ISO 3744:1981

International Standard @ 3744
INTERNATIONAL ORGANIZATION FOR STANDARDIZATIONOMEWYHAPO~HAR OPTAHM3AUMR no CTAHnAPTH3AUHMOORGANlSATlON INTERNATIONALE DE NORMALlSATlON
L Acoustics - Determination of sound power levels of
noise sources - Engineering methods for free-field
conditions over a reflecting plane
Acoustique - Détermination des niveaux de puissance acoustique émis par les sources de bruit - Méthodes d’expertise pour les
conditions de champ libre au-dessus d’un plan réfléchissant
First edition - 1981-05-01
UDC 534.6 Ref. No. IS0 3744-1981 (E)
-
Descriptors : acoustics, acoustic measurement, sound power, noise (sound), testing conditions, test equipment, sound pressure, rules of
@
calculation.
s F)
O
v,
Price based on 25 pages
Foreword
IS0 (the International Organization for Standardization) is a worldwide federation of
national standards institutes (IS0 member bodies). The work of developing Inter-
national Standards is carried out through IS0 technical committees. Every member
body interested in a subject for which a technical committee has been set up has the
right to be represented on that committee. International organizations, governmental
d
and non-governmental, in liaison with EO, also take part in the work.
Draft International Standards adopted by the technical committees are circulated to
the member bodies for approval before their acceptance as International Standards by
the IS0 Council.
International Standard IS0 3744 was developed by Technical Committee ISO/TC 43,
Acoustics, and was circulated to the member bodies in February 1977.
It has been approved by the member bodies of the following countries :
Australia Hungary
Portugal
Austria India Romania
Belgium Israel South Africa, Rep. of
Canada Japan Spain
Czechoslovakia Mexico Sweden
Denmark Netherlands
Switzerland
Finland New Zealand United Kingdom
France
Norway USA
Germany, F. R. Poland USSR
No member body expressed disapproval of the document.
This International Standard cancels and replaces IS0 Recommendation R 495-1966, of
which it constitutes a technical revision.
@ International Organization for Standardization, 1981 O
Printed in Switzerland
ii
Contents
Page
O . 1 Related International Standards . 1
.
Synopsis of IS0 3744 . 2
0.2
*.v'
0.3 Introduction . 2
1 Scope and field of application . 2
2 References . 3
3 Definitions . 4
4 Acoustic environment . 4
Instrumentation . 5
6 Installation and operation of source . 5
7 Sound pressure levels on measurement surface . 6
8 Calculation of surface sound pressure level.
sound power level and directivity factor . 9
_-
Information to be recorded . 10
'L
10 Information to be reported . 11
Annexes
A Qualification procedures for the acoustic environment . 12
B Microphone array on the hemisphere . 16
C Microphone array on the parallelepiped measuring surface . 19
D Microphone array on the conformal measurement surface . 21
Examples of suitable instrumentation systems . 24
E
F Guidelines for the detection of impulsive noise . 25
G Calculation of directivity index and directivity factor using
hemispherical microphone array . 25
...
INTERNATIONAL STANDARD IS0 3744-1981 (E)
Acoustics - Determination of sound power levels of
noise sources - Engineering methods for free-field
conditions over a reflecting plane
decide which one is most appropriate for the conditions and
0.1 Related International Standards
purposes of the test. The operating and mounting conditions of
the machine or equipment to be tested must be in accordance
This International Standard is one of a series specifying various
with the general principles stated in these basic documents.
methods for determining the sound power levels of machines
and equipment. These basic documents specify only the
Guidelines for operating and mounting conditions are provided
acoustical requirements for measurements appropriate for dif-
in IS0 3740. If no sound test code is specified for a particular
ferent test environments as shown in table 1.
machine, the mounting and operating conditions shall be fully
described in the test report.
When applying these basic documents, it is necessary to
Table 1 - International Standards specifying various methods for determining the sound power levels of
machines and equipment
Inter-
Sound power Optional
national Classification Test Volume Character
levels information
Standard of method ** environment of source of noise
obtainable available
Steady,
Reverberation broad-band
In one-third
A-weighted
Precision room meeting
Steady, octave or sound power
(grade 1) specified
discrete- octave bands level
requirements
Preferably less frequency or
I 3742
than 1 % of test
narrow-band
room volume
Steady,
Special broad-band
Engineering
A-weighted and Other weighted
3743 reverberation narrow-band
(grade 2)
in octave bands sound power levels
test room or discrete-
frequency
Greatest
Outdoors or Directivity
Engineering dimension
3744 in large information;
(grade 2) less than A-weighted
room
sound pressure
15 m
and in one-third
levels as a
octave or
Preferably less function of time;
Anechoic or octave bands
Precision 0.5 % of
than other weighted
3745 semi-anechoic
(grade 1)
test room sound power levels
room
volume
Sound pressure
No restrictions :
levels as a
Survey No special test limited only by
3746 A-weighted function of time;
(grade 3) environment available test
other weighted
environment
sound power levels
See clause 2.
