S/A) dB
Figure 5.1 can be used to find C. Lp2
This addition assumes equal energy in all three octave
(b) inside-to-outside
bands.
— R+ 10
Sound transmission coefficient, T: The sound transmis- L2 = log10 5—20 loglo r— 17 +
sion coefficient is a measure ofthe incidentsound energy DI dB
that 'passes through' a wall, partitionor any barrier. The
sound doesnot actuallypass through the barrier. Incident (c) outside-to-inside
sound energy causes the barrier to vibrate and the =
vibrating barrier then radiates sound into the receiving L1—R+10log10(S/A)—K+6dB
space. The sound transmission coefficient is defined as is the soundpressure level on the sourceside (dB),
L2 is the sound pressure level on the receiver side
Intensity incident upon a partition (dB),
Intensity transmitted by partition S is the area of the partition (m2),
R is the sound reduction index of the partition (dB),
Transmission loss A is the absorption in the receiving room (m2),
Coincident effect critical frequenc When sound waves DI is the directivity index of the facade,
strike a partition, bending waves are excited in it, the r is the distance of the receiver from the partition.
velocity of which depends upon the frequency. As the K is a constant, the value of which
frequency of the bending wave increases, the bending external to the depends on where,
wavevelocity increases and at some frequency, known as partition, the sound pressure level was
the critical frequency,it is equal to thevelocityofsoundin measured:
air. At this frequency the wavelengthof the bendingwave K = 6 dB if measured very close to the partition,
is equal to the wavelength of the sound in air and if the K = 2.5 dB if measurementposition is about 1 m,
sound wave impinging on the partition is of the same K = 0 dB if measurement position is far from facade.
frequency, resonant excitation can occur and the sound
transmission of the panel is increased and the sound Sound transmission class, STC: This is a single figure
reduction index decreased. descriptor used for rating the sound transmission of a
Matching of the wavelengths at the critical frequency partition obtained by laboratory measurements. The
occurs at grazing incidence and little energy is actually measured sound reduction indices in one-third octave
transferred to the partition. However, as the frequency bands are comparedwith a set of standard curves and at
increases above the critical frequency, matching occurs at each frequency the difference between the measured
increasing angles ofincidence and energy transference to valuesand the standardcurve valuesis obtained. The STC
the partitionis significantlyincreased. value corresponds to that curve for which the sum of the
The reduction in performanceof the panel is known as deficiencies is less than or equal to 32 dB and the
the coincidence effect and the performance of the maximum deficiency at any one frequency is less than
partitioncan be significantlyreducedat frequencies in the 8dB.
regionofthe critical frequency althoughat an octaveorso The reference curve for an STC value of 52 is
beyond the critical frequency the panel will again
approachits expected performance. Frequency (Hz) Level (dB)
125 36
Field sound transmission class, FSTC: This the STC 160 39
obtained from values of the sound reduction index 200 42
measuredunder field conditions. 250 45
flanking transmission: The transmission of sound energy 315 48
via paths which bypass a partition is known as flanking 400 51
transmission. 500 52
630 53
Noiseisolation class, MC: This is the STC obtainedfrom 800 54
values of the sound level differences measuredbetween 1 000 55
two spaces under laboratory conditions. 1 250 56
Normalizednoise isolation class, NNIC: This is the STC 1 600 56
obtained from values of the sound level differences 2000 56
measured between two spacesunderfield conditions. The 2500 56
sound levels in the receiving room are corrected to a 3150 56
standardreverberation time of 0.5s. 4000 56
142 Acoustics in the Built Environment
The value of the STC is equal to the dB value of the curve correctedto a standard receiving room reverberation time
at 500 Hz. Other standardcurves are obtainedby moving of 0.5 s. Hence
this curve up or down in increments of 1 dB.
DnT = measured level difference + 10 log T/0.5 dB
SRIofcomposite construction: A composite construction T = reverberation time in receiving room.
is one havingareas with different sound reduction indices,
Structureborne sound: Sound which travels from one
e.g. a wall with windows.
For a facade consisting of areas S1, £2 S with space to another not through the air but through the
sound reduction indices R1, R2 fabric ofthe
R,1, the value of the is one formbuilding
is known as structurebornesound. It
sound reduction index for the facade is given by of flanking transmission. Structureborne
sound can travel long distances with little attenuationand
be re-radiated, causing problems far from the original
S1 + + ... + S, source of noise.
R= 10 log10 i
Lb0R1b0 x S1 + 1o-R10 X £2 Transmission loss, TL: Transmission loss is an alternative
name for the sound reduction index.
+. + . X Sn
Weighted apparent sound reduction index, R: This is
Il
similar to R.but is usedifit is thoughtthat the value of
The composite sound reduction index may also be was obtained with flankingtransmission.
obtained using Figure 5.5, taking two areas at a time. sound level difference, D: The weightedlevel
Small areas of very low insulation can drastically Weighted
difference is obtainedfrom level differences, measured in
reduce the overall performance of a facade, e.g. an one-third octave bands, in
exactly the same way as the
opening of area 0.1 m2 and SRI 0dB in a facade of area airborne sound insulation index
25 m2 and SRI 50 dB reduces the overall SRI value to 50 rating, R, is obtained.
— 26 = 24dB.
