Vous êtes sur la page 1sur 46

AMCA

Standard 300-05
Reverberant Room Method for
Sound Testing of Fans

AIR MOVEMENT AND CONTROL


ASSOCIATION INTERNATIONAL, INC.

The International Authority on Air System Components


AMCA Standard 300 - 05

Reverberant Room Method for


Sound Testing of Fans

AIR MOVEMENT AND CONTROL ASSOCIATION INTERNATIONAL


30 WEST UNIVERSITY DRIVE
ARLINGTON HEIGHTS, IL 60004-1893 U.S.A.
PHONE: (847) 394-0150
fax: (847) 253-0088
web: WWW.AMCA.ORG
© 2005 by Air Movement and Control Association International, Inc.

All rights reserved. Reproduction or translation of any part of this work beyond that permitted by Sections
107 and 108 of the 1976 United States Copyright Act without the permission of the copyright owner is
unlawful. Requests for permission or further information should be addressed to the Executive Director,
Air Movement and Control Association International, Inc.

ii
Forward/Authority
AMCA Standard 300-05 was adopted by the membership of the Air Movement
and Control Association International, Inc. on 30 July 2005. The effective date of
this standard is 01 November 2005.
Tung Nguyen (Chair) Emerson Ventilation Products
Dr. John Cermak Acme Engineering & Manufacturing Corporation
Joseph Langford American Coolair Corp.
David Wolbrink Broan-Nutone LLC
Jeff Hill Cleanpak International
Dr. W.T.W. Cory Flakt Woods Ltd.
Iain Kinghorn (Alt.) Flakt Woods, Ltd.
Pete Neitzel Greenheck Fan Corporation
Max Clarke (Alt.) Greenheck Fan Corporation
Thomas Gustafson Hartzell Fan, Inc.
Ralph Sussey Howden Buffalo, Inc.
Dr. John Murphy JOGRAM, Inc.
Tan Tin Tin Kruger Ventilation Industries Pte. Ltd.
Ralph Sexton Matthews & Yates
Boyd Kunze The New York Blower Company
Scott Hausmann The Trane Co.
Scott Williamson Twin City Fan Companies, Ltd.

Disclaimer
AMCA International uses its best efforts to produce standards for the benefit of the
industry and the public in light of available information and accepted industry prac-
tices. However, AMCA does not guarantee, certify or assure the safety or perform-
ance of any products, components or systems tested, designed, installed or oper-
ated in accordance with AMCA standards or that any tests conducted under its
standards will be non-hazardous or free from risk.

Objections
Air Movement and Control Association International, Inc. will consider and decide
all written complaints regarding its standards, certification programs, or interpreta-
tions thereof. For information on procedures for submitting and handling com-
plaints, write to:
AIR MOVEMENT AND CONTROL ASSOCIATION INTERNATIONAL
30 WEST UNIVERSITY DRIVE
ARLINGTON HEIGHTS, IL 60004-1893 USA

iii
Foreward
This standard was originally developed in response to the need for a reliable and
accurate method of determining the sound power levels of fan equipment. The
original document was written by the AMCA P158NB Sound Test Code
Committee. Where possible, it was based on ASHRAE Standard 36-62, and com-
bined state-of-the-art with practical considerations. It was first published as a
Recommended Practice in February 1962, and adopted as a Standard Test Code
in October 1963. The sound power reference level now used in this standard was
changed in January 1965, from 10-13 watts to 10-12 watts. The third edition
(January 1967) AMCA 300-67 Test Code for Sound Rating included minor revi-
sions. In 1974, minor editorial changes were made, and size-speed conversions
were transferred to AMCA 301 Methods for Calculating Fan Sound Ratings From
Laboratory Test Data. The 1985 edition continued the original philosophy of com-
bining the theoretical and the practical. The 1996 edition was improved by
increasing the accuracy of Reference Sound Source values through improve-
ments in calibration requirements and procedure, and where appropriate, calling
for units of measure in SI (I-P) sequence. Where there have been successful
improvements in state-of-the-art, full advantage has been taken. This latest edi-
tion refines the duct end correction factors to values whose source can be traced
to its origin.
Introduction
This standard establishes a method of determining the sound power levels of a
fan. The method is reproducible in all laboratories that are qualified to the require-
ments of this standard.
The method employs standard sound measurement instrumentation, applied to
rooms that are restricted to certain acoustic properties. The test setups are
designed generally to represent the physical orientation of a fan as installed, fol-
lowing ANSI/AMCA 210 Laboratory Methods of Testing Fans for Aerodynamic
Performance Rating. Sound is defined as radiant mechanical energy that is trans-
mitted by pressure waves in air; it is the objective cause of hearing. Sound pres-
sure level is described mathematically as a logarithmic quantity derived from
sound pressure. The unit of sound pressure level is the decibel, referenced to a
base of 20 micropascals, or 20 microbar. The sound pressure level at any given
point in space depends on the distance between the source and the receiver,
reflection if in an enclosed room, proximity of the source to other sound sources,
etc.
Sound in a room is the result of one or more active sound power sources within
that room. Sound power is the total sound energy radiated per unit time. Sound
power level is described mathematically as a logarithmic quantity derived from the
sound power. The unit of sound power level is the decibel referenced to 1 picow-
att (1.0E-12 watt). Sound power levels determined through use of this standard
are useful for comparison between fans and in acoustical design.
Since sound power is independent of acoustic environment, two or more fans pro-
posed for a specific aerodynamic performance condition may be evaluated by
comparison to determine whether one is more suitable for an application than
another. Moreover, fan sound power levels establish an accurate base for esti-
mating the acoustical outcome of the fan installation in terms of sound pressure
levels. A successful estimate of sound pressure levels requires extensive informa-
tion on the fan and the environment in which it is to be located.
It is often advantageous for the fan equipment user to employ acoustical consul-
tation to ensure that all factors that affect the final sound pressure levels are con-
sidered. Additional information on the complexity of this situation may be found in
other documents available elsewhere.

iv
Contents Page
1. Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
2. Normative references . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3. Definitions / units of measure / symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1
3.2 Symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
4. Instruments / methods of measurement . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
4.1 Sound level meter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
4.2 Microphone system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
4.3 Frequency analyzer and weighting system . . . . . . . . . . . . . . . . . . . . . . .3
4.4 Data recording equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
4.5 Reference sound source (RSS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3
4.6 Test method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
4.7 Accuracy of results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
5. Equipment / setups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
5.1 Reverberant room . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
5.2 Setup categories . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
5.3 Aerodynamic performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5
5.4 Mounting methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
5.5 Duct length . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
5.6 Microphone travel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
5.7 Calibration of system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6
5.8 Equations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
6. Observations and conduct of test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
6.1 Observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10
6.2 Information to be recorded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11
7. Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11
7.1 Background correction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
7.2 Sound power level (Lw) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
8. Results and report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
8.1 Test subject . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12
8.2 Laboratory and instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13
8.3 Acoustical data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13
Annex A (normative) Room qualification: full and one-third octave . . . . . . . . . .15
A.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
A.2 Instruments and quipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
A.3 Test procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

v
A.4 Computation procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15
A.5 Qualification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16
Annex B (informative) Room qualification: pure tones / narrow-band . . . . . . .17
B.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17
B.2 Instruments and equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17
B.3 Test procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17
B.4 Computation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18
B.5 Qualification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18
Annex C (informative) Uncertainties analysis . . . . . . . . . . . . . . . . . . . . . . . . . . .21
C.1 Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21
C.2 Uncertainties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21
C.3 Room response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21
C.4 Fan operating points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23
C.5 Instrument error . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23
C.6 Reference sound source (RSS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24
C.7 Estimated standard deviation for determination of sound power levels 24
C.8 Duct end reflection corrections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .24
C.9 Octave band vs. One-third octave band . . . . . . . . . . . . . . . . . . . . . . . .25
C.10 Accuracy of the 63 hz octave band . . . . . . . . . . . . . . . . . . . . . . . . . . . .26
Annex D (informative) Alternative procedure for reference sound source calibra-
tion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27
D.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27
D.2 Equipment and facilities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27
D.3 Qualification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27
D.4 Procedure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .27
D.5 RSS sound power levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .28
Annex E (normative) Duct end reflection correction . . . . . . . . . . . . . . . . . . . . . .29
E.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29
E.2 End reflection curves . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .29
Annex F (informative) Filter-weighted measurements . . . . . . . . . . . . . . . . . . . .34
Annex G (informative) Radiation of sound by fan casing . . . . . . . . . . . . . . . . . .35
G.1 General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
G.2 Instruments and equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
G.3 Setup and test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
G.4 Observations and calculations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35
Annex H (informative) Total fan sound testing with attached ducts . . . . . . . . . .36
Annex J (informative) References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .38

vi
AMCA INTERNATIONAL INC. AMCA 300-05

America, 120 Wall St., 32nd Floor, New York, NY


REVERBERANT ROOM 10005-3993 U.S.A., 1985 (AMCA #2315-83-AO)
METHOD
ANSI S1.11-2004 Specification for Octave Band
FOR SOUND TESTING and Fractional Octave Band Analog and Digital
Filters, Acoustical Society of America, 120 Wall
OF FANS St., 32nd Floor, New York, NY 10005-3993
U.S.A., 1986 (AMCA #1727-86-AO)
1. Scope
ANSI S1.40-1984 Standard Specification for
This standard applies to fans of all types and Acoustical Calibrators, Acoustical Society of
sizes. This standard is limited to the America, 120 Wall St., 32nd Floor, New York, NY
determination of airborne sound emission for the 10005-3993 U.S.A., 1984 (AMCA #1895-84-AO)
specified setups. Vibration is not measured, nor
is the sensitivity of airborne sound emission to ANSI S12.5-1990 Requirements for the
vibration effects determined. Performance and Calibration of Reference
Sound Sources, Acoustical Society of America,
The size of a fan that can be tested in 120 Wall St., 32nd Floor, New York, NY 10005-
accordance with this standard is limited only by 3993 U.S.A., 1990 (AMCA #1863-90-AO)
the practical aspects of the test setups.
Dimensional limitations, test subject dimensions, ANSI S12.12-1992 Engineering Method for the
and air performance will control the test room Determination of Sound Power Levels of Noise
size and power and mounting requirements for Sources Using Sound Intensity, Acoustical
the test subject. Society of America, 120 Wall St., 32nd Floor,
New York, NY 10005-3993 U.S.A., 1992 (AMCA
The test setup requirements in this standard #1850-92-AO)
establish the laboratory conditions necessary for
a successful test. Rarely will it be possible to ANSI/IEEE/ASTM SI 10-1997 Standard for Use
meet these requirements in a field situation. of the International System of Units (SI): The
This standard is not intended for field Modern Metric System, Institute of Electrical and
measurements. Electronic Engineers, 345 east 47th Street, New
York, NY 10017 U.S.A., 1997 (AMCA #2924-97-
AO)
2. Normative references