** See IS0 2204.
IS0 3744-1981 (E)
0.2 Synopsis of IS0 3744 Data obtained in accordance with this International Standard
may be used for the following purposes amongst others :
Applicability
comparison of machines which are similar in size and
a)
Test environment kind;
Free field over a reflecting plane (indoors or outdoors) b) comparison of machines which are different in size and
kind;
Type of source
c) rating of apparatus in terms of its sound power output;
Device, machine, component, subassembly.
d) prediction of the sound pressure levels produced by a
Size of noise source device or machine at a given point in a given enclosure or
environment.
Greatest linear dimension less than 15 m.
In this International Standard the computation of sound power
Character of noise radiated by the source from sound pressure level measurements is based on the
premise that the mean-square sound pressure averaged over
All types (as defined in IS0 2204 : steady, non-steady, the enveloping measuring surface and over time is
broad-band, discrete-frequency, narrow-band).
d
a) directly proportional to the sound power output of the
Accuracy
source;
Engineering or grade 2 (as defined in IS0 22M)(standard
inversely proportional to the area S of the measurement
b)
deviation for determining sound power levels for 1 kHz oc-
surface;
tave band is about 1,5 dB).
otherwise dependent only on the physical constants of
c)
Quantities to be measured
air density and sound velocity.
Sound pressure levels (weighted and in frequency bands) at
This International Standard, together with the others in this
prescribed microphone positions.
series (see table I), supersedes ISO/R 495.
Quantities to be determined
Weighted sound power levels (A-weighting is required;
1 Scope and field of application
other weightings are optional); sound power levels in
frequency bands; directivity characteristics of the source
1.1 General
(optional).
This International Standard specifies engineering methods for
measuring the sound pressure levels on a measurement surface
0.3 Introduction
enveloping the source and for calculating the sound power
This International Standard specifies engineering methods for level produced by the source. It gives requirements for the test
-A
environment and instrumentation as well as techniques for ob-
measuring the sound pressure levels on a given surface which
envelops the noise source under investigation. The sound taining the surface sound pressure level from which the
power level of the source is calculated from the results of these A-weighted sound power level of the source and octave or one-
third octave band sound power levels are calculated.
measurements. The method may be applied in laboratory
rooms which provide a free field over a reflecting plane or in
field installations whose acoustical characteristics meet the re-
This International Standard is concerned with specification of
quirements of this International Standard. If the test room does
the basic acoustic requirements for the measurement of noise
not meet the requirements of this International Standard,
under free-field conditions over a reflecting plane. It is impor-
measurements made in accordance with IS0 3746 might never- tant that specific test codes for various types of equipment be
theless be possible.
established and used as the basis for measurements according
to this International Standard. Such test codes will, for each
Free-field conditions are usually not encountered in typical type of equipment, give detailed requirements on mounting,
machine rooms where sources are normally installed. If loading and operating conditions for the equipment under test
measurements are made of such installations, corrections may as well as descriptions of the measurement surface, the
number of microphone positions, and the measurement
be required to account for background noise or undesired
reflections. These corrections reduce the accuracy of the sound distance to be used.
power determination. -
The use of different measurement surfaces may yield differing
The methods described in this International Standard conform estimates of the sound power level of a source. Measurements
to the general recommendations included in IS0 2204.1 The on a given family of equipment should all be based on the same
methods specified in this International Standard permit the shape of measurement surface. The test code for a particular
determination of sound power level both in frequency bands
type of equipment should give detailed information on the par-
and directly as an A-weighted value. ticular measurement surface that is selected.
