Minimizing Excessive Sound

in Ventilation System Design

s
Minimizing Excessive
Sound in Ventilation
System Design
Application Guide
125-1929 Rev. 4, June, 2009

Siemens Building Technologies, Inc. ii

Rev.4, June, 2009

NOTICE

Document information is subject to change without notice by Siemens Building Technologies, Inc. Companies, names, and various
data used in examples are fictitious unless otherwise noted. No part of this document may be reproduced or transmitted in any form
or by any means, electronic or mechanical, for any purpose, without the express written permission of Siemens Building
Technologies, Inc.

All software described in this document is furnished under a license agreement and may be used or copied only in accordance with
license terms.

For further information, contact your nearest Siemens Building Technologies, Inc. representative.

Copyright 2004 by Siemens Building Technologies, Inc.

TO THE READER

Your feedback is important to us. If you have comments about this manual, please submit them to:
[email protected]

CREDITS
APOGEE is a trademark of Siemens Building Technologies, Inc.
Other product or company names mentioned herein may be the trademarks of their respective owners.
Printed in U.S.A.

Siemens Building Technologies, Inc. ii

Table of Contents
About this Application Guide I

Purpose of this Guide I

How this Guide is Organized I
Suggested Reference Materials II
Symbols III
Getting Help III
Where to Send Comments III

Chapter 1–Introduction 1

Scope of This Guide 1

HVAC Sound Transmission 2
Background Sound 2
Laboratory Applicability 2
Computer Program Sound Analysis 2

Chapter 2–Physics of Sound 5

Sound Wave Propagation 5

Sound Wave Parameters 6
Sound Measurement Parameters 8
Sound Power Level 8
Decibels 9
Sound Pressure Level 11
Octave Bands 13
A-Weighted Sound Level 16
NC Curves 17
RC Curves 18
Determining an RC Rating 20
Step 1. Measure Existing Sound Pressure 20
Step 2. Mark Average Sound Pressure 20
Step 3. Plot Curve of Octave Band 21
Example RC Analysis 21
Chapter 3–HVAC Sound Sources 23

Sources of Sound in HAVC Systems 23

Fan Sound Components 24
Fan Aerodynamic Sound 24

Siemens Building Technologies, Inc. i

Purpose of this Guide

Blade Frequency Increment 24

Fan Efficiency 25
Fan Sound Power Level Data 25
Fan Sound Power Level Calculation 26
Step 1. Actual Operating Conditions Increase 26
Step 2. Blade Frequency Increment (BFI) 26
Step 3. Efficiency Correction 27
Example Fan Sound Power Level Calculation 28
Step 1. Actual Operating Conditions Increase 28
Step 2. Blade Frequency Increment (BFI) 29
Step 3. Efficiency Correction 29
Damper Airflow Noise 30
U (Velocity Factor) 30
Calculate Pressure Loss Coefficient C 30
Calculate Damper Blockage Factor BF 31
Calculate the Velocity Factor U 31
K Factor 32
Example Damper Sound Power Level Calculation 33
Elbow Airflow Noise 34
K Factor 35
Example Elbow Sound Power Level Calculation 36
Junction and Takeoff Airflow Noise 38
K Factor 38
JC Factor 39
Example Duct Takeoff Sound Power Level Calculation 40
Air Delivery Device Sound 43
Flexible Duct Connection to Diffusers 44
Discharge Sound and Radiated Sound 44
Sound Breakout and Break-in 45
Laboratory Elements 45
Chapter 4–HVAC Sound Attenuation 47

Introduction to HVAC Sound Attenuation 47

Plenums 48
Example Plenum Attenuation Calculation 49
Duct Attenuation 51
Rectangular Unlined Sheet Metal Ducts 51
Example Rectangular Duct Attenuation Calculation 51
Rectangular Unlined, Externally Insulated, Sheet Metal Ducts 53
Rectangular Acoustically Lined Sheet Metal Ducts 54
Round Unlined Sheet Metal Ducts 57

ii Siemens Building Technologies, Inc.

About this Application Guide

Round Acoustically Lined Sheet Metal Ducts 57

Duct Elbows 58
Example Rectangular Duct Elbow Attenuation Calculation 60
Duct Takeoffs and Divisions 60
Duct Silencers 62
End Reflection 62
Environment Adjustment Factor 63
Space Effect 63
Radiated Sound Attenuation 64

Chapter 5–HVAC System Sound Analysis 67

Introduction to HVAC System Sound Analysis 67

Example HVAC System Sound Analysis 67
Step 1. Actual Operating Conditions Increase 68
Step 2. Blade Frequency Increment (BFI) 69
Step 3. Efficiency Correction 69
Duct Section A 70
Duct Elbow B 70
Duct Section C 72
Junction D 73
Duct Section E 75
Junction F 76
Duct Section G 76
Duct Takeoff/Junction H 77
Duct Section I 80
Duct Elbow J 80
Reheat Terminal 82
Duct Sections L 82
Perforated Diffuser 83
End Reflection 83
Space Effect 83
Commentary on HVAC System Sound 88
Laboratory Room Sound Analysis 88
Laboratory Room Ambient Sound 89
Fume Hood Sound 89
Terminal Radiated Sound -Example Analysis 90
Radiated Sound 91
Discharge Sound 93
Terminal Radiated Sound -Example Analysis 2 94

Siemens Building Technologies, Inc. iii

Purpose of this Guide

Chapter 6–Minimizing HVAC Sound 95

Introduction to Minimizing HVAC Sound 95

Basic System Design Criteria 95
Fans 96
Duct Configurations 97
Terminal Equipment 97
Sound Attenuation Devices 100
Passive Sound Attenuation Devices 100
Linings 100
Duct Silencers and Attenuators 100
Ceiling and Wall Absorbers 101
Enclosures 101
Active Sound Attenuation Devices 101
Sound Measurement Instrumentation 103
Sound Measurement Procedure 103
Appendix 105

NC and RC Curves, Tabular Listing 105

NC Curve 106
RC Curve 107
Sound Analysis Worksheet 108
Sound Measurement Worksheet 109

Glossary 111

Index 115

iv Siemens Building Technologies, Inc.

About this Application Guide
This section discusses the following topics:

• Purpose of this guide

• How this guide is organized

• Suggested reference materials

• Conventions and symbols used

It also provides information on how to access help and where to direct comments about this
guide.

Purpose of this Guide

This application guide explains the nature of sound generation and attenuation within air
movement components of HVAC systems, and is intended to help the reader understand
how to achieve a ventilation system design that does not generate excessive or objectionable
sound. It is the intent of this guide to provide a working level of HVAC sound dynamics
knowledge for the benefit of those who may not have yet acquired a sufficient technical
background in the subject of HVAC sound analysis.

How this Guide is Organized

This application guide contains the following chapters:

• Chapter 1, Introduction, discusses laboratory control and safety solutions. It includes

a scope of this guide and discusses HVAC sound transmission, background sound,
laboratory applicability, and computer program sound analysis.

• Chapter 2, Physics of Sound, discusses the properties of sound and how sound is
measured. It includes sound wave propagation and parameters; measurement
parameters; NC and RC curves; and how to determine an RC rating.

• Chapter 3, HVAC Sound Sources, discusses sources of sound associated with

HVAC systems. It includes fan sound components and power level calculation;
damper and elbow airflow noise; junction and takeoff airflow noise; and air delivery
device noise.

• Chapter 4, Ventilation Systems Classification, discusses the attenuating effect of

common HVAC system elements (also referred to as transmission loss or insertion
loss).

• Chapter 5, HVAC System Sound Analysis, provides examples of how to analyze the
components of a specific HVAC system.

Siemens Building Technologies, Inc. I

About this Application Guide

• Chapter 6, Minimizing HVAC Sound, offers general guidance on minimizing

excessive or objectionable HVAC sound.

• The Appendix contains blank copies of certain graphs and forms that appear in this
document. They are intended to be copied and used for sound measurement and
analysis.

• The Glossary describes the terms and acronyms used in this manual.

• The Index helps you locate information presented in this application guide.

Suggested Reference Materials

In addition to this application guide, the following publications are recommended sources of
detailed technical information associated with minimizing sound in ventilation systems:

• American Society Of Heating, Refrigeration, & Air Conditioning Engineers, Inc.:

− A Practical Guide To Noise and Vibration Control

− HVAC Applications, 1991 (Chapter 42 - Sound and Vibration Control)

− Fundamentals, 1993 (Chapter 7 Sound and Vibration)

• Sheet Metal and Air Conditioning Contractors National Association Inc. (SMACNA):

− HVAC Systems Duct Design

• Air Movement and Control Association, Inc.:

− Laboratory Method of Testing in-Duct Sound Power Measurement Procedure for

Fans ANSI/AMCA 330-86

− Methods for Calculating Fan Sound Ratings from Laboratory Test Data AMCA
301-90

− Reverberant Room Method for Sound Testing of Fans AMCA 300-85

− Application of Sound Power level Ratings for Fans AMCA 303-79

• Air Conditioning & Refrigeration Institute:

− Procedure for Estimating Occupied Space Sound Levels in the Application of Air
Terminals and Air Outlets (ARI 885-90)

− Standard for Air Terminals (ARI 880-89)

• American Society of Mechanical Engineers United Engineering Center:

− Measurement of Industrial Sound ANSI/ASME PTC 36-1985

II Siemens Building Technologies, Inc.

 Symbols

• Department of Labor, Occupational Safety & Health Administration, Superintendent

of Documents, U.S. GPO:

− Occupational Exposure to Hazardous Chemicals in Laboratories; Final Rule, 29

CFR Part 1910, 1990

Symbols
The following table lists the symbols used in this guide to draw your attention to important
information.

Notation Symbol Meaning

WARNING: Indicates that personal injury or loss of life may occur to the
user if a procedure is not performed as specified.

CAUTION: Indicates that equipment damage, or loss of data may occur if

the user does not follow a procedure as specified.

Note Provides additional information or helpful hints that need to be

brought to the reader's attention.
Tip Suggests alternative methods or shortcuts that may not be
obvious, but can help the user better understand the
capabilities of the product.

Getting Help
For more information about minimizing sound in ventilation systems, contact your local
Siemens representative.

Where to Send Comments

Your feedback is important to us. If you have comments about this guide, please submit them
to: [email protected]

Siemens Building Technologies, Inc. III

About this Application Guide

IV Siemens Building Technologies, Inc.

Chapter 1–Introduction
Chapter 1 introduces laboratory control and safety solutions. It includes the following topics:

• Scope of this guide

• HVAC sound transmission

• Background sound

• Laboratory applicability

• Computer program sound analysis

Scope of This Guide

This application guide focuses on HVAC air movement and distribution system generated
sound. It does not specifically address sound or vibration problems of other related
mechanical system components such as boilers, chillers, cooling towers, pumping, and
piping systems. It is the intent of this document to provide sufficient background information
in the basics of sound and its application to air systems to enable the reader to properly use
equipment manufacturer’s sound rating data in the design of a ventilation system.

Sound and vibration are a science in themselves and an all-inclusive study is beyond the
scope of this guide. Additionally, it is believed that the reader need not delve too deep into
the theory to achieve a practical working knowledge of the subject.

For these reasons, this guide will limit its approach to only the essential elements of
acoustics theory and will attempt to emphasize practicality rather than theory whenever
possible. For those readers who want more background on the subject or need additional
information, the Suggested Reference Materials section in the About this Application Guide
lists a number of books, and other sources of more detailed and specialized technical
information on the subject of HVAC sound and vibration.

The information in this guide should assist in handling typical ventilation system design
applications for offices, laboratories, classrooms, and the like. However, the reader is
cautioned that more specific and detailed knowledge is warranted if the system design is
intended for applications where sound is a much more critical issue. This includes acoustical
laboratories, recording studios, and any location where maintaining a very low or specific
sound level is crucial. If any of these types of applications are a part of an HVAC design
project, it is recommended that the designer consult an appropriate acoustical or sound
specialist for guidance.

Siemens Building Technologies, Inc. 1

Chapter 1–Introduction

HVAC Sound Transmission

Ventilation system ductwork conducts or transmits sound in the same way that any conduit
can convey sound. We’re all familiar with how effectively a hose or pipe can conduct the
sound waves of the human voice. In the same way, ductwork conducts fan noise, and other
component sounds to the areas served. If the sound level is excessive or the sound pattern is
annoying, it can cause dissatisfaction with an HVAC system that otherwise does an excellent
job of maintaining comfort and providing the proper level of ventilation.

Background Sound
It is important to understand that the typical goal of a properly designed ventilation system is
not to obtain the least possible amount of sound, but to achieve a specific sound level and
profile. In most applications, a specific background sound level and sound profile are
desirable since it helps cover or mask other objectionable sounds. In the workplace, a good
ventilation system provides just enough background sound to prevent other sounds
(telephone conversations, keypad clicking, copy machines, etc.) from being excessively
annoying. This desirable background sound level and profile is sometimes termed white
noise and is usually very noticeable when not present. (Recall how much louder common
office sounds seem to be if the ventilation system is shut down and the white noise is not
present.) However, there is a sound level threshold that is dependent upon the room and its
activity, and when exceeded, results in excessive and objectionable sound.

Laboratory Applicability
Of the many HVAC applications where ambient sound level is an important design
component, laboratory applications are a particular challenge to the HVAC designer, due to
the necessity for providing high room ventilation rates to ensure the health and safety of the
occupants.

Computer Program Sound Analysis

Computer programs that assist the HVAC designer in producing optimum duct design layouts
often include a sound analysis capability that saves having to manually perform sound
analysis calculations. As with all such design aids, it is important that the user be
knowledgeable about the fundamentals of the subject in order to properly use the program. In
addition, having this background knowledge enables the user to recognize whether a specific
program incorporates an acceptable analytical approach.

2 Siemens Building Technologies, Inc.

 Computer Program Sound Analysis

Siemens Building Technologies, Inc. 3

Chapter 2–Physics of Sound
Chapter 2 discusses the properties of sound and how sound is measured. It includes the
following topics:

• Sound wave propagation

• Sound wave parameters

• Sound measurement parameters

• NC Curves

• RC Curves

• Determining an RC rating

Sound Wave Propagation

The human ear hears or senses sound when oscillations or vibrations occur within its hearing
mechanism. Under normal circumstances, these oscillations are transmitted to our ear as
sound waves that are really air pressure waves. These air pressure waves impact upon the
ear’s sensing or hearing mechanism and cause it to oscillate or vibrate.

As sound waves travel to the ear, they may travel not only through air but also use different
mediums as well. Recall that a basic physics classroom experiment on sound consists of
putting a sound generating device (sometimes an alarm clock) under a large glass container
(typically a bell jar). The vibrations of the sound generating device cause sound waves in the
air within the bell jar that travel outward until they reach the glass wall of the bell jar. There,
they cause the wall of the bell jar to vibrate that in turn causes sound waves to be generated
in the air outside of the bell jar. These sound waves then continue and eventually reach the
ears of those in the classroom. As long as the bell jar contains room air at normal
atmospheric pressure and density, the above scenario takes place and the sound is easily
heard.

However, after a vacuum pump removes most of the air from inside the bell jar, the sound
made by the sound generating device is dramatically reduced since the density of the air
within the bell jar has been substantially reduced and thus has a much more limited impact
on the wall of the bell jar. This experiment shows that sound waves are highly dependent
upon having an adequate medium for their transmission.

Siemens Building Technologies, Inc. 5

Chapter 2–Physics of Sound

This experiment also shows another important element of sound wave transmission; that
sound waves traveling through air are not dependent upon movement of the air itself.
Although the air in the bell jar could not leave the jar, the sound traveled outward from the jar
without involving any physical movement of the air out from the inside of the bell jar.
Likewise, sound movement in a ventilation system is not dependent upon the movement or
direction of the airflow. Not only will sound generated in the supply side of a ventilation
system travel in the direction that the air happens to be moving to the areas served, but also
sound generated in an exhaust system will travel opposite the direction of airflow and also be
heard in the areas served by the exhaust system.

Sound Wave Parameters

Any analysis or study of sound (acoustics) is especially concerned with the generation and
reception of sound waves. It is necessary to first understand the fundamental concepts of
sound wave generation and how this relates to the overall science of acoustics. Once these
fundamentals are understood, actual quantifiers or measurement parameters can be applied
and used in actual acoustical design practice.

Figure 1 shows a diagram of the major parameters that apply to the analysis of sound. At the
left side of the diagram is the sound source shown as a solid dot. Anytime sound is produced,
there must be a sound source. When we speak, our vocal chords create sound and are the
sound source. As a sound source creates sound in the air, it radiates energy outward in the
form of compressed air waves or sound waves.

Figure 1. Sound Wave Parameters.

6 Siemens Building Technologies, Inc.

 Sound Wave Parameters

Unless there is a barrier, the sound waves continue to travel outward in all directions in a
spherical manner, until they either are absorbed by an object or their energy level is
dissipated by the surrounding air.

With regard to understanding the science of sound and its effects, it is necessary to have an
understanding of two fundamental terms: sound power and sound pressure. These terms are
not interchangeable and it is important to have a clear understanding of each term.

The intensity of the sound at the source is expressed in terms of sound power and
establishes the energy level of the sound. Sound power is the parameter that indicates the
total energy or power output of the sound. It is universally expressed in terms of watts. The
sound power spectrum that we are familiar with ranges from a high point of 10,000 (104)
watts of sound power for a jetliner takeoff or gun fire, to a low of 0.000000009 (10-9) watts for
a soft whisper.

Ultimately, sound waves, which in our context, are really compressed air waves, will
ultimately impinge on a receiver and at that point their effect is expressed in terms of the
sound pressure.

For our purposes, the most common receiver will normally be the eardrum of a person who
hears the sound. Another common example of a sound receiver is a microphone that is part
of a sound amplification system. Sound power itself, however, does not really establish
whether a sound will be interpreted as loud or soft by the receiver. That is entirely dependent
upon the amount of energy loss or attenuation of the sound waves that occurs prior to
impinging upon the receiver as sound pressure.

Attenuation, which is simply a decrease in the sound power before it gets to the listener,
occurs primarily due to two factors: distance and physical barriers. When sound is generated
in an open or unconfined space, as in Figure 1, the primary attenuation factor is the distance
between the sound source and the receiver.

When a sound source generates sound, the sound power energy is radiated outward in all
directions, as shown in Figure 1, and the sound power energy is dissipated over a rapidly
increasing area. This can be likened to an ever expanding sphere surrounding the source of
the sound. An analogy of the effect of sound power radiation would be like having a fixed
quantity of paint and the task of achieving a uniform thickness paint coating on the surface of
the sphere. In this analogy the paint quantity represents the available sound power level,
while the resulting thickness of the paint coating on the surface of the sphere represents the
sound pressure level.

As the radius of a sphere increases, the surface area also increases and the thickness of the
surface coating must be decreased. If a sphere having a radius of 1 foot (surface area 12.56
square feet) is expanded until the radius is doubled (becomes 2 feet), the surface area would
have increased to 50.24 square feet or four times the original surface area. This of course
means that the paint coating on the surface could then only be 1/4 of the previous thickness.

Likewise, each time the distance between a sound source and the receiver is doubled, the
effect at the receiver that is the sound pressure level is reduced by a factor of 4.

Siemens Building Technologies, Inc. 7

Chapter 2–Physics of Sound

The sound pressure level is the most widely used parameter in the field of acoustical
engineering since it is the closest thing to what we experience in terms of loudness or
softness of a sound. In the previous analogy, the thickness of the paint coating on the
spherical surface represented the effect that a certain quantity of paint could have on the
surface of the sphere. Therefore ,sound pressure expresses the effect that the sound power
energy has when impinging upon a unit area of the receiver.

Sound Measurement Parameters

To use sound generation and attenuation data in the design of HVAC systems, it is
necessary to understand the measurement parameters with which sound power, sound
pressure, and other factors involved are quantified. These key factors are each listed below
with an explanation of their units of rating or measurement.

Sound Power Level

As discussed above, sound power expresses the overall sound energy of the sound source
and sound power level is represented in terms of watts. However, as was previously
indicated, the range between the highest and lowest sound power levels is too large to
conveniently use actual watt values. In addition, the actual watt values that apply to the
typical sound power levels encountered in HVAC are so small (that is, 10-3 to 10-11 watts) that
too many zeros would be needed after the decimal point to express specific watt values.

A more convenient scale is to represent sound power level so the decibel (dB) unit is used.
With decibels, sound power low level begins at 0 decibels, which is just enough power for the
human ear to begin hearing something that is right next to the ear. At the upper end of the
scale is the sound power level of a jetliner taking off, which could be 160 dB. (An explanation
of decibels and how specific physical measurements are expressed in terms of decibels
follows.)