Weighted sound reduction index, R: This is a weighted
single figure descriptor of the sound insulation perform-
Standardized sound level difference, DflT: The standar- ance of a partition measured under laboratory condi-
dized sound level difference is used to assess airborne tions. The sound reduction index in each of the one-
,
sound insulation between rooms in buildings. As the third octave bands from 100Hz to 3150Hz is compared
sound level difference across a partitionwill depend upon with a standard set of curves. The value of I? for a given
the absorption in the receiving room it is recommended partition is obtained from the standard curve which
(BS 5821: 1984) that the measured level difference is when comparedwith the measured SRI values produces
,z
60 1.0
1.
50 .1
0
.lJ
U
OH
40 .01.
.IJW
43
JL
1ii1:
(flW
00) 30 • 001
0
I
4E0 20 .0001
JIj
rl
0
> 1.0
0
0 10 20 30 40 50 60
Difference in Sound Reduction
Indices (Higher — Lower)
Figure5.5 Sound reduction indices loss
Technical information 143
an adverse deviation as close to —32dB as possible. Only Direct sound field: The direct sound field refers to the
the SRI values which fall below a particular standard acoustic energy that arrives at a listener directly from the
curve are considered in the sum. Positive deviations source without any reflections from nearby surfaces or
from the standardcurve are not taken into account. The objects.
standardvalues for the curve corresponding to an R of
52 are Early decay time, EDT: This is the time in seconds,
multiplied by six, which the sound in an enclosure takes to
Frequency (Hz) Reference value (dB) decay by 10dB from its equilibrium value. It is thoughtto
100 33
be important in determining the quality of auditoria,
125 36 especiallyfor music. The EDT is sometimes referred to as
the subjective reverberation time.
150 39
200 42 Noise reduction coefficient, NRC: This is the average, to
250 45 the nearest multiple of0.05, ofthe absorption coefficients
315 48 measured in the octave bands centred on 250, 500, 1000
400 51 and 2000Hz.
500 52 Normalroom modes: Sound waves in an enclosure travel
630 53 around in all directions being reflected obliquely off the
800 54 walls. Some paths repeat themselves continuously, form-
1 000 55
1 250 56 ingwhat are known as normal modes.
The normalmodesoccurat specificfrequencies related
1 600 56 to the dimensions of the room. For a rectangular room
2000 56 the frequencies are given by
2500 56
\1i (2
(2
3150 56
fn = 1/ + + (n2
The R value is the value in decibels of the reference \'i
curve at 500 Hz.
To obtain other reference curves the one-third octave
band values are changedin 1 dB steps up or down.
2 V
where c= velocity ofsound (mis),
dimensions (m), and n,
\LLJ
n.7 and
, ç and are the room
n7 can independently
Reverberation time: The time which is taken for the Sound absorption coefficient: The sound absorption
reverberantsoundenergy in an enclosure to decay to one coefficient is the quantity used to describe how well a
millionth of its equilibrium value, i.e. by 60dB, after the particular material absorbs sound. It is denoted by a and
source is turned off, is known as the reverberation time. is defined as
The reverberation time is frequency dependent and it is
customary to measure its value in octave or one-third Sound energy not reflected from material
octave bands. There are a number ofsimple equations for Sound energy incident upon material
predicting reverberation times.
Room radius: For a source operating in an enclosure For a perfect absorber a would have a value of 1 while for
there are two soundfields, the reverberant and the direct, a perfect reflector a would equal zero.
and the value of the sound pressure level at any point is The absorption coefficient varies with frequency and
the sum, that is the decibel sum, of the direct and also with the angleatwhich the sound strikes the material.
reverberantsoundpressurelevels. Far from the source the Because ofthe angulardependenceit is usual to measure
reverberantfield will dominatewhile close to the source the absorption coefficient of materials in diffuse sound
the direct field will be greatest. fields so that sound effectivelystrikes the material at all
angles of incidence. The absorption coefficient measured
The distance, r, from the source where the direct and
reverberant sound pressurelevels are equal is known as
the room radius and it can be found from
under theseconditions is known as the randomincidence
sound absorption coefficient and is denoted by It is
usually measured in one-third octave or octave bands.
.
There is no accepted way of obtaining the octave band
r=-J value from the one-thirdoctave band values. The average
y l6lT of the three values or the highestvalue are both used.
where Q = source directivity factor, and A = room
Assessinginternal spaces
absorption. Articulation index, Al: The articulation index is a weigh-
Sabine equation: The Sabine equation gives the reverbera- ted fraction representing, for a given speech channeland
tion time in terms of the room volume and total room noise condition, the effective proportion of the normal
absorption as speech signal that is available to a listener for conveying
0.161 V speech intelligibility.It is obtainedfrom measurements or
T= s
estimates of the speech spectrum and of the effective
A
if Vis in cubic metres and A in square metres.
The equation is valid for diffuse sound fields only and UI
80
gives the best results when the average absorption a
coeffficient is less than 0.1. However,it is often usedwhen
this condition is not met. For large enclosures air 0'I 70
absorption is included so that x
0.161 V UI
L 60
T=
A + 4 mV U)
;;—;
:::
—
L S — — —
JR too
curveswhich have been proposedas an update ofthe NC to S S S —
— —
NA go
curves. They are not asyet generally accepted. Details can
be found in a paper by Beranek, Balanced Noise Criterion
Curves,j Acoust.Soc.Am., 86(2),August 1989,pp.650—664.
'-I
>
80
70 \ ,,,,
S
c :::_ — —
NA .5
•0
NP 75
go
10 60
Noisecriterion, NC: The conceptofthe noise criterionwas
s _ .. ..--
,. .'.r 70
NR 05
originally developed in the United States specificallyfor
It
( 50 S
.