IS0 3744-1981 (El
1.2 Field of application
Table 2 - Uncertainty in determining sound power levels
for engineering measurements indoors or outdoors
1.2.1 Types of noise
One-third octave Standard
Octave band
I deviation
band
This International Standard applies to sources which radiate
centre frequencies
centre frequencies of mean value
broad-band noise, narrow-band noise, discrete tones and com-
Hz Hz dB
binations of such components. The procedures given are ap-
I
plicable to steady noise. These procedures may also be applied
to non-steady noise as defined in IS0 2204, with the exception
200to 630
250to 500
of an isolated burst of sound energy or a burst train with a
1000to4000 moto 5000
repetition rate less than 10 per second.
8000 6 300 to 10 O00
1.2.2 Size of source
This International Standard is not applicable to noise sources
with any linear dimension (length, width or height) which ex-
2 References
ceed 15 m.
IS0 266, Acoustics - Preferred frequencies for
measurements.
'4J' 1.3 Measurement uncertainty
ISO/R 354, Measurement of absorption coefficients in a
Measurements made in conformity with this International Stan-
re verbera tion room.
dard should result in standard deviations which are equal to or
less than those given in table 2. The uncertainties in table 2 de-
IS0 2204, Acoustics - Guide to International Standards on
pend not only on the accuracies with which sound pressure
the measurement of airborne acoustical noise and evaluation of
levels and measurement surface areas are determined but also
its effects on human beings.
on the "near field error" which increases for smaller measure-
ment distances and lower frequencies (i.e. those below
IS0 3740, Acoustics - Determination of sound power levels of
250 HA. The contribution to the uncertainty of the error caused
noise sources - Guidelines for the use of basic standards and
by small measurement distances may be checked by repeating
for the preparation of noise test codes.
the measurements at a larger measurement distance and com-
paring the resulting sound power levels. The near field error
IS0 3741. Acoustics - Determination of soundpowerlevels of
always leads to sound power levels which are higher than those
noise sources - Precision methods for broad-band sources in
determined with larger measurement distances.
reverberation rooms.
NOTES
IS0 3742, Acoustics - Determination of sound power levels of
1 If the methods specified in this International Standard are used to
noise sources - Precision methods for discrete- frequency and
compare the sound power levels of similar machines that are
narrow-band sources in reverberation rooms.
omnidirectional and radiate broad-band noise, the uncertainty in this
comparison tends to result in standard deviations which are less than
IS0 3743, Acoustics - Determination of sound power levels of
that given in table 2, provided that the measurements are performed in
L' - Engineering methods for special reverberation
noise sources
the same environment with the same shape of measurement surface.
test rooms.
2 The standard deviations given in table 2 reflect the cumulative ef-
fects of all causes of measurement uncertainty, excluding variations in
IS0 3745. Acoustics - Determination of sound power levels of
the sound power level from test to test which may be caused, for ex-
noise sources - Precision methods for anechoic and semi-
ample, by changes in the mounting or operating conditions of the
anechoic rooms.
source. The reproducibility and repeatability of the test results may be
considerably better (i.e., smaller standard deviations) than the uncer-
IS0 3746, Acoustics - Determination of sound power levels of
tainties given in table 2 would indicate.
noise sources - Survey method.
3 For a source which emits noise with a relatively "flat" spectrum in
the 100 to 10 O00 Hz frequency range, the A-weighted sound power
IS0 6926, Acoustics - Determination of sound power levels of
level is determined with a standard deviation of approximately 2 dB.
noise sources - Characterization and calibration of reference
For outdoor measurements, the standard deviations in the octave band
centered on 63 Hz will be approximately 5 dB. sound sources. 1)
1) At present at the stage of draft.
IS0 3744-1981 (E)
I E C Pu bl ica t i on 50 ( 08 1, In terna tional elec tro technical
octave bands with centre frequencies between IO0 and
vocabulary - Electro-acoustics.
10 O00 Hz. Any band may be excluded in which the level is
more than 50 dB below the highest band pressure level. For
IEC Publication 225, Octave, half-octave and third-octave band
special purposes, the frequency range of interest may be ex-
filters intended for the analysis of sound and vibrations. tended at either end, provided the test environment and instru-
ment accuracy are satisfactory for use over the extended fre-
IEC Publication 651, Sound level meters.
quency range. For sources which radiate predominantly high
(or low) frequency sound, the frequency range of interest may
be limited in order to optimize the test facility and procedures.
3 Definitions
3.9 measurement surface : A hypothetical surface of area
For the purposes of this International Standard, the following
S enveloping the source on which the measuring points are
definitions apply :
located.