Be sure not to confuse sound power level with sound pressure level (even though both are
expressed in terms of decibels). Remember:

• Sound power level expresses the power or energy of the sound source.

• Sound pressure level expresses the loudness or effect of the sound at the receiver.

When we hear a soft whisper, we are experiencing a sound power level of approximately
0.000000001 (10-9) watts. When we converse in a normal voice, we are experiencing a
sound power at about 0.00001 (10-5) watts. If we had the unfortunate experience to be just
below a jetliner taking off, we could later tell everyone that we experienced a sound power
level of around 10,000 (104) watts. (However, it may take us a few days before we could
again hear their reply.)

As previously discussed, wattage based direct sound measurements are non-linear and vary
over a very large range, thus it becomes clumsy to stay with watts as the basic unit of
measurement.

8 Siemens Building Technologies, Inc.

 Sound Measurement Parameters

We could improve the situation by using a direct comparison or ratio between the two
different sound power levels. For example, when comparing a normal voice to a whisper, we
could divide 10-5 by 10-9 that would yield 10,000. In other words, a normal voice has about
10,000 times more sound power than a whisper. This same approach would also tell us that
the sound power from a jetliner takeoff is 1,000,000,000 times more sound power than a
normal voice.

Unfortunately, these numbers are still awkward to work with because of the large number of
zeros. We can remove these zeros by using logarithms. Recall that a logarithm to the base
10 (or log for short), means that the power of 10 would be raised to become the number that
we’re concerned with. In other words, if we’re working with the number 10,000, the log is
simply 4, since 104 = 10,000. If we’re working with 1,000,000,000 the log would be 9 since
109 = 1,000,000,000.

Decibels
Using a comparison or ratio approach for large numbers tends to make it easier to relate to
the data. In addition, converting large numbers to logarithms further reduces the amount of
digits (and possible errors) when handling numbers comprised of many digits. This is where
the use of decibels offers a practical approach to quantifying sound parameters since a
decibel is based upon both a ratio and numbers converted into logarithms.

The first thing to note with regard to a decibel is that it expresses a ratio or makes a
comparison between two values; it is not a specific unit of measurement such as a watt,
pound, or even a foot of length. (Since decibels are based upon a ratio, they can be applied
to many other different scientific parameters besides sound.) Therefore, with regard to
applying decibels (dB), a reference point must be established as one component of the ratio.

With regard to sound power and sound pressure values, a bel is simply the logarithm of the
ratio of two different sound power or sound pressure levels. A decibel is 10 bels. (The reason
for using decibels instead of just staying with bels is that we do ourselves a favor by getting
rid of any decimals in the final values.) This may sound complicated and possibly somewhat
confusing, but you’ll understand it better after going through the process of establishing
decibels (dB) for the previous examples of a whisper, a normal voice, and the jetliner takeoff.
Then (hopefully) you’ll see the advantage of using dB instead of the large decimal numbers
that are required to express values in wattage.

Since we’re really only concerned with the sound that humans hear, we’ll use the threshold of
hearing as the common point of the comparison ratio. The sound power at the threshold of
-12
hearing is generally accepted as 0.000000000001 watts (10 watts), so this will always be
the reference point or one of the two parts to each sound power ratio.

Siemens Building Technologies, Inc. 9

Chapter 2–Physics of Sound

The following formula will yield the decibels for any absolute value of sound power that we
compare to the threshold of hearing:

Lw = 10 x Log (W ÷ Wref)

Where:

Lw = the sound power level in dB.

W = the power of the specific sound in watts.
Wref = the reference point and is always 10-12 watts.

Using this formula, let’s determine the sound power level of a whisper, which produces a very
tiny amount of sound power around 0.000000001 watts (10-9 watts). Using the above formula
this becomes:

Lw = 10 x Log (10-9 ÷ 10-12)

= 10 x Log (103)

= 10 x 3

= 30 dB

Therefore, the sound power level of a whisper is approximately 30 dB.

Using the formula again, let’s determine the sound power level of a normal conversational
voice that is around 0.00001 watts (10-5 watts).

Lw = 10 x Log10 (10-5 ÷ 10-12)

= 10 x Log10 (107)

= 10 x 7

= 70 dB

And, using this same formula for the jetliner takeoff sound power level of 1,000.0 watts (104
watts) becomes:

Lw = 10 x Log10 (104 ÷ 10-12)

= 10 x Log10 (1016)

= 10 x 16

= 160 dB

It’s much easier having the numbers determined above (30, 70, and 160 dB) instead of
having to refer to the actual wattage values when comparing sound power levels. Note that
when we say that a whisper is 30 dB, a regular voice is 70 dB and a jetliner takeoff is 160, we
are really comparing these individual levels with respect to the threshold of hearing that is 0
dB.

10 Siemens Building Technologies, Inc.

 Sound Measurement Parameters

Again, there are no units associated with decibels since they are a comparison between two
values, (or more scientifically, a ratio between different magnitudes). Also, decibels are used
for different parameters besides sound power level. Decibels are also used to express sound
pressure level, which is discussed below, and as we are keenly aware, is a different sound
parameter than the sound power level.

Sound Pressure Level

As previously stated, sound pressure is concerned with the effect that a specific sound power
level has on a receiver that is usually some distance away from the sound source. Recall in
our analogy about covering the surface of an expanding sphere with a fixed quantity of paint,
a receiver is like a limited area of the sphere, it will receive only a small portion of the paint
(sound power).

Therefore, a receiver is only exposed to a portion of the total sound power. In other words,
the effect of the sound power becomes less and less (is attenuated more and more) on the
receiver. Therefore, the sound power level and sound pressure level are different parameters
and cannot be used interchangeably. However, decibels also are used to express the ratios
of the relative sound intensity or loudness at the receiver.

The basic unit of acoustic pressure is the Pascal (Pa). (One PSI is equivalent to 6,895
Pascals.) Even though a Pascal is a very small unit of pressure measurement, the specific
values of Pascals that are encountered with sound pressure are so small and vary over such
a wide range that the “decibel” approach is applied to express sound pressure levels in a
more practical manner.

The basic formula to determine a specific sound pressure level in decibels is:

Lp = 10 x Log (P ÷ Pref)2

Where:

Lp = the sound pressure level in dB.

P = the pressure of a specific sound at the receiver in Pascals.

Pref = the reference pressure and is always.

2 x 10-5 Pascals is approximately the sound pressure on an eardrum at the hearing

threshold.

A person speaking in a normal voice, about three feet away from a listener, will produce a
sound pressure of around 0.02 Pa. Using this formula, let’s determine the sound pressure
level in dB that the listener would experience.

Lp = 10 x Log (2 x 10-2 ÷ 2 x 10-5)2

= 10 x Log (103)2

= 10 x Log (106)

Siemens Building Technologies, Inc. 11

Chapter 2–Physics of Sound

= 60 dB

Therefore, the sound pressure level of normal conversation for the listener is approximately
60 dB.

To give a “feel” for the common range of values for both sound power and sound pressure,
Table 1 gives some typical values for the common levels in our environment.

Table 1. Common Sound Power & Sound Pressure Levels.

Source Sound Power Level Sound Pressure Level

dB dB
Jetliner takeoff 160 140* (at 100 ft away)
Very loud sound (race car engine, gun fire) 120 -130 110*
Very noisy room (loud machinery) 100 -110 100*
Somewhat noisy room (computer printout 80 80
room)
Normal voice conversation 70 60 (at 3 feet)
General office 45 45
Soft whisper 30 30 (at 5 feet)
Leaf rustling 20 20
Hearing threshold 0 0
* Hearing protection should be used in sound pressure levels of 90 dB or more since permanent
hearing loss will occur after exposure of 8 or more hours at 90 dB. OSHA limits the permissible
exposure time for unprotected workers in a sound pressure level that averages 90 dB or more.

Remember, even though the decibel values in Table 1 are almost the same for the sound
power level and the sound pressure level, they represent different physical parameters. In
HVAC design, we are mostly concerned with the sound pressure level experienced by an
occupant in an area served by an HVAC system.

Even though you will find that the sound power level of a sound source such as an HVAC
system supply fan may be considerably high (such as, 90 or more dB), the objective is to
design and configure the HVAC system so the sound pressure level will be attenuated down
to an acceptable level, perhaps 35 dB, when heard by an occupant of the area served by the
HVAC system.

Table 2 lists a few rules that generally predict how a person perceives loudness of sound as
changes occur in the sound pressure level.

Table 2. Effects of Sound Pressure Level Changes.

dB Change Effect
0 to 2 None
3 Just noticeable
8 to 10 Increase Twice as loud
8 to 10 Decrease Half as loud

12 Siemens Building Technologies, Inc.

 Sound Measurement Parameters

Figure 2 illustrates another important rule regarding sound pressure levels and distance.
Whenever the distance between a sound source and a receiver is doubled, the sound
pressure level at the receiver is reduced by 6 dB from its previous value. This is a very
important relationship. For instance, if a sound source produces a sound pressure level of 40
dB at a receiver 15 feet away, the sound pressure level would be reduced to 34 dB if the
distance away were doubled to 30 feet. (Note that this is a non-linear relationship and the
results cannot be interpolated. Therefore, in this example, it would be incorrect to assume
that the dB level drops at 2 dB for every 5 foot increase in distance.

If the distance were again doubled from 30 feet to 60 feet, the sound pressure level at 60 feet
would be 28 dB. This relationship continues each time the previous distance is doubled.

If the sound pressure level was 60 dB at a 12 foot distance between the sound source and
the receiver, what would the sound pressure level be at a distance of 100 feet? Using the
distance doubling rule, the sound pressure levels at various distances would be:

• 60 dB at 12 ft

• 54 dB at 24 ft

• 48 dB at 48 ft

• 42 dB at 96 ft

• 36 dB at 192 ft

Since 100 feet is slightly more than 96 feet, the dB level would be perhaps just a bit less than
42 dB.

Figure 2. Sound Pressure Level Decrease Due to Distance.

Octave Bands
Sound can vary in pitch or frequency from a very low base sound to a very high pitch sound
such as a squeak. In terms of actual frequency, human hearing ranges from about 20 cycles
per second (Hz) at the low end to around 20,000 Hz at the high end. The actual frequency
span of hearing varies from person to person and tends to decline somewhat as we age with
the upper frequency end of our hearing being the portion mostly affected by age.

Siemens Building Technologies, Inc. 13

Chapter 2–Physics of Sound

Previously, we discussed the terms sound power level and sound pressure level and arrived
at how their intensity was expressed in decibels. If you recall how the screen of an
oscilloscope looks when it’s monitoring the audio output of a speaker, you can visualize that
sounds are usually composed of a multitude of tones at different frequencies. To scientifically
describe a particular sound accurately, a curve should be plotted showing the sound power
level or sound pressure level in decibels with reference to the frequency.

Since the normal audible spectrum covers the frequency range of 20 Hz to 20,000 Hz, it
would be totally impractical to deal with each individual frequency. For this reason, it has
become customary in sound analysis to divide the overall audible spectrum into 8 frequency
bands called octave bands. (These are often referred to as 1/1 Octave Bands.) In each band
the highest frequency is twice the lowest frequency, and the mid frequency of each band is
used for identifying the octave band and as the specific frequency for expressing the sound
power level or sound pressure level in decibels.

Figure 3 illustrates how sound curves can be shown on a graph that plots the sound pressure
level at each of the standard octave band mid frequencies. The resulting curves establish
what’s referred to as a sound criterion curve for the particular sound.

Figure 3. Sound Pressure Level vs. Octave Band - Sound Criterion Curves.

With reference to Figure 3, the dB scale ascends from 0 to 90 along the vertical axis and the
center frequencies of 10 bands are along the horizontal axis. Note that the frequency scale is
not linear but increases rapidly in moving from left to right.

14 Siemens Building Technologies, Inc.

 Sound Measurement Parameters

Note that a solid line curve in the lower left portion of the graph is labeled as the approximate
threshold of a hearing curve. This represents the dB sound pressure level that must be
present in a person’s eardrum in order for the person to hear a particular sound frequency.
Recall that in Table 1, the threshold for hearing is listed as 0 dB sound pressure level. With
reference to Figure 3, this really applies to sound frequencies above 4,000 Hz that are in the
area of the high pitched beep of a computer speaker. At the lower frequencies, the sound
pressure level must be considerable higher to be audible.

The sound pressure level of a particular sound such as a fan running, a transformer hum, or
car horn can be measured with a sound level meter at a specific distance from the sound.
The sound pressure in dB at each frequency band can be plotted on the graph and the
resulting curve will show the profile of the sound similar to the dotted line and dashed line
curves shown in Figure 3.

The dotted line curve is a predominantly lower frequency curve since it has a high dB level in
the lower frequency bands and a lower dB level in the higher frequency bands. This sound is
characterized as rumbly or somewhat like a drumming sound. The dashed line curve is just
the opposite and is characterized as a hissy type sound or somewhat like an air leak. In order
for a sound to be acceptable for sound masking (white noise) or as an acceptable
background, it must be fairly well balanced across the audible sound spectrum. Since neither
of these two sound curves are well balanced, they would not be acceptable for sound
masking and instead would probably be very annoying.

Table 3. Adding Sound Pressure or Sound Power Levels.

Difference between the highest and Add this dB to the highest dB of the
lowest dB of multiple sounds at a specific sounds to obtain the resultant dB at the
octave band center frequency octave band’s center frequency
0 3.0
1 2.6
2 2.1
3 1.8
4 1.5
5 1.2
6 1.0
7 0.8
8 0.6
9 0.5
10 0.4
12 0.3
14 0.2
16 0.1

For example, in Figure 3 at the 500 Hz frequency, one sound pressure level is at 24 dB and
another at 50 dB. The difference between them is 26 dB. With reference to Table 3, this is
well beyond the 16 dB difference. As a result, 0 dB is added to the higher one (50 dB) that
results in no change to the total sound pressure level.

Siemens Building Technologies, Inc. 15

Chapter 2–Physics of Sound

Figure 4 shows the resulting sound pressure level when combining the two curves of Figure 3
using Table 3. Note that where the individual curves of Figure 1 are more than 16 dB apart,
the resultant always equals the higher dB value of the individual curves.

Incidentally, the resulting sound produced by the combined sound curve of Figure 4 would be
a combination of a rumble and hiss and would still be objectionable as an ambient sound.

Figure 4. Two Sound Pressure Levels Combined.

A-Weighted Sound Level

In an effort to come up with a simpler method to address sound ratings for equipment,
A-weighted sound levels that also use decibels, are sometimes used particularly when
compliance with OSHA noise limits is the issue. However, the A-weighted criterion is limited
to only being a reference of the overall loudness and does not represent the full frequency
distribution characteristics of a sound. In particular, it does not specifically indicate the
presence of the low frequency level sound component, which is the most important area of
sound analysis.

16 Siemens Building Technologies, Inc.

 NC Curves

Avoid using A-Weighted sound criterion when designing HVAC systems or conducting a
detailed analysis of the sound pressure level in a room with the intent of improving the room
ambient sound profile. Using A-Weighted values should be limited only to general noise level
comparison measurements or when involved in ensuring against exceeding permissible
occupational sound levels.

NC Curves
In an effort to come up with ambient sound pressure level curves that provide a good balance
between the sound frequency spectrum and the acceptable loudness for various room
applications, standard Noise Criterion (NC) curves have been developed. Figure 5 shows the
family of NC curves.

Until now, this document has avoided using the term noise since there is no scientific way to
define it. It is merely a term that each individual subjectively applies to a sound profile that for
them ranges from unwelcome or bothersome to very annoying. With regard to the Noise
Criterion curves, they establish balanced sound (noise) levels that are generally acceptable
for specific room applications. In other words, the sound produced by an HVAC system
serving a specific application would be acceptable by the vast majority of occupants if the
sound pressure level that it produces does not exceed the dB level of the appropriate NC
curve at any point, and it also has the same general shape as the referenced NC curve.

When analyzing a given room sound profile, it is also acceptable to visually interpolate
between the NC curves. For instance, if the highest penetration of a listed curve (NC 45 in
this case) is 52 dB at 500 Hz, then the measured sound can be stated as having an NC 48
rating.

The NC curves were developed in 1957 and are still widely used today. However, note that
they do not include any dB values for frequencies below 63 Hz. In general, the most
objectionable HVAC noise is the low frequency rumble that is produced by HVAC fans. The
bulk of this sound occurs below the 63 Hz octave band.

Siemens Building Technologies, Inc. 17

Chapter 2–Physics of Sound

(See the Appendix for a copy of this graph that is suitable for reproduction.)
Figure 5. Noise Criterion Curves.

RC Curves
The Room Criterion (RC) rating is a more recent development for analyzing and rating the
sound present in a room. The RC rating should be used, whenever possible, in specific
design applications since it is superior to the NC curves for the following reasons:

1. The RC curves extend down to 16 HZ that covers the low frequency sound spectrum
more completely than the NC curves.

2. Establishing the applicable RC curve that applies to an actual sound profile is dependent
on the overall shape or profile of the actual sound curve, rather than merely the highest
penetration of the sound into the NC family of curves.

3. Since a given sound curve is likely to have a unique curvature or profile, each RC sound
curve is further annotated as to its actual characteristics:

• A curve with a rumbly (low frequency) component is also given an R suffix.

• A curve with a hissy (high frequency) component is also given an H suffix.

• A more neutral curve without a rumbly or hissy component is given an N suffix.

18 Siemens Building Technologies, Inc.

 RC Curves

• If an identifiable predominant tone exists in the sound (such as, clicking, whining,
whistle, etc.), a T is also added to the above suffix.

• If excessive vibration is present, a V suffix is also added to the above.

Figure 6 shows the standard family of Room Criterion curves. Table 4 lists specific
applications and the maximum acceptable sound criterion that apply when referencing these
curves. Note that the Criterion level is always the dB level of the particular RC curve as it
passes through 1,000 Hz.

Utilizing the NC curves for design purposes typically results in background sound
characteristics having a noticeable rumble or hiss. Although the dB level may be acceptable
for speaking, the overall sound profile is less likely to be as acceptable as a design that is
based upon the RC Criterion.

Table 4. Applicable NC and RC Sound Criterion Curves for Various Applications.

Application Criterion Level

General Office 35 - 40
Private Office 30 - 35
Conference Room 25 - 30
Corridors 40 - 45
Hospital Room 25 - 30
Surgical Room 30 - 35
Classroom 25 - 35
Cafeteria 45 - 50
Library 30 - 40
Lobby 40 - 45
Auditorium 25 - 30
Washroom 40 - 50
Research Laboratory 35 - 45
Educational Laboratory 35 - 40
Sound Studio 15 - 20

Siemens Building Technologies, Inc. 19

Chapter 2–Physics of Sound

(See the Appendix for a copy of this graph that is suitable for reproduction.)
Figure 6. Room Criterion Curves.

Determining an RC Rating
To determine what RC rating should be applied to an existing room, follow the steps listed
below.

Step 1. Measure Existing Sound Pressure

Measure the existing sound pressure level in the room in decibels at all of the octave band
center frequencies. Calculate an average dB value from the room dB values obtained at 500,
1,000, and 2,000 Hz.

Step 2. Mark Average Sound Pressure

Mark the average obtained in Step 1 on an RC Criterion graph on the 1,000 Hz vertical scale.
Create an RC reference curve by drawing a line through this point that parallels the standard
RC curves. (Note that the standard slope of an RC curve is a loss of 5 dB per frequency band
as it goes from left to right.)

20 Siemens Building Technologies, Inc.

 Determining an RC Rating

Step 3. Plot Curve of Octave Band

Plot an actual curve of all of the octave band frequencies obtained in Step 2 on the graph,
and compare this curve with the reference curve drawn in Step 2.

• If the actual curve does not depart from the reference curve throughout all octave
bands by more than 5 dB, the actual curve is considered to be neutral. The suffix N is
added to the value obtained in Step 1.

• If the actual curve is above the reference curve by more than 5 dB at any octave
frequency less than 500 Hz, the actual sound is considered to be rumbly. The suffix
R is added to the value obtained in Step 1.

• If the actual curve is above the reference curve by more than 3 dB at any octave
frequency greater than 500 Hz, the actual sound is considered to be hissy. The suffix
H is added to the value obtained in Step 1.

• If the actual sound has an identifiable predominant tone such as a clicking, whining,
whistle etc., the actual sound is considered to have a tonal character. The T suffix is
also added to the N, R, and H suffixes.

Example of RC Analysis
If an existing room has an actual measured sound profile as listed in the following chart, what
RC Criterion would apply?

The average dB at 500, 1,000, and 2,000 Hz is calculated as: (43 + 35 +30) / 3 = 36.