— —
—
•0
application in commercial buildings. Its calculation is
—
—
S
S — — —
NA 55
based onan octaveband analysis ofa noise andreference is 40
- ,-.,4 50
madeto a setofcurves whichare shown in Figure 5.6. II S S S
•. - NA 45
—
S
—
-
30 -
: NI 40
The noise criterion is obtained by plotting the octave S S — — —
NA 35
band soundpressurelevelsonto the reference curves and 20 NA 30
determining the lowest curve which is nowhere exceeded 10
NA 25
by the plotted octave band levels. NA 20
D NA 15
Noise rating number, NR: The noise rating number is a c 0 NA 10
U
single figure index obtainedfrom an octave band analysis to —10 NA 5
of a noise. To obtain the NR number the octave band
sound pressure levels are plotted onto a set of reference
) I
L) (. II U U 0 II a.) a) 0 0 0 0 0 0 0 0 0 0 0 0
II•Eewoo'coe.,oopoonooonoooo
NA 0
1
curves which are shown in Figures 5.7 and 5.8. The In
highestNRcurve that is intersected by the curve forming Third Octave Band Centre Frequency (Hz)
the plotted sound pressure levels gives the noise rating
number. The octave band sound pressure levels are Figure 5.8 Noiserating curves (one-thirdoctave band)
normallyjoined by straight lines.
r1
a
>
a
50
'I II PNC65
PNC no
PNC 55
PNC
Impact noise level, L: This is the sound pressure level,
measured in a one-third octave band, when a standard
tapping machine is operating on the floor above the
room.
OJ
40 directly on a structure.
L The frequency content of the soundwill depend upon
U: PNC 40 the duration of the impact; a short sharp event giving a
ao PNC n
broadband frequency content while a longer duration
L
a event caused, for example, by having a resilient layerover
C
PNC 30 the structure, will contain mainly low frequency sound
C 20 and will be subjectivelyless disturbing.
PNC25
0
U)
PNC20 Normalized impact noise level, L,: The impact noise level
10 as measured in the laboratory will depend upon the
C PNC is acoustic characteristics of the receiving room so to
B
normalize results the measured noise levels are corrected
to a constant10 m2 of absorption.
0
Ui
.-I
(1
('1
U)
l Ui
(U
0
Ui
(U
0
0
U)
0
0
0
•1 (
0
0
0
0
0
0
0
0
0
B
Hence I = L— 10 log A/A0
A = actual sound absorption in the receiving room in the
Octave Band Centre Frequency (Hz)
one-thirdoctave band under consideration.
5.9 noisecriterioncurves
Figure Preferred A,1, = 10m2
L'0 is used if flanking transmission cannot be
eliminated.
Speech transmission index, STI: The speech transmission See BS 5821, Part2.
indexquantifies the effect ofa sound transmission system
on speech intelligibility.It is based upon an analysisof the Resonant frequency, J: The resonant frequency of an
reduction in intensitymodulationofa signal which occurs isolator of stiffness K (Nm1) which supports a mass M
along the transmission path from source to receiver. The (kg) is
analysis is carriedout for 8 octavebands ofnoise, typically
125 Hz to 8kHz, and 14modulation frequencies. The
results of this analysis are then combinedandnormalized f=—/— 2'rr V M
.Hz
to give thespeech transmissionindexwhich hasavalue of1
for perfect transmission and a value of 0 for no signal This is often re-written as
recognition.
Relationshipsexist between the STI values,the signal-to- 15.8
noise ratio and the reverberation time of an enclosure, r fl-
\/ d
allowing theoretical calculations to be made of the STI
value. This is most useful at the design stage ofaproject. dbeing the static deflection (inmm) ofthe isolator when
Details may be found in Bruel and Kjaer Technical the mass Mis placed on it.
Review 3. For rubber isolators the constantbecomes 19.5.
For an isolator to be effective its resonant frequency
Impact soundand vibration must be at least three times lower than the lowest
Impact insulation class, IIC: This is a single number frequency to be isolated.
rating, used in the United States, which permits easy
comparison of the impact isolation performanceoffloor/ Standardized impact sound level, LT: This is the impact
ceiling assemblies. Impact sound levels normalized to a sound level measured between two rooms under field
room absorption of 10 m2 are compared with a set of conditions and standardized to a reverberation time of
standardcurves to producethe impactinsulation class. 0.5s,
The calculation is exactlythe same as that described to
determine the weighted normalized impact sound pres- T
sure level, L,, except that in additionto the total adverse i.e. L'OT = L' — 10 log
0.5
deviation being less than or equal to 32dB, no one
deviation must exceed 8 dB. L' = measured impact noise level.
Technical infonnation 147
See BS 5821, Part 2: 1986 and BS 2750, Parts 6 and 7: Acceleration is what is commonly measured to quantifr
1980. the vibration of a surface.
Standard tappingmachine: A standardtapping machineis Vibration displacement, d: When an object vibrates, its
used to rate the impact noise isolation of floors. The surface will oscillate about its stationary position. These
machinehas five hammers, each of mass 0.5kg, equally
changes represent the vibration displacement.
spaced along a line. The hammers are dropped, from a
height of 4cm, successively to give 10 impacts per Vibration isolation: Vibration isolation is a means of
second.
reducingthe transmission of vibrating motions or forces
Transmissibility:The effectivenessofa vibration isolator is from one structure to another. It is usually achieved by
measured in terms of its transmissibility. Two types of separating the two structures by an elastic element,known
transmissibilityare generally defined: (1) Force transmissi- as a vibration isolator.
bility which is the ratio of the force transmitted by the
isolator to the force applied to the structureon top ofit; Vibration velocity, v The velocityof a vibrating surface is
(2) displacement transmissibilitywhich is the ratio of the related to the displacement by
displacement transmitted by the isolator to the displace- v=
ment appliedat its base. In both cases the transmissibility,
7 is, for lightly dampedsystems, given by v(m/s) = d(m) X o(rad/s)
wherew = angular frequency.