3.1 free field : A sound field in a homogeneous, isotropic
3.10 reference box : A hypothetical reference surface
medium free of boundaries. In practice, it is a field in which the
which is the smallest rectangular parallelepiped that just
effects of the boundaries are negligible over the frequency
encloses the source and terminates on the reflecting plane.
range of interest.
3.11 measurement distance : The minimum distance from
3.2 free field over a reflecting plane : A sound field in the
the reference box to the measurement surface.
presence of one reflecting plane on which the source is located.
3.3 anechoic room : A test room whose surfaces absorb
4 Acoustic environment
essentially all the incident sound energy over the frequency
range of interest, thereby affording free-field conditions over
the measurement surface.
4.1 General
The test environments that are suitable for measurements ac-
3.4 semi-anechoic room : A test room with a hard reflec-
cording to this International Standard include :
ting floor whose other surfaces absorb essentially all the inci-
dent sound energy over the frequency range of interest,
a laboratory room which provides a free field over a
a)
thereby affording free-field conditions above a reflecting plane.
reflecting plane;
3.5 surface sound pressure : The sound pressure averaged
b) a flat outdoor area that meets the requirements of 4.2
in time on a mean-square basis, averaged over the measure-
and annex A:
ment surface using the averaging procedures specified in 8.1
and corrected for the effects of background noise and the in-
cl a room in which the contributions of the reverberant
fluence of reflected sound at the measurement surface.
field to the sound pressures on the measurement surface
are small compared with those of the direct field of the
source.
3.6 surface sound pressure level, in decibels, : Ten
--c
times the logarithm to the base 10 of the ratio of the square of
Conditions described under c) above are usually met in very
the surface sound pressure to the square of the reference
large rooms as well as in smaller rooms with sufficient sound-
sound pressure. The weighting network or the width of the fre-
absorptive materials on their walls and ceilings.
quency band used shall be indicated : for example, A-weighted
sound pressure level, octave band sound pressure level, one-
third octave band sound pressure level, etc. The reference
4.2 Criteria for adequacy of the test environment
sound pressure is 20 pPa.
Ideally, the test environment should be free from reflecting ob-
jects other than a reflecting plane SO that the source radiates
3.7 sound power level, in decibels, Lw: Ten times the
into a free-field over a reflecting plane. Annex A describes pro-
logarithm to the base 10 of the ratio of a given sound power to
cedures for determining the magnitude of the environmental
the reference sound power. The weighting network or the
correction (if any) to account for departures of the test environ-
width of the frequency band used shall be indicated; for exam-
ment from the ideal condition. Test environments which are
ple, A-weighted sound power level, octave band sound power
suitable for engineering measurements permit the sound power
level, one-third octave band sound power level, etc. The
reference sound power is 1 pW (= 10-12 W). level to be determined with an uncertainty that does not exceed
the values given in table 2.
NOTE - The mean sound pressure level at some reference radius is
numerically different from the sound power level and its use in lieu of NOTE - If it is necessary to make measurements in spaces which do
the sound power level is not recommended. not meet the criteria of annex A, standard deviations of the test results
may be greater than those given in table 2. In those cases, IS0 3746
describes applicable test procedures. In spite of this limitation, the
3.8 frequency range of interest : For general purposes,
sound power level determined according to this International Standard
the frequency range of interest includes the octave bands with
may be useful for obtaining a valid upper limit for the sound power level
centre frequencies between 125 and 8 O00 Hz and the one-third
of a source.
IS0 3744-1981 (E)
4.3 Criterion for background noise The microphone and its associated cable shall be chosen so
that their sensitivity does not change over the temperature
At the microphone positions, the sound pressure levels of the range encountered during the measurement. If the microphone
background noise shall be at least 6 dB and preferably more
is moved, care shall be exercised to avoid introducing
than 10 dB below the sound pressure level to be measured in acoustical noise (for example, wind noise) or electrical noise
each frequency band within the frequency range of interest. (for example, from gears, flexing cables, or sliding contacts)
that could interfere with the measurements.
NOTE - Levels of background noise less than 6 dB below the sound
pressure levels to be measured are too high for the purposes of this
5.3 Frequency response of the instrumentation
International Standard. Under such circumstances, the survey method
system
of IS0 3746 should be used.
The frequency response of the instrumentation system for the
Care shall be taken to minimize the effects of wind which may
angle of incidence specified by the manufacturer shall be flat
increase the apparent background noise. The appropriate in-
over the frequency range of interest within the tolerances given
structions provided by the microphone manufacturer shall be
for a type 1 instrument in IEC Publication 651.
followed.