With respect to Figure 7, the RC reference curve is plotted as the dashed line and the actual
sound curve is plotted as a solid line. Note that the actual curve does not exceed the
reference curve by more than 5 dB below 500 Hz, nor more than 3 dB above 500 Hz. Thus,
the sound RC criterion for this particular room sound would be classified as neutral and is
summarized as: 36 (N).

Although this particular sound has a slight rumble as indicated by the rise above the
reference curve in the lower frequencies, it would still be very acceptable as an overall sound
level for applications requiring an RC 35 level.

Siemens Building Technologies, Inc. 21

Chapter 2–Physics of Sound

Figure 7. Actual Room Sound Profile Curve vs. RC Reference Curve.

22 Siemens Building Technologies, Inc.

Chapter 3–HVAC Sound Sources
Chapter 3 discusses sources of sound associated with HVAC systems. It includes the
following topics:

• Sources of sound in HVAC systems

• Fan sound components

• Fan sound power level calculation

• Damper airflow noise

• Elbow airflow noise

• Junction and takeoff airflow noise

• Air delivery device noise

Sources of Sound in HAVC Systems

Sound associated with HVAC systems and equipment is generated from multiple sources. All
operating equipment generates sound by the inherent vibration of its components. This
includes HVAC fan systems, pumps, and the primary mechanical equipment (boilers, chillers,
air compressors, etc.). All of these units contain mechanically rotating components that
generate operational sound. This sound travels both as sound waves through the air and by
transmission of vibrations through adjoining elements of the building structure including walls,
floors, pipes etc.

In addition to the sound generated by the mechanical components of rotating equipment, the
air movement produced by a fan generates aerodynamic sounds due to interaction with the
distribution system components including dampers, duct fittings, junctions, terminal units, air
diffusers and inlet grilles.

Rotational equipment sound is primarily attenuated by isolating the equipment from occupied
areas of a building, incorporating physical barriers to sound waves and utilizing vibration
isolation to prevent vibrations from being transmitted through the building structure.
(Information on attenuating equipment operational sound is given in a later section.

Siemens Building Technologies, Inc. 23

Chapter 3–HVAC Sound Sources

HVAC aerodynamic sound is somewhat harder to attenuate since the ductwork provides a
direct conduit for its transmission to the conditioned spaces. In addition, some aerodynamic
sound is generated locally by HVAC system supply and exhaust components associated with
the room served by the system. On the supply side, this includes VAV box dampers, reheat
or cooling coils, air diffusers, and associated duct fittings. On the exhaust side, this primarily
involves the room exhaust terminals, laboratory fume hoods and other specialized room
exhaust units. Other sources of locally generated sound associated with HVAC systems
includes water flow through reheat coil valves, fan powered terminal units, and sometimes
even sound caused by bleeding or exhausting compressed air from the HVAC control
system.

Fan Sound Components

Fans are the predominant source of HVAC system sound. The fan sound power level must
be known to determine its contribution to the sound pressure level in a given space served by
the fan system. Fan sound is made up of several components. However, before we discuss
how to determine the overall fan sound power level, it will help to understand the nature of
each individual component affecting fan sound.

Fan Aerodynamic Sound

Aerodynamic sound is generated by air in motion. As you blow out a candle, an aerodynamic
sound is produced by the air rapidly passing through your lips. Since a fan imparts a high
level of motion to the air, it also results in significant aerodynamic sound.

Fans are tested for the sound power level produced by the manufacturers according to
standard tests covered by ASHRAE Standard 68-1986, and also by AMCA Standard 330-
1986. Virtually all fan manufacturers also send their fans to the AMCA laboratory for
certification of their test data. For greater accuracy of data, these tests cover the sound levels
produced in 1/3 octave bands. (Each of the eight octave bands is further divided into three
bands thus making 24 bands in all for the test. Three sound power level values are thus
obtained for each of the eight octave band. This data is then converted into the sound power
level for each of the eight octave bands and becomes the published data.)

Although fan manufacturers provide sound power level data for each of their different sizes
and types of fans, the data cannot cover each possible combination of operating conditions
(airflow, static pressure, etc.) in which a given fan may be applied. Therefore, fan sound
power level data is typically given at one set of standard operating conditions that also is a
common denominator for all fans. This consists of an airflow of 1 cfm and a static pressure of
1.00 in. WC. With this data, the fan sound power level at other operating conditions can be
determined through a calculation process that includes additional fan sound components.

Blade Frequency Increment

Before the arrival of electronic sound producing equipment, emergency warning sounds were
commonly produced by a mechanical device such as the unmistakable wailing sound of a fire
truck siren. The mechanical siren was very similar in design to a fan in that it had a rotor with
blades or slots that produced vibrations as it rotated.

24 Siemens Building Technologies, Inc.

 Fan Sound Components

Although a fan is not designed with the intention of deliberately creating sound, a sound
component, in addition to aerodynamic sound, is nonetheless generated as the result of fan
blade vibration during rotation. This component of fan sound is referred to as the blade
frequency increment (BFI). In some situations this component of the fan sound can be the
major source of annoyance. Therefore, it is a very important part of calculating the sound
power level of a fan in a given application.

Fan Efficiency
Another factor that affects the actual sound power level of a fan is its operating efficiency.
The closer a fan operates to its peak efficiency conditions, the lower the aerodynamic sound
that will be produced.

Fan Sound Power Level Data

To calculate the expected sound power level of a particular fan in a specific application, we
need the fan manufacturer’s certified sound power level data for the size and type of fan
we’re interested in. When selecting a fan for a given set of operating conditions, the ideal
situation is to use the fan manufacturer’s actual certified sound data.

However, since designers typically don’t want to design a system around one specific fan
manufacturer, they can use the sound power level data for typical fans that ASHRAE has
published. Table 5 reproduces this data for your convenience.

Table 5. Typical Sound Power Levels of Various Fans at 1 cfm and 1.00 in. WC.

63H 125H 250H 500H 1kH 2kH 4kH 8kH

Fan Type BFI
z z z z z z z z
Over 36-inch 40 40 39 34 30 23 19 17 3
Centrifugal Airfoil
Wheel Diameter
Backward Curved Up to 36-inch 45 45 43 39 34 28 24 19 3
Backward Inclined Wheel Diameter
Centrifugal All Diameters 53 53 43 36 36 31 26 21 2
Forward Curved
Centrifugal Radial 4 to 10 in. WC 56 47 43 39 37 32 29 26 8
6 to 15 in. WC 58 54 45 42 38 33 29 26 8
15 to 60 in. WC 61 58 53 48 46 44 41 38 8
Vaneaxial .3 to .4 Hub
49 43 43 48 47 45 38 34 6
Ratio
.4 to .6 Hub
49 43 46 43 41 36 30 28 6
Ratio
.6 to .8 Hub
53 52 51 51 49 47 43 40 6
Ratio
Tubeaxial Under 40-inch 51 46 47 49 47 46 39 37 7
Wheel Diameter
Over 40-inch 48 47 49 53 52 51 43 40 7
Wheel Diameter

Siemens Building Technologies, Inc. 25

Chapter 3–HVAC Sound Sources

63H 125H 250H 500H 1kH 2kH 4kH 8kH

Fan Type BFI
z z z z z z z z
Propeller All Sizes 48 51 58 56 55 52 46 42 5

In Table 5, the sound power levels listed are quite low with respect to those listed in Table 1
for sources of sound. However, the values in Table 5 are only a starting point. They only
cover fan operation at 1 cfm and 1.00 inches water static pressure (wsp). Since no fan would
ever be applied under this set of conditions, we must calculate the expected sound power
level at our actual design conditions.

However, just looking at this table tells us something about which fans tend to produce less
sound than others. For instance, centrifugal fans are quieter than axial fans, and the top
group of centrifugal fans has the lowest sound power level. Radial type centrifugal fans
produce the most sound and really are really intended for industrial applications and not
HVAC.

Note that the BFI column refers to the blade frequency increment component of the sound
power level. The lower the number, the lower this sound component will be. Also, the fewer
the number of blades the fan has, the lower the frequency of the sound.

Fan Sound Power Level Calculation

Determining the actual sound power level at each octave band for a fan at its design
conditions, is a three-step process.

Step 1. Actual Operating Conditions Increase

Use the following formula to calculate the actual sound pressure level increase for the fan’s
design operating conditions:

FLw = 10 Log Q + 20 Log P

Where:

FLw = the fan sound power level increase.

Q = the design airflow rate (cfm).

P = the design fan static pressure (in. WC).

Step 2. Blade Frequency Increment (BFI)

Determine that the octave band the Blade Frequency Increment (BFI) should be added. The
formula below yields the frequency (Hz) at which this blade sound component will occur.

Hz = Fan RPM ÷ 60 x Number of Blades

26 Siemens Building Technologies, Inc.

 Fan Sound Power Level Calculation

When the frequency is determined, refer to Table 6 to determine the octave band to which
this frequency applies. The BFI dB value from the manufacturer (or from Table 5) is then
added to the dB obtained in Step 1 for this particular octave band.
Table 6. Octave Band Frequency Division.
(Octave band is listed above the applicable frequency range)

Step 3. Efficiency Correction

Make an appropriate efficiency correction to the sound power level of each octave band
when actual fan operation is intended to be less than 90% of peak efficiency. Table 7 gives
the decibels that must be added to each octave for various efficiency ranges.

When the actual operational efficiency is an unknown (as is typically the situation in the
design stage), determine where the fan will be operating on its fan curve. Figure 8 shows a
typical centrifugal and axial fan curve and the required dB correction for various operational
regions.

Table 7. Efficiency Correction.

% of Peak Static Add to all

Efficiency Octaves
90 to 100 0
85 to 89 3 dB
75 to 84 6 dB
65 to 74 9 dB
55 to 64 12 dB
50 to 54 15 dB

For details, see Example Fan Sound Power Level Calculation in this section.

Siemens Building Technologies, Inc. 27

Chapter 3–HVAC Sound Sources

Figure 8. Fan Sound Power Level Correction for Off-Peak Efficiency Operation.

Example Fan Sound Power Level Calculation

Assume an HVAC system design will use a 30-inch diameter backward inclined 10 blade
centrifugal fan to deliver 15,000 cfm at a static pressure of 4.00 in. WC. According to
manufacturer’s data, this requires the fan to rotate at 1,300 rpm.

Determine the actual sound power level to be expected under those conditions assuming the
actual manufacturer’s data is the same as Table 5.

Step 1. Actual Operating Conditions Increase

The second horizontal line of values (up to 36-inch Wheel Diameter) in Table 5 applies to a
30” backward inclined centrifugal fan. Using the FLw formula, calculate the dB level increase
for the eight octave bands.

FLw = 10 Log + 20 Log P

= 10 Log 15,000 + 20 Log 4

= 10 (4.2) + 20 (0.6)

= 42 + 12.0

= 54

Table 8 summarizes Step 1, 2, and 3 sound calculations beginning with the initial dB values
obtained from Table 5, and resulting in the final actual operating conditions dB values.

28 Siemens Building Technologies, Inc.

 Fan Sound Power Level Calculation

Step 2. Blade Frequency Increment (BFI)

Using the Hz formula, the frequency at which the blade sound component will occur is:

Hz = Fan RPM x (Number of Blades ÷ 60)

Hz = 1300 x (10 ÷ 60)

Hz = 1300 x (0.167)

Hz= 217

Referring back to Table 6, since 217 Hz is within the 250 Hz octave band, the BFI value of 3
dB (obtained from the rightmost column of Table 5) is added to the 250 Hz octave band
column in our summary chart on the next page.

Step 3. Efficiency Correction

Let’s assume that our overall HVAC system design is based upon an operating fan efficiency
of at least 90%. In other words, with reference to Figure 8, the operating point will be on the
centrifugal fan curve slightly to the right of the peak. In this area, there is no need to add any
decibels to correct off peak fan efficiency. However, since there’s always the likelihood of
actual conditions resulting in operation at less than our theoretical efficiency, in this case,
we’ll add 3 dB to be on the safe side with our calculations. The sound power level decibels in
the above chart are more in line with what we might expect with reference to the fan typical
sound power levels as indicated back in Table 1. In fact, they may seem to be quite high and
pose a real problem, but they’re actually quite normal in view of typical HVAC fans.

Table 8. Sound Calculation Summary for Actual Fan Operating Conditions.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
Table 5 45 45 43 39 34 28 24 19
Step 1 54 54 54 54 54 54 54 54
(FLw)
Step 2 — — 3.0 — — — — —
(BFI)
Step 3 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0
(EFF)

Final 102 dB 102 dB 103 dB 96 dB 91 dB 85 dB 81 dB 76 dB

The resulting sound pressure level in rooms served by the HVAC system will be attenuated
to some extent by the ductwork itself and other duct components and may result in an
acceptable sound pressure level in the rooms served. However, any good designer must
determine what attenuation will likely occur in the system and if additional attenuation must
be added.

Siemens Building Technologies, Inc. 29

Chapter 3–HVAC Sound Sources

Note that although a fan is normally the major sound producer, you should also analyze the
entire system of ductwork, terminal devices and the room diffusers to determine what their
contribution to the HVAC system aerodynamic sound they may make, as well as what
attenuation they may provide.

Analyzing a duct system’s components for the purpose of determining the attenuation factors
of each separate element is covered later in this document.

Damper Airflow Noise

Dampers generate aerodynamic sound and also attenuate a certain amount of aerodynamic
sound. The result can be either a net gain or reduction in sound pressure level, depending
upon the airflow velocity through the damper, pressure drop and several other factors. Like
the previous discussion on determining a fan’s sound power level, it is necessary to
determine the actual sound power level at each octave band for a damper at the HVAC
system’s design operating conditions.

The following formula provides the actual sound power level increase or decrease at each
octave band for a damper at specific operating conditions:

DLw = K + 10 Log F + 50 Log U + 10 Log S + 10 Log D -107 dB

Where:

DLw = the net damper sound power level increase (or decrease if it is negative).

K = a factor (characteristic spectrum) dependent upon the damper’s operating conditions.

F = the octave band center frequency in Hz.

U = the damper velocity factor dependent upon the damper pressure drop.

S = the duct cross sectional area in square feet at the damper location.

D = the duct height in feet.

107 dB = a constant for any damper and relates to a damper’s ability to attenuate sound.

U (Velocity Factor)
Determining the U (velocity factor) requires a separate series of calculations. The following
procedure will yield the U value. However, in lieu of following this lengthy procedure, Figure 9
can also be used to obtain approximate U values for common airflows and damper pressure
drops.

Calculate Pressure Loss Coefficient C

To calculate pressure loss coefficient C, use the following formula:

30 Siemens Building Technologies, Inc.

 Damper Airflow Noise

C = 15.9 x 106 DP S2 ÷ Q2

Where:

DP = the pressure drop across the damper from the manufacturer’s data.

S = the duct area at the damper in square feet

Q = the CFM.
In lieu of S2/Q2, the term 1/V2 may be substituted, where V is the air velocity in feet/minute.

Calculate Damper Blockage Factor BF

To calculate damper blockage factor BF for a multi-blade damper, use the following formula:

If C ≠ 1, BF = (C½ -1) ÷ (C - 1)

If C = 1, BF = 0.5

To calculate damper blockage factor BF for a single-blade damper, use the following formula:

If C < 4, BF = C0.5 - 1) ÷ (C - 1)

If C > 4, BF = (0.68 x C-0.15) - 0.22

Calculate the Velocity Factor U

To calculate the velocity factor U, use the following formula:

U = (Q ÷ S) ÷ BF
In lieu of Q/S, the term V may be substituted, where V is the air velocity in feet/minute.

The graph in Figure 9 provides Velocity Factor values for dampers within common airflow
ranges and pressure drops. Using the graph eliminates the need to go through the foregoing
three equations to calculate U.

Siemens Building Technologies, Inc. 31

Chapter 3–HVAC Sound Sources

Figure 9. Velocity Term U, For Dampers.

K Factor

The K factor is determined by first calculating the Strouhal number, and then referring to the
Strouhal Graph in Figure 10, or using the following separate K equations. Since the Strouhal
number is dependent upon the octave band frequency, a separate Strouhal number (St) and
K factor must be determined for each of the eight octave bands.

Strouhal number (St) = 60 F D ÷ U

Once the Strouhal numbers are calculated for each octave band, refer to Figure 10 to
determine the K factors or use the appropriate equation below to calculate the K factor
directly from the Strouhal number.

For St = ≤ 25: K = -36.3 - 10.7 Log (St)

For St = ≥ 25: K = -1.1 -35.9 Log (St)

32 Siemens Building Technologies, Inc.

 Damper Airflow Noise

Figure 10. Characteristic Spectrum K vs. Strouhal Number for Dampers.

Example of Damper Sound Power Level Calculation

Assume a rectangular opposed blade control damper is to be used to control duct static
pressure in a 48 inches wide x 24 inches high supply system duct. The system is designed to
provide 20,000 cfm at a 0.3 in. WC gauge pressure drop across the damper. Determine the
sound power level that the damper will produce under these conditions.

Determine the values for the terms in the Damper Sound Power Level (DLw) formula:

DLw = K + 10 Log F + 50 Log U + 10 Log S+ 10 Log D -107 dB

Where:

F = the octave band center frequency in Hz: (63, 125, 250, 500, 1,000, etc.).

U = the damper Velocity Factor: (Duct Air Velocity = 20,000 cfm/8 sq. ft.
= 2,500 ft/min)
From Figure 10, U = 4,650.

S = the duct area square feet at the damper location: (4 ft x 2 ft = 8 sq ft).

D = the duct height in feet (2 ft).

Siemens Building Technologies, Inc. 33

Chapter 3–HVAC Sound Sources

K requires solving for the Strouhal number at each octave band center frequency and then
using the graph in Figure 10 to determine K.

St = 60FD÷ U

St= 60 x 63 Hz x 2 ft. ÷ 4,650 = 1.6, K = -39

St= 60 x 125 Hz x 2 ft. ÷ 4,650 = 3.2, K = -42
St= 60 x 250 Hz x 2 ft. ÷ 4,650 = 6.4, K = -45
St= 60 x 500 Hz x 2 ft. ÷ 4,650 = 12.8, K = -48
St= 60 x 1,000 Hz x 2 ft. ÷ 4,650 = 25.6, K = -52
St= 60 x 2,000 Hz x 2 ft. ÷ 4,650 = 51.2, K = -63
St= 60 x 4,000 Hz x 2 ft. ÷ 4,650 = 102.4, K = -75
St= 60 x 8,000 Hz x 2 ft. ÷ 4,650 = 204.8, K = -86

Note that after the first St number is calculated, the remaining seven values can be quickly
determined by just doubling the previous St value.

Table 8 summarizes the terms and results of using the DLw formula to determine the net
sound power level of this damper at each octave band.

Note in Table 8 that since the DLw values are all positive, the damper has the potential for
adding to the net sound power level produced by the HVAC system. However, if this damper
was located within a short distance from the fan example of a few pages earlier, the net
change in the total sound power level would not perceptibly change since the fan’s sound
power level was significantly higher than this damper by approximately 25 to 35 dB
throughout all of the octave bands. See Table 3 for adding sound power or sound pressure
levels.

On the other hand, if this damper were near a room served by the HVAC system, and the fan
sound had already been substantially attenuated (lessened), the sound power level
generated by this damper could have a significant impact on the sound pressure level in the
room.

Keep in mind that control devices such as dampers should be located as far as possible
upstream to minimize their effect on the overall HVAC system sound power level.

Elbow Airflow Noise

Elbows generate aerodynamic sound and also attenuate a certain amount of aerodynamic
sound. The resultant will either be a net gain or reduction in sound pressure level, depending
upon the airflow velocity through the elbow, pressure drop and several other factors. Like the
previous discussion on fan and damper sound power, it is necessary to determine the actual
sound power level at each octave band for an elbow at the HVAC system’s design operating
conditions.

The following formula will provide the actual sound power level increase or decrease at each
octave band for an elbow between two sections of the same size of duct:

ELw = K + 10 Log F + 50 Log U + 10 Log S+ 10 Log D + EC

34 Siemens Building Technologies, Inc.

 Elbow Airflow Noise

Where:

ELw = the net elbow sound power level increase (or decrease if it is negative).

K = a factor that is dependent upon the elbow operating conditions.

F = the octave band center frequency in Hz.

U = the duct airflow velocity.

S = the duct cross sectional area in square feet.

D = the height of the elbow in feet for elbows without turning vanes.

D = the cord length of a vane in feet for elbows with turning vanes.

EC = a constant that depends upon the type of elbow and relates to the elbow’s ability to
attenuate sound.

For an elbow without turning vanes, EC = -107.