T=
J(f/f)2 Weightednormalized impact sound level,
a obtained
I, L: This is
from one-thirdoctave
wherefis the frequency of vibrating motion andJr is the
resonant frequency of the isolator together with the
single figure descriptor
values of the normalized impact sound levels L, (labo-
ratory) or L' (field). The normalized levels are compared
structure mounted on it. The variation of the transmissi- with a set of weighting curves and the curve found for
bility with frequency is shown in Figure 5.10. which the total adverse difference between the normal-
Vibration acceleration, a: The acceleration of a vibrating ized levels and the curves is less than but as close to 32dB
surface is related to the displacement and velocityby as possible. Adverse differences occur when the normal-
ized levels fall above the rating curve.
a = dw2 = The weighted normalized impact soundpressure level is
the sound pressure level at 500 Hz on the standardcurve,
jo which meets the above criterion. For an LNW or L'NW of 60
the curve is defined by
iJ \
>' 5 Frequency Level (dB)
4J 100 62
125 62
160 62
2 200 62
m
m 250 62
E 1 315 62
m
C 400 61
ii
C. 500 60
I- .5 630 59
mm 800 58
C
m
E
ioo 57
.2 1250 54
U
m 1 600 51
Ii
ri
a 2000 48
U
.1
0
C
.1
N .5 2500
3150
45
42
0 05 Other curves are obtainedby moving the one-third octave
S
U
band values up or down in 1 dB steps.
C- See BS 5821, Part 2 and BS 2750, Parts 6 and 7.
0 .1
Li.
.02
.01
______
05 ,
Weightedstandardized impact soundlevel, L',1. Thisis a
single figure descriptor obtained from one-third octave
0 band values of the standardized impact sound levels,
.1 .2 .5 1 2 5 10
L'IIT, w
Ratto of Forcing Frequency It is obtained in exactly the same way as the weighted
to Natural Frequency normalized impact sound level, L.,.
Figure 5.10 Transmissibility See BS 5821, Part 2 and BS 2750, Parts 6 and 7.
148 Acoustics in the Built Environment
Equivalent Standards
Assessinge,wironmental noise
BS 4142: 1990 ASTM E1014—84 ISO 1966: 1986 ANSI S3.23: 1980
BS 5228: 1984/1994 ISO 2204: 1979 ANSI S12.4: 1986
ISO/DP 7196
DIN 18005: 1987
DIN 45641: 1976/1987 NF S30—010: 1974
DIN 45642: 1974 NF S30—008: 1984
DIN 45643: 1984 NF S31—010: 1987
DIN VDI 2714: 1988 NF S31—050: 1987
DIN YDI 3723: 1982 NF S31—110: 1985
DIN VDI 2718: 1975
Measurement ofimpactinsulation
BS 2750: 1980 ASTM E492—77 ISO 140: 1980
ASTM E1007—84
DIN 52210: 1984
NF S31—052: 1979
NF 531—053: 1979
NF S31—056: 1982
Measurement ofsoundpower
BS 4196: 1986 ASTM E1124—86 ISO 3740: 1980 ANSI 81.23: 1976
ISO 3741: 1988 ANSI 81.30: 1985
ISO 3742: 1988 ANSI S1.31: 1986
ISO 3743: 1988 ANSI S1.32: 1986
ISO 3744: 1981 ANSI S1.33: 1982
ISO 3745: 1977 ANSI S1.35: 1985
ISO 3746: 1979 ANSI S1.36: 1979
ISO 3747: 1987 ANSI S2.34: 1988
DIN 45635: 1985 NF S31—025: 1977
NF S31—026: 1978
NF S31—027: 1977
NF S31—022: 1989
NF S31—024: 1989
NF 831—067: 1986
Sound jorcementsystems
BS 6259: 1982 IEC 268: 1985
BS 6840: 1987
DIN 45589: 1979
Technical informabon
Speech andnoise
ISO/TR 3352: 1974 ASTM E1041—85 ANSI S3.2: 1960
ASTM E1110—86 ANSI S3.5: 1969
ASTM E1130—88 ANSI S3.14: 1977
DIN 18041: 1968
DIN 45621: 1973 NF S31—047: 1975
NF S32—001:1975
NFXS5—108: 1987 IEC 84(CO)2: 1986
-
Vibration measurementandresponse
BS 6177: 1982 ISO 2017: 1982 ANSI S3.18: 1979
BS 6472: 1984 ISO 2631: 1985 ANSI S3.29: 1983
BS 6841: 1987 ISO 5805: 1981 ANSI S3-W-39
BS 6611: 1985 ISO 6897: 1984
ISO 7849: 1987
ISO 4866: 1986
ISO 8569: 1989
DIN 4150: 1986
DIN 45669: 1981/1989 VDI 2057: 1987
Mechanical services
BS 848: 1985 ASTM E477-84 ANSI S12.11: 1987
BS 4718: 1971
DIN 45646: 1988 NF E51—701: 1980
DIN VDI 2081: 1983 NF P50—402: 1985
DIN 45635: 1986 NF E51—706: 1988
DIN VDI 3731: 1982 NF S31—021: 1982
NF S31—046: 1988
Soundinsulation in buildings
BS 8233: 1987 ISO/DIS 6242: 1989
DIN 4109: 1989
DIN 18165: 1987
DIN VDI 2569: 1990
DIN VDI 2571: 1976
DIN VDI 2711: 1978
DIN VDI 2719: 1987
DIN VDI 3728: 1987
Acousties qrbuildings
NF P90—207: 1987
Instrumentation
BS 2475: 1964 IEC 196: 1965 ISO 266: 1975 ANSI S1.4: 1983
BS 3593: 1986 IEC 225: 1966 ANSI S1.4: 1985
BS 5969: 1981 IEC 651: 1979 ANSI S1.6: 1984
BS 6698: 1986 IEC 804: 1985 ANSI S1.11: 1986
ANSI S1.13: 1971
DIN 45401: 1985 NF S30—002: 1972 ANSI S1.42: 1986
DIN 45651: 1964 NF S31—109: 1983
DIN 45652: 1964
Terminology
BS4727: 1985 ASTM C634-86 ISO 31—7: 1978 ANSI S3.32: 1982
BS 5775: 1979 ISO 131: 1979
IEC 50(801): 1984
DIN 1320: 1990 NF S30—004: 1966
DIN 45630: 1971 NF S30—101: 1973
DIN 52217: 1984 NF S30—102: 1973
NF S30—103: 1973
NF S30—106: 1975
NFXO2—207: 1985
150 Acousticsin the Built Environment
InternationalElectrotecilnicalCommission(IEC)
lEG 50 Internationalelectrotechnical vocabulary.