5.4 Weighting network, frequency analyser
5 Instrumentation
An A-weighting network fulfilling at least the requirements for
Lx'
a type 1 instrument in accordance with IEC Publication 651 and
5.1 General
an octave band or one-third octave band filter set fulfilling the
requirements of IEC Publication 225 shall be used. The centre
The instrumentation shall be designed to measure the mean-
frequencies of the frequency bands shall correspond to those
square value of the weighted sound pressure level and the oc-
tave or one-third octave band levels, averaged over time and of IS0 226.
over the measurement surface. Surface averaging is usually
carried out by measuring the time-averaged sound pressure NOTE - If other weighting networks are used in addition to
A-weighting, the characteristics of such networks shall be reported.
levels with a prescribed time constant for a fixed number of
microphone positions (7.1 and computing the average value
according to 8.1.
5.5 Calibration
The instrumentation used can perform the required time-
During each series of measurements, an acoustical calibrator
averaging in two different ways :
with an accuracy of I 0,5dB shall be applied to the
microphone for checking the calibration of the entire measuring
a) By continuous averaging of the squared signal using
system at one or more frequencies over the frequency range of
RC-smoothing with a time-constant SA. Such continuous
interest. The calibrator shall be checked at least once every year
averaging provides only an approximation of the true time-
to verify that its output has not changed. In addition, an
average, and it places restrictions on the "settling" time and
acoustical and an electrical calibration of the instrumentation
observation time (see 7.4.3).
system over the entire frequency range of interest shall be car-
ried out at least every 2 years.
NOTE - An example of an instrument employing such averaging is
a sound level meter fulfilling at least the requirements for a type 1
\/
instrument in accordance with IEC Publication 651 with the time
"S". For noise which is impulsive in
weighting characteristic
6 Installation and operation of source
character, refer to annex F.
6.1 General
b) By integrating the squared signal over a fixed time-
interval 'SD. This integration may be performed by either
In many cases, the sound power emitted by a source depends
digital or analogue means.
upon its support or mounting conditions as well as the manner
in which the source is operated. This clause gives general
Examples of suitable instrumentation systems are given in
recommendations concerning the installation and operation of
annex E.
sources. Reference should be made to specific test codes for
more detailed information concerning installation and operation
5.2 The microphone and its associated cable
of specific classes of sources (for example, rotating electrical
machines).
A condenser microphone, or the equivalent in accuracy, sta-
bility and frequency response, shall be used. The microphone
Particularly for large machines, it is necessary to make a de-
shall have a flat frequency response, over the frequency range
cision as to which components, subassemblies, auxiliary equip-
of interest, for the angle of incidence specified by the manufac-
ment, power sources, etc., belong to the source under test.
turer.
6.2 Source location
NOTE - This requirement is met by a microphone of a standardized
sound level meter fulfilling at least the requirements for a type 1 instru-
The source to be tested shall be installed and mounted with
ment in accordance with IEC Publication 651 and calibrated for free
field measurements. respect to the reflecting plane in one or more positions that are

IS0 3744-1981 (E)
typical of normal usage, if practicable. If several possibilities The sound power levels of sources may be determined for any
exist, or if typical installation and mounting conditions are desired set of operating conditions (i.e. temperature, humidity,
unknown, special arrangements are to be made and described device speed, etc.). These test conditions shall be selected
in the test report. In locating the source within the test environ- beforehand and shall be held constant during the test. The
ment, it is important that sufficient space be allowed so that the source shall be in a stable operating condition before any noise
measurement surface can envelop the machine according to measurements are made.
the requirements of 7.1.
The source shall be located at a sufficient distance from any
7 Sound pressure levels on measurement
reflecting wall or ceiling or any reflecting object so that the re-
surface
quirements given in annex A are satisfied on the measurement
surface.
7.1 Measurement surface
NOTE - The typical installation conditions for some sources involve
two reflecting surfaces (for example an appliance installed against a
7.1 .I Reference surface and measurement surface
wall) or free space (for example, a hoist) or an opening in an otherwise
(so that radiation may occur on both sides of the ver-
reflecting plane
To facilitate the location of the microphone positions, a
tical plane). Detailed information on installation conditions and the
configuration of microphone arrays should be based on the general re- hypothetical reference surface is defined. This reference sur-
quirements of this International Standard and specified in specific test
face is the smallest possible rectangular box, i.e., rectangular
codes for such sources. -
parallelepiped, that just encloses the source and terminates on
the reflecting plane. When defining the dimensions of this
reference box, elements protruding from the source which are
6.3 Source mounting
not significant radiators of sound energy may be disregarded.