The following standards contain provisions that, 3. Definitions / units of measure /


through specific reference in this text, constitute symbols
provisions of this American National Standard.
At the time of publication the editions indicated 3.1 Definitions
were valid. All standards are subject to revision,
and parties to agreements based on this 3.1.1 Blade Passage Frequency (BPF): The
American National Standard are encouraged to frequency of fan impeller blades passing a
investigate the possibility of applying the most single fixed object, per the following formula:
recent editions of the standards listed below.
BPF = (number of blades)(fan rotational
ANSI/AMCA 210-99 / ANSI/ASHRAE 51-1999 speed, rev/min) / 60, in Hz.
Laboratory Methods of Testing Fans for
Aerodynamic Performance Rating, Air 3.1.2 Chamber: An enclosure used to
Movement and Control Association regulate airflow and absorb sound; it may also
International, Inc., 30 W. University Drive, conform to air test chamber conditions given in
Arlington Heights, IL 60004-1893 U.S.A, 1999 ANSI/AMCA 210.
ANSI S1.4-1983; S1.4A-1985 Specification for 3.1.3 Decibel (dB): A dimensionless unit of
Sound Level Meters, Acoustical Society of
AMCA 300-05

level in logarithmic terms for expressing the ratio


of a power, or power-like, quantity to a similar 3.1.10 Octave Band: The interval between
reference quantity. (See 3.1.13 and 3.1.14) any two frequencies having a ratio of two. Fan
sound power levels are reported in eight
3.1.4 Ducted Fan: A fan having a duct standardized octave bands shown in Table 1.
connected to either its inlet, its outlet, or to both. Fan sound power levels may also be reported in
one-third octave bands, also shown in Table 1.
3.1.5 End Reflection: A phenomenon that
occurs whenever sound is transmitted across an 3.1.11 Reverberant Room: An enclosure
abrupt change in area, such as at the end of a meeting the requirements of Annex A, or Annex
duct in a room. When end reflection occurs A and Annex B.
some of the sound entering the room is reflected
back into the duct and does not escape into the 3.1.12 Shall and Should: The word shall is to
room. be understood as mandatory; the word should
as advisory.
3.1.6 Frequency: The number of times in
one second that a periodic function repeats 3.1.13 Sound Power Level: The value,
itself. expressed in decibels (dB), of ten times the
logarithm (base 10) of the ratio of the sound
3.1.7 Informative: A term that indicates that power W to the reference sound power Wref,
the referenced material is provided as advice to according to:
the reader but does not constitute a mandatory
requirement. LW, in dB = 10 log10 ( W / Wref ) (3-1)

3.1.8 Non-ducted Fan: A fan without a duct 3.1.14 Sound Pressure Level: The value,
connected to either its inlet or outlet. expressed in decibels (dB), of twenty times the
logarithm (base 10) of the ratio of the sound
3.1.9 Normative: A term that indicates that pressure p to the reference sound pressure pref,
the referenced material, if applied, constitutes a according to:
mandatory requirement.

Table 1 - Standardized octave and one-third octave bands [5]

Octave Bands

Band no. 1 2 3 4 5 6 7 8
ANSI Band no. 18 21 24 27 30 33 36 39
Center frequency f, Hz 63 125 250 500 1000 2000 4000 8000

One-Third Octave Bands

Band 1 Band 2 Band 3 Band 4


ANSI Band no. 17 18 19 20 21 22 23 24 25 26 27 28
Center freq. f, Hz 50 63 80 100 125 160 200 250 315 400 500 630
Band 5 Band 6 Band 7 Band 8
ANSI Band no. 29 30 31 32 33 34 35 36 37 38 39 40
Center freq. f, Hz 800 1000 1250 1600 2000 2500 3150 4000 5000 6300 8000 10000

2
AMCA 300-05

Lp, in dB = 20 log10 ( p / pref ) (3-2)


4.3 Frequency analyzer and weighting
3.1.15 Wavelength: The distance between system
two points having the same phase in two
consecutive cycles of a periodic wave, along a An octave band or one-third octave band filter
line in the direction of propagation [5]. set is required and shall meet the Order 3 Type
Wavelength (O) is determined by frequency and 3-D requirements of ANSI S1.11. An A-
the speed of sound in the air through which the weighting network shall meet the requirements
wave propagates: of ANSI S1.4 and S1.4A. Other weighting
networks may be used to improve the accuracy,
O=c/f (3-3) as outlined in Annex F.

where: 4.4 Data recording equipment

f = frequency, Hz This standard does not attempt to set limitations


c = 343 m/s @ 20°C (1125 ft/s @ 68°F) on data recording equipment. Considerations
include long-term stability, ease of use, and the
The value for c is acceptable for use in this method of averaging the sound pressure signal.
standard within the limits of ± 5°C (9°F) for Modern integrating-type analyzers that comply
standard air. with IEC 804 are recommended because they
produce Lp values eliminating any need for
3.1.16 Standard Air: Air having a density of visual averaging. Graphic level recorders can
3 3 be used to make permanent records and ease
1.2 kg/m (0.075 lbm/ft ). Standard air has a
ratio of specific heats of 1.4 and a viscosity of the problem of making visual averages from
1.8185E-03 Pa•s (1.222E-05 lbm/ft-s). Air at sound level meter indications.
20°C (68°F), 50% relative humidity, and 101.325
kPa (14.696 lbm/in.2, 29.92 in. Hg) barometric 4.5 Reference sound source (RSS)
pressure has these properties, approximately).
The reference sound source should comply with
3.2 Symbols the requirements of ANSI S12.5.

(See Table 2.) 4.5.1 The RSS shall be a small, modified, direct-
driven centrifugal fan having maximum overall
dimensions of 610 mm (2 ft) or less.
4. Instruments / methods of
measurement 4.5.2 The RSS shall produce steady broad-
band sound over at least the frequency range
4.1 Sound level meter from 50 Hz to 10,000 Hz. It shall comply in all
respects with the performance requirements of
The sound level meter shall meet the ANSI S12.5.
requirements of ANSI S1.4 and S1.4A. The
sound level meter should be capable of 4.5.3 The RSS shall be equipped with vibration
accepting a microphone extension cable. isolators that minimize transmitted vibration.
The degree of isolation should be 20 dB or
4.2 Microphone system more. If metal springs are used as vibration
isolators, rubber pads shall be used between the
The microphone system (transducer and any isolator and the structure of the reverberant
associated components and cable) shall meet room.
the requirements for use in a Type 1 precision
sound level meter according to ANSI S1.4 and 4.5.4 To ensure compliance with the stability
S1.4A. A microphone with a nominal diameter requirements of ANSI S12.5, all operating parts
of 13 mm (0.5 in.) is recommended. of the RSS shall be rigidly and permanently

3
AMCA 300-05

Table 2 - Symbols

UNIT OF MEASURE
SYMBOL DESCRIPTION SI I-P

Amin Minimum distance to reverberant field m ft


c Speed of sound m/s ft/s
D Duct diameter m ft
Eo End reflection factor, at duct outlet dB dB
Ei End reflection factor, at duct inlet dB dB
EW End reflection factor, adjustment to sound power level dB dB
f Frequency Hz Hz
J1 Bessel function of the first kind, first order
k Wave number
K1 Modified Bessel function of the second kind, first order
Lp Sound pressure level, re 20 ȝPa (20 ȝbar) dB dB
Lpc Corrected fan sound pressure level dB dB
Lpb Sound pressure level of room background, measured over the dB dB
normal microphone path
Lpm Sound pressure level of fan + room background, measured over dB dB
the normal microphone path
Lpq Sound pressure level of the RSS, corrected dB dB
Lpqm Sound pressure level of RSS + room background, measured dB dB
over the normal microphone path
LW Sound power level re 1 picowatt (1.0E-12 W) dB dB
LWi Sound power level; transmitted to inlet duct from fan dB dB
LWm Sound power level measured at the open inlet and outlet of the dB dB
fan
LWmi Sound power level measured at the open inlet of the fan dB dB
LWmo Sound power level measured at the open outlet of the fan dB dB
LWo Sound power level transmitted to the outlet duct from fan dB dB
LWr Sound power level of RSS dB dB
p Sound pressure Pa bar
pref Sound pressure reference level, 20 ȝPa Pa bar
Ps Fan static pressure Pa in. wg
Pt Fan total pressure Pa in. wg
r Ratio (of Duct area / Orifice area) dimensionless
R Reflection coefficient dimensionless
s Standard deviation dB dB
W Sound power (in watts) W W
Wref Reference sound power (1 picowatt) W W
ZM Mechanical impedence N˜s/m
D Ratio of transmitted to reflected sound dimensionless
J Ratio of specific heats dimensionless
O Wavelength m ft
Z Angular frequency rad/s rad/s