For an elbow with turning vanes, EC = 10 Log n -107, where n is the number of turning
vanes.

K Factor

The K factor is determined by first calculating the Strouhal number. Since the Strouhal
number is dependent upon the octave band frequency, a separate Strouhal number (St) and
K factor must be determined for each of the eight octave bands.

Strouhal number (St) = 60 F D ÷ U

Siemens Building Technologies, Inc. 35

Chapter 3–HVAC Sound Sources

Figure 11. K versus Strouhal Number for Elbows.

Once the Strouhal numbers are calculated for each octave band, see Figure 12 to obtain the
K factors. The K factors may also be calculated using the following formulae:

For elbows with turning vanes:

K = - 47.5 - 7.69 (Log St)2.5

For elbows without turning vanes:

K = - 9.22 - 16.48 (Log St) - 5.05 (Log St)2

Example Elbow Sound Power Level Calculation

Determine the net sound power level that a mitered rectangular elbow will produce in a 36
inches wide x 24 inches high supply system duct with 12,000 cfm. The elbow has 15 single-
thickness turning vanes with a 9.5-inch cord length.

36 Siemens Building Technologies, Inc.

 Elbow Airflow Noise

Determine the values for the terms in the Elbow Sound Power Level (DLw) formula:

ELw = K + 10 Log F + 50 Log U + 10 Log S + 10 Log D – EC

Where:

F = the octave band center frequency in Hz: (63, 125, 250, 500, 1,000, etc.).

U = the airflow velocity (12,000 cfm/6 sq ft = 2,000 fpm).

S = the duct area sq. ft. (36 in. x 24 in. =3 ft x 2 ft = 6 sq ft).

D = the vane cord length in feet (0.8 ft.).

For an elbow with turning vanes, EC = 10 Log n – 107 = 10( 1.18) - 107 = 11.8- 107 = -95.2.

K requires solving for the Strouhal number at each octave band center frequency and then
using the graph in Figure 12 to determine K.

St = 60FD÷ U St=60 x 63 Hz x0.8 ft.÷ 2,000 = 1.5, K = -48

St =60 x 125 Hz x 0.8 ft. ÷ 2,000 = 6.0, K = -52
St = 60 x 250 Hz x 0.8 ft. ÷ 2,000 = 12.0, K = -58
St = 60 x 500 Hz x 0.8 ft. ÷ 2,000 = 24.0, K = -65
St = 60 x 1,000 Hz x 0.8 ft. ÷ 2,000 = 48.0, K = -76
St = 60 x 2,000 Hz x 0.8ft. ÷ 2,000 = 96.0, K = -91
St = 60 x 4,000 Hz x 0.8 ft. ÷ 2,000 = 192.0, K = -111
St = 60 x 8,000Hz x 0.8 ft.÷ 2,000 = 384.0, K = -130

Note that after the first St number is calculated, the remaining seven values can be quickly
determined by just doubling the previous St value.

Table 9 summarizes the terms and results of using the ELw formula to determine the net
sound power level of this elbow at each octave band.

Table 9. Sound Calculation Summary for Actual Fan Operating Conditions.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
K -48 -51 -52 -58 -65 -76 -91 -111
10 Log F 18 21 24 27 30 33 36 39
50 log U 204 204 204 204 204 204 204 204
10 Log S 8 8 8 8 8 8 8 8
10 Log D -1 -1 -1 -1 -1 -1 -1 -1
EC -95 -95 -95 -95 -95 -95 -95 -95
ELw 86 86 90 85 81 73 61 44
dB dB dB dB dB dB dB dB

Siemens Building Technologies, Inc. 37

Chapter 3–HVAC Sound Sources

In Table 9, the ELw values are all positive, so this elbow adds to the sound power level
produced by the HVAC system. If this elbow was located within a short distance from the
Example Fan Sound Power Level Calculation, the net change in the total sound power level
would not perceptibly change since the fan’s sound power level was significantly higher by a
range of 16 to 51 dB throughout the octave bands. On the other hand, if this elbow were near
a room served by the HVAC system, the sound power level could have a very significant
impact on the sound pressure level in the room since this elbow generates significant sound
power in the lower frequency octaves.

Junction and Takeoff Airflow Noise

Just as dampers and elbows have the ability to add to or attenuate the sound power level, so
do junctions and takeoffs. The resultant may be a net gain or reduction in sound pressure
level, depending upon the airflow velocity, pressure drop and several other factors. Like the
previous discussions on fan, damper, and elbow sound power levels, it is necessary to
determine the actual sound power level at each octave band for a junction or takeoff at the
HVAC system’s design operating conditions.

The following formula will provide the actual sound power level increase or decrease at each
octave band for a duct junction or takeoff:

JLw = K + 10 Log F + 50 Log UB + 10 Log S + 10 Log D - JC

Where:

Elw = the net sound power level increase (or decrease if it is negative).

K = an factor dependent upon design conditions.

F = the octave band center frequency in Hz.

UB = the branch duct airflow velocity.

S = the branch duct cross sectional area in square feet.

D (for Takeoffs) = branch duct height in feet.

D (for Junctions) = (4S/π)0.5 feet.

K Factor
The K factor is determined by first calculating the Strouhal number. Since the Strouhal
number is dependent upon the octave band frequency, a separate Strouhal number (St) and
K factor must be determined for each of the eight octave bands.

Strouhal number (St) = 60 F D ÷ UB

38 Siemens Building Technologies, Inc.

 Junction and Takeoff Airflow Noise

When the Strouhal numbers are calculated for each octave band, use Figure 12 to obtain the
K factors. Before using Figure 12, the velocity ratio M, between main duct airflow velocity
(UM) and branch duct airflow velocity (UB) is calculated as:

M = UM ÷ U B

The K factors may also be calculated using the following very lengthy formula:

K = -21.6 + 12.4 M0.67 - 16.5 M-0.3 (Log St) 5.0 M-0.25 (Log St)2

Figure 12. K versus Strouhal Number for Junctions.

JC Factor
JC is a constant that depends upon the configuration of the junction and relates to the
junction’s ability to attenuate sound.

JC = -107 + Δr + ΔT

Siemens Building Technologies, Inc. 39

Chapter 3–HVAC Sound Sources

Where:

Δr = Junction Radius/Branch Duct Diameter.

= r ÷ DBR

Note that for duct junctions and takeoffs without a radius, Δr is 0.0. However for duct
junctions and takeoffs with a radius, Δr is greater than zero, and reduces the sound power
level generated at the junction.

For junctions and takeoffs with a radius, the Δr values must be corrected for each octave
band. This requires taking the Strouhal number (already calculated for the K factors) for each
octave band and also the value of r ÷ DBR, then using Figure 13 to arrive at the final corrected
Δr values for each octave band.

The turbulence factor of ΔT applies if there are other duct elements within five duct diameters
upstream of the main junction or takeoff. After calculating M, Table 10 gives values for ΔT.

Table 10. Upstream Turbulence Factor Values for Junctions and Takeoffs

Example Duct Takeoff Sound Power Level Calculation

Determine the net sound power level that will result from a 12-foot square duct takeoff from a
36 inch wide x 24 inch high main supply system duct. The main duct supplies 14,500 cfm
while the takeoff duct is intended to provide a maximum of 1,600 cfm to a VAV box. The
takeoff uses is a standard 45 degree converging tee (no radius) duct element. There is also a
normally open smoke damper approximately eight feet upstream from this takeoff.

40 Siemens Building Technologies, Inc.

 Junction and Takeoff Airflow Noise

Figure 13. Final Corrected Δr Factors for Duct Junctions and Takeoffs with a Radius.

Determine the values for the terms in the Junction Sound Power Level DLw) formula:

JLw = K + 10 Log F + 50 Log UB + 10 Log S+ 10 Log D – JC

Where:

F = the octave band center frequency in Hz: (63, 125, 250, 500, 1,000, etc.).

UB = the branch duct air velocity: (1,600 cfm / 1 sq ft = 1,600 fpm).

S = the branch duct area: (1 sq ft).

D = the branch duct height: (1 ft).

Determining the K Factor requires first calculating M and then solving for the Strouhal
number at each octave band center frequency. Then the graph in Figure 12 is used to
determine K.

M = UM ÷ U B

UM = 14,500 cfm ÷ 6 sq ft

= 2,420 fpm

UB = 1,600 fpm

Siemens Building Technologies, Inc. 41

Chapter 3–HVAC Sound Sources

Therefore:

M = 2,420 ÷ 1,600
M = 1.5
St = 60 F D ÷ UB
St = 60 x 63 Hz x 1 ft ÷ 1,600 = 2.4, K = -11
St =60 x 125 Hz x 1 ft ÷ 1,600 = 4.7, K = -18
St =60 x 250 Hz x 1 ft ÷ 1,600 = 9.4, K = -24
St = 60 x 500 Hz x 1 ft ÷ 1,600 = 18.8, K = -30
St = 60 x 1,000 Hz x 1 ft ÷1,600 = 37.5, K = -40
St = 60 x 2,000 Hz x 1 ft ÷1,600 = 75.0, K = -49
St = 60 x 4,000 Hz x 1 ft ÷ 1,600 = 150.0, K = -57
St = 60 x 8,000 Hz x 1 ft ÷ 1,600 = 300.0, K = -68

JC is a constant that depends upon the configuration of the junction and relates to the
junction’s ability to attenuate sound.

JC = -107 + Δr + ΔT

Δr = Junction Radius/Branch Duct Diameter

= r ÷ DBR r = 0 and r ÷ DBD = 0

=0÷0

=0

Since there is upstream turbulence (a smoke damper) within five main duct diameters, Table
10 is used with the 1.5 value for M to obtain a ΔT value of approximately 1 dB.

Therefore, JC = -107 + 0 + 1 = -106

Table 11 summarizes the terms and results of the JLw formula to determine the net sound
power level of this junction at each octave band.

Table 11. Sound Calculation Summary for Actual Fan Operating Conditions.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
K -11 -18 -24 -30 -40 -49 -57 -68
10 Log F 18 21 24 27 30 33 36 39
50 log U 160 160 160 160 160 160 160 160
10 Log S 0 0 0 0 0 0 0 0
10 Log D 0 0 0 0 0 0 0 0
JC -106 -106 -106 -106 -106 -106 -106 -106
JLw 61 57 54 51 44 38 33 25
dB dB dB dB dB dB dB dB

42 Siemens Building Technologies, Inc.

 Air Delivery Device Sound

In Table 11, the JLw values are all positive, so this takeoff adds to the sound power level
produced by the HVAC system. Since a takeoff is normally near an occupied area, its sound
power level will likely have an impact on the HVAC sound in the area.

Note that the higher the Strouhal number, the lower the net sound power level will be. The
Strouhal number increases as the duct size increases and also as the airflow velocity is less,
so the general approach to maintaining an acceptable sound level is always to avoid high air
velocities.

Air Delivery Device Sound

Just as dampers, elbows, junctions and takeoffs have the ability to add to or attenuate the
sound power level, so do the terminals and diffusers that function as the final air volume
control and distribution devices.

Sound power levels at rated airflow’s are normally available for these devices from the
manufacturers. Manufacturers typically provide tables for these devices that give the NC
rating at specific airflow’s (cfm) through the device. Recall from the earlier discussion in this
document on noise criteria (NC) curves, that a NC rating means that the sound power level
does not exceed the dB values of the respective NC curve at any of the octave band center
frequencies. Thus to determine the octave band dB values associated with a specific NC
rating, one may refer to a plot of the respective NC curve and note the corresponding dB
values at each of the octave band center frequencies. To save having to do this, Table 12
lists the dB values associated with each of the common NC curves for the octave band
center frequencies.

However, note that the dB values associated with the NC curve (and listed in the table) are
the maximum levels to be expected. Actual device sound levels in many of the octave bands
may be less thereby resulting in a lower room sound level.

Table 12. NC Curves vs. Sound Pressure Level Decibels.

NC 63 125 250 500 1,000 2,000 4,000 8,000

CURVE Hz Hz Hz Hz Hz Hz Hz Hz
15 47 36 29 22 17 14 12 11
20 51 40 33 26 22 20 17 16
25 54 45 38 31 27 24 22 21
30 57 48 42 35 31 30 28 27
35 60 53 46 40 36 34 33 32
40 64 57 51 45 41 39 38 37
45 67 60 54 49 46 44 43 42
50 71 64 59 54 51 49 48 47
55 74 67 62 58 56 54 53 52
60 77 71 67 63 61 59 58 57
65 80 75 71 68 66 64 63 62

Siemens Building Technologies, Inc. 43

Chapter 3–HVAC Sound Sources

There will be little opportunity for sound attenuation after the diffuser. Many air diffusers are
available with an integral throttling damper as part of the diffuser. Since the throttling damper
has the potential for generating airflow sound, this must also be taken into account with
reference to the air diffuser’s sound rating. Therefore, whenever a diffuser has a throttling
damper in its collar, the following increases to the dB values in Table 9 apply:

+5 dB for a 0.05-inch throttling damper pressure drop

+10 dB for a 0.15-inch throttling damper pressure drop

+15 dB for a 0.25-inch throttling damper pressure drop

Since a device such as a VAV terminal, is normally located above a room’s ceiling and also
uses additional ductwork to connect to the air diffusers, there is some potential for sound
attenuation of a terminal’s sound pressure level by its diffuser ductwork. However, since air
diffusers are typically the last element on the HVAC system, they should be selected based
upon an NC rating that does not exceed the room’s NC rating, since there will be little
opportunity for sound attenuation after the diffuser.

Flexible Duct Connection to Diffusers

Designers should incorporate a notation on the mechanical plans wherever flexible duct is
permitted to connect air terminal units to room supply diffusers, to avoid creating additional
diffuser sound due to the air turbulence caused by the duct offset. Centerline offsets greater
than 1/8 of the diffuser collar diameter for each equivalent collar length will begin to
appreciably add to the diffuser NC rating, and can ultimately result in increases of up to 12
dB.

Likewise, any turbulence inducing device such as a balancing damper should be located as
far as practical (that is, 10 duct diameters) upstream of the diffuser to avoid possible
increased diffuser sound.

Discharge Sound and Radiated Sound

So far, we have discussed how sound travels through a duct system and is eventually
discharged into a space. Aside from this type of sound transmission, sound can also be
transmitted as HVAC radiated sound into a space. Radiated sound is when a sound source
produces sound waves that travel directly to the listener, (through the air or through ceilings
and walls) without using the ductwork as a conduit. The whine and rumble of a HVAC fan that
is heard when inside of an equipment room is sound that is radiated directly to the listener.

Room terminal units, especially fan powered boxes, also radiate sound that can be heard by
a room occupant if the acoustical attenuation of the ceiling material and the room is not
sufficient to absorb the radiated sound power. Manufacturer’s data for terminal units should
provide sound ratings for radiated sound and discharge sound. The section on sound
attenuation will cover how to account for the effects of both radiated and discharge sound.

44 Siemens Building Technologies, Inc.

 Air Delivery Device Sound

Sound Breakout and Break-in

Aerodynamic sound that has sufficiently high energy within a duct can also be heard outside
of the duct. If you were on a ladder extending above a ceiling where HVAC equipment is
located, you might hear the sound of the air passing across a damper or even the rumble of
an upstream fan. The sound is radiated directly to us and is referred to as breakout sound.

Similarly, if we were doing some hammering nearby or talking loudly next to a duct, the
sound being produced could radiate to the inside of the duct system and travel along with the
sound already there. Sound that is radiated into a duct system is referred to as break-in
sound. Breakout sound can have an adverse impact on the overall HVAC system sound in a
space particularly if there is no drop ceiling between the space and the duct, or if the
acoustical absorption capability of the ceiling is limited.

While breakout sound can adversely impact the sound level in a space where there is an
inadequate acoustical barrier, breakout sound also reduces the overall sound power level in
the duct since the breakout sound carries off a portion of the total sound power level energy.
Thus, the breakout sound phenomenon has an attenuating effect on the sound that would
otherwise be present in the duct system. However, due to the complexity of attempting to
quantify this effect, it is typically omitted from HVAC sound analysis calculations.

In applications where there are very long duct runs, the breakout sound phenomenon can
have a significant positive effect. For details on breakout and break-in sound transmission
loss and related calculations, consult the ASHRAE HVAC Applications Manual, Chapter 42.

Laboratory Elements
The primary source of a chemical laboratory’s HVAC ambient sound is on the exhaust side of
the ventilation system. The principles that apply to HVAC supply side produced sound
(airflow’s through duct fittings, dampers, etc.) also apply to the exhaust side and are
calculated in the same manner. Since fume hoods are the terminal point of the exhaust
system, the NC factors that apply to a particular fume hood operating at specific airflow’s
must be obtained from the manufacturer and can then be used as part of the overall exhaust
system analysis.

Later in this document, you will learn how to perform a sound analysis of varying HVAC
system configurations including a laboratory fume hood exhaust system.

Siemens Building Technologies, Inc. 45

Chapter 4–HVAC Sound Attenuation
Chapter 4 discusses the attenuating effect of common HVAC system elements (also referred
to as transmission loss or insertion loss). It includes the following topics:

• Plenums

• Duct Attenuation

• Duct Takeoffs and Divisions

• Duct Silencers

• End Reflection

• Environment Adjustment Factor

• Space Effect

• Radiated Sound Attenuation

Introduction to HVAC Sound Attenuation

Chapter 3, HVAC Sound Sources, established the means to calculate the sound power levels
generated as air passes through the duct system and some of its energy was converted into
sound pressure due to the resistance caused by damper blades, elbows, takeoffs, etc. Not
only do these duct elements cause additional generation of sound, but they also have an
attenuating effect on sound that was already generated by other duct system elements.
Therefore, at any point in a duct system, we have the phenomenon of an ever changing
sound power level and profile due to the simultaneous process of sound generation and
attenuation taking place.

So, even though the net sound power level at a given point in a duct system, such as
immediately upstream or downstream of an exhaust or supply fan may be quite high, the
sound power level will undergo a natural attenuation (reduction) with reference to more
distant areas of the duct system.

With respect to attaining an acceptable sound level in a given space or room, the key issue is
whether there will be sufficient natural attenuation to reduce the sound power to the proper
level, or whether the HVAC system needs to incorporate additional sound attenuation
elements to provide the required attenuation.

This chapter covers the attenuating effect of common HVAC system elements (also referred
to as transmission loss or insertion loss). With this information and the information provided
in Chapter 3, a given HVAC system or a portion of an HVAC system can be analyzed to
obtain the resultant sound power level at any given point. The next section provides
examples of how to perform this type of analysis.

Siemens Building Technologies, Inc. 47

Chapter 4–HVAC Sound Attenuation

As previously discussed, sound travel is independent of the direction of airflow. Therefore, all
calculations involving sound generation or attenuation apply to the exhaust portion of an
HVAC system as well as to the supply side.

Plenums
Plenums that are constructed of concrete will have virtually negligible attenuation effect on
sound generated by HVAC components. Plenums of unlined sheet metal will provide only a
little more attenuation effect than concrete. On the other hand, plenums that are fully lined
with at least two inches of sound adsorbing material can provide very significant sound
attenuation. The specific analysis of a given plenum is very complex from an acoustical
standpoint. Reference books provide mathematical procedures to calculate the attenuation of
certain plenum configurations, however these are very time consuming and are still only an
approximation. In addition, since there are almost endless arrangements possible with
plenum designs, it is extremely difficult to come up with precise results regardless of which
calculation procedure is used.

For practical considerations, it is best to assume that an unlined sheet metal plenum will have
a minimal effect on fan sound attenuation and therefore one can normally disregard its effect
on the fan sound generation. (Note that unlined sheet metal plenums are typically used in
centralized laboratory exhaust systems. However in laboratory exhaust systems, the
relatively long duct runs and large number of junctions usually provides adequate attenuation
of exhaust fan sound for the majority of areas served by such systems.)

On the other hand, supply fans discharging into fully lined plenums with 2-inch thick (or more)
sound absorbing material will typically reduce the low frequency (63 Hz to 125 Hz) sound
power by at least 5 dB, and will also reduce the upper frequencies (2,000 Hz to 8,000 Hz) by
at least 15 dB or more.

Figure 14 gives the sound absorption coefficient of different plenum materials. The higher the
plenum sound absorption coefficient, the greater will be the sound attenuation. Note how the
attenuation effect of the thicker fiberglass type liners is maximized at the lower frequencies.

Table 13 also provides specific values of plenum sound absorption coefficients.

Table 13. Plenum Lining Material vs. Sound Absorption Coefficient.