These protruding elements should be identified in specific test
Many small sound sources (for example, ballasts for fluores-
codes for different types of equipment. The microphone pos-
cent lamps, electrical clocks), although themselves poor
itions lie on the measurement surface, a hypothetical surface of
radiators of low frequency sound, may, as a result of the
area S which envelops the source as well as the reference box
method of mounting, radiate more low-frequency sound when
and terminates on the reflecting plane.
their vibrational energy is transmitted to surfaces large enough
to be efficient radiators. Resilient mounting should be inter-
posed if possible between the device to be measured and the
7.1.2 Co-ordinate system and definition of distance do
supporting surfaces so that the transmission of vibration to the
support and the reaction on the source are both minimized. In
The location of the source under test, the measurement surface
this case, the mounting base should have a sufficiently high
and the microphone positions are defined by a co-ordinate
mechanical impedance to prevent it from vibrating and
system with the horizontal axes x and y in the ground plane
radiating sound excessively. However, such resilient mounts
parallel to the length and width of the reference box and with
shall not be used if the device under test is not resiliently
the vertical axis z passing through the geometric centre of the
mounted in typical field installations.
box. The characteristic distance do is the distance from the
origin of the co-ordinate system to one of the upper corners of
the reference box :
6.4 Auxiliary equipment
If practicable, all auxiliary equipment necessary for the opera-
tion of the device under test that is not a part of the source
(see 6.1 shall be located outside the test environment.
where II, l2 and l3 are the length, width and height of the
reference box.
6.5 Operation of source during test
During the acoustical measurements, the source shall be
7.1.3 Shape of measurement surface
operated in a specified manner typical of normal use. One or
more of the following operational conditions may be ap-
Any one of the following three shapes may be used for the
propriate :
measurement surface :
a) device under specified load and operating conditions; a hemispherical surface of radius r;
1)
b) device under full load [if different from ail;
2) a rectangular parallelepiped whose sides are parallel to
those of the reference box; in this case, the measurement
device under no load (idling);
c) distance, d, is the distance between the measurement sur-
face and the reference box;
di device under operating condition corresponding to
maximum sound generation representative of normal use;
3) a conformal surface which is the same as the rec-
tangular parallelepiped except that the corners are rounded,
e) device with simulated load operating under carefully
the corners being formed by portions of cylinders and
defined conditions.
spheres. The measurement distance, d, is the distance from
IS0 3744-1981 (E)
the measurement surface to the reference box. Each point ple many reflecting objects and high levels of background
on the conformal surface is the same distance, d, from the noise), the selection of a small measurement distance is ap-
closest point on the reference box. propriate and usually dictates the selection of a parallelepiped
or conformal measurement surface. For machines usually
For measurements on a series of similar sources (for example, mounted and/or to be measured in large open areas under
machines of the same type), the use of the same shape of satisfactory acoustical conditions, a large measurement
measurement surface is recommended. The specific test code distance is usually selected and in this case the hemispherical
pertinent to the particular source under investigation should be measurement surface is preferred.
consulted for detailed information. The construction of the
reference box, the size and shape of the measurement surface If the dimensions of the reference box, I,, /2 and /3, are all less
as well as the measurement distance, d, or the radius of the than 1,0 m, the hemispherical surface is preferred. If one or
hemisphere, r, shall be described in the test report. more dimensions of the reference box exceeds 1.0 m, but the
reference box is "approximately cubical" in shape, the
hemisphere is still preferred. "Approximately cubical" means
7.1.4 Selection of measurement surface
that the largest dimension is less than twice the smallest dimen-
sion. If any dimension of the reference box exceeds 1,0 m and
7.1.4.1 General
the reference box is not approximately cubical in shape, then
the parallelepiped (or conformal) measurement surface is
When environmental conditions permit, use of the hemi-
preferred.
spherical measurement surface is usually preferred. When
L- '
measurements are to be made close to the source because of
adverse environmental conditions or other constraints, either a
7.1.5 Key microphone positions on the hemispherical
rectangular parallelepiped or a conformal surface may be used
measurement surface
for the measurement surface. In general, the conformal surface
is expected to yield a more accurate value of the sound power
The microphone positions lie on the hypothetical hemispherical
level than the rectangular parallelepiped, but use of the confor-
surface of area S = 2 nr2, enveloping the source and ter-
mal surface requires more effort to position the microphones.
minating on the reflecting plane. The centre of the hemisphere
The microphone positions on the conformal surface are at a
is the projection of the geometric centre of the reference box
uniform distance from the reference box which may be an ad-
on the reflecting plane. The key microphone positions of the
vantage under some environmental conditions.
hemispherical surface are shown in figures 2 and 3 in annex B.