4
AMCA 300-05

attached. No rubber or wearing parts shall be 5. Equipment/setups


permitted (except lubricated bearings) and
protection against corrosion shall be provided. 5.1 Reverberant room

4.5.5 The RSS calibration shall consist of a An enclosure meeting the requirements of
determination of the sound power level radiated Annex A is mandatory for the purposes of this
by the RSS (including vibration isolators) when it standard. An enclosure meeting the
is in operation on a reflecting plane with requirements of Annex B is recommended for
radiation into a free field above that plane. The broad-band sound testing and is mandatory for
calibration shall be in accordance with ANSI the purpose of investigating pure tones and
S12.5 or as provided in Annex D. The maximum narrow bands.
time interval since calibration shall not exceed
that specified by the manufacturer or three 5.2 Setup categories
years, whichever is shorter.
A number of specific fan test setups are allowed.
4.6 Test method They are determined by the airflow direction
and the particular mounting arrangement of the
The test method is based on a Reference Sound test subject. The test setups fall into two
Source (RSS) substitution for the determination general categories.
of sound power. The reference document for
this method is ANSI S12.51. The first category is for a free-standing unit that
would be placed entirely in the test room (see
Application of the test method requires that the Figure 1). Results of this arrangement yield total
test subject fan be set in position in a test room sound power LW of the test subject, non-ducted.
that is qualified according to the requirements of For the total sound power of a ducted test
Section 5.1. subject located entirely in the test room, see
Annex H.
Once the test room has been qualified, sound
pressure levels are recorded with the RSS The second category includes those fans that
operating. The fan is then operated, without the would be tested on a chamber or two-room
RSS in operation, at various performance points system where only the inlet or outlet terminate in
of interest for the given test speed and the the test room (see Figures 2 and 3). These
sound pressure levels are recorded. Since the arrangements result in the determination of inlet
sound power levels of the RSS are known, the (LWi) or outlet (LWo) sound power. Section 5.6
substitution method is used to determine the discuses the limitations that must be imposed
sound power levels of the fan for each operating on the test room for determining the position of
point. the test subject and the location of the
microphone. The choice of test setup for a
Current ANSI and ASA documents on sound specific test will depend on the way the fan is
testing, facilities and equipment are useful expected to be applied in the field.
references. See Annex J.
5.3 Aerodynamic performance
4.7 Accuracy of results
Where an aerodynamic performance test is
Accuracy of test results is addressed in Annex C necessary to determine the point of operation of
and depends upon several variables, including a test subject, the test shall be performed in
the room qualification and the type of test setup accordance with ANSI/AMCA 210 or other fan
utilized. aerodynamic performance test standard having
a demonstrated accuracy equivalent to
ANSI/AMCA 210.

5
AMCA 300-05

5.4 Mounting methods Where:

The method of mounting a test subject, or Amin = the minimum distance between the
connecting it to a non-integral driver, or sound source and the microphone,
connecting it to an airflow test facility is not m(ft),
specified. Any conventional method may be
used including vibration isolation devices and C2 = 0.61 (if using SI units), (2.0 if using
short flexible connectors. Other than these, IP units), and
sound and vibration absorptive material may not
be incorporated in the test subject unless it is a (LWr-Lpq) = is the maximum value for
standard part of the fan. Ducts shall be of metal Octave Bands 2 to 7.
or other rigid, dense non-absorptive material
and have no exposed sound absorption material
on the interior or exterior surfaces. If the test room and test setup have been
qualified in accordance with Annex A, the
The driving motor and drive, when not an continuous microphone traverse used for the
integral part of the test subject, may be damped qualification shall also be used for the sound
or enclosed in any manner that does not expose pressure measurements.
sound absorption material to the test room.
When a driving motor and drive are an integral If a microphone traverse is used, it shall meet
part of the test subject, they may not be treated the following requirements:
in any manner, and normal belt tensions,
bearings, and lubricants shall be used. When a a) no point on the traverse shall be any
fan and its drive are both in the reverberant closer than Amin from the sound source;
room, the test results may contain sound
contributions from flanking paths as well as b) no point on the traverse shall be any
mechanical and/or electrical sound from the closer than 1.0 m (3.333 ft) to any
drive system. surface of the test room;

5.5 Duct length c) no point on the traverse shall, at any


time, be closer than 0.5 m (1.67 ft) to any
On a chamber or two-room setup, the length of surface of a rotating diffuser;
duct shall be consistent with acceptable practice
per ANSI/AMCA 210 necessary to accurately d) the microphone traverse should not lie in
establish the point of rating. any plane within 10° of a room surface;

The length of duct shown in Figures 2 and 3 is e) the microphone shall swing or move on a
consistent with the procedures of ANSI/AMCA normal path of an arc or straight line with
210. Care must be exercised to ensure that no a minimum distance of 3 m (10 ft)
duct resonances exist in close proximity to between the extreme points of travel;
specific frequencies of interest such as the
Blade Pass Frequency. f) the maximum air velocity over the
microphone shall be 1 m/s (200 fpm);
5.6 Microphone traverse and room
requirements g) room volume is not specified but the
room must be large enough in volume
When using the substitution method, the such that the volume of the test fan and
minimum distance between the sound source associated ductwork does not exceed
and the nearest microphone position may be 1% of the room volume;
calculated from:
h) neither the RSS nor fan shall be within
Amin
L L / 20
C210 Wr pq (5-1)
300 mm (1 ft) of any room centerline.

6
AMCA 300-05

5.7 Calibration of system measurement. If the microphone is moved, care


shall be exercised to avoid introducing
Before each sound power determination, the acoustical or electrical noise (for example, from
following calibration checks shall be performed. gears, flexing cables, or sliding contacts) that
A calibration check shall be made of the entire could interfere with measurement.
measurement system at one or more
frequencies within the frequency range of The frequency response of the instrument
interest. An acoustical calibrator conforming to system shall be flat over the frequency range of
ANSI S1.40 and with an accuracy of ± 0.5 dB interest within the tolerances given in Table 3,
shall be used for this purpose. In conformance and applied as outlined in ANSI S12.51.
with ANSI S1.40, the calibrator shall be checked
at least once every year to verify that its output Table 3 - Tolerances for the instrument
has not changed. In addition, an electrical system
calibration of the instrumentation over the entire
frequency range of interest shall be performed One-third Octave Band Tolerance
periodically, at intervals of not more than one Center Frequency (Hz) (dB)
year. 40-80 ±1.5
100-4000 ±1.0
The microphone and its associated cable shall 5000-8000 ±1.5
be chosen so that their sensitivity does not 10000 ±2.0
change by more than 0.2 dB over the 12500 ±3.0
temperature range encountered during the

SOUND POWER CALCULATIONS

Installation Type LW Equation

A: Free Inlet LWm=Lpc+(LWr-Lpq)


Free Outlet

This test procedure and the above calculations are based on the following assumptions:

1. Directivity from the fan is averaged by the reverberant room and the microphone location is
such that it is sensing total averaged sound pressure levels.

2. No resonances are present on either the fan structure, supporting devices, or driving devices
that provide any significant pure tones that may add to the fan recorded sound pressure
levels.

Section 5, Figure 1 - Fan total sound testing

7
AMCA 300-05

1 to 3D

*May require acoustical treatment.

SOUND POWER CALCULATIONS

Installation Type LW Equations


A or B: Free Inlet LWmi=Lpc+(LWr-Lpq)
C or D: Ducted Inlet LWi=Lpc+(LWr-Lpq)+Ei

This test procedure and the above calculations are based on the following assumptions:

Acoustical energy in an outlet duct which terminates in a second room or chamber does not
contribute to fan test sound pressure levels. This requires adequate transmission loss between
adjourning rooms and the addition of absorptive material within a chamber to absorb this energy.
1. Adequate absorption takes place at the discharge of a duct in a second room or chamber so
that any energy passing down that duct is adequately attenuated.
2. Directivity from the fan is averaged by the reverberant room and the microphone location is
such that it is recording total averaged sound pressure levels.
3. Duct construction is such that the transmission loss through the duct wall is large enough to
eliminate any addition to measured room sound pressure levels.
4. No resonances are present on either the fan structure, supporting devices, or driving devices
that provide any significant pure tones that may add to the recorded fan sound pressure
levels.
5. Inlet orifices to control the operating point are not permitted, unless integral to the fan.

Section 5, Figure 2 - Fan inlet sound testing

8
AMCA 300-05

2 to 3D

*May require acoustical treatment.


SOUND POWER CALCULATIONS

Installation Type LW Equations


A or C: Free Outlet LWmo=Lpc+(LWr-Lpq)
B or D: Ducted Outlet LWo=Lpc+(LWr-Lpq)+Eo

This test procedure and the above calculations are based on the following assumptions:

1. Acoustical energy in an inlet duct that terminated in a second room or chamber does not
contribute to fan test sound pressure levels. This requires adequate transmission loss
between adjoining rooms and the addition of absorptive material within a chamber to absorb
this energy.
2. Adequate absorption takes place at the inlet of a duct in a second room or chamber so that
any energy passing down that duct is adequately attenuated.
3. Directivity from the fan is averaged by the reverberant room and the microphone location is
such that it is recording total averaged sound pressure levels.
4. Duct construction is such that the transmission loss through the duct wall is large enough to
eliminate any addition to measured room sound pressure levels.
5. No resonances are present on either the fan structure, supporting devices, or driving devices
that provide any significant pure tones that may add to the recorded fan sound pressure
levels.
6. Outlet orifices to control the operating point are not permitted, unless integral to the fan.