Material 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
Bare Concrete 0.01 0.01 0.01 0.02 0.02 0.02 0.03 0.04
Bare Sheet Metal 0.04 0.04 0.04 0.04 0.05 0.05 0.07 0.09
1-in. Thick Fiberglass 0.02 0.03 0.22 0.69 0.91 0.96 0.99 1.00
2-in. Thick Fiberglass 0.18 0.22 0.82 1.00 1.00 1.00 1.00 1.00
3-in. Thick Fiberglass 0.48 0.53 1.00 1.00 1.00 1.00 1.00 1.00
4-in. Thick Fiberglass 0.76 0.84 1.00 1.00 1.00 1.00 0.97 0.91

48 Siemens Building Technologies, Inc.

 Plenums

Figure 14. Relative Sound Absorbing Capabilities of Various Plenum Lining Material.

The following is a simplified formula that can be used to determine the approximate
attenuation a plenum will provide:

Plenum attenuation dB = Lf + {-10 Log [Ae (1 ÷ 6.283 d2) + Ae (1 -a)/(a * Aw)]}

Where:

Lf = Low frequency factor: 7 at 63 Hz, 6 at 125 Hz, and 1 at 250 Hz.

Ae = Area of plenum exit in square feet.

Aw = Area of sound absorption material on plenum walls in square feet.

d = Distance between the plenum inlet and exit in feet.

a = Material sound absorption coefficient (from Table 13).

Example Plenum Attenuation Calculation

Assume a plenum that is 10 feet long by 6 feet high and 6 feet wide is lined with 2-inch thick
fiberglass. If the fan inlet at one end is 3 feet square and the supply duct outlet at the
opposite end is 3 feet by 4 feet, determine what the expected attenuation would be.

Siemens Building Technologies, Inc. 49

Chapter 4–HVAC Sound Attenuation

Area of the outlet is: 3 ft x 4 ft = 12 ft2

Area of the inlet is: 3 ft x 3 ft = 9 ft2
Area of each plenum end is: 6 ft x 6 ft = 36 ft2
Net Area of outlet is: 36 ft2 - 12 ft2 = 24 ft2
Net Area of inlet end is: 36 ft2 - 9 ft2 = 27 ft2
Area of plenum sides is: 12 ft x (6 ft + 6 ft + 6 ft + 6 ft) = 12 ft x (24 ft) = 288 ft2
Total plenum acoustically lined area is: 24 ft2 + 27 ft2 + 288 ft2 = 339 ft2

From the 2-inch Thick Fiberglass row in Table 13 we get the sound absorption coefficients for
each octave band as indicated in Table 14.

Table 14. Plenum Sound Absorption Coefficients for 2-inch Thick Fiberglass.

Material 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
2-in. Thick Fiberglass 0.18 0.22 0.82 1.00 1.00 1.00 1.00 1.00

The approximate dB attenuation for each octave band is determined by taking the sound
absorption coefficient for each octave band and using the preceding formula. The calculation
is shown for the 63 Hz band and the results for the other bands are listed in Table 15.

dB = Lf + {-10 Log [Ae (1 ÷ 6.283 d2) + Ae (1 - a)/(a * Aw)]}

Where:

Lf = Low frequency factor: = 7 at 6 Hz.

Ae = Area of plenum exit: = 12 ft2.

Aw = Area of sound absorption material: = 339 ft2.

d = Distance between the plenum inlet: = 10 ft.

a = Material sound absorption coefficient at 63 Hz: = 0.18.

Therefore:

dB =7 + {-10 Log [12 x (1 ÷ 6.283 x 102) +12 x (1 -0.18) ÷ (0.18 x 339)]}

= 7 + {-10 Log [12 x (1 ÷ 628.3) +12 x (0.82 ÷ 61)}

= 7 + {-10 Log [0.0191 + 0.1613]}

= 7 + {-10 Log [0.18]} = 7 + (7.439)

= 14.4

50 Siemens Building Technologies, Inc.

 Duct Attenuation

Table 15. Approximate dB Attenuation for Each Octave Band.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
14.4 dB 14.4 dB 16.7 dB 17.2 dB 17.2 dB 17.2 dB 17.2 dB 17.2 dB

Note that when the sound absorption coefficients are the same (as they are for 500 Hz and
higher in this example), the attenuation level is the same.

Duct Attenuation
The most efficient approach to sound attenuation is to attain the necessary level of
attenuation by means of the duct runs. While this may not always be practical, or even
attainable, it is the optimum solution from a first cost and operating cost standpoint.

The attenuation or insertion loss of sheet metal ducts is highly dependent upon the size and
shape of the duct. Smaller ducts attenuate more dB per foot than larger ducts, and for a
given effective area, ducts that are more rectangular attenuate more dB than those that are
closer to being square or round. Also, ducts that have an internal lining attenuate sound
much more than unlined ducts.

Rectangular Unlined Sheet Metal Ducts

To determine the attenuation of this type of duct, first calculate the Perimeter to Area ratio:

Perimeter to Area ratio = P ÷ A

Where:

P = the duct perimeter in feet.

A = the duct cross sectional area in square feet.

Next, use the appropriate formula below to determine the dB attenuation per foot of duct
length or refer to Table 16.

• If P/A is equal to or greater than 3:

Attenuation per foot = 17.0 x (P ÷ A)-0.25 x Hz-0.85

• If P/A is less than 3:

Attenuation per foot = 1.64 x (P ÷ A)0.73 x Hz-0.58

Example Rectangular Duct Attenuation Calculation

An unlined duct is 48 inches wide by 30 inches high. Determine the attenuation for each
octave band for an 80 foot run.

Siemens Building Technologies, Inc. 51

Chapter 4–HVAC Sound Attenuation

First calculate the P/A ratio:

P = (48 in. + 36 in. + 48 in. + 36 in. ) ÷ 12 in./ft

= 168 ÷ 12

= 14 feet

A = (48 in. x 36 in.) ÷ 144 sq in./sq ft

= 1728 ÷ 144

= 12 sq ft

P/A = 14 ÷ 12 = 1.17

With reference to Table 16, the following dB attenuation factors per foot can be interpolated:

P/A 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
1.17 .166 .112 .075 .050 .033 .022 .015 .010

Multiplying the values obtained by 80 feet yields the following attenuation in each octave
band:

— 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
— 13.3 dB 9.0 dB 6.0 dB 4.0 dB 2.6 dB 1.8 dB 1.2 dB 0.8 dB

Table 16. Unlined Rectangular Duct Attenuation Per Foot of Length.

P/A 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
5.00 .336 .188 .104 .058 .032 .018 .010 .006
4.00 .355 .198 .110 .061 .034 .019 .010 .006
3.75 .361 .202 .112 .062 .034 .019 .010 .006
3.50 .367 .205 .114 .063 .035 .019 .011 .006
3.25 .374 .209 .116 .064 .036 .020 .011 .006
3.00 .382 .213 .118 .066 .036 .020 .011 .006
2.75 .310 .209 .140 .093 .062 .042 .028 .018
2.50 .290 .195 .130 .087 .058 .039 .026 .017
2.25 .268 .180 .120 .081 .054 .036 .024 .016
2.00 .246 .165 .111 .074 .049 .033 .022 .015
1.90 .237 .159 .107 .071 .048 .032 .021 .014
1.80 .228 .153 .102 .069 .046 .031 .021 .014

52 Siemens Building Technologies, Inc.

 Duct Attenuation

P/A 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
1.75 .223 .150 .100 .067 .045 .030 .020 .013
1.70 .219 .147 .098 .066 .044 .029 .020 .013
1.60 .209 .140 .094 .063 .042 .028 .019 .012
1.50 .199 .134 .090 .060 .040 .027 .018 .012
1.40 .190 .127 .085 .057 .038 .026 .017 .011
1.30 .180 .121 .081 .054 .036 .024 .016 .011
1.25 .175 .117 .078 .053 .035 .024 .016 .010
1.20 .169 .114 .076 .051 .034 .023 .015 .010
1.15 .164 .110 .074 .049 .033 .022 .015 .010
1.10 .159 .107 .071 .048 .032 .021 .014 .010
1.05 .154 .103 .069 .046 .031 .021 .014 .009
1.00 .148 .100 .067 .045 .030 .020 .013 .009
0.95 .143 .096 .064 .043 .029 .019 .013 .009
0.90 .137 .093 .062 .041 .028 .019 .012 .008
0.85 .132 .089 .059 .040 .026 .018 .012 .008
0.80 .126 .085 .057 .038 .025 .017 .011 .008
0.75 .120 .081 .054 .036 .024 .016 .011 .007
0.70 .114 .077 .051 .034 .023 .015 .010 .007
0.65 .108 .073 .049 .033 .022 .015 .010 .007
0.60 .102 .069 .046 .031 .021 .014 .009 .006
0.55 .096 .064 .043 .029 .019 .013 .009 .006
0.50 .089 .060 .040 .027 .018 .012 .008 .005

Rectangular Unlined, Externally Insulated, Sheet Metal Ducts

Sheet metal ducts that have an unlined interior but are thermally insulated with about one
inch of fiberglass on the exterior, will attenuate sound at approximately twice the rate of
uninsulated, unlined rectangular ducts. (Be careful not to confuse external insulation with
ducts that are internally lined with sound absorbing material. Internally lined ducts have a
much higher rate of sound absorption than externally insulated ducts.)

To determine the attenuation of externally insulated rectangular ducts, multiply the final
attenuation figures obtained by the procedure for unlined (non-insulated) ducts, by 2.0.

In the previous example, the sound attenuation of the 80 foot run of 48-inch by 30-inch duct
would thus be approximately twice the values listed if the exterior of the duct were thermally
insulated.

Siemens Building Technologies, Inc. 53

Chapter 4–HVAC Sound Attenuation

Rectangular Acoustically Lined Sheet Metal Ducts

Rectangular ducts internally lined with at least one-inch thick fiberglass or similar material are
very effective at attenuating sound, particularly in the mid frequencies. Recall from Figure 14
on plenum lining, that fiberglass dramatically increases sound attenuation in the upper octave
bands. Also 2-inch thick fiberglass is significantly superior to the 1-inch thick material
particularly in the lower octave bands.

Figure 15. FW Values vs. dB Attenuation For Rectangular and Round Duct Elbows.

Equations have been developed over the last several years to calculate the specific dB
attenuation levels for various lining thickness; however, these are complex and would
consume extensive time. Table 17 and Table 18 provide values for rectangular ducts with 1-
inch and 2-inch internal acoustical linings based upon the Perimeter to Area (P/A) ratio of the
duct. The P/A ratio is calculated by dividing the duct perimeter (P) in feet by the area (A) in
square feet. Also, due to other factors involved in duct sound transmission, no more than 40
dB should be used when determining the attenuation values of a specific length of lined duct.

Table 17. One-Inch Thick Fiberglass Lined Rectangular Duct Attenuation Per Foot of Length.

P/A 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
12.00 2.00 2.06 2.18 3.66 10.10 10.36 4.80 2.87
11.00 1.75 1.86 2.05 3.51 9.51 9.66 4.61 2.81
10.00 1.49 1.63 1.89 3.33 8.90 8.95 4.41 2.74

54 Siemens Building Technologies, Inc.

 Duct Attenuation

P/A 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
9.00 1.28 1.43 1.75 3.16 8.27 8.22 4.21 2.67
8.40 1.16 1.32 1.66 3.05 7.80 7.78 4.07 2.63
8.00 1.08 1.24 1.60 2.98 7.62 7.48 3.98 2.60
6.40 0.82 0.96 1.35 2.66 6.52 6.26 3.59 2.46
6.00 0.77 0.90 1.29 2.57 6.23 5.94 3.49 2.42
5.33 0.69 0.79 1.18 2.42 5.74 5.41 3.31 2.36
5.00 0.65 0.74 1.12 2.34 5.49 5.13 3.21 2.32
4.80 0.63 0.71 1.09 2.29 5.34 4.97 3.15 2.30
4.33 0.60 0.67 1.04 2.22 5.10 4.72 3.06 2.26
4.00 0.56 0.60 0.96 2.09 4.70 4.29 2.90 2.20
3.90 0.55 0.59 0.94 2.06 4.62 4.21 2.86 2.19
3.60 0.53 0.55 0.89 1.98 4.37 3.94 2.76 2.15
3.33 0.51 0.52 0.85 1.90 4.14 3.71 2.67 2.11
3.20 0.51 0.51 0.82 1.87 4.02 3.59 2.62 2.09
3.00 0.50 0.48 0.79 1.81 3.85 3.41 2.54 2.06
2.70 0.40 0.44 0.75 1.71 3.57 3.12 2.42 2.01
2.66 0.40 0.43 0.74 1.70 3.54 3.10 2.41 2.00
2.40 0.36 0.39 0.68 1.61 3.29 2.85 2.29 1.96
2.19 0.33 0.35 0.64 1.54 3.09 2.65 2.20 1.91
2.13 0.32 0.34 0.62 1.52 3.03 2.59 2.17 1.90
2.00 0.30 0.32 0.59 1.47 2.90 2.24 2.00 1.82
1.77 0.27 0.28 0.54 1.38 2.67 2.24 2.00 1.82
1.66 0.25 0.26 0.51 1.34 2.55 2.13 1.94 1.80
1.50 0.23 0.24 0.47 1.27 2.37 1.95 1.85 1.76
1.33 0.21 0.21 0.43 1.20 2.19 1.78 1.75 1.71
1.20 0.19 0.19 0.39 1.13 2.03 1.63 1.67 1.67
1.14 0.18 0.18 0.38 1.11 1.96 1.57 1.63 1.65
1.11 0.18 0.17 0.37 1.09 1.93 1.54 1.61 1.64
1.00 0.16 0.16 0.34 1.03 1.79 1.41 1.54 1.60
0.95 0.16 0.15 0.32 1.01 1.72 1.35 1.50 1.58
0.86 0.14 0.14 0.30 0.96 1.61 1.25 1.43 1.55
0.83 0.14 0.13 0.29 0.94 1.58 1.22 1.42 1.54

Siemens Building Technologies, Inc. 55

Chapter 4–HVAC Sound Attenuation

Table 18. Two-Inch Thick Fiberglass Lined Rectangular Duct Attenuation Per Foot of Length.

P/A 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
12.00 2.00 2.06 2.18 3.66 10.10 10.36 4.80 2.87
11.00 1.75 1.86 2.05 3.51 9.51 9.66 4.61 2.81
10.00 1.49 1.63 1.89 3.33 8.90 8.95 4.41 2.74
9.00 1.28 1.43 1.75 3.16 8.27 8.22 4.21 2.67
8.40 1.16 1.32 1.66 3.05 7.80 7.78 4.07 2.63
8.00 1.08 1.24 1.60 2.98 7.62 7.48 3.98 2.60
6.40 0.82 0.96 1.35 2.66 6.52 6.26 3.59 2.46
6.00 0.77 0.90 1.29 2.57 6.23 5.94 3.49 2.42
5.33 0.69 0.79 1.18 2.42 5.74 5.41 3.31 2.36
5.00 0.65 0.74 1.12 2.34 5.49 5.13 3.21 2.32
4.80 0.63 0.71 1.09 2.29 5.34 4.97 3.15 2.30
4.33 0.60 0.67 1.04 2.22 5.10 4.72 3.06 2.26
4.00 0.56 0.60 0.96 2.09 4.70 4.29 2.90 2.20
3.90 0.55 0.59 0.94 2.06 4.62 4.21 2.86 2.19
3.60 0.53 0.55 0.89 1.98 4.37 3.94 2.76 2.15
3.33 0.51 0.52 0.85 1.90 4.14 3.71 2.67 2.11
3.20 0.51 0.51 0.82 1.87 4.02 3.59 2.62 2.09
3.00 0.50 0.48 0.79 1.81 3.85 3.41 2.54 2.06
2.70 0.40 0.44 0.75 1.71 3.57 3.12 2.42 2.01
2.66 0.40 0.43 0.74 1.70 3.54 3.10 2.41 2.00
2.40 0.36 0.39 0.68 1.61 3.29 2.85 2.29 1.96
2.19 0.33 0.35 0.64 1.54 3.09 2.65 2.20 1.91
2.13 0.32 0.34 0.62 1.52 3.03 2.59 2.17 1.90
2.00 0.30 0.32 0.59 1.47 2.90 2.24 2.00 1.82
1.77 0.27 0.28 0.54 1.38 2.67 2.24 2.00 1.82
1.66 0.25 0.26 0.51 1.34 2.55 2.13 1.94 1.80
1.50 0.23 0.24 0.47 1.27 2.37 1.95 1.85 1.76
1.33 0.21 0.21 0.43 1.20 2.19 1.78 1.75 1.71
1.20 0.19 0.19 0.39 1.13 2.03 1.63 1.67 1.67
1.14 0.18 0.18 0.38 1.11 1.96 1.57 1.63 1.65
1.11 0.18 0.17 0.37 1.09 1.93 1.54 1.61 1.64
1.00 0.16 0.16 0.34 1.03 1.79 1.41 1.54 1.60

56 Siemens Building Technologies, Inc.

 Duct Attenuation

P/A 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
0.95 0.16 0.15 0.32 1.01 1.72 1.35 1.50 1.58
0.86 0.14 0.14 0.30 0.96 1.61 1.25 1.43 1.55
0.83 0.14 0.13 0.29 0.94 1.58 1.22 1.42 1.54

Round Unlined Sheet Metal Ducts

While unlined round ducts provide the least amount of sound attenuation, nevertheless their
attenuating effect should be included when designing a duct system. Table 19 provides dB
per foot values that can be applied to unlined round ducts.

Table 19. Unlined Round Duct Per Foot of Length.

Diamete 63 125 250 500 1,000 2,000 4,000 8,000

r
Inches Hz Hz Hz Hz Hz Hz Hz Hz
6 or Less 0.03 0.03 0.05 0.05 0.10 0.10 0.10 0.11
7 to 10 0.03 0.03 0.04 0.05 0.09 0.09 0.09 0.10
11 to 15 0.03 0.03 0.03 0.05 0.07 0.07 0.07 0.08
16 to 22 0.02 0.02 0.02 0.04 0.06 0.06 0.06 0.07
24 to 30 0.02 0.02 0.02 0.03 0.05 0.05 0.05 0.06
32 to 48 0.01 0.01 0.01 0.02 0.03 0.03 0.03 0.04
50 and 0.01 0.01 0.01 0.02 0.02 0.02 0.02 0.03
Above

Round Acoustically Lined Sheet Metal Ducts

Round spiral wound sheet metal ducts lined with at least one-inch thick fiberglass or similar
material are more effective at attenuating sound, particularly in the mid frequencies. As with
rectangular ducts, the equations to calculate the specific dB attenuation levels for various
lining thickness are very complex and require much effort to work through. Table 20 and
Table 21 provide values that can be used for round ducts with 1-inch and 2-inch internal
acoustical linings.

Table 20. One-Inch Thick Fiberglass Lined Round Duct Attenuation Per Foot of Length.

Diamete 63 125 250 500 1,000 2,000 4,000 8,000

r
Inches Hz Hz Hz Hz Hz Hz Hz Hz
6 0.38 0.59 0.93 1.53 2.17 2.31 2.04 1.26
8 0.32 0.54 0.89 1.50 2.19 2.17 1.83 1.18
10 0.27 0.50 0.85 1.48 2.20 2.04 1.64 1.12

Siemens Building Technologies, Inc. 57

Chapter 4–HVAC Sound Attenuation

Diamete 63 125 250 500 1,000 2,000 4,000 8,000

r
Inches Hz Hz Hz Hz Hz Hz Hz Hz
12 0.23 0.46 0.81 1.45 2.18 1.91 1.48 1.05
14 0.19 0.42 0.77 1.43 2.14 1.79 1.34 1.00
16 0.16 0.38 0.73 1.40 2.08 1.67 1.21 0.95
18 0.13 0.35 0.69 1.37 2.01 1.56 1.10 0.90
20 0.11 0.31 0.65 1.34 1.92 1.45 1.00 0.87
22 0.08 0.28 0.61 1.31 1.82 1.34 0.92 0.83
24 0.07 0.25 0.57 1.28 1.71 1.24 0.85 0.80
26 0.05 0.22 0.53 1.24 1.59 1.14 0.79 0.77
28 0.03 0.19 0.49 1.20 1.46 1.04 0.74 0.74
30 0.02 0.16 0.45 1.16 1.33 0.95 0.69 0.71
32 0.01 0.14 0.42 1.12 1.20 0.87 0.66 0.69
34 0 0.11 0.38 1.07 1.07 0.79 0.63 0.66
36 0 0.08 0.35 1.02 0.93 0.71 0.60 0.64
38 0 0.06 0.31 0.96 0.80 0.64 0.58 0.61
40 0 0.03 0.28 0.91 0.68 0.57 0.55 0.58
42 0 0.01 0.25 0.84 0.56 0.50 0.53 0.55
44 0 0 0.23 0.78 0.45 0.44 0.51 0.52
46 0 0 0.20 0.71 0.35 0.39 0.48 0.48
48 0 0 0.18 0.63 0.26 0.34 0.45 0.44
50 0 0 0.15 0.55 0.19 0.29 0.41 0.40
52 0 0 0.14 0.46 0.13 0.25 0.37 0.34
54 0 0 0.12 0.37 0.09 0.22 0.31 0.29
56 0 0 0.10 0.28 0.08 0.18 0.25 0.22
58 0 0 0.09 0.17 0.08 0.16 0.18 0.15
60 0 0 0.08 0.06 0.10 0.14 0.09 0.07

Duct Elbows
Duct elbows are quite effective at attenuating sound in the mid frequency levels. The most
effective attenuation is in rectangular elbows, which are lined elbows and do not have turning
vanes. The minimum attenuation occurs in unlined round elbows. More data has been
derived for rectangular elbows. Round elbows that are lined present a very complex scenario
and, for that reason, are not discussed here.