801: 1984 Vocabulary: acoustics and electoacoustics.
lEG 84 (GO) 2: 1986 sound system equipment; report on the RASTI method for the objective rating of
speech intelligibilityin auditoria; (Central Office) 2.
lEG 196: 1965 lEG standard frequencies.
lEG 225: 1966 Octave, half-octaveandthird octaveband filters intendedfor the analysis ofsoundsand
vibrations.
lEG 268 Sound system equipment
Part 1: 1985 General
Part 2: 1971 Explanation of general terms
Part 4: 1972 Microphones
Part 5: 1989 Loudspeakers
Part 7: 1984 Headphones
Part 16 Draft Report on the RASTI — method for the objective rating of speech intelligibility in
auditoria.
lEG 651: 1979 Soundlevel meters
lEG 804: 1985 Integrating —
averaging sound level meters.
ISOStandards
ISO 31—7: 1978 Quantitites and units of acoustics.
ISO 131: 1979 Acoustics expression of physical and subjective magnitudes of sound or noise in
—
air.
— measurement of sound insulationin
ISO 140 Acoustics buildings and building elements.
Part 1: 1978 Requirements for laboratories.
Part 2: 1978 Statementof precision requirements.
Part 3: 1995 Laboratory measurements of airborne sound insulation of building elements.
Part 4: 1978 Field measurements of airbornesound insulation between rooms.
Part 5: 1978 Field measurements of airbornesound insulation of facade elements and facades.
Part 6: 1978 Laboratory measurements of impact sound insulationoffloors.
Part 7: 1978 Field measurements of impact sound insulation of floors.
Part 8: 1978 Laboratory measurements of the reduction of transmitted impact noise by floor
coverings on a standard floor.
Part 9: 1985 Laboratory measurements ofroom-to-room airborne sound insulation ofsuspended
ceiling with a plenum above it.
Technical information 151
ISO/TR 7849: 1987 Acoustics — estimation of airborne noise emitted by machinery using vibration
measurement.
Iso 8569: 1989 Mechanical vibration: shock-and-vibration-sensitiveelectronicequipment; methods of
measurement and reportingdata of shock and vibration effects in buildings.
ISO 9296: 1988 Acoustics — declared noise emission values of computer and business equipment.
GemianNational Standards
DIN 1320: 1990 Acoustics: terminology.
DIN 1800: 1987 Teil 1. Noise abatementin town planning; calculation methods.
Teil 1, Beiblatt 1. Noise abatement in town planning; acoustic orientationvalues in
town planning.
DIN 1804: 1968 Acousticalquality in small to medium size rooms.
DIN 18165: 1987 Teil 2. Fibre insulatingbuilding materials: impact sound insulating materials.
DIN 4150: 1986 Teil 3. Structural vibration in buildings: effects on structures.
DIN 4109: 1989 Sound insulation in buildings: requirements and verifications.
Beiblatt 1 Soundinsulation in buildings: construction examples and calculation methods.
Beiblatt 2 Soundinsulation in buildings: guidelines for planningandexecution; proposals for
increased sound insulation; recommendations for sound insulation in personalliving
and working areas.
DIN 45035: 1980 Teil 14. Noise measurement on machines: airborne noise measurement, enveloping
surfacemethod, air cooled heatexchangers (air coolers).
DIN 45401: 1985 Acoustic, electroacoustic: standard frequencies for measurements.
DIN 45589: 1979 Requirements for congress microphones.
DIN 45621: 1973 Teil 2. Word listsfor intelligibilitytest.
Teil 2. Sentence lists for intelligibilitytest.
DIN 45630: 1971 Teil 1. Physical and subjective magnitudes of sound.
DIN 45635: 1985 Teil 3. Measurement of airborne noise emitted by machines: engineeringmethod for
special reverberation test rooms.
DIN 45635: 1986 Teil 38. Measurement of noise emitted by machines; airborne noise emission;
enveloping surface method, reverberation room method and indirect methods; fans.
DIN 45635: 1985 Teil 46. Measurement of noise emitted by machines; airborne noise emission;
enveloping surface method; cooling towers.