Section 5, Figure 3 - Fan outlet sound testing

9
AMCA 300-05

5.8 Equations measured in the test room with the test subject
and the RSS off. The background noise
The type of fan and its test setup determine the includes all noise sources not directly
calculations required to determine the sound associated with fan sound. Examples of
power levels (LW, LWm, LWi, LWmi, LWo, LWmo) of background noise sources are: noise due to the
the test subject. Equations for each test setup motion of the microphone and noise due to any
are included under the specific arrangement other external source. Efforts should be made
along with any qualifying statements or to keep the background noise level at a
limitations. Also included are any assumptions minimum. For a test, or set of determinations, at
that were made regarding these specific setups. various points of test subject operation,
End reflection factors (Ei) and (Eo), when background sound pressure levels need to be
required, shall be calculated from Annex E Duct observed once.
End Reflection Correction, using the appropriate
duct and orifice size. 6.1.2.2 Sound pressure levels, RSS (Lpqm)

It cannot be assumed that the inlet and outlet RSS sound pressure levels are those measured
sound powers are always equal. Therefore, in the test room with the RSS operating and the
total sound power levels shall not be used to test subject off. RSS sound pressure levels
derive inlet or outlet sound power levels. include background sound pressure levels. For
a test, or set of determinations, at various points
of test subject operation, RSS sound pressure
6. Observations and conduct of test levels need to be observed once.

6.1 Observations 6.1.2.3 Sound pressure levels, fan (Lpm)

6.1.1 Point of operation Fan sound pressure levels are those measured
in the test room with the test subject operating
Although the acoustical observations necessary and the RSS off. Fan sound pressure levels
to determine sound power output are the same include background sound pressure levels. Fan
for all types of fans, the non-acoustical sound pressure levels must be observed for
observations necessary to determine the each operating point.
aerodynamic point of operation differ. This
standard provides different test setups for the Note: The observations above are valid only
testing of various fan types. Regardless of the when taken in a room that is qualified per the
test setup, the point of operation shall be procedures defined in Annex A or B.
determined. If the sound test setup also
conforms to one of the test setups in 6.1.3 Test conditions
ANSI/AMCA 210, then the point of rating can be
established with sufficient accuracy. If the The test conditions shall, as nearly as possible,
sound test setup does not conform to one of the be the same for all sound pressure level
test setups in ANSI/AMCA 210, steps must be measurements. Operation of the microphone
taken to ensure that the fan rotational speed is traverse and any rotating vanes shall be the
known within ± 1% and the point of operation same for all measurements. Observers and
can be established within ± 5% along a system operators should not be in the test room during
curve. measurements, but if it is absolutely necessary
for them to be present, they shall be away from
6.1.2 Sound pressure levels the test subject and remain in the same position
during the test. Readings should be a time
6.1.2.1 Sound pressure levels, background weighted average over an integral number of
(Lpb) microphone swings. The time span used shall
be sufficient to provide a stable value and shall
Background sound pressure levels are those be a minimum of 30 seconds for frequency

10
AMCA 300-05

bands ” 160 Hz, and 15 seconds for frequency


bands • 200 Hz. 6.2.3 Laboratory and instruments

A) Laboratory name
6.2 Information to be recorded
B) Laboratory location
As applicable, the following information shall be
compiled and recorded for all observations C) Technician(s) conducting test
made in accordance with this standard.
D) List of test equipment used, with
6.2.1 Test subject calibration information

A) Description of the test subject E) Scope of room qualification. Data shall


indicate whether the room is qualified for
1) Manufacturer full octaves or one-third octaves, and in
2) Model the case of pure tone testing, the one-
3) Nominal size third octaves for which the qualification
4) Impeller diameter, mm (in.) applies.
5) Number of impeller blades
6) Blade angle setting (adjustable or 6.2.4 Acoustical data
variable pitch fans only)
7) Number of stator vanes A) Background sound pressure levels Lpb
2 2
8) Inlet area, m (ft )
9) Outlet area, m (ft2)
2
B) RSS sound pressure levels Lpqm

B) Operating conditions C) Background corrections for the RSS

1) Fan rotational speed, rev/min D) Fan sound pressure levels Lpm


2) Fan airflow rate, m3/s (ft3/min)
3) Fan static pressure or total pressure E) Background corrections for the fan
at actual test conditions, Pa (in. wg)
4) Fan air density, kg/m3 (lbm/ft3) F) Un-weighted fan sound power levels
LWmi or LWmo
C) Mounting conditions
G) End reflection correction data
1) Test figure per this standard
2) Test Installation Type 1) End reflection correction values Ei
3) Sketch showing the test room setup, or Eo
including the dimensional locations 2) Duct length
of the test subject and points or path 3) Flush or non-flush mounting of the
of acoustical measurements duct into the test room
4) Orifice plate inside diameter, m (ft)
6.2.2 Test environment
H) Test date
A) Barometric pressure, kPa (in. Hg)

B) Ambient dry-bulb temperature, °C (°F) 7. Calculations

C) Ambient wet-bulb temperature, °C (°F) Calculations are affected by the Installation Type
and setup. See Section 5.8 in addition to the
D) Fan inlet dry-bulb temperature, °C (°F) following.

E) Static pressure at the fan inlet, Pa (in.


wg)

11
AMCA 300-05

7.1 Background correction 7.2 Sound power level (LW)

The observed RSS or test subject sound A sound power level is calculated using
pressure levels include both the sound source equations given in Section 5. The equations
and background noise. The effect of vary with product type and test setup. The
background noise level is termed background sound power level of a full octave band may be
correction and must be subtracted from the calculated from one-third octave band values by
observed sound pressure level. Background using the formula:
correction values depend on the difference
between the observed sound pressure levels
and the background noise levels. § LW 1 LW 2 LW 3 ·
LW 10 log10 ¨¨10 10  10 10  10 10 ¸¸ (7-3)
When the difference between the observed © ¹
sound pressure levels (RSS – background) in a
frequency band is less than 6 dB, the Where:
corresponding sound pressure level from the
source cannot be determined accurately by this LW1, LW2, and LW3 are one-third octave sound
standard. For any band for which the difference power level values.
between the background and the (background +
source) sound pressure level is less than 6 dB,
Lpc shall be reported as 1.3 dB less than Lpm. 8. Results and report
The data for each such band shall be clearly
marked as upper boundary levels. Test results are presented as the sound power
level, in dB, in each of the eight full octave
A sound pressure level reading shall be bands for each fan test speed and point of
corrected for background noise level by operation. Full octave bands are given in Table
logarithmic subtraction using the following 1. The report shall also include data defined in
formulae: Sections 8.1 through 8.3. This standard does
not require that pure tone effects be isolated
Test subject (fan) sound pressure level: from broad-band sound. However, a laboratory
equipped with suitable instrumentation is
encouraged to investigate and report pure tones
§ §¨ Lpm ·¸ § Lpb · ·
¨ ¸ separately.
¨ © 10 ¹ 10 ¸
Lpc 10 log10 ¨ 10  10© ¹ ¸ (7-1)
¨ ¸ 8.1 Test subject
© ¹
A) Description of the test subject
RSS sound pressure level:
1) Manufacturer
2) Model
3) Nominal size
§ Lpqm Lpb ·
4) Impeller diameter, mm (in.)
Lpq ¨
10 log10 10 10  10 10 ¸ (7-2)
¨ ¸ 5) Number of impeller blades
© ¹ 6) Blade angle setting (adjustable or
variable pitch fans only)
Example: The sound pressure level of a fan in a
given frequency band is observed to be 58 dB. B) Operating conditions
The background sound pressure level in the
same band is observed to be 51 dB. The 1) Aerodynamic performance test
background value is subtracted logarithmically standard
from the fan sound pressure level using 2) Fan rotational speed, rpm
3 3
Equation 7-1, which results in 57 dB (rounded). 3) Fan airflow rate, m /s (ft /min)

12
AMCA 300-05

4) Fan static pressure or total pressure nearest whole decibel


at actual test conditions, Pa (in. wg)
5) Fan air density, kg/m3 (lbm/ft3) B) Test date

C) Mounting conditions C) Background sound pressure level in


each reported band
1) Test Figure per this standard
2) Installation Type D) Background correction for the RSS for
each reported band
8.2 Laboratory and instruments
E) RSS sound pressure level in each
A) Laboratory name reported band

B) Laboratory location F) Background correction for test subject,


in each reported band
8.3 Acoustical data
G) Test subject sound pressure level, in
A) Un-weighted fan sound power level, in each reported band
each reported band, reported to the

13
AMCA 300-05

This page intentionally left blank.

14
AMCA 300-05

Annex A
(normative)

Room qualification: full and one-third octave

A.1 General

This annex covers the procedures for a broad-band qualification of a test room for full and one-third
octave bands. If pure tone qualification is required, refer to Annex B Room Qualification: Pure Tones /
Narrow Band.

A.2 Instruments and equipment

The instruments and microphone traverse shall be the same as those used during the actual testing of a
fan. The instruments shall conform to the requirements given in Sections 4.1. through 4.4, inclusive. The
microphone traverse shall conform to the requirements of Section 5.6. The test procedure given in this
annex requires the use of a Reference Sound Source (RSS) having the characteristics specified in
Section 4.5.

A.3 Test procedure

Eight or more measurements shall be made of the reverberant field sound pressure levels in the room,
each with the RSS placed at a different location within the room, under the following conditions:

A.3.1 Each location for the RSS shall be selected on the floor and shall not be closer than 1 m (3 ft)
from a wall and not closer to any microphone than permitted by equation Amin (Section 5, Figure 4). The
distance between any two RSS locations shall be greater than 0.9 m (3 ft). No source location shall lie
within ± 300 mm (1 ft) of a room centerline. The RSS locations shall be in the general vicinity of the
locations intended for the test subject as seen in a plan view of the test room.