58 Siemens Building Technologies, Inc.

 Duct Attenuation

Table 21. Two-Inch Thick Fiberglass Lined Round Duct Attenuation Per Foot of Length.

Diamete 63 125 250 500 1,000 2,000 4,000 8,000

r
Inches Hz Hz Hz Hz Hz Hz Hz Hz
6 0.56 0.80 1.37 2.25 2.17 2.31 2.04 1.26
8 0.51 0.75 1.33 2.23 2.19 2.17 1.83 1.18
10 0.46 0.71 1.29 2.20 2.20 2.04 1.64 1.12
12 0.42 0.67 1.25 2.18 2.18 1.91 1.48 1.05
14 0.38 0.63 1.21 2.15 2.14 1.79 1.34 1.00
16 0.35 0.59 1.17 2.12 2.08 1.67 1.21 0.95
18 0.32 0.56 1.13 2.10 2.01 1.56 1.10 0.90
20 0.29 0.52 1.09 2.07 1.92 1.45 1.00 0.87
22 0.27 0.49 1.05 2.03 1.82 1.34 0.92 0.83
24 0.25 0.46 1.01 2.00 1.71 1.24 0.85 0.80
26 0.24 0.43 0.97 1.96 1.59 1.14 0.79 0.77
28 0.22 0.40 0.93 1.93 1.46 1.04 0.74 0.74
30 0.21 0.37 0.90 1.88 1.33 0.95 0.69 0.71
32 0.20 0.34 0.86 1.84 1.20 0.87 0.66 0.69
34 0.19 0.32 0.82 1.79 1.07 0.79 0.63 0.66
36 0.18 0.29 0.79 1.74 0.93 0.71 0.60 0.64
38 0.17 0.27 0.76 1.69 0.80 0.64 0.58 0.61
40 0.16 0.24 0.73 1.63 0.68 0.57 0.55 0.58
42 0.15 0.22 0.70 1.57 0.56 0.50 0.53 0.55
44 0.13 0.20 0.67 1.50 0.45 0.44 0.51 0.52
46 0.12 0.17 0.64 1.43 0.35 0.39 0.48 0.48
48 0.11 0.15 0.62 1.36 0.26 0.34 0.45 0.44
50 0.09 0.12 0.60 1.28 0.19 0.29 0.41 0.40
52 0.07 0.10 0.58 1.19 0.13 0.25 0.37 0.34
54 0.05 0.08 0.56 1.10 0.09 0.22 0.31 0.29
56 0.02 0.05 0.55 1.00 0.08 0.18 0.25 0.22
58 0 0.03 0.53 0.90 0.08 0.16 0.18 0.15
60 0 0 0.53 0.79 0.10 0.14 0.09 0.07

To approximate the attenuation provided by either a rectangular or round elbow, first

calculate the Frequency Width (FW) factor for the elbow at each octave band as follows:

FW = (Frequency x Width) ÷ 1,000

Siemens Building Technologies, Inc. 59

Chapter 4–HVAC Sound Attenuation

Frequency is in Hz Width is in inches.

When the FW values are determined, refer to the appropriate curve in Figure 15 to determine
the dB attenuation at each octave band.

Example Rectangular Duct Elbow Attenuation Calculation

An unlined duct is 36 inches wide by 28 inches high. Determine the attenuation at each
octave band provided by a square elbow that has turning vanes.

First calculate the FW factor for each octave band:

FW = (Frequency x Width) ÷ 1,000

FW = (Frequency x 36 inches) ÷ 1,000

FW = Frequency x 0.036

Table 22 lists the calculated FW factors and the corresponding attenuation dB values using
the “RECTANGULAR, SQUARE, UNLINED, WITH VANES” curve.

Table 22. Calculated FW Factors and Attenuation dB Values.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
FW 2.3 4.5 9.0 18.0 36.0 72.0 144 288
ATTN: 2 dB 5 dB 6 dB 3 dB 3 dB 2 dB 2 dB 2 dB

Since round ducts are more efficient in conveying air than the same area of rectangular duct,
they are also more efficient at conveying sound, and will provide less attenuation whether
they are lined or unlined.

In Figure 15, the curves that apply to lined elbows, are based upon at least a 1-inch thick
lining, and the lining extending at least one to two duct diameters ahead of and after the
elbow.

Duct Takeoffs and Divisions

When a takeoff occurs on a main duct or the main duct divides, the sound energy is also
divided between the resulting duct runs after the takeoff, so that neither duct at the point
where it leaves a junction has all of the sound power level energy entering the junction. The
following formula will estimate the resulting dB attenuation that occurs with reference to the
junction and a specific duct leaving the junction:

Attenuation = 10 Log [Branch Area ÷ Total Area]

Branch Area = Branch Duct Area (Square Inches)

Total Area = Total Duct Area Leaving Junction (Square Inches)

60 Siemens Building Technologies, Inc.

 Duct Takeoffs and Divisions

Once the term enclosed within the parenthesis is calculated, Figure 17 provides dB values.

For example, assume that a 36-inch x 18-inch rectangular duct main has an 8-inch round 90
degree takeoff that is followed by a 32-inch x 18-inch continuation of the duct main. What
attenuation would this junction provide for the 8-inch and the reduced sized main?

Attenuation for an 8-inch diameter duct:

= 10 Log [8 in. Duct Area ÷ Total Area Lv. Jct.]

= 10 Log [50.3 in.2 ÷ (50.3 in.2 32 in. x 18 in.)]

= 10 Log [50.3 in.2 ÷ (50.3 in.2 + 576 in.2)]

= 10 Log [50.3 in.2 ÷ 626.3 in.2]

= 10 Log [0.0803]

= - 11 dB at each octave band

Attenuation for a 32-inch x 18-inch rectangular duct:

= 10 Log [(32 in. x 18 in.) ÷ Total Area Lv. Jct.]

= 10 Log [(576 in.2) ÷ (626.3 in.2)]

= 10 Log [0.9197]

= - 0.4 dB at each octave band

Table 23. Takeoff/Junction Attenuation.

Branch Duct Area dB Branch Duct Area dB

Total Area Lv. Jct. Attenuated Total Area Lv. Jct. Attenuated
0.95 to 0.91 0.4 0.089 to 0.071 11
0.90 to 0.88 0.5 0.070 to 0.057 12
0.88 to 0.71 1 0.056 to 0.045 13
0.70 to 0.57 2 0.044 to 0.036 14
0.56 to 0.45 3 0.037 to 0.029 15
0.44 to 0.36 4 0.028 to 0.023 16
0.35 to 0.29 5 0.022 to 0.018 17
0.28 to 0.23 6 0.017 to 0.015 18
0.22 to 0.18 7 0.014 to 0.012 19
0.17 to 0.15 8 0.011 to 0.009 20
0.14 to 0.12 9 0.0089 to 0.0071 21
0.11 to 0.09 10 0.0070 to 0.0057 22

Siemens Building Technologies, Inc. 61

Chapter 4–HVAC Sound Attenuation

Duct Silencers
Conventional duct silencers consist of prefabricated arrangements of sound absorbing
material intended for insertion within a duct run to attenuate sound. They offer only limited
attenuation in the low frequency (125 Hz and below) octave bands, and moderate attenuation
at the high frequency bands. Their maximum attenuation is in the mid frequency (that is,
1,000 Hz) octave bands. Apart from the extra cost of silencers, they also require a certain
amount of physical space and create additional pressure drop.

If a duct silencer is added to a duct run, the silencer will provide a certain attenuation or
insertion loss of the sound power level generated upstream. However, like any duct fitting, a
silencer also generates some sound power of its own. Therefore, aside from applying the
attenuation provided by a silencer to the sound power level, also determine if the sound
power generated by the silencer will have an appreciable impact on the net sound power
level.

Consult the manufacturer’s data sheets for the attenuation and sound power level generation
data for a given silencer. Also be sure to follow the manufacturer’s instructions regarding the
proper installation and location for a silencer. For proper functionality, there should be a
certain minimum distance, equivalent to a number of duct diameters, between the discharge
of a fan and the silencer, and between the silencer and other duct elements (elbows, etc.).

End Reflection
When ducts terminate into a ceiling air diffuser that discharges air into a room, a significant
amount of low frequency sound energy is reflected back into the ductwork. This phenomenon
is referred to as end reflection. The effect of duct end reflection is estimated by the following
formula:

Attenuation = 10 Log [1 + (3453 ÷ F D)1.88]

D = Diameter (inches) for round ducts

D = (1.27 x Area in.2)½ for rectangular ducts

Table 24 lists end reflection attenuation values for round and square ducts based upon this
formula.

Table 24. End Reflection Attenuation for Typical Room Discharge Size Ducts.

Duct 63 125 250 500 1,000 2,000 4,000 8,000

Size Hz Hz Hz Hz Hz Hz Hz Hz
6 in. dia. 18.1 12.7 7.6 3.6 1.3 0.4 0 0
8 in. dia. 15.8 10.5 5.7 2.5 0.8 0.2 0 0
10 in. dia. 14.0 8.9 4.5 1.8 0.6 0.1 0 0
12 in. dia. 12.6 7.6 3.6 1.3 0.4 0 0 0

62 Siemens Building Technologies, Inc.

 Environment Adjustment Factor

14 in. dia. 11.5 6.6 3.0 1.0 0.3 0 0 0

16 in. dia. 10.5 5.8 2.5 0.8 0.2 0 0 0
18 in. dia. 9.6 5.1 2.1 0.7 0.2 0 0 0
20 in. dia 8.8 4.5 1.8 0.6 0.1 0 0 0
6 in. sq. 17.2 11.8 6.8 3.1 0.3 0.1 0 0
8 in. sq. 14.9 9.7 5.1 2.1 0.2 0 0 0
10 in. sq. 13.2 8.1 3.9 1.5 0.1 0 0 0
12 in. sq. 11.7 6.9 3.1 1.1 0 0 0 0
14 in. sq. 10.6 5.9 2.5 0.8 0 0 0 0
16 in. sq. 9.6 5.1 2.1 0.7 0 0 0 0
18 in. sq. 8.7 4.4 1.7 0.5 0 0 0 0
20 in. sq. 8.0 3.9 1.4 0.4 0 0 0 0

Environment Adjustment Factor

The Air Conditioning & Refrigeration Institute (ARI) has established ARI Standard 880-89 (Air
Terminals) for testing, rating and certifying HVAC air terminals and outlets. The device noise
rating is part of the data that the tests provide.

ARI Standard 885-90 (Procedure for Estimating Occupied Space Sound Levels in the
Application of Air Terminals and Outlets) states that an Environmental Adjustment Factor
should be subtracted from the sound power rating of terminals and outlets if the sound power
test data has been measured under free field (open space) conditions rather than
reverberant (room like) conditions.

When utilizing a manufacturer’s sound rating data, verify whether the data has been obtained
under a reverberant or free field test setup. If the data has been obtained under a free field
test condition, then subtract the Environmental Adjustment Factor from the free field sound
rating data:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
-7 dB -3 dB -2 dB -1 dB -1 dB -1 dB -1 dB -1 dB

Environmental adjustment factors for noise rating data obtained under Free Field test
conditions.

Space Effect
The last factor affecting attenuation of the sound power level to an acceptable room sound
pressure level is the ability of the room to absorb and attenuate the sound power.

Siemens Building Technologies, Inc. 63

Chapter 4–HVAC Sound Attenuation

A room’s ability to attenuate the sound to an acceptable level is dependent upon its size,
shape, ceiling height, acoustical properties, and other factors. The following equation
(referred to as the Schultz equation) provides the means to calculate the attenuation and the
resulting sound pressure level at a given location from a sound source.

RLp = 5 Log (V) + 10 Log (r) - 25 + 3 Log (f)

Where:

V = Room Volume.

R = Distance to the sound source (feet).

f = Octave band frequency Hz.

For example, assume that a laboratory room is 24 feet long by 12 feet wide by 10 feet high.
What attenuation would this room to provide at 10 feet from a supply air diffuser?

RLp = 5 Log (V) + 10 Log (r) - 25 + 3 Log (f)

= 5 Log (2,400) + 10 Log (10) - 25+ 3 Log (f)

= 16.9 + 10 - 25 + 3 Log (f)

= 1.9 + 3 Log (f)

Substituting each octave band frequency in the (f) term will yield the following room dB
attenuation values at each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
7.3 8.2 9.1 10.0 10.9 11.8 12.7 13.6

The attenuation values determined for the room are then subtracted from the sound power
level value at the sound source to get the sound pressure level at the point of concern.

When multiple sound sources are present (as is the case with several supply air diffusers or
fume hoods), separate calculations are done to arrive at the sound pressure level due to
each separate sound source. Then the sound pressure levels can be combined according to
Table 3 to determine the net resulting sound pressure level at the point of concern.

Radiated Sound Attenuation

HVAC components served, such as VAV and CAV air terminal units that are typically located
close to the area served along with chemical fume hood exhaust terminals and similar
devices, have both a radiated sound and a discharge sound rating. A discharge sound rating
applies to the sound power level generated by the device that may add to the existing sound
power level already within the ductwork. This sound energy ultimately enters the room
through the air diffusers (or exhaust grilles).

64 Siemens Building Technologies, Inc.

 Radiated Sound Attenuation

Radiated sound is essentially the breakout component of the sound generated by the device
and has the potential of entering the room primarily on a direct path. Most often this
component of sound energy must penetrate a dropped ceiling or other architectural features
before entering a room.

For air terminal units located above a dropped ceiling, the radiated sound component is quite
often (and ideally) attenuated by the dropped ceiling and the plenum effect of the space
above the ceiling. Table 25 gives the attenuation effect of the combination of a typical
dropped ceiling and the “plenum” space above it. Use these values to determine the net
sound power level reduction in the radiated sound power level of devices located a few feet
above a dropped ceiling.

Table 25. Attenuation of a Dropped Ceiling and Plenum on Radiated Sound Ceiling.

Ceiling 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
1/2-inch 4 dB 7 dB 8 dB 9 dB 10 dB 11 dB 14 dB 18 dB
Thick Tiles
Fiberglass
5/8-inch 5 dB 8 dB 10 dB 12 dB 13 dB 14 dB 16 dB 19 dB
Thick Tiles
Fiberglass

5/8-inch 10 dB 15 dB 22 dB 26 dB 30 dB 28 dB 30 dB 30 dB
Thick
Gypsum
Board

After the radiated sound attenuation has been deducted from the radiated sound power level
of a unit, the resulting sound power level value should be compared with the discharge sound
power level that will typically occur at the supply air diffuser(s) present in the room. Table 3 in
Chapter 2 indicates the resulting sound power level of two sound power levels.

Note that in some instances, a sound producing device may be within the occupied room
itself, or there may not be a dropped ceiling. In such cases, there is no attenuation as
indicated in Table 25, and the radiated sound power level must be directly compared with the
discharge sound power level in the room. In this type of situation, it is possible that the
combined sound power level is mainly due to the radiated sound pressure level.

Siemens Building Technologies, Inc. 65

Chapter 5–HVAC System Sound
Analysis
Chapter 5 provides examples of how to analyze the components of a specific HVAC system.
It includes the following topics:

• Introduction to HVAC system sound analysis

• Commentary on HVAC system sound

Introduction to HVAC System Sound Analysis

Earlier in this document, we discussed the sound generated by and attenuated by individual
HVAC components. This section provides examples of how to analyze the components of a
specific HVAC system to determine what sound level can be expected in a space served by
the HVAC system.

Figure 16 shows a portion of an HVAC system including Room 101, which is served by the
system. Although there are computer programs that can provide an analysis of a given
system configuration, it is advantageous to understand the process and be able to determine
the effect of a single element on a given system. The following example illustrates the
analysis of this system with respect to the sound level that can be expected to result in Room
101 due to the HVAC system.

Example HVAC System Sound Analysis

In Figure 16, the sound analysis for the HVAC system begins with the supply fan that is
typically the element that produces the highest sound level in the system.

With reference to the fan sound power level calculation procedure from Chapter 3, the fan’s
sound power level rating of 1 cfm and 1.00 in. WC, as shown in Table 5. These values are
part of the fan’s total sound power and are listed in the first horizontal row of Table 26.

Siemens Building Technologies, Inc. 67

Chapter 5–HVAC System Sound Analysis

Figure 16. HVAC System Layout for Sound Analysis Example.

Step 1. Actual Operating Conditions Increase

This calculation provides the sound pressure level increase for the fan at the actual operating
cfm and static pressure. Using the FLw formula, will yield the dB level increase that must be
added to each of the eight octave bands.

FLw = 10 Log Q + 20 Log P

= 10 Log 15,000 + 20 Log 3

= 10 (4.2) + 20 (0.477)

= 42 + 9.5

= 51.5

This value is in the second row of Table 26.

68 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

Step 2. Blade Frequency Increment (BFI)

This calculation uses the Hz formula to determine the frequency at which the blade sound
component occurs.

Hz = Fan rpm x (Number of Blades ÷ 60)

Hz = 1,225 x (10 ÷ 60)

Hz = 1,225 x 0.167

Hz = 204

Referring back to Table 6, the 204 Hz is within the 250 Hz octave band. Therefore, the BFI
value of 3 dB (from Table 5) is added as a contributing component of sound power level of
the 250 Hz octave band.

Step 3. Efficiency Correction

Based upon an operating fan efficiency between 85 and 90%, Table 7 gives a 3 dB value that
must be added to all octave bands.

Table 26. Fan Sound Level Summary.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
Table 5 45 45 43 39 34 28 24 19
Step 1 51.5 51.5 51.5 51.5 51.5 51.5 51.5 51.5
(FLw)
Step 2 — — 3.0 — — — — —
(BFI)
Step 3 3.0 3.0 3.0 3.0 3.0 3.0 3.0 3.0
(EFF)

Final 99.5 99.5 100.5 93.5 88.5 82.5 78.5 73.5

dB dB dB dB dB dB dB dB

The final values from Table 26 represent the expected supply fan generated noise level
(GNL). At the end of this analysis, these values are entered into the HVAC system - sound
analysis form, that is shown in Figure 17. This form provides a means to systematically
tabulate the individual element sound levels for an HVAC system or a portion of a system to
determine the resultant sound power level. A blank copy of this form (along with other forms)
is provided in the Appendix. You can use these forms when making sound level analysis of
actual systems.

Siemens Building Technologies, Inc. 69

Chapter 5–HVAC System Sound Analysis

Duct Section A
Airflow in the velocity range typically used in HVAC systems does not generate appreciable
sound within straight duct runs; therefore, no generated noise level (GNL) calculation
procedure was given in an earlier section. Rather, ducts may be assumed to only attenuate
the sound generated by other duct elements.

Following the procedure given on HVAC Sound Attenuation for Sheet Metal Ducts, first
calculate the P/A (Perimeter to Area ratio) for the HVAC system’s Duct Section A.

P = Duct perimeter in feet

= (36 in. + 48 in. +36 in. + 48 in.)

= (3 ft +4 ft +3 ft + 4 ft)

= 14 feet

A = Duct area in square feet

= (3 ft x 4 ft)

= 12 square feet

Therefore, the P/A ratio is 14 ÷ 12 = 1.16

With reference to Table 16, we can assume the following attenuation per foot of duct length:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
.165 .111 .074 .049 .033 .022 .015 .010
dB dB dB dB dB dB dB dB

Multiplying these values by 14 feet yields the following attenuation in each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
2.3 1.5 1.0 0.7 0.5 0.3 0.2 0.1
dB dB dB dB dB dB dB dB

These values are also entered into the HVAC system - sound analysis form shown in Figure
17.