DIN 45635: 1986 Teil 56. Measurement of noise emitted by machines; airborne noise emission;
enveloping surface methodand indirectmethod;fan assistedwarmair generators, fan
assisted air heaters and fan units of air handlingdevices.
DIN 45641: 1976 Averaging of time varying sound level; ratinglevels.
DIN 45641: 1987 Averaging of sound levels; single eventlevel.
DIN 45642: 1974 Measurement of traffic noise.
DIN 45643: 1984 Teil 1. Measurement and assessment of aircraft noise; quantities and parameters.
Teil 3. Measurement andassessment of aircraft noise; determinationofrating level of
aircraft noise exposure.
DIN 45646: 1988 Measurement procedures for ducted silencers; insertion loss, transmission loss, flow
noise, total pressureloss.
DIN 45654: 1964 Octave filters for electroacoustical measurements.
DIN 45652: 1964 Third octave filters for electroacoustical measurements.
Technical infonuation 153
DIN 45669: 1981 Teil 1. Measurementof vibration emission; requirements on vibration meter.
DIN 45669: 1989 Teil 2. Measurement of vibration emissions; measuring method; amendment1.
DIN 52210: 1984 Teil 1. Tests in building acoustics; airborne and impact sound insulation; measuring
methods.
Teil 2. Tests in building acoustics; airborneand impact sound insulation; laboratories
for measuring of the sound reduction of building elements.
DIN 52210: 1987 Teil 3. Testing of acoustics in buildings, airborne and impact sound insulation;
laboratory measurements of sound insulation of building elements and field
measurements between rooms.
DIN 52210: 1984 Teil 4. Tests in building acoustics;airborneand impact sound insulation; determination
of single-number quantities.
DIN 52210: 1985 Teil 5. Testing in building acoustics; airborne and impact sound insulation; field
measurements of airborne sound insulation of exterior building elements.
DIN 52210: 1989 Teil 6. Testsin building acoustics;airborneandimpactsound insulation; determination
of the level difference.
DIN 52210: 1989 Teil 7. Tests in building acoustics; airborneandimpact sound insulation; determination
of the lateral sound reduction index.
DIN 52212:
1961 Testing of architectural acoustics; measurement of sound absorption coefficient in a
reverberation room.
DIN 52215: 1963 Testing of architectural acoustics; determinationof sound absorption coefficient and
impedance in a tube.
DIN 52216: 1965 Testing of architectural acoustics; measurementof reverberation time in auditoria.
DIN 52217: 1984 Test in building acoustics; flanking transmission; terms and definitions.
VDI 2057 Blatt 4.1: 1987 Effect of mechanical vibrations on human beings; measurements and assessment for
workshop places in buildings.
VDI 2081: 1983 Noise generation and noise reduction in air-conditioning systems.
VDI 2566: 1988 Noise reduction on lifts.
VDI 2569: 1990 Sound protection and acoustical design in offices.
VDI 2571: 1976 Sound radiation from industrial buildings.
VDI 2711: 1978 Noise reduction by enclosures.
VDI 2714: 1988 Outdoor soundpropagation.
VDI 2718: 1975 Noise abatementin town planning.
YDI 2719: 1987 Sound isolation of windowsand theirauxiliary equipment.
VDI 3720 Blatt 1: 1980 Noise abatementby design; general fundamentals.
VDI 3720 Blatt 2: 1982 Noise abatementby design; compilation of examples.
BDI 2723 Blatt 1: 1982 Application of statistical methods for the description of variating ambient noise
levels.
VDI 3728: 1987 Airborne sound isolation of doors and movable walls.
VDI 3729 Blatt 1: 1982 Characteristic noise emissionvalues oftechnical sound sources; office machines; basic
directions.
VDI 3729 Blatt 2: 1982 Characteristic noise emission values of technical sound sources; office machines,
typewriters.
VDI 3729 Blatt 3: 1982 Characteristic noise emission values of technical sound sources; office machines,
duplicating machines and copiers.
VDI 3729 Blatt 5: 1982 Characteristic noise emission values of technical sound sources; office machines; mail
processing (preparation) machines.
VDI 3729 Blatt 6: 1990 Characteristic noise emissionvalues of technical sound sources; computer andbusiness
equipment; computer.
VDI 3731 Blatt 1: 1982 Characteristic noise emission values of technical sound sources; compressors.
VDI 3731 Blatt 2: 1988 Characteristic noise emission values of technical sound sources; fans.
154 Acoustics in the Built Environment
French Standards
NF E51—701: 1980 Controlledmechanical ventilation components. Code for aerodynamic and acoustic
testing of extract air terminaldevices.
NF E51—706: 1988 Controlled mechanical ventilation components. Code for aerodynamic and acoustic
testings of extraction units for private houses. Simple flux.
NF P05—321: 1986 Performance standardin building. Presentation of the performances of facades made
ofcomponents from the samesource.
NF P50—402: 1985 Ventilation components. Code for aerodynamic and acoustic testing of facade air
inlets.
NF P90—207: 1986 Sporthalls. Acoustics.
NF S30—002: 1972 Acoustics.Standard frequencies for acoustic measurement.
NF S30—004: 1966 Acoustics.Expressing the physical andpsychophysiologicalproperties of a sound or a
noise.
NF S30—008: 1984 Acoustics. Guide to standards on the measurement of airborne acoustical noise and
evaluation of its effects on human beings.
NF S30—010: 1974 Acoustics. NR curve for the assessment of noise.
NF S30—101: 1973 Acoustics.Terminology: generaldefinitions.