Figure 3 prescribes the locations of ten key microphone pos-
itions each associated with equal areas on the surface of the
7.1.4.2 Hemispherical measurement surface
hemisphere of radius r. The hemispherical array of figures 2
and 3 has been selected to minimize the errors which can be
The hemisphere shall be centred on the projection on the
caused by interference between the sound wave reaching the
reflecting plane of the geometric centre of the reference box.
microphone directly and the wave reflected by the reflecting
The radius, r, of the hemispherical measurement surface shall
plane.
be equal to or greater than twice the characteristic distance, do,
or four times the average distance of the geometric centre of
the source from the reflecting plane, whichever is the larger,
7.1.6 Key microphone positions on the parallelepiped
and not less than 1 m. The radius of the hemisphere is
measurement surface
preferably in the series 1 - 2 - 4 - 6 - 8 - 10 - 12 - 14 or
16 m. Some of these radii may be so large that the environmen-
The microphone positions lie on the measurement surface, a
tal requirements given in annex A of this International Standard
hypothetical surface of area S enveloping the source whose
cannot be satisfied; such large values of the radii may not be
sides are parallel to the sides of the reference box and spaced
used.
out a distance d (measurement distance) from the box. The key
microphone positions on the parallelepiped measurement sur-
7.1.4.3 Parallelepiped and conformal measurement surfaces
face are shown in figure 4 in annex C. The area S of the
measurement surface is given by the formula
If a hemispherical measurement surface is not selected, either a
rectangular parallelepiped or a conformal surface may be
S = 4 (ab + bc + CU)
selected as the shape of the measurement surface. The
measurement distance, d, is the perpendicular distance bet-
where
ween the reference box and the measurement surface. The
preferred value of dis 1 m and it must be at least 0,25 m. The
U = 0.5 I, + d
value of d shall preferably be one of the following : 0,25 - 0,5
- 1 - 2 - 4 or 8 m. Measurement distances larger than 1 m
b = 0,5 I, + d
may be used only if the environmental requirements given in
annex A of this International Standard are satisfied.
c=i3+d
7.1.4.4 Additional considerations concerning choice of I,, 12 and I, are the length, width and height of the reference
measurement surface box.
For machines usually mounted and/or to be measured in rooms Figure 4 prescribes the locations of nine key microphone pos-
or spaces under unfavourable acoustical conditions (for exam- itions.
IS0 3744-1981 (E)
7.1.7 Key microphone positions on the conformal z-axis of the new array is coincident with the top point of the
measurement surface original array. The total number of microphone positions is in-
creased from ten to twenty.
The microphone positions lie on the measurement surface, a
hypothetical surface of area S enveloping the source which
7.2.2.2 Rectangular parallelepiped
consists of an enclosure formed by a rectangular parallelepiped
with rounded corners, the corners being formed by portions of
For the rectangular parallelepiped array, the additional
cylinders and spheres. The measurement distance d is the
microphone positions are shown in figure 4. The total number
distance from the perpendicular side of the measurement sur-
of microphone positions is increased from nine to seventeen.
face to the reference box. The conformal surface is that surface
which is defined as being located everywhere a distance d from
7.2.2.3 Conformal surface
the nearest point on the envelope of the reference box (see
figure 5 in annex DI. The eight key microphone positions are
For the conformal surface, the additional microphone positions
shown in figure 6 in annex D.
are shown in figure 6. The total number of microphone pos-
itions is increased from eight to sixteen.
The area S of the measurement surface is given approximately
by the formula
7.2.3 Sufficiency of number of microphone positions
a+b+c
S = 4 (ab + bc + ca) x
In general, the key microphone positions and the additional
a + b + c + 2d
microphone positions defined in 7.2.2.1, 7.2.2.2 or 7.2.2.3 are
sufficient for the purposes of measurements obtained accor-
where a, 6, c and d are the same as for the rectangular
ding to this International Standard. The only exception is for
parallelepiped.
very large sources, i.e., those for which the characteristic
distance do exceeds 5.0 m. In this case, additional microphone
7.2 Additional microphone positions on positions are required. They should be uniformly distributed
over the measurement surface and preferably not more than 2d
measurement surface
apart.