A.3.2 With the RSS at each of the eight (or more) above locations, determine the average sound
pressure levels in accordance with the procedures of Section 6.

A.3.3 The microphone traverse, sound diffuser (if any), instruments and observation times shall be
identical to those to be used for a test subject.

A.4 Computation procedure

For each frequency band for which the test room is to be qualified, the standard deviation s, in dB, shall
be computed using the formula:

1/ 2
­ nRSS 2½
° 1 °
s ®
n  1
¦ j
ªL
«¬ pq
 Lpq º ¾
»¼
(A.4-1)
°̄ RSS j 1 °
¿

Where:

(Lpq)j = the sound pressure level, in dB, averaged over all microphone positions, when the RSS is in the
jth location
___
Lpq = arithmetic mean of (Lpq)j values, in dB, averaged over all RSS locations

15
AMCA 300-05

nRSS = number of RSS locations, a minimum of eight

A.5 Qualification

For each frequency band, the test room qualifies for the measurement of broad-band sound if the
computed standard deviation s, in dB, does not exceed the limits given in Table A1.

Table A1 - Maximum allowable standard deviation S, (dB)

Octave One-Third Maximum


Band Octave Band Allowable
Center Center Standard
Frequencie Frequencies Deviation (dB)
s (Hz) (Hz) s
63 50 to 80 3.0
125 100 to 160 1.5
250 and 500 200 to 630 1.0
1000 and 800 to 2500 0.5
2000
4000 and 3150 to 10000 1.0
8000

16
AMCA 300-05

Annex B
(informative)

Room qualification: pure tones / narrow-band

B.1 General

This annex covers the procedure for the qualification of a test room to investigate pure tones. The
reference document for this procedure is ANSI S12.51. Qualification testing applies only to those one-
third octave bands having mid-frequencies from 100 Hz to 2500 Hz, inclusive, as shown in Table B1.
Qualification excludes those bands having mid-frequencies below 100 Hz and is not required for those
bands having a mid-frequency greater than 2500 Hz. The qualification testing applies to a specific
location in the test room and determines which of the one-third octave bands the test room location is
qualified for. A sound test based on such qualification must state the mid-frequency of the one-third
octave band(s) qualified for the test by this procedure.

B.2 Instruments and equipment

The instruments shall be as specified in Section 4 with the following substitutions / additions.

a) The signal analyzer shall be a one-third octave band analyzer conforming to ANSI S1.11.
b) The sound source will consist of:
1) A loudspeaker / horn: one or more, each having a sufficiently smooth frequency response
within the range of frequencies to be qualified.
2) A frequency generator, tunable to and meeting the tolerances given for the frequencies given
in Table B1. A digital frequency synthesizer is recommended for ease of setting frequency.
3) A frequency counter accurate within ± 0.05 Hz over the pertinent frequency range.
4) A power amplifier of suitable power and having an output impedance compatible with the
loudspeaker(s) / horn(s).
5) A voltmeter capable of monitoring within ± 0.05% of the voltage across the loudspeaker(s) /
horn(s) at all test frequencies.

B.3 Test procedure

Qualification testing consists of two sections, the first being concerned with the near-field characteristics
of the loudspeaker / horn and the second with the test room itself. In both sections, measurements are
made for each of the discrete frequencies associated with the one-third octave band being qualified. The
same test equipment must be used for both sections of the qualification testing.

B.3.1 Loudspeaker / horn test

The loudspeaker / horn shall be located on the horizontal surface of a hemi-anechoic field with the open
cone facing upward. A microphone with diaphragm horizontal is located over the center of the
loudspeaker / horn 10 to 20 mm (0.375 to 0.75 in.) above the plane of the loudspeaker / horn rim. The
input voltage to the loudspeaker / horn must be sufficient to overcome background noise but must in no
case be permitted to cause physical distortion of the loudspeaker / horn components. The sound
pressure levels for the discrete frequencies of a one-third octave band are then measured. The
loudspeaker / horn is suitable only if the sound pressure levels at adjacent frequencies do not differ by
more than 1 dB. This test determines the near-field characteristics of the loudspeaker / horn and gives
calibration sound pressure levels for the loudspeaker / horn.

B.3.2 Room test

The loudspeaker / horn shall be positioned in the room at the horizontal and vertical coordinates intended

17
AMCA 300-05

for the test subject and placed so that the open cone faces away from the nearest room surface. Using
the same input voltage to the loudspeaker(s) / horn(s) as for the loudspeaker / horn test, space and time
averaged sound pressure levels Lps are measured for the discrete frequencies of the one-third octave
band.

B.4 Computation

The room test sound pressure level is then corrected to remove the effect of the loudspeaker’s / horn’s
near-field characteristic by subtracting the loudspeaker / horn test sound pressure level. The arithmetic
mean for the room sound pressure level is then calculated, and the standard deviation s of the difference
between the average sound pressure level and the arithmetic mean sound pressure level is determined
by:

1/ 2
­ n 2½
° 1 °
s ® >
¦ Lps
°̄ n  1 k 1
k @
 Lps ¾
°
¿
(B.4-1)

Where:

(Lps)k = the corrected sound pressure level, in dB, averaged over all microphone positions, of the kth
discrete frequency,
___
Lps = the arithmetic mean of (Lps)k values averaged over all n test frequencies within the one-third
octave band,

n = the number of discrete frequencies within the one-third octave band.

B.5 Qualification

A test room is accepted as qualified for pure tone testing within a given one-third octave band if the
standard deviation s, in dB, for that band does not exceed the values given in Table B2. If a one-third
octave band does not qualify, some modification will be required to the microphone location, to the test
position, or to the room absorption [7] [8].

18
AMCA 300-05

Table B1 - Test frequencies for alternative qualification of reverberant room facility for measuring
sound power levels of noise sources containing significant discrete frequency components (from
ANSI S12.51-2002)

Center frequency of one-third octave bands, Hz


100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500
-- -- 147 -- -- -- 361 -- -- -- -- -- 1470 -- --
-- 113 148 -- 226 -- 364 -- -- -- -- 1130 1480 -- 2260
-- 114 149 -- 228 -- 367 445 564 712 -- 1140 1490 -- 2280
90 115 150 180 230 285 370 450 570 720 900 1150 1500 1800 2300
91 116 151 182 232 288 373 455 576 728 910 1160 1510 1820 2320
92 117 152 184 234 291 376 460 582 736 920 1170 1520 1840 2340
93 118 153 186 236 294 379 465 588 744 930 1180 1530 1860 2360
94 119 154 188 238 297 382 470 594 752 940 1190 1540 1880 2380
95 120 155 190 240 300 385 475 600 760 950 1200 1550 1900 2400
96 121 156 192 242 303 388 480 606 768 960 1210 1560 1920 2420
97 122 157 194 244 306 391 485 612 776 970 1220 1570 1940 2440
98 123 158 196 246 309 394 490 618 784 980 1230 1580 1960 2460
99 124 159 198 248 312 397 495 624 792 990 1240 1590 1980 2480
100 125 160 200 250 315 400 500 630 800 1000 1250 1600 2000 2500
101 126 161 202 252 318 403 505 636 808 1010 1260 1610 2020 2520
102 127 162 204 254 321 406 510 642 816 1020 1270 1620 2040 2540
103 128 163 206 256 324 409 515 648 824 1030 1280 1630 2060 2560
104 129 164 208 258 327 412 520 654 832 1040 1290 1640 2080 2580
105 130 165 210 260 330 415 525 660 840 1050 1300 1650 2100 2600
106 131 166 212 262 333 418 530 666 848 1060 1310 1660 2120 2620
107 132 167 214 264 336 421 535 672 856 1070 1320 1670 2140 2640
108 133 168 216 266 339 424 540 678 864 1080 1330 1680 2160 2660
109 134 169 218 268 342 427 545 684 872 1090 1340 1690 2180 2680
110 135 170 220 270 345 430 550 690 880 1100 1350 1700 2200 2700
111 136 171 222 272 348 433 555 696 888 1110 1360 1710 2220 2720
-- 137 172 -- 274 -- 436 -- 702 -- -- 1370 1720 -- 2740
-- 138 173 -- 276 -- 439 -- -- -- -- 1380 1730 -- 2760
Increment, 1 1 1 2 2 3 3 5 6 8 10 10 10 20 20
Hz

Tolerance of ±0.3 ±0.3 ±0.3 ±0.5 ±0.5 ±1 ±1 ±1.5 ±2 ±3 ±3 ±5 ±5 ±5 ±5


Increment,
Hz

Number of 22 26 27 22 26 22 27 23 24 23 22 26 23 22 26
test
frequencies,
n

Table B2 - Maximum allowable sample standard deviation, s

One-third Octave Maximum


Band Center Allowable
Frequencies (Hz) Standard
Deviation s (dB)
100 to 160 3.0
200 to 315 2.0
400 to 630 1.5
800 to 2500 1.0

19
AMCA 300-05

This page intentionally left blank.

20
AMCA 300-05

Annex C
(informative)

Uncertainties analysis

C.0 General

The analysis of the uncertainty associated with measurements made in accordance with this standard
provides identification of certain critical points so as to recognize the limitations of the results.
Furthermore, it provides an approximation, in real values, of the imprecision in the recorded results.

C.1 Definitions

Precision error is an error that causes readings to take random values on either side of some mean
value.

Systematic error is an error that persists and cannot be considered as due entirely to chance.

Uncertainty is an estimated value for error, i.e., what we think an error would be if we could and did
measure it by calibration. Although uncertainty may be the result of both precision and systematic errors,
only precision errors can be treated by statistical methods.