NF S30—102: 1973 Acoustics. Terminology: transmission systems and transducers for sound and
vibrations.
156 Acoustics in the Built Environment
British Standards
BS 648: 1964 Schedules ofweights ofbuilding materials.
BS 848 Fans for general purpose.
Part2: 1985 Methods of noise testing.
Part6: 1989 Methods of measurement offan vibration.
BS 1042
Part 1: Various dates by Pressure differential devices.
section 1981 to 1993
BS 2475: 1996 Specification for octave and one-thirdoctave band pass filters.
BS 2750: Measurement ofsound insulation in buildings and of building elements.
Part 1: 1980 Recommendations for laboratories
Part 2: 1993 Determination, verification and application of precision data.
Part 3: 1995 Laboratory measurements of airborne sound insulation of building elements.
Part 4: 1980 Field measurements of airbornesound insulation between rooms.
Part 5: 1980 Field measurements of airbornesound insulation of facade elements and facades.
Part 6: 1980 Laboratory measurements ofimpact sound insulation of floors.
Part 7: 1980 Field measurements of impact sound insulation of floors.
Part 8: 1980 Laboratory measurements of the reduction of transmitted impact noise by floor
coverings on a standardfloor.
Part 9: 1987 Method for laboratorymeasurementof room-to-room airborne sound insulation of
a suspended ceiling with a plenum above it.
BS 3593: 1986 Recommendation on preferredfrequencies for acoustical measurements.
BS 3638: 1987 Method for measurement of sound absorption in a reverberantroom.
BS EN ISO 3746: 1996 Acoustics —Determinationofsound power levels of noise sources usingsound pressure:
survey methodusing an enveloping measurement surface over a reflecting plane.
BS4142: 1990 (under review) Method of rating industrial noise affecting mixed residential and industrial areas.
BS 4196 Soundpower levels ofnoise sources.
Part 0: 1986 Guide for the use ofbasic standards and for the preparationof noise test codes.
Part 1: 1991 Precision methodsfor determinationof sound power levels for broad-band sources
in reverberation rooms.
Part 2: 1991 Precision methods for determinationof sound power levels for discrete-frequency
and narrow-band sources in reverberation rooms.
Part 3: 1991 Engineering methodsfor determination ofsound power levels for sources in special
reverberation test rooms.
Part 4: 1981 Engineering methods for determinationof sound power levels for sources in free
field conditions over a reflecting plane.
Part 5: 1981 Precision methods for determinationof sound power level for sources in anechoic
and semi-anechoic rooms.
Part7: 1988 Survey method for determination of sound power levels of noise sources using a
reference sound source.
Part8: 1991 Specificationfor the performance and calibration of reference sound sources.
BS 4718: 1971 Methods of test for silencers for air distribution systems.
BS 4727: Part 3 Acoustics and electroacoustics terminology.
Group 08: 1985
BS 4773: 1989 Methods for testing and rating air terminaldevices for air distribution systems.
Part 2: 1989 Acoustic testing.
BS 4856 Methods for testing and rating fan coil units, unit heaters and unit coolers.
Part 4: 1978 Acousticperformance,withoutadditionalducting.
Part 5: 1979 Acousticperformance,with ducting.
158 Acoustics in the Built Environment
Part 6: 1987 Methods for specifying and measuring the characteristics of auxiliary passive
elements.
Part 8: 1988 Methods for specifyingand measuring the characteristics of automatic gain control
devices.
Part 9: 1987 Methods for specifyingand measuring the characteristics of artificial reverberation,
time delay and frequency shift equipment.
Part 11: 1988 Specification for application of connectors for the interconnection of sound system
components.
Part 12: 1995 Specification for applications of connectors for broadcast and similar use.
Part 13: 1987 Guide for listening tests on loudspeakers.
Part 14: 1987 Guide for circularand elliptical loudspeakers; outer frame diameters andmounting
dimensions.
Part 15: 1988 Specification for matching values for the interconnection of sound system
components.
Part 16: 1989 Guide to the 'RASTI' method for the objective rating of speech intelligibility in
auditoria.
Part 18: 1996 Guide for digital peak level indicator.
BS 6841: 1987 (under review) Guide to measurementand evaluation of human exposure to whole body mechanical
vibration and repeated shock.
BS 6912: Part 3: 1990 Safetyof earth-moving machinery — sound test methodfor machine-mounted forward
and reverse warning signal.
BS 6926: 1995 Determination of sound powerlevels of noise sources.
BS 7385 Evaluation and measurement for vibration in buildings.
Part 1: 1990 Guide for measurement ofvibrations and evaluation of their effects on buildings.
Part 2: 1993 Guide to damage levels from groundbornevibration.
BS 7443: 1991 Specificationfor sound systemsfor emergency purposes (IEC 849) (largely updatesBS
5839 'Fire detectionand alarm systems for buildings').
BS 7445 Description and measurementof environmental noise.
Part 1: 1991 Guide to quantities and procedures.
Part 2: 1991 Guide to the acquisition of data pertinent to land use.
Part 3: 1991 Guide to the application to noise limits.
BS 7458 Test code for the measurement of airborne noise emitted by rotating electrical
machinery.
Part 1: 1991 Engineeringmethod for the free-field conditions over a reflecting plane.
Part 2: 1991 Survey method.
BS 7594: 1993 Code of Practice for Audio-FrequencyInductionLoop Systems(AFILS).
BS 7580, Part 2: 1995 Shortened procedurefor type 2 sound level meters.
BS 7636: 1993 Method of determination of thresholds of hearingusing sound field audiometrywith
pure tone and narrow-band test signals.