7.2.1 General
7.2.4 Additional localized microphone positions on
In 7.1, the key microphone positions on the three alternative measurement surface
measurement surfaces are described. Sound pressure level
measurements shall always be obtained at all of the key If conditions c) or d) of 7.2.1 exist, a detailed investigation may
microphone positions unless the criterion of 7.3 is satisfied. be necessary of the sound pressure levels over a restricted por-
Sound pressure level measurements may be required at ad- tion of the measurement surface. The purpose of this detailed
ditional microphone positions if investigation is to determine the highest and lowest values of
the sound pressure level in the frequency bands of interest. The
additional microphone positions will usually not be associated
a) the range of sound pressure level values measured at
the key microphone positions (i.e., the difference in with equal areas on the measurement surface. In this case, the
decibels between the highest and lowest sound pressure calculation procedure of IS0 3745, sub-clause 7.7.1.2 (unequal
areas), for the determination of Lw shall be used.
levels) exceeds the number of key measurement points, or
if any dimension of the reference box is larger than 2d.
b)
7.3 Reduction of number of microphone positions
or
For sources that produce a symmetrical radiation pattern, it
c) if the source radiates noise with a high directivity, or
may be sufficient to distribute the measurement points over
only a portion of the measurement surface. This is permissible
d) if the noise from a large source is radiated only from a
if preliminary investigation shows that the surface sound
small portion of the source, for example, the openings of an
pressure levels so determined according to the calculation pro-
otherwise enclosed machine.
cedures of clause 8 do not deviate by more than 1 dB from
those determined from measurements over the entire measure-
If conditions a) or b) exist, additional microphone positions
ment surface.
defined in 7.2.2 shall be used. If conditions c) or d) exist, ad-
ditional measurement positions on the measurement surface in
the region of high noise radiation shall be used (see 7.2.4).
7.4 Conditions of measurement
7.2.2
Location of additional microphone positions
7.4.1 General
7.2.2.1 Hemispherical measurement surface Environmental conditions may have an adverse effect on the
microphone used for the measurements. Such conditions (for
For the microphone array on the hemisphere, an additional ten- example, strong electric or magnetic fields, wind, impingement
point array is defined by rotating the original array of figure 3 of air discharge from the equipment being tested, high or low
through 180" about the z-axis. Note that the top point on the temperatures) must be avoided by proper selection or
IS0 3744-1981 (E)
Table 3 - Corrections for background noise
positioning of the microphone. The microphone shall always be
oriented in such a way that the angle of incidence of the sound
waves is that for which the microphone is calibrated.
Corrections to be subtracted
Difference between sound
from sound pressure level
pressure level measured
measured with sound source
with sound source
The sound pressure level shall be observed over a typical period
operating to obtain sound
operating and background
of operation of the source. Readings of the sound pressure
pressure level due to
sound pressure
level (corresponding to the level of the mean-square sound
sound source alone
level alone
pressure) shall be taken at each measurement point with
dB dB
A-weighting and for each frequency band within the frequency
measurements invalid
<6
range of interest. The instrumentation used must comply with
the requirements of clause 5.
6 1 .O
1 ,O
The following data shall be obtained :
1 .O
a) the A-weighted sound pressure levels and band 9
pressure levels during operation of the source under test;
10 0,5
> 10 0.0
b) the A-weighted sound pressure levels and band
pressure levels produced by the background noise.
L
for the frequency bands centred on or below 160 Hz, the period
8 Calculation of surface sound pressure level,
of observation shall be at least 30 s. For A-weighting and for
sound power level and directivity factor
the frequency bands centred on or above 200 Hz, the period of
observation shall be at least 10 s
8.1 Calculation of sound pressure level averaged
over the measurement surface
7.4.2 Measurements with sound level meter
For the A-weighted sound pressure level and the level in each
If the indicating meter of a sound level meter is used, the time
frequency band of interest, an average sound pressure level
weighting "S" characteristic shall be used. When the fluctua-
over the measurement surface, L,,, is calculated from the
tions of the indicating pointer on the sound level meter are less
measured sound pressure levels LPi (after corrections for
than f 3 dB using the time weighting "S" characteristic, the
background noise are applied according to 7.4.4, if necessary)
noise is considered to be steady for the purposes of this Inter-
by using the following equation :
national Standard
...