The uncertainty in a researched value is described by specifying the measured value followed by the
uncertainty interval at the desired confidence level:

LW = m ± w at P confidence level (C.1-1)

Where:
m = measured value
w = uncertainty
P = percent

C.2 Uncertainties

The uncertainties associated with the determination of sound power levels through measurements
performed in accordance with this standard are room response (C.3), fan operating points (C.4),
instrument error (C.5), and RSS (C.6). Uncertainties associated with duct end reflection corrections
involve the accuracy of estimating the losses from orifice plates (C.8). Other areas of interest involve the
use of octave or one-third octave bands (C.9) and the problems associated with testing in the 63 Hz band
(C.10).

C.3 Room response

A reverberant room is an appropriate place for determining the acoustical power of a source, such as a
fan, that emits a steady sound power. The reverberant room must be diffuse enough to produce a
reverberant field.

When a sound source is operated inside a reverberant room, the sound waves are reflected by the walls
and are propagated in all directions. If the paths of all the waves could be seen, we would notice a
number of repetitions, (e.g., the path followed by a wave between two parallel walls). These paths are
called normal modes. The greater the number of normal modes, the better the sound dispersion in the
room. The modes must be sufficiently numerous in any measurement band so that the microphone
traverse will serve to average the sound pressure. The number of normal modes in a given space

21
AMCA 300-05

increases with frequency. Hence, it is usually more precise to measure higher frequencies. When the
number of modes are few, it helps to measure the sound in many locations and average the results. Two
important sources of error may affect the measurements made in a reverberant room: 1) the error
introduced by measuring the sound field at a limited number of points, and 2) variations in sound power
due to the location of the sound source. Many sources radiate sound that is not entirely broad-band, but
contains significant discrete-frequency components, or pure tones. Some fans generate a pure tone at
the blade passage frequency and sometimes at harmonic frequencies.

In a reverberant room, a pure tone tends to excite certain modes that will dominate all others. This
noticeably increases the variability of the pressure field due to an insufficient dispersion of the sound field.
Due to the consequent inaccuracy of sound pressure averaging, the precision of the results is reduced.

C.3.1 Broad-band measurement in a reverberation room

Broad-band sound is uniformly distributed in frequency with relatively steady levels and with no prominent
discrete-frequency or narrow-band components. Measurement of broad-band sound may be made in a
test room qualified per Annex A.

C.3.2 Pure-tone measurement in a reverberation room

When a discrete-frequency component is present in the sound spectrum of a source, the spatial
variations in sound pressure level usually exhibit maxima separated by minima having an average
spacing of approximately 0.8 O, where O is the wavelength corresponding to the discrete frequency of
interest.

The presence of a significant discrete-frequency component in the sound produced by a source can often
be detected by a simple listening test. If such a component is audible, or detectable by narrow-band
analysis, the qualification procedure described in Annex B is recommended.

If the test room is not qualified for pure-tone measurement, the measurement uncertainty will most
probably be higher in the bands containing the blade passage frequency and its harmonics than if
measured in a qualified test room.

Discrete-frequency components may be present in the sound spectrum even when these components
are not audible. A conclusion that no discrete-frequency components are present can only be reached by
performing the test described in C.3.3.

C.3.3 Test for discrete-frequency components

The following procedure can be used to estimate the spatial standard deviation of the sound pressure
levels produced by the test subject in the test room.

Select an array of six fixed microphones (or a single microphone at six positions) spaced at least O/2
apart, where O is the wavelength of the sound corresponding to the lowest band mid-frequency of interest
and meeting all the requirements for microphone positions in Annex A. Locate the sound source at a
single position in the test room in accordance with Annex A.

Obtain the time-averaged sound pressure level Lpj at each microphone position according to the
techniques described in Annex A.

For each one-third octave band within the frequency range of interest, calculate the standard deviation s,
in dB, from the following equation:

22
AMCA 300-05

1/ 2
ª 1 nm 2º
s «
¦ Lpcj  Lp j
«¬ nm  1 j 1
»»¼ (C.3-1)

Where:

Lpcj = sound pressure level, corrected for the background sound level in accordance with the procedures
of Section 6.2.1 for the jth microphone position, dB
__
Lpj = arithmetic mean of (Lpc)j values, averaged over all microphone positions, dB

nm = number of microphone positions = 6

The magnitude of s depends upon the properties of the sound field in the test room. These properties are
influenced by the characteristics of the room as well as the characteristics of the sound source (i.e.,
directivity and spectrum of the emitted sound). In theory, a standard deviation of 5.57 dB corresponds to
a spectral component of zero bandwidth, i.e., a discrete tone.

Table C1 - Characterization of the presence of discrete-frequency or narrow-band components,


based upon the spatial variation of the sound field

Standard Characterization
Deviation, s
(dB)
s<1.5 Assume broad-band source (use
procedures of Annex A).
1.5<s<3 Assume that a narrow-band of
noise is present. Recommend use
of the qualification procedure in
Annex B.
s>3 Assume that a discreet tone is
present. Test room must qualify
per Annex B.

C.4 Fan operating points

When the sound power levels of a fan are determined, each measurement must relate to one point of
operation of the fan. Uncertainty in identifying this point thus affects the global uncertainty of the results.
Therefore it is recommended that the procedures of ANSI/AMCA 210 or other recognized fan
aerodynamic performance test standard be used as a guideline in identifying the test subject’s operating
points. The sensitivity of the sound levels to a change in point of operation is a function of the test
subject’s performance characteristics, and this will dictate how accurately the point of operation must be
determined. A fan that exhibits a large change in sound power level as airflow is changed (at a given fan
rotational speed) is of more concern than one that shows a small change in sound power level for the
same airflow change.

C.5 Instrument error

The frequency response of the instrument system shall be flat over the frequency range of interest to
within the tolerances given in Table C2.

23
AMCA 300-05

Table C2 - Tolerances for the instrument system

Frequency Tolerance
(Hz) (dB)
100 to 4000 ±1.0
5000 to 8000 ±1.5
10000 ±2

C.6 Reference sound source (RSS)

The sound power produced by the RSS shall be determined in octave and one-third octave bands within
the tolerances specified in Table C3.

Table C3 - Calibration accuracy for RSS

One-Third Tolerance
Octave Band (dB)
Center
Frequency (Hz)
100 to 160 ±1.0
200 to 4000 ±0.5
5000 to 10000 ±1.0

C.7 Estimated standard deviation for determination of sound power levels

The determination of sound power levels through measurements made in accordance with this standard
will result, with very few exceptions, in standard deviations that are less than or equal to those given in
Table C4. The standard deviations in Table C4 take into account the cumulative effects of all causes of
measurement uncertainty noted in C.3 through C.6 above, except for duct end reflection corrections and
the testing in an unqualified test room of fans containing pure-tones.

Table C4 - Estimated deviation of sound power level determinations

Octave One-Third Standard


Band Octave Band Deviation
Center Center (dB)
Frequency Frequency
(Hz) (Hz)
125 100 to 160 3.0
250 200 to 315 2.0
500 to 4000 400 to 5000 1.5
8000 6300 to 10000 3.0

C.8 Duct end reflection corrections

Table C5 gives the uncertainties for duct end reflection correction E for various 0.5 kD and r values.

24
AMCA 300-05

Table C5 - Uncertainties in duct end reflection correction E

Duct r Uncertainty in E (dB)


Configuration Range of 0.5 kD
<0.25 0.25-1 >1
Flush 1 ±3 ±2 ±0.5
Free Space 1 ±3 ±2 ±0.5
1-2 ±3 ±2 ±0.5
2-5 ±4 ±3 ±1
Note: When pure tones are present, uncertainties will be substantially greater.

C.9 Octave band vs. one-third octave band

According to this standard, the frequency analysis of sound may be performed either in full octave bands
or in one-third octave bands. Qualification of a reverberant test room for pure tones can only be effected
in the one-third octave bands. Full octave band analysis takes less time because fewer numerical values
are treated. However, this analysis supplies little information on the shape of a sound spectrum.
Furthermore, full octave band analysis does not allow isolation of pure tones in a spectrum; the poor
resolution of an octave band gives little information about a steeply sloping spectrum. The pure-tone
value produced by a test subject may be reduced by 1 to 2 dB without changing the octave band reading.

For certain test conditions, this standard uses a duct end reflection correction factor that is frequency
dependent. Because of this dependence, analysis in full octave bands instead of one-third octave bands
may cause an error of up to ± 2 dB.

Example:
Test Conditions: A fan having a 508 mm (20 in.) diameter inlet, no orifice plate, and low airflow.

There is a significant difference between the two methods of determining the octave band values. This
difference is a function of two things:

1) The shape of the sound spectrum determined by one-third octave band analysis, and
2) The slope of the duct end reflection attenuation curve at the point where the attenuation value is
evaluated.

The error made in using octave band analysis can overestimate or underestimate the real values.
Therefore, the use of one-third octave band analysis is recommended. Refer to Figure C1.

If full octave band analysis is performed, a precaution would be to adjust the fan rotational speed to
caused the blade passage frequency to fall in the central one-third octave band of any full octave band.
Care should also be taken to keep the blade passage frequency from falling on the border between
bands, thus avoiding the problems associated with the characteristics of filter skirts.

Table C6 - Example using full octave band analysis

1/3 Octave Lp Combined +E = (Lp+E)


Center Measured dB
Frequency,
(Hz)
50 80
63 65 80.2 +10.2 =90.4
80 64

25
AMCA 300-05

Table C7 - Example using one-third octave band analysis

1/3 Octave Lp Combined +E =(Lp+E)


Center Measured dB
Frequency,
(Hz)
50 80 +12.1 =92.1
63 65 +10.2 =75.2 =92.2
80 64 +8.3 =72.3

C.10 Accuracy of the 63 Hz octave band

At low frequencies, the sound power output of a source depends upon its position in the test room. At
low frequencies, very few modes are excited, and because of reflections from test room surfaces, the
reflected pressure at the source combines with the direct sound pressure field produced by the source.
This affects the radiation impedance seen by the source, and therefore its sound power output. This is
particularly true of the 63 Hz octave band. Most standards do not discuss this band, although it is
important to fan manufacturers and users alike. Measurements in this band must be reported. However,
the measured sound pressure values, and therefore the determined sound power level values, have an
uncertainty of ± 6 dB at best.