Duct Elbow B
While airflow through an elbow generates sound, it also attenuates sound already generated
(in this case the fan sound). Following the procedure given for Elbow Airflow Noise in the
HVAC Sound Sources section, the basic formula is:

ELw = K + 10 Log F + 50 Log U + 10 Log S + 10 Log D + EC

70 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

Where:

ELw = the net elbow sound power level increase (or decrease if it is negative).

K = a factor that is dependent upon the elbow operating conditions.

F = the octave band center frequency in Hz.

U = the duct airflow velocity = 15,000/12 = 1,250 ft/min.

S = the duct cross sectional area in square feet = 12 sq ft as calculated previously.

For elbows with turning vanes:

D = the cord length of a vane in feet (9.6 in. = 0.8 ft).

EC = a constant.

For elbows with turning vanes:

EC = 10 Log n - 107 (n =number of vanes) = 10 Log 15 - 107 = -95.

The K factor is determined by first calculating the Strouhal for each of the eight octave bands.
Figure 12 gives K factor values for the Strouhal numbers.

St = 60FD ÷ U

St = 60 x 63 Hz x 0.8 ft ÷ 1,250 = 2.4,K = 48

St = 60 x 125 Hz x 0.8 ft ÷ 1,250 = 4.8,K = 50
St = 60 x 250 Hz x 0.8 ft ÷ 1,250 = 9.6,K = -55
St = 60 x 500 Hz x 0.8 ft ÷ 1,250 = 19.2,K = -62
St = 60 x 1,000 Hz x 0.8 ft ÷ 1,250 = 38.4,K = -72
St = 60 x 2,000 Hz x 0.8 ft ÷ 1,250 = 76.8,K = -85
St = 60 x 4,000 Hz x 0.8 ft ÷1,250 = 153.6,K = -102
St = 60 x 8,000 Hz x 0.8 ft ÷1,250 = 307.2,K = -123*

* Calculated by formula (K = - 47.5 - 7.69 (log St)2.5

Table 27 sums up the factors comprising the ELw values for each octave band.

Table 27. Duct Elbow B ELw Values.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
K -48 -50 -55 -62 -72 -85 -102 -123
10 Log F 18 21 24 27 30 33 36 39
50 Log U 155 155 155 155 155 155 155 155
10 Log S 11 11 11 11 11 11 11 11
10 Log D -1 -1 -1 -1 -1 -1 -1 -1

Siemens Building Technologies, Inc. 71

Chapter 5–HVAC System Sound Analysis

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
EC -95 -95 -95 -95 -95 -95 -95 -95
ELw 40 41 39 35 28 18 4 0
dB dB dB dB dB dB dB dB*

* Although this value adds up to -14 dB, a zero is entered as the net result since it is considered that a GNL
cannot be less than zero.

These values are now entered into the HVAC system - sound analysis form (Figure 17) as
Elbow B “GNL” entries.

Next, you must estimate the attenuation that the elbow will provide. To estimate the
attenuation provided by either a rectangular or round elbow, you must first calculate the
Frequency Width (FW) factor for the elbow at each octave band.

FW = (Frequency x Width) ÷ 1,000

Where:

Frequency = the octave band Hz.

Width = the nominal duct width in inches. (48 in. is used since the elbow is in the vertical
plane.)

FW = (Frequency x 48 in.) ÷ 1,000.

After the FW values are determined, refer to Figure 15 for the dB attenuation at each octave
band. (The rectangular, square, unlined with vanes curve applies.) The following chart lists
the calculated FW factors and the corresponding attenuation values from the
“RECTANGULAR, SQUARE, UNLINED, WITH VANES” curve.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
FW 3.0 6.0 12.0 24.0 48.0 96.0 192 384
ATTN: 3.1 dB 6.2 dB 7.1 dB 7.0 dB 6.9 dB 6.8 dB 6.8 dB 6.8 dB

These values are now entered into the HVAC system - sound analysis form (Figure 17) as
Elbow B “ATTENUATION” entries.

Duct Section C
Since this is the same size duct as Section A, the same attenuation per foot of duct length
applies:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
.165 .111 .074 .049 .033 .022 .015 .010

72 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

dB dB dB dB dB dB dB dB

Multiplying these values by the 6 feet yields the following attenuation in each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
1.0 0.7 0.4 0.3 0.2 0.1 0.1 0.1
dB dB dB dB dB dB dB dB

These values are entered into the Figure 17 form as Duct Section C, ATTENUATION.

Junction D
Following the procedure given for Junction and Takeoff Airflow Noise in Chapter 3, HVAC
Sound Sources, the basic formula is:

JLw = K + 10 Log F + 50 Log UB + 10 Log S +10 Log D – JC

Where:

Elw = the net sound power level increase (or decrease if it is negative).

K = a factor dependent upon design conditions.

F = the octave band center frequency in Hz.

UB = the branch duct airflow velocity (1,250 fpm).

S = the branch duct cross sectional area in square feet (16 in. x 36 in. ÷ 144) = 4.

D for Junctions = (4S/π)0.5 ft = (4x4/π)0.5 ft = 2.3.

Determining the K values requires first calculating M.

M = UM ÷ UB.

Since all ducts leaving the junction have the same airflow velocity, UM and UB = 1,250 FPM.
Therefore, M = 1.0.

Next, the Strouhal number is determined for each octave band center frequency. Then the
graph in Figure 13 will provide the K values.

Strouhal number (St) = 60 F D ÷ UB

Siemens Building Technologies, Inc. 73

Chapter 5–HVAC System Sound Analysis

St = 60 x 63 Hz x 2.3 ft. ÷ 1,250 = 7.0 K = - 27

St = 60 x 125 Hz x 2.3ft. ÷ 1,250 = 13.8, K = - 34
St = 60 x 250 Hz x 2.3ft. ÷ 1,250 = 27.6, K = - 42
St = 60 x 500 Hz x 2.3ft. ÷ 1,250 = 55.2, K = - 51
St = 60 x 1000 Hz x 2.3 ft. ÷ 1,250 = 110.4, K = - 60
St = 60 x 2000 Hz x 2.3 ft. ÷ 1,250 = 220.8, K = - 72
St = 60 x 4000 Hz x 2.3 ft. ÷ 1,250 = 441.6, K = - 90*
St = 60 x 8000 Hz x 2.3 ft. ÷ 1,250 = 883.2, K = - 150*

*Estimated Values

JC is a constant that depends upon the configuration of the junction and relates to the
junction’s ability to attenuate sound.

JC = -107 + Δr + ΔT

Where:

Δr = Junction Radius/Branch Duct Diameter

= 16 in. ÷ 16 in.

= 1.0.

ΔT = a turbulence factor that applies since there is another duct element (elbow B) within
five duct diameters upstream. However since M = 0.0, Table 10 indicates that ΔT will be
0.0.

Therefore, JC =-107 + 1.0 + 0.0 = -106

The following chart summarizes the terms and results of the JLw formula to determine the
GNL of this junction at each octave band.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
K -27 -34 -42 -51 -60 -72 -90 -150
10 Log F 18 21 24 27 30 33 36 39
50 log U 155 155 155 155 155 155 155 155
10 Log S 6 6 6 6 6 6 6 6
10 Log D 3.6 3.6 3.6 3.6 3.6 3.6 3.6 3.6
JC -106 -106 -106 -106 -106 -106 -106 -106
JLw 49.60 45.6 40.6 34.6 28.6 19.6 4.6 0
dB dB dB dB dB dB dB dB

These values are now entered into the HVAC system - sound analysis form (Figure 17) as
Junction D GNL entries.

74 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

Next, we need to come up with the attenuation that Junction D will provide. When a takeoff
occurs on a main duct, the available sound energy must divide between the resulting duct
runs after the takeoff. As a result, neither duct at the point where it leaves a junction can have
all of the sound power level energy that was available at that point. The following formula
estimates the resulting dB attenuation occurring at a junction:

Attenuation = 10 Log [Branch Area ÷ Total Area]

Branch Area = Branch Duct Area (square inch)

Total Area = Total Duct Area Leaving Junction (square inch)

Attenuation = 10 Log [8 sq ft ÷ 12 sq ft] = 10 Log [0.67] = 1.8

The 1.8 dB attenuation is entered into the HVAC system - sound analysis form (Figure 17) for
Junction D.

Duct Section E
Duct Perimeter = (32 in. + 36 in. + 32 in. + 36 in.) ÷ 12 = 11.33

Duct Area = (32 in. x 36 in.) ÷ 144

=8

P/A = 11.33 ÷ 8

= 1.42

With reference to Table 16, the following attenuation applies per foot of duct length:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
.192 .128 .086 .057 .038 .026 .017 .011
dB dB dB dB dB dB dB dB

Multiplying these values by the 16 feet yields the following attenuation in each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
3.1 2.0 1.4 0.9 0.6 0.4 0.3 0.2
dB dB dB dB dB dB dB dB

These values are entered into the HVAC system - sound analysis form shown in Figure 17.

Siemens Building Technologies, Inc. 75

Chapter 5–HVAC System Sound Analysis

Junction F
Since the takeoff at Junction F has the same physical dimensions and airflow velocity as the
takeoff at Junction D, calculating the GNL is identical to that previously done for Junction D,
and the same GNL values apply. Therefore, the same values are entered into the HVAC
system - sound analysis form (Figure 17) as Junction F GNL entries.

Note that there is a difference in determining the attenuation values for Junction F since the
ratio between the branch duct and the main duct are different at junction F than they were for
Junction D.

Attenuation = 10 Log [Branch Duct ÷ Total Duct Area Leaving Junction]

= 10 Log [4 sq ft ÷ 8 sq ft]

= 10 Log [0.50] = -3.0

3.0 dB attenuation is entered into the HVAC system - sound analysis form (Figure 17) for
Junction F.

Duct Section G
Duct Perimeter = (16 in. + 36 in. + 16 in. + 36 in.) ÷ 12

= 8.67

Duct Area = (16 in. x 36 in.)/144

=4

P/A = 8.67/4

= 2.17

With reference to Table 16, the following attenuation applies per foot of duct length:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
.260 .175 .117 .079 .052 .035 .023 .016
dB dB dB dB dB dB dB dB

Multiplying these values by the 14 feet yields the following attenuation in each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
3.6 2.5 1.6 1.1 0.7 0.5 0.3 0.2
dB dB dB dB dB dB dB dB

76 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

Duct Takeoff/Junction H
Following the procedure given for Junction and Takeoff Airflow Noise in Chapter 3, HVAC
Sound Sources, the basic formula is:

JLw = K + 10 Log F + 50 Log UB + 10 Log S + 10 Log D – JC

Where:

Elw = the net sound power level increase (or decrease if it is negative).

K = a factor dependent upon design conditions.

F = the octave band center frequency in Hz.

UB = the branch duct airflow velocity.

12 in. Diameter = 0.79 sq ft

1,200 cfm ÷ 0.79 = 1,528 fpm

S = the branch duct cross sectional area in 0.79 square feet

D for Takeoffs = Branch duct height in feet = 1.0 ft.

Determining the K values requires first calculating M.

M = UM ÷ UB (UM = 5,000 ÷ 4 = 1,250 fpm)

M = 1,250 ÷ 1,528

= 0.8

Next, the Strouhal number is determined for each octave band center frequency. Then the
graph in Figure 13 provides the K values.

Strouhal number (St) = 60 F D ÷ UB

St = 60 x 63 Hz x 1.0 ft ÷ 1,528 = 2.5, K = -21

St = 60 x 125 Hz x 1.0 ft ÷ 1,528 = 4.9, K = -28
St = 60 x 250 Hz x 1.0 ft ÷ 1,528 = 9.8, K = -35
St = 60 x 500 Hz x 1.0 ft ÷ 1,528 = 19.6, K = -44
St = 60x1000 Hz x 1.0 ft ÷ 1,528 = 39.3, K = -53
St = 60x2000 Hz x 1.0 ft ÷ 1,528 = 78.5, K = -63
St = 60x4000 Hz x 1.0 ft ÷ 1,528 = 157.1, K = -74
St =60 x 8000 Hz x 1.0 ft÷1,528=314.1, K = -95*

*Estimated Value

JC is a constant that depends upon the configuration of the junction and relates to the
junction’s ability to attenuate sound.

Siemens Building Technologies, Inc. 77

Chapter 5–HVAC System Sound Analysis

JC = -107 + Δr + ΔT

Δr = Junction Radius/Branch Duct Diameter

Since there is no radius at the junction:

Δr = 0.0 and r ÷ DBR = 0.0

ΔT is a turbulence factor that doesn’t apply since there is no element within five duct
diameters upstream of Junction H.

Therefore:

JC = -107 + 0.0 + 0.0

= -107

The following chart summarizes the terms and results of the JLw formula to determine the
GNL of Junction H at each octave band.

Next, the Strouhal number is determined for each octave band center frequency. Then the
graph in Figure 13 will provide the K values.

Strouhal number (St) = 60 F D ÷ UB

St = 60 x 63 Hz x 1.0 ft ÷ 1,528 = 2.5, K = -21

St = 60 x 125 Hz x 1.0 ft ÷ 1,528 = 4.9, K = -28
St =60 x 250 Hz x 1.0 ft ÷ 1,528 = 9.8, K = -35
St =60 x 500 Hz x 1.0 ft ÷ 1,528 = 19.6, K = -44
St =60 x 1,000 Hz x 1.0 ft ÷ 1,528 = 39.3, K = -53
St =60 x 2,000 Hz x 1.0 ft ÷ 1,528 = 78.5, K = -63
St =60 x 4,000 Hz x 1.0 ft ÷ 1,528 = 157.1, K = -74
St =60 x 8,000 Hz x 1.0 ft ÷ 1,528 = 314.1, K = -95*

*Estimated Value

JC is a constant that depends upon the configuration of the junction and relates to the
junction’s ability to attenuate sound.

JC = - 107 + Δr + ΔT

Δr = Junction Radius/Branch Duct Diameter

Since there is no radius at the junction,

Δr = 0.0 and r ÷ DBR = 0.0

ΔT is a turbulence factor that doesn’t apply since there is no element within five duct
diameters upstream of Junction H.

78 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

Therefore:

JC = -107 + 0.0 + 0.0

= -107

The following chart summarizes the terms and results of the JLw formula to determine the
GNL of Junction H at each octave band.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
K -21 -28 -35 -44 -53 -63 -74 -95
10 Log F 18 21 24 27 30 33 36 39
50 Log U 159 159 159 159 159 159 159 159
10 Log S -1 -1 -1 -1 -1 -1 -1 -1
10 Log D 0 0 0 0 0 0 0 0
JC -107 -107 -107 -107 -107 -107 -107 -107
JLw 48.0 44.0 40.0 34.0 28.0 21.0 13.0 0
dB dB dB dB dB dB dB dB

These values are entered into the HVAC system - sound analysis form (Figure 17) as
Junction H GNL entries.

Whenever a takeoff occurs on a main duct, the sound energy is also divided between the
resulting duct runs after the takeoff, so that neither duct at the point where it leaves a junction
or takeoff has all of the sound power level energy. The following formula approximates the
resulting dB attenuation occurring at a junction:

Attenuation = 10 Log [Branch Duct ÷ Total Duct Area Leaving Junction]

= 10 Log [0.79 sq ft ÷ 3.9 sq ft]

= 10 Log [0.2] = 6.9

The 6.9 dB attenuation is entered into the HVAC system - sound analysis form (Figure 17) for
Junction H.

Siemens Building Technologies, Inc. 79

Chapter 5–HVAC System Sound Analysis

Duct Section I
Duct Diameter = 12 inches

With reference to Table 16, the following attenuation applies per foot of duct length:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
0.03 0.03 0.03 0.05 0.07 0.07 0.07 0.08
dB dB dB dB dB dB dB dB

Multiplying these values by the 8 feet yields the following attenuation in each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
0.2 0.2 0.2 0.4 0.6 0.6 0.6 0.6
dB dB dB dB dB dB dB dB

Duct Elbow J
Following the procedure for determining Elbow Airflow Noise, the basic formula is:

ELw = K + 10 Log F + 50 Log U + 10 Log S + 10 Log D + EC

Where:

ELw = the net elbow sound power level increase (or decrease if it is negative).

K = a factor that is dependent upon the elbow-operating conditions.

F = the octave band center frequency in Hz.

U = the airflow velocity = 1,528 fpm (from H).

S = the duct cross sectional area in square feet = 0.79 sq ft as calculated previously.

For elbows without turning vanes, D is the height of the elbow in feet (12 in. = 1 ft).

EC is a constant. For elbows without turning vanes EC = -107.

The K factor is determined by first calculating the Strouhal for each of the eight octave bands.
Figure 12 gives K factor values for the Strouhal numbers.

St = 60 F D ÷ U

St = 60 x 63 Hz x 1.0 ft ÷ 1,528 = 2.5, K = -21

St = 60 x 125 Hz x 1.0 ft ÷ 1,528 = 4.9, K = -28
St = 60 x 250 Hz x 1.0 ft ÷ 1,528 =9.8, K = -35
St = 60 x 500 Hz x 1.0 ft ÷ 1,528 = 19.6, K = -44

80 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

St = 60 x 1000 Hz x 1.0 ft ÷ 1,528 = 39.3, K = -53

St = 60 x 2,000 Hz x 1.0 ft ÷ 1,528 = 78.5, K = -63
St = 60 x 4,000 Hz x 1.0 ft ÷ 1,528 = 157.1, K = -74
St = 60 x 8,000 Hz x 1.0 ft ÷ 1,528 = 314.1, K = -95*

*Estimated Value

The chart below sums up the factors comprising the ELw values for each octave band.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
K -21 -28 -35 -44 -53 -63 -74 -95
10 Log F 18 21 24 27 30 33 36 39
50 Log U 159 159 159 159 159 159 159 159
10 Log S -1 -1 -1 -1 -1 -1 -1 -1
10 Log D 0 0 0 0 0 0 0 0
JC -107 -107 -107 -107 -107 -107 -107 -107
JLw 48.0 44.0 40.0 34.0 28.0 21.0 13.0 0
dB dB dB dB dB dB dB dB

These values are now entered into the HVAC system - sound analysis form (Figure 17) as
Elbow J GNL entries.

To approximate the attenuation provided by either a rectangular or round elbow, it is

necessary to first calculate the Frequency Width (FW) factor for the elbow at each octave
band.

FW = (Frequency x Width) ÷ 1,000

Frequency is the octave band Hz. Width is the duct width (12 in.)

FW = (Frequency x 12) ÷ 1,000

When the FW values are determined, Figure 16 gives the dB attenuation at each octave
band from the ROUND ELBOWS curve.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
FW 0.76 1.5 3.0 6.0 12.0 24.0 48.0 96.0
ATTN: 0.2 dB 0.5 dB 1.8 dB 2.8 dB 3.3 dB 3.3 dB 3.0 dB 2.7 dB

These values are entered into the HVAC system - sound analysis form (Figure 17) as Elbow
J ATTENUATION entries.

Siemens Building Technologies, Inc. 81

Chapter 5–HVAC System Sound Analysis

Reheat Terminal
The reheat terminal GNL values are obtained from the manufacturer and are restated in the
chart below. (Note also that in this example, the short duct section after elbow J and before
the Reheat Terminal is considered to have a negligible effect on the discharge sound and is
therefore not evaluated.)

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
Terminal 73 70 66 62 56 54 49 44
GNL* dB dB dB dB dB dB dB dB

* Discharge Noise Data Based Upon Reverberant Test.

The COMBINATION sound pressure level values that result due to the reheat terminal GNL
discharge sound and the preceding sound pressure level are derived with reference to Table
3, which covers adding sound pressure levels.

Duct Sections L
Two flexible 12-inch diameter ducts with 1-inch thick lining. Since the outlet from the reheat
terminal is divided between two 12-inch ducts, the sound power in at the beginning of each is
divided between the two duct runs by the following junction attenuation formula:

Attenuation = 10 Log [Branch Area ÷ Total Area]

= 10 Log [0.79 sq ft ÷ 1.58 sq ft]

= 10 Log [0.5] = 3.0

The 3.0 dB attenuation is entered into the HVAC system - sound analysis form (Figure 17) as
Duct Division L.

With reference to Table 17, the following attenuation applies per foot of 12-inch diameter, 1-
inch lined duct:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
0.23 0.46 0.81 1.45 2.18 1.91 1.48 1.05
dB dB dB dB dB dB dB dB

Multiplying these values by the 10 feet yields the following attenuation in each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
2.3 4.6 8.1 14.5 21.8 19.1 14.8 10.5
dB dB dB dB dB dB dB dB

These values are now entered into the HVAC system - sound analysis form (Figure 17) as
Duct Section L ATTENUATION.