BS 7643, Part 3: 1993 Building construction: expression of users' requirements: Acousticalrequirements.
BS 7698, Part 9: 1996 Measurement and evaluation of mechanical vibrations.
BS 7703, Part 1: 1993 Acoustics — Determination of sound power levels of noise sources using sound
intensity.
BS 7827: 1996 Code of Practice for designing, specifying, maintaining, and operating emergency
sound systems at sports venues
BS 8233: 1987 Code of Practice for sound insulation and noise reduction for buildings.
BS 8297: 1995 Determination of sound power levels of multi-source industrial plants.
Page blank
in original
Berlin Philharmonic, 58, 62, 64
Index Bestpractical means, 33
Blockwork, 39, 43
Boilers,92, 102
Brickwork, 43, 45
BridgewaterHail, Manchester,54, 58, 59
British Gypsum,43, 44, 45, 47, 56
British Standards, 157—9
British Standards Institution, 107
Broadcasting,64, 67
Broadcastingauthorities criteria, 88
Builder's work enclosure, 102
A-weighting, 136 Builder's work penetrations, 97, 99, 102
Abatement notice, 27, 33 Building control, 3
Absorbers, fibrous,50 Building damage, 29, 30, 89
Absorption, 48, 53, 54, 143 Building regulations, 3, 5, 24, 37, 48, 70
Absorption coefficients,50, 51—2 Building services, 26
Acceleration, 29 Building vibration:
Accelerometer, 12 humanresponse, 31
Acoustic appraisal, 13
Acoustic canopies, 64
Acoustic curtains, 102 Cable routes, 115
Acoustic doorsets, 99 Calculation ofrailway noise (Dept ofTransport), 18
Acoustic laboratories, 52 Calculation ofroadtraffic noise (CORTN),14
Acoustic louvres, 101, 102 Calibrator, 12
Acoustic modeltesting, 66 Cardiff, St David'sHall, 61
Acoustic plaster, 52 Cavitywalls, 41
Acoustical parameters, 118 Ceiling panels, soundreflecting, 72
Air-handlingunits, 91, 93, 104 Ceiling reflectors, 65
Mr space, 48 Ceiling tiles, 52
Aircraft flight path, 22 Ceiling voids, 43
Aircraft frequency spectra, 21 Ceilings, 43, 51
Aircraft noise, 11, 19, 21 suspended, 43, 44
Albert Hall, seeRoyal Albert Hall Centralized plant, 98
Ambient noise level, 120 Channel tunnel, 10
American National Standards Institute, 154 Chartered Institute ofBuilding ServicesEngineers (CIBSE),87, 91, 92,
American Societyfor Testingand MaterialsStandards (ASTMS), 154—5 94, 109
American Societyof Heating, Refrigeration and Mr-Conditioning Cinemas,28, 56, 57
Engineers (ASHRAE), 87, 91, 92, 107 Cladding:
Amsterdam,Concertgebauw,57 ductsand pipes, 40, 101
Anechoic chambers, 53, 107 Clarity, 61
Annoyance, 78 ClarityindexC50, 64
Appraisals: ClarityindexC80, 64
environmental, 7, 31 Clay pigeon shooting, 28, 29
Articulation index, 128, 144 Clubs, 28, 33
ArupAcoustics,54, 59 Coincidence, 35, 42, 48
Assessment procedure: Commissioning:4
industrial noise, 28 sound systems, 116
Association ofNoise Consultants (ANC), 7, 109 Commissioningtests, 108
Atkins,W. 5., 14 Compositeconstruction, 35, 42
Atria, 75 Compression,119
Attenuation, 16, 22, 96 Compressors,104, 105
cross talk, 96 Computer-aided theatre technique (CAn),54
definition, 133 Computer rooms, 78
train noise, 18 Concert halls, 30, 57—68
Attenuators, 94, 100, 101, 107 shape, 57
Audience, 52 Concertgebauw,Amsterdam, 57
Audio Frequency Induction Loop System (AFILS), 125, 131 Condenser units, 91, 99
Auditoria modelling, 54 Conference rooms, 72, 73
Auditorium ventilation, 100 Construction noise, 11, 13, 23—6
Auralization, 54, 56 prediction of, 24
Authorities, regulatory, 3 Construction plant noise, 23
Average sound insulation index ratingRw, 38 Contours, 16
Axial fan, 92, 104 Control ofPollution Act, 23, 24
Conversions,57
Cooling towers, 91
Background noise levels, 10, 26, 33, 45, 78, 88, 108, 116, 140 Council chambers, 67, 72
Balconyfronts, 81 Courtorder, 33
Band limitfrequencies, 138 Cross-talk, 97, 93
Banners, side wall absorbers,65 Curtains, 51
Barrier attenuation, 15 Curtains, acoustic, 102
Barriers, 23
Bass traps, 52
BBC, 80 Damping, 89
BDPAcoustics,55, 56 Damping, edge, 48
Belfast, Waterfront Hall, 58, 60, 61 Daytime, 10
Beranek and Ver, 5 'Deafaid loop systems, 125
162 Index
Valves, 94
Velocity, 29 Young's modulus, 42
Ventilation,64, 108
auditorium, 100
natural, 2, 14, 75 Zoning, 78
ARCHITECTURE ACOUSTICS
Its broad range of subject matter and its ease of use make
ISBN 0—7506—3644—0
ARCHITECTURALPRESS
_____ Ankir*d.,flk-HàL.....IIm
I
JttiUU
9 780750 636445
hupi/www.bh.com
oC