OCTAVE OCTAVE OCTAVE


BAND BAND BAND

OVER ESTIMATION NO ERROR UNDER ESTIMATION

Figure C1 - Effect of summing one-third octave bands

26
AMCA 300-05

Annex D
(informative)

Alternative procedure for reference sound source calibration

D.1 General

Calibration of a Reference Sound Source (RSS) in conformance with the requirements of ANSI S12.5
requires a hemi-anechoic room qualified for measurements over the entire frequency range of interest.
Laboratories that otherwise would be able to perform the required calibration but which are not qualified
for measurements in the first octave band may use the alternative procedure of this Annex. This
alternative procedure is based on sound intensity measurements per ANSI S12.12.

D.2 Equipment and facilities

Equipment and facilities shall be as required for RSS calibration in conformance with ANSI S 12.5, with
the exception that the hemi-anechoic chamber need not be qualified below the 125Hz full octave band
(100 Hz one-third octave band). Sound intensity measuring equipment shall comply with the
requirements of ANSI S 12.12.

Additional RSS units may be sound power level calibrated by comparing the sound power levels of the
source to another unit that was calibrated in accordance with Sections D.1 through D.5. It is not
necessary that each and every reference sound source be calibrated directly in accordance with the
procedures described below. It may be possible to transfer a calibration from one unit to another by
using a simpler type of test. For example, the Substitution Method of the present standard might be used
to calibrate (secondary calibration) one reference sound source relative to another, similar, reference
sound source that has been calibrated as described below (primary calibration). In order that such a
secondary calibration does not result in an unacceptable degradation of accuracy, it normally will be
necessary to use more source locations and microphone positions than the minimum requirements of the
present standard and to exercise additional caution in carrying out the measurements.

D.3 Qualification

The RSS calibration procedure of ANSI S12.5 shall be carried out over the 50 Hz through 10,000 Hz one-
third octave band frequency range and 63 Hz through 8000 Hz full octave band frequency range. If the
calibration is in conformance with ANSI S12.5 in all respects except for the qualification of the test facility
below the 100 Hz one-third octave band, the alternative calibration procedure below may be used. If the
calibration is not in complete conformance with ANSI S12.5 for any other reason, the alternative
calibration procedure is not applicable.

D.4 Procedure

The requirements of ANSI S12.5 are duplicated in the lowest three full octave (nine one-third octave)
bands, with the substitution of sound intensity level measurements, made in compliance with ANSI
S12.12, for the sound pressure level measurements required by ANSI S12.5. For all measurements,
sound intensity shall be measured in the outward radial direction. The sound power levels determined
from these measurements shall be compared with those determined from the corresponding sound
pressure level measurements. If in all frequency bands the determined sound power levels differ by no
more than the tolerances given in Table D1, the calibrated sound power levels for the RSS are reported
as specified in Section D.5. The directivity index is not calculated from the intensity measurements.

27
AMCA 300-05

Table D1 - Tolerance for measured sound power level difference

Octave Band One-third Tolerance


(Hz) Octave (dB)
Band (Hz)
63 50-80 ±4.0
125-250 100-315 ±1.0

D.5 RSS sound power levels

The reported RSS sound power levels and directivity index shall be as determined by the ANSI S12.5
procedure for the 100 Hz through 10,000 Hz one-third octave bands and the 125 Hz through 8,000 Hz full
octave bands. For the 50 Hz through 80 Hz one-third octave bands and the 63 Hz full octave band, the
reported RSS sound power level(s) shall be as determined from the sound intensity measurements, and
the directivity index is not to be reported. The calibration report shall be marked to indicate the levels
determined from sound intensity measurements, and shall indicate whether the calibration was performed
in full compliance with this Annex.

28
AMCA 300-05

Annex E
(normative)

Duct end reflection correction

E.1 General

Conditions at the end of a test duct will prevent some of the sound energy from being transmitted into the
test room. Therefore, the sound power measured in the room will be less than the true sound power in a
duct. Unless an anechoic termination is used, correction factors must be added to the fan sound
pressure measured in the test room in order to account for the reduction caused by end reflection.

The prediction of the duct end reflection is difficult. Theoretical solutions exist only for round ducts with
highly idealized end conditions and are based on the assumption that the frequency is low enough that
only plane waves exist (which implies that ka<S). Actual fan test setups rarely, if ever, conform to the
conditions under which the theoretical solutions are valid. Using the methods suggested in this Annex
will result in predicted values that are reasonably close to the actual values. Nonetheless, the test setup
should be selected to minimize the potential error by using components that most closely reproduce the
theoretical conditions.

For open ducts (i.e., no orifice) theoretical solutions exist for two cases: a thin-walled round duct
terminating in an infinite space [On the Radiation of Sound from an Unflanged Circular Pipe, Levine,
H., and Schwinger, J. – Physical Review, Vol. 72, No. 4, February 15, 1948] and a round duct terminating
in an infinite wall [Fundamentals of Acoustics, 3RD Edition, Kinsler, Frey, Coppens and Sanders, Wiley,
New York, 1982 , equations 9.13 and 9.14]. Most test setups incorporate terminations that use a flanged
duct terminating in a large space, which would make the solution provided by Levine & Schwinger more
appropriate, assuming no orifice is used.

For ducts with orifices, no theoretical solution exists for the case of a duct terminated in infinite space.
For the flush-mounted duct (duct terminated in an infinite wall) the effect of an orifice plate with a round,
centrally located hole can be calculated [Acoustics, Beranek, L., McGraw-Hill, New York, 1950, Section
5.2].

For most test setups, when the test is conducted using an orifice on the tested end, there is no theory to
predict the end correction values.

E.2 End reflection curves

It is strongly recommended that, whenever possible, sound test setups be chosen so that there is no
requirement to apply duct end correction. In the event that circumstances require a setup indicating the
presence of a duct end correction there are four cases to be considered. The four cases are considered
separately below.

E.2.1 Open ducts in a large space

To determine the end reflection values, it is necessary to first calculate the reflection coefficient R, which
gives the fraction of the energy reflected back into the duct. Levine and Schwinger reduced the exact
solutions to manageable forms, one for ka<1 and one for ka>1. Note: k = Z/c = 2ʌ/O, a = D/2, and
Z = 2 Sf.

The two equations are:

29
AMCA 300-05

ª  ka 2 º ª ka 4 § § 1 · 19 · º
R exp« » «1 ¨ log10 ¨ ¸  ¸» for ka < 1 Eq. E-1
«¬ 2 »¼ «¬ 6 © © Jka ¹ 12 ¹ »
¼

ª 3 1 º
R Ska exp(ka)«1 32 » for ka > 1 Eq. E-2
«¬ ka 2 »¼

The ratio between the transmitted sound and the reflected sound is D = 1 - ~R~2 and thus the end
correction (in dB) is E 10 log10 D . These equations shall be used to calculate E as a function of ka
(0.5kD). The resulting curve is shown for illustrative purposes in Figure E1 (r=1). Values are presented
up to ka = 4, even though the equations are strictly limited to ka < 3.832.

r=5

r=2

r=1

Figure E1 - End correction for open ducts in large space

30
AMCA 300-05

E.2.2 Open ducts terminated in a wall

For the case of a round duct terminated at a large wall, the end correction can be determined using
equations 9.13 and 9.14 from Kinsler, Frey, Coppens and Sanders with the impedence calculated
using equations 5.1, 5.2 and 5.3 from Beranek. It should be noted that there is no transition at the wall-
duct interface. The equations to be used to calculate E as a function of ka are given below.

ª J (2ka) º SUc
ZM Sa2Uc «1 1 » j K1 2ka Eq. E-3
«¬ ka »¼ 2k 2

W W3 W5 W7
J1 W  2  2 2  2 2 2 ˜ ˜˜ Eq. E-4
2 2 ˜4 2 ˜4 ˜6 2 ˜4 ˜6 ˜8

2 §W3 W5 W7 ·
K1 W ¨  2  2 2 ˜ ˜˜¸ Eq. E-5
S© 3 3 ˜5 3 ˜5 ˜7 ¹

R
B Z Sa Uc  1
M
2

Eq. E-6
A
Z Sa Uc  1
M
2

2
D 1 R Eq. E-7

E 10 log10 D Eq. E-8

The series for the Bessel functions J1 and K1 converge rapidly (at least for values of ka < 3.6), so the
computation of E vs. ka is straightforward. The resulting curve for illustrative purposes is shown in Figure
E2 (r=1). As before, values are shown up to ka = 4, but for ka >3.6, the value of D is defined to be 1.

E.2.3 Orificed ducts terminated in a large wall

If a round duct terminating into a large wall is fitted with an orifice plate with a centrally located round
hole, the equations in Section E.2.2 may be easily modified to predict the end reflection. Continuing with
the assumption of plane wave propagation, the end reflection may be calculated by calculating R using ka
based on the orifice radius, and calculating the transmission coefficient Į by assuming that the orifice
reduces the transmission coefficient by a factor of 1/r, where r is the ratio of duct area to orifice area. End
reflection values for r = 2 and r = 5 are shown for illustrative purposes in Figure E2.

The curves in Figure E2 are drawn to ka = 4, even though their range of applicability may be limited to
much lower values. For the open duct (r = 1) the end reflection is clearly seen to be zero for all values of
ka > 3 since the failure to meet the plane wave criteria is not critical. For the orificed cases, the end
correction values for ka>1 are questionable due to the failure to meet the plane wave criteria, and are
0.5
very suspect for ka > (Sr) since for these values of ka the wave length is smaller than the orifice
diameter.

31
AMCA 300-05

r=5

r=2

r=1

Figure E2 - End correction for open ducts terminated in a large wall

E.2.4 Orificed ducts terminating in a large Space

Although there is no theory applicable to these cases, it is reasonable to argue on physical grounds that
the effect of the orifice must be reasonably similar to the flush-mounted case. Adopting this approach,
the curves for r = 2 and r = 5 have been added to Figure E1 by merely adding the orifice effect
determined from Figure E2. The same qualifications to the accuracy at ka > S apply here also. Values
can be found in Table E1

32
AMCA 300-05

Table E1 – End corrections for orificed ducts terminating in a large space

ka r=2 r=5 ka r=2 r=5


0.14 18 18.6 1.5 3.9 7.4
0.15 17.4 18 1.6 3.7 7.4
0.16 16.9 17.5 1.7 3.6 7.3
0.17 16.5 17.1 1.8 3.5 7.3
0.18 16 16.7 1.9 3.5 7.3
0.19 15.6 16.3 2.0 3.4 7.3
0.20 15.2 15.9 2.1 3.3 7.2
0.25 13.5 14.3 2.2 3.3 7.2
0.30 12.2 13.1 2.3 3.3 7.2
0.35 11.1 12.1 2.4 3.2 7.2
0.40 10.2 11.3 2.5 3.2 7.2
0.45 9.4 10.6 2.6 3.2 7.1
0.50 8.8 10.1 2.7 3.2 7.1
0.55 8.2 9.6 2.8 3.2 7.1
0.60 7.7 9.2 2.9 3.1 7.1
0.65 7.3 8.9 3.0 3.1 7.1
0.70 7 8.6 3.1 3.1 7.1
0.75 6.7 8.4 3.2 3.1 7.1
0.80 6.4 8.2 3.3 3.1 7.1
0.85 6.1 8.0 3.4 3.1 7.0
0.90 5.9 7.9 3.5 3.1 7.0
0.95 5.7 7.8 3.6 3.1 7.0
1.0 5.4 7.7 3.7 3.0 7.0
1.1 4.8 7.6 3.8 3.0 7.0
1.2 4.5 7.5 3.9 3.0 7.0
1.3 4.52 7.5 4.0 3.0 7.0
1.4 4 7.4

33
AMCA 300-05

Annex F
(informative)

Filter-weighted measurements

In certain sound measurement situations, the presence of high amplitude sound at frequencies ” 45 Hz
can reduce the effective dynamic range of the analyzer in the measurement frequency range of interest
for this standard (45 Hz to 11,200 Hz). While use of an analyzer with a large dynamic measurement
range can solve this problem, it may sometimes be necessary to use another approach.

Sound pressure level readings may be made with the sound level meter or signal amplifier set for a well-
defined filter weighting effect in order to improve the dynamic range and measurement quality, provided
that any effect in the frequency range 45 Hz to 11,200 Hz is compensated and the equipment satisfies all
the requirements of Section 4 of this standard. The weighting filter shall be the same for all
measurements (background, RSS, and fan).

34
AMCA 300-05

Annex G
(informative)

Radiation of sound by fan casing

G.1 General

The sound radiated by a fan casing may be determined by the following method. Except as provided for
below, all the requirements of this standard apply.

G.2 Instruments and equipment

Shall be as required in Section 4.

G.3 Setup and test

The fan inlet and fan outlet shall be ducted to termination points outside the test room. Ducts and
connections should be constructed and secured such that the acoustic energy radiated through this
equipment is no more than 10% of the total energy radiated by the fan casing into the test room. The test
room sound pressure levels may be affected by sound radiating from the inlet and discharge ductwork
connected to the test subject, causing measured sound pressure levels to be somewhat higher than the
true casing radiated sound pressure levels. This effect can be minimized by using internally lined round
ductwork. No correction for duct-radiated sound power is allowed. NOTE: If there is any doubt
concerning the contribution of extraneous sound transmitted by ductwork, the importance of same can be
checked by increasing the transmission loss of the ductwork.

G.4 Observations and calculations

Sound pressure levels Lpq and Lpk shall be observed as provided for in Section 6. The sound pressure
levels Lpq and Lpk are observed and subject to the provisions for Lp in Section 6. For possible pure tones
and additional testing, the results of the test of a fan casing are subject to the same requirements as the
test of a fan.

LWk = Lpk + (LWr - Lpq ) in each frequency band (G.4-1)

Where:

LWk = sound power radiated through the fan casing,


Lpk = fan casing sound pressure level.

35
AMCA 300-05

Annex H
(informative)

Total fan sound testing with attached ducts

It is intended that the fan sound power levels determined by this standard reflect the sound produced at a
known fan operating point. The length of test ducts used to determine sound power would, therefore, be
identical to the duct length defined an ANSI/AMCA 210. It has been determined that shorter duct lengths
are also acceptable and may be used. Care must be taken to ensure that for the actual duct lengths
used, no duct resonances exist in close proximity to specific frequencies of interest, e.g., the blade
passage frequency.

Although it is recognized that the inlet and outlet sound power levels of a fan are generally not equal, it is
necessary to make some assumptions about the relationship between these levels to apply duct end
reflection correction. The equations in Figure H1 are based upon the assumption that the inlet and outlet
sound power levels of a fan are equal.

36
AMCA 300-05

FAN

OPTIONAL
ORIFICE
FAN FAN

B: FREE INLET C: DUCTED INLET D: DUCTED INLET


DUCTED OUTLET FREE OUTLET DUCTED OUTLET

Installation Type LW Equations

B: Free Inlet, LW = Lp + (LWr – Lpq) + [3 – 10 log10 (1 + 10(Eo/10))] + Eo


Ducted Outlet

C: Ducted Inlet, LW = Lp + (LWr – Lpq) + [3 – 10 log10 (1 + 10(Ei/10))] + Ei


Free Outlet

D: Ducted Inlet, LW = Lp + (LWr – Lpq) + Ei + Eo + [3 – [10 log10 (10(Eo/10) + 10(Ei/10))]]


Ducted Outlet

This test procedure and the above calculations are based on the following:

1. Directivity from the fan is averaged by the reverberant test room and the microphone location is such that it is
sensing total averaged sound pressure levels.
2. Duct construction is such that the transmission loss through the duct wall is large enough to eliminate any addition
to the measured sound pressure levels.
3. No resonances are present on either the fan structure, supporting devices or driving devices that provide any
significant pure tones that may add to the measured sound pressure levels.
4. The factor of 3 in the above equations is based on the assumption that fan sound power is equally distributed
between inlet and outlet.

Figure H1 - Fan total sound testing with ducts attached

37
AMCA 300-05

Annex J
(informative)

References

[1] AMCA Standard 300-67 Test Code for Sound Rating, Air Movement and Control Association
International, Inc., Arlington Heights, IL, 1967.

[2] AMCA Standard 301-90 Methods for Calculating Fan Sound Power Levels from Laboratory Test
Data, Air Movement and Control Association International, Inc., Arlington Heights, IL, 1990.

[3] Harris, C.M., Editor, Handbook of Noise Control, 2nd Edition, McGraw-Hill, New York, NY, 1979

[4] Parker, S.P., Dictionary of Scientific and Engineering Terms, 4th Edition, McGraw-Hill, New York,
NY, 1989

[5] ANSI S1.6-1984 (R1990) Preferred Frequencies, Frequency Levels and Band Numbers for
Acoustical Measurements, Acoustical Society of America, New York, NY, 1990
(AMCA #1108-84-AO)

[6] Sepmeyer, L.W., Computed Frequency and Angular Distribution of the Normal Modes of
Vibration in Rectangular Rooms, Journal of the Acoustical Society of America, New York, NY,
Vol. 37 – No. 3, March, 1985 (AMCA #1891-65-AO)

[7] AMCA #1901-85-A1 List of References on Room Calibration, Air Movement and Control
Association International, Inc., Arlington Heights, IL, 1985.

[8] Crocker, M. J., w/ Pande, L. and Sandbakken, R., Investigation of End Reflection Coefficient
Accuracy Problems with AMCA 300-67, Herrick Laboratories Report HL 81-16, Purdue
University, West Lafayette, IN, 1981. (AMCA #1184-81-A6)

[9] Noise Control Engineering, Vol. 7, No. 2, Noise Measurement Facilities, and ANSI S1.21-1972,
Methods for the Determination of Sound Power Levels of Small Sources in Reverberant Rooms.

[10] ANSI S12.11-1987 (R1993) Methods for the Measurement of Noise Emitted by Small Air Moving
Devices, Acoustical Society of America, New York, NY, 1993.

[11] Baade, P.K., 1977, Effects of acoustic loading on axial flow fan noise generation, Noise Control
Engineering, 8(1):5-15

[12] ANSI S12.51-2002 Nationally Adopted International Standard (NAIS Standard), Acoustics –
Determination of sound power levels of noise sources using sound pressure – Precision method
for reverberation rooms, Acoustical Society of America, New York, NY, 1993.

38
AIR MOVEMENT AND CONTROL
ASSOCIATION INTERNATIONAL, INC.
30 West University Drive
Arlington Heights, IL 60004-1893 U.S.A.
Tel: (847) 394-0150 Fax: (847) 253-0088
E-Mail : info@amca.org Web: www.amca.org

The Air Movement and control Association International, Inc. is a not-for-profit international association of the
world’s manufacturers of related air system equipment primarily, but limited to: fans, louvers, dampers, air
curtains, airflow measurement stations, acoustic attenuators, and other air system components for the industrial,
commercial and residential markets.