82 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

Perforated Diffuser
The perforated diffuser GNL NC rating is listed as NC 35 by the manufacturer and are
expanded in the chart below based on the values from Table 12 and are then entered into
Figure 17.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
DIFFUSER 60 53 46 40 36 34 33 32
NC 35 dB dB dB dB dB dB dB dB
(GNL)

End Reflection
When ducts terminate into a ceiling air diffuser, a significant amount of low frequency sound
energy is reflected back as end reflection.

From Table 24, the end reflection attenuation values for 12-inch diameter round duct is
shown in the following chart.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
12-inch 12.6 7.6 3.6 1.3 0.4 0 0 0
Diameter dB dB dB dB dB dB dB dB

These values are now entered into the HVAC system - sound analysis form (Figure 17) as
End Reflection ATTENUATION.

Space Effect
The last factor regarding attenuation of the sound power level is the ability of the room to
absorb and attenuate the sound power.

The following equation (referred to as the Schultz equation) provides the means to calculate
the resultant room sound pressure level (Rlp) at a given octave band level five feet above the
floor, from the sound power level present at the diffuser(s).

RLp = 27.6 Log (H) + 5 Log (A) + 3 Log (f) -1.3 Log (N) - 30

Where:

H = ceiling height in feet = 10 ft.

A = (floor area ft2÷ number of diffusers) ÷ H2.

= (12 ft x 24 ft ÷ 2) ÷ 100.

= (144) ÷ 100

Siemens Building Technologies, Inc. 83

Chapter 5–HVAC System Sound Analysis

= 1.44

f = octave band frequency Hz.

N = number of ceiling diffusers = 2.

RLp = 27.6 Log (10) + 5 Log (1.4) + 3 Log (Hz) -1.3 Log (2) – 30

= 27.6 +0.7+ 3 Log (Hz) -0.4 – 30

= 3 log (Hz) -2.1

Substituting each octave band frequency in the Hz term will yield the following room dB
attenuation values at each octave band:

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
3.3 4.2 5.1 6.0 6.9 7.8 8.7 9.6
dB dB dB dB dB dB dB dB

The attenuation values determined for the room are then subtracted from the sound power
level values that are present at the diffusers to yield the sound pressure level five feet above
the floor in the vicinity of the respective diffuser.

The last row of entries on the second part of Figure 17 show the dB sound pressure level
values calculated for Room 101. These are reproduced in the chart below and plotted on the
RC curve of Figure 18.

63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
54.8 55.1 54.9 44.8 34.0 31.9 31.0 29.9
dB dB dB dB dB dB dB dB

84 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

HVAC system -sound analysis Sheet 1 of 2

Analysis By: Noise Associates Inc.
Date: 8/15/95
System/Room Identity: Supply System to Room 101
Notes:

Figure 17. Example of an HVAC System Sound Analysis – Page 1 of 2.

Siemens Building Technologies, Inc. 85

Chapter 5–HVAC System Sound Analysis

HVAC system -sound analysis Sheet 2 of 2

Analysis By: Noise Associates Inc.
Date: 8/15/95
System/Room Identity: Supply System to Room 101
Notes:

Figure 17. Example of an HVAC System Sound Analysis – Page 2 of 2.

86 Siemens Building Technologies, Inc.

 Introduction to HVAC System Sound Analysis

Figure 18. Room 101 Sound Pressure Level Curve.

As Figure 18 illustrates, the sound pressure level in the room, due to the supply system, is
acceptable for a laboratory (NC 45), and except for the relatively moderate amount of lower
frequency sound in the 125 to 500 Hz range, would even be acceptable in an office.

With reference to the sound analysis data in Figure 17, we can see that the sound power
level in the 125 to 500 Hz range continues to go down until it reaches the 74 to 69 dB level
just prior to the reheat terminal. At this point, the combined sound power with that of the
reheat terminal results in a slight increase to about 76 to 70 dB.

Note that if the sound power level in this Hz range were brought down to no more than 60 dB
by using 1-inch lined ductwork throughout the system, the resultant sound power level at that
point would remain at the reheat terminal GNL level of 70 to 62 dB. This reduction of 6 to 8
dB would result in the room curve lying more closely to the NC 30 and NC 35 curves and also
significantly reduce the lower frequency rumble.

Siemens Building Technologies, Inc. 87

Chapter 5–HVAC System Sound Analysis

Commentary on HVAC System Sound

Based upon the preceding example of a sound analysis for an HVAC supply system to a
room, we can form these general observations regarding HVAC system sound sources and
attenuation:

• The major source of sound in a system is typically the supply fan. In this example, it
has a peak sound power level of 100.5 dB at 250 Hz.

• Duct fittings such as rectangular elbows, and takeoffs (such as, B & H), provide a
major attenuating effect on sound power.

• Unlined sheet metal duct, especially the larger sizes (such as, A, C, & E), offers very
limited sound attenuation.

• Although duct fittings generate sound, their GNL effect typically does not adversely
impact the sound power level at moderate airflow rates (up to 2,000 fpm).

• Unless a fan is relatively close to a room, the major source of sound in rooms with
higher air change rates (such as laboratories with fume hoods open) will be caused
by the terminal unit and/or the air diffusers.

• Lined ducts offer considerable sound attenuation. (Compare the unlined duct section
L to lined duct section L.)

Laboratory Room Sound Analysis

Laboratory rooms with high air change rates and high chemical fume hood exhaust rates are
particularly prone to higher ambient sound levels. The sound generated by HVAC
components such as supply and exhaust terminal units that must be located in close
proximity to the room, and supply air diffusers is mainly dependent on the airflow velocity
through these units. The higher the airflow velocity, the higher the sound power level
generated by the units.

Because it is necessary for higher ventilation airflow rates in rooms such as chemical
laboratories, it is not typically feasible to maintain a ventilation ambient sound level on a par
with an office environment (RC or NC Criteria of 30 to 35). Depending upon the purpose of
the laboratory, somewhat higher sound levels are normally acceptable. A two person
laboratory intended for occupancy by individuals who normally concentrate on their separate
research projects may be acceptable with a 45 to 50 Noise Criteria sound level. On the other
hand, academic laboratories where the instructor must almost continuously communicate and
be heard by the students should preferably not exceed 40 dB.

As discussed earlier, HVAC generated sound that has sufficiently high energy, will not only
become part of the sound within the duct system (discharge sound), but will also be audible
outside of the device that is generating the sound. This externally audible sound is radiated
sound and is also referred to as breakout sound. Breakout sound can have an adverse
impact on the overall HVAC system sound in a space particularly if there is no drop ceiling
between the room space and the HVAC components, or if the acoustical absorption
capability of the ceiling is limited.

88 Siemens Building Technologies, Inc.

 Commentary on HVAC System Sound

Laboratory Room Ambient Sound

Aside from the sound caused by the supply system serving a laboratory room, the major
source of a chemical laboratory’s HVAC ambient sound is often due to the exhaust side of
the ventilation system particularly chemical fume hood exhaust. As a result, a laboratory
room sound analysis should also address the exhaust side of the ventilation system. The
principles that apply to HVAC supply side produced sound (airflow’s through duct fittings,
terminal units, etc.) also apply to the exhaust side and are calculated in the same manner.

Figure 19 shows a portion of a chemical laboratory room with potential HVAC sound sources
identified. Note that because of the probable high airflow rates, the radiated sound potential
of the SUPPLY VAV TERMINAL and the EXHAUST TERMINALS should be part of the
room’s sound analysis.

Fume Hood Sound

Since a fume hood will typically exhaust at least 100 cfm for each square foot of sash
opening, it follows that the large exhaust airflow’s that leave by means of the fume hood will
be a very significant potential source of sound in a laboratory.

Figure 19. Chemical Laboratory - Sources of Ventilation System Sound.

To make a proper system analysis, NC factors that apply to a particular fume hood operating
at specific exhaust airflow’s would need to be provided by the manufacturer. However, since
fume hood manufacturers do not normally test their product for sound, a precise means of
analyzing the resulting sound at varying sash openings does not (as of this time) exist.

It is recommended that particular attention be given to the sound level ratings of VAV fume
hood exhaust terminals, especially the radiated component, since, in many cases, these
components are located in the laboratory room directly atop the fume hood.

Siemens Building Technologies, Inc. 89

Chapter 5–HVAC System Sound Analysis

When a fume hood sash is open, exhaust terminal discharge sound will emanate from the
fume hood opening and be the major source of ventilation sound heard by the fume hood
user. In this case, the discharge sound level rating provided by the exhaust terminal
manufacturer (at various airflows) should be reviewed for acceptability.

For a central laboratory exhaust system, the exhaust fan sound power level will typically be
substantially attenuated due to the length of the connecting ductwork and the numerous
junctions and takeoffs on the central exhaust system. Therefore, except for laboratory units
located close to the exhaust fans themselves, the sound power level of the exhaust fans can
be largely disregarded since they will likely be attenuated to a value less than that of the
fume hood exhaust terminals.

Since fume hood exhaust ducts cannot utilize an internal sound absorbing lining (such as,
fiberglass), there is typically only a minimal amount of attenuation possible for exhaust
terminals that are located within laboratory rooms or very close to the fume hoods. The only
alternative (which is not always possible) is to locate fume hood exhaust terminals above the
ceiling, and at a significant distance downstream from the fume hood.

Terminal Radiated Sound - Example Analysis

A 240 square foot laboratory room with two fume hoods will have an acoustical tile type of a
dropped ceiling 10 feet above the floor. It is decided to locate the two 8-inch diameter fume
hood exhaust terminals, that control the VAV fume hood face velocity, in a horizontal position
about 3 feet above the laboratory room ceiling and approximately 10 feet apart.

Figure 20. Example of Radiated Sound.

With reference to Figure 20, determine if the radiated sound pressure level and also the
discharge sound pressure level this arrangement is likely to produce for someone who
happens to be standing in the laboratory room midway between the two fume hoods. Assume
each fume hood sash is about 50% open. Also, consider how this design arrangement will
impact a desired room Noise Criteria of NC-40.

The following tables show the manufacturer’s exhaust terminal sound ratings at 50% of
maximum airflow (450 cfm).

90 Siemens Building Technologies, Inc.

 Commentary on HVAC System Sound

Exhaust Terminal Reverberant Radiated Sound Power Level at 50% Maximum Airflow

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
47 dB 41 dB 36 dB 32 dB 28 dB 23 dB

Exhaust Terminal Reverberant Discharge Sound Power Level at 50% Maximum Airflow

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
61 dB 56 dB 53 dB 49 dB 45 dB 39 dB

NOTE: Exhaust terminal manufacturer’s sound ratings typically cover only the 125 to 4,000 Hz octave bands.

See Environmental Adjustment Factor in Chapter 4 to determine if the manufacturer’s sound

rating needs to be adjusted based upon the type of sound rating test (Free Field or
Reverberant).

Radiated Sound
First, we’ll determine the effect of the radiated sound pressure level of the two exhaust
terminals. With reference to the text covering Radiated Sound Attenuation, Table 25 gives
expected attenuation of typical dropped ceiling material on radiated sound that originates
above the ceiling.

Table 28. Attenuation of a Dropped Ceiling and Plenum on Radiated Sound (repeated).

Ceiling 63 125 250 500 1,000 2,000 4,000 8,000

Hz Hz Hz Hz Hz Hz Hz Hz
1/2-inch 4 dB 7 dB 8 dB 9 dB 10 dB 11 dB 14 dB 18 dB
Thick Tiles
Fiberglass
5/8-inch 5 dB 8 dB 10 dB 12 dB 13 dB 14 dB 16 dB 19 dB
Thick Tiles
Fiberglass

5/8-inch 10 dB 15 dB 22 dB 26 dB 30 dB 28 dB 30 dB 30 dB
Thick
Gypsum
Board

Subtracting the Table 28 values for 5/8 inch-thick fiberglass tiles from the manufacturer’s
rating data for the 125 Hz through 4,000 Hz bands, leaves the following radiated sound level
at the ceiling just below each terminal:

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
39 dB 31 dB 24 dB 19 dB 14 dB 7 dB

Siemens Building Technologies, Inc. 91

Chapter 5–HVAC System Sound Analysis

To determine the attenuation and subsequently the sound power level a person would
experience standing mid way between the two terminals (about a 7 foot distance from the
person’s ears to each point in the ceiling just below the terminals), the Space Effect (Schultz)
equation is used:

RLp = 5 Log (V) + 10 Log (r) - 25 + 3 Log (f)

Where:

RLp = Attenuation due to distance and room size.

V = Room Volume (240 sq ft x 10 ft).

R = Distance from the sound source (7 ft).

F = Octave band frequency Hz.

RLp = 5 Log (2,400) + 10 Log (7) - 25 + 3 Log (f)

= 16.9 + 8.5 - 25 + 3 Log (f)

= 0.4 + 3 Log (f)

Substituting each octave band frequency in the equation’s (f) term will yield the following
room dB attenuation values at each octave band:

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
6 dB 7 dB 8 dB 9 dB 10 dB 10 dB

Subtracting these values from the radiated sound level at the ceiling yields the following room
sound pressure level at 5 feet above the floor from each terminal:

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
33 dB 24 dB 16 dB 10 dB 4 dB 0 dB

Since each of the two exhaust terminals generates a sound pressure level of equal value that
converge at the same point, add 3 dB to the individual sound pressure level value (per Table
3), to determine the resulting sound pressure level at that point. The result is listed in the
following table:

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
36 dB 28 dB 21 dB 13 dB 7 dB 0 dB

The values determined for the radiated sound pressure level are quite low and will not
adversely affect the laboratory noise criteria level of 40 dB. Remember that these values are
for the radiated sound pressure level component; there is also the discharge sound power
component that must be addressed.

92 Siemens Building Technologies, Inc.

 Commentary on HVAC System Sound

Discharge Sound
Since the terminals are 8 inches in diameter and will be horizontal, we can assume that there
will be one 90 degree radius type of elbow and some modest length of unlined round duct
between each of the exhaust terminals and the fume hood.

With respect to an 8 inch diameter 90 degree radius type of elbow, there will be some small
amount of sound power generated as 450 cfm of air (1,290 fpm) passes through it, and there
will also be some attenuation. Using the procedures in Chapter 3 and Chapter 4, we would
find that the sound power level generated by the elbow is too low to impact the exhaust
terminal sound power. However, the elbow would still attenuate a few decibels at the upper
frequencies and would result in the following discharge sound power level:

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
61 dB 55 dB 51 dB 46 dB 42 dB 36 dB

Since a round unlined duct provides very little attenuation, we can assume that the above
discharge sound power level will exist at the fume hood connection.

The primary attenuation of the fume hood exhaust air terminal will occur at the junction with
the fume hood due to end reflection, and also the baffle arrangement inside the fume hood.
As previously stated, there is very little information on the properties of fume hoods with
respect to sound power generation or attenuation, so we will base our final values on
attenuation by means of end reflection alone.

Table 24 lists end reflection attenuation values. Subtracting the appropriate end reflection
values for an 8-inch round duct from the preceding chart results in the following final
discharge sound power level values:

Fume Hood Sound Discharge Sound at 50% Maximum Airflow

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
50 dB 49 dB 48 dB 45 dB 42 dB 36 dB

We can expect this sound power level to exist at the open sash of each fume hood. With
respect to the Noise Criterion curves (Figure 5), these dB values would not exceed those of
the NC-40 curve. With respect to the person standing some distance away, the resulting
sound pressure level would further decrease in accordance with the Space Effect equation.

Note that this analysis assumed that the fume hoods were 50% open. The sound power level
would increase as fume hood sashes are opened further. But the 50% open value is a
reasonable level to use for general ambient noise analysis, so the conclusion would be that
the room noise level would be very acceptable for a chemical laboratory.

Siemens Building Technologies, Inc. 93

Chapter 5–HVAC System Sound Analysis

Terminal Radiated Sound -Example Analysis 2

If it is decided to locate the two exhaust terminals, from the previous example, directly atop
the fume hoods (approximately seven feet above the floor in the laboratory), you should ask
yourself the following questions:

• What sound pressure level would a person then experience when standing at the
front of a fume hood with the sash 50% open?

• Would the new sound pressure level experienced by a person at the fume hood be
noticeably different from the previous example?

The difference in this arrangement is that the radiated sound power level will not be
attenuated by the drop ceiling, and the discharge sound power level will not be attenuated by
a duct elbow.

As a result, the person standing at the fume hood, which would probably be within only 3 feet
or so from the exhaust terminal, would be exposed to the following full radiated sound level:

Exhaust Terminal Radiated Sound Power Level at 50% Maximum Airflow

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
47 dB 41 dB 36 dB 32 dB 28 dB 23 dB

In addition, the sound pressure level at the fume hood sash would only be attenuated by the
fume hood’s End Reflection and not by the exhaust duct elbow, so it would be:

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
50 dB 50 dB 50 dB 48 dB 45 dB 39 dB

Since these two sources of sound are physically very close together, we use the procedure of
Table 3 to come up with the combined results that become a bit higher at the lower frequency
as listed in the following chart:

Combined Fume Hood Exhaust Terminal Discharge and Radiated Sound

at 50% Maximum Airflow

125 250 500 1,000 2,000 4,000

Hz Hz Hz Hz Hz Hz
52 dB 51 dB 50 dB 48 dB 45 dB 39 dB

This is the sound pressure level that a user at the fume hood would experience while the
fume hood sash is at about 50% open.

Recall that a 3 dB increase in sound pressure level will be noticeable. Since all of the decibel
values listed above are about 2 to 3 dB higher than in the previous example (where
everything was located above the ceiling), the resulting sound pressure level would be
noticeable to the user.

94 Siemens Building Technologies, Inc.

Chapter 6–Minimizing HVAC Sound
Chapter 6 offers general guidance as a convenient summation on minimizing excessive or
objectionable HVAC sound. It includes the following topics:

• Introduction to minimizing HVAC sound

• Basic system design criteria

• Fans

• Duct configurations

• Terminal equipment

• Sound attenuation devices

• Sound measurement instrumentation

Introduction to Minimizing HVAC Sound

In the previous sections, procedures were established for calculating the sound power levels
generated, and the attenuation effect for nearly all of the elements that comprise an HVAC
system. Based upon the formulas, tables, charts, and graphs in the preceding sections, the
following guidelines are offered as a convenient summary on minimizing excessive or
objectionable HVAC sound.

Following these guidelines will help attain an HVAC system design with the lowest practical
ambient sound level. But remember, the objective is not a “nearly silent” system, but one that
will provide a desirable background sound level that is conducive to the type of occupancy or
activity to be performed in the conditioned space.

See Table 29 for suggestions regarding reducing annoying sound in existing ventilation
systems.

Basic System Design Criteria

• Mechanical rooms should never be directly above or below noise sensitive rooms.

• Mechanical equipment (fans, pumps, etc.) should be installed utilizing vibration

absorbing concrete pads, adequate vibration isolating equipment, and ample spacing
from walls ceilings, and structural members.

• In general, HVAC systems utilizing efficient air delivery designs will generate and
transmit less sound than system configurations with higher pressure drops, higher
airflow velocities, and lower operating efficiencies.

Siemens Building Technologies, Inc. 95

Chapter 6–Minimizing HVAC Sound

• Anticipate potential noise problems when difficult situations or configurations leave

no choice but to depart from good design practice.

• Consider utilizing additional sound isolating equipment enclosures and additional

sound absorbing elements.

• Rooms without a dropped ceiling and exposed ductwork are especially prone to
radiated sound from terminal units, and other duct elements. Consider lowering
airflow velocity by using larger sized duct and acoustical covering over terminal units
and sharp elbows to reduce sound breakout.

• Require ductwork to be well sealed. Much sound is due to air leaking out of supply
ducts and also leakage into exhaust ducts. In addition, reducing air leakage will
reduce the amount of the overall required airflow in the system, which increases
system efficiency and lessens the sound.

• Room size and furnishings have a large effect on the final ambient sound level.
Larger rooms absorb and dissipate sound better than smaller rooms. It is difficult to
achieve a low sound in smaller rooms since there is less sound energy absorption by
the furnishings and less overall Space Effect (ambient sound reduction due to the
volumetric effect).

• Low frequency noise is the hardest to attenuate and typically the most annoying.
Effective low frequency sound attenuation normally requires duct linings of at least
two inches thick.

Fans
• Backward inclined centrifugal fans are recommended for the lowest noise generation.

• Axial fans also have a higher BFI (blade frequency increment) component that
typically contributes the most to annoying fan sound.

• Always select the most efficient fan for the purpose and strive for operation near the
top of the fan curve.

• Variable speed drives are also a contributing source of noise in equipment rooms
and consequently to overall fan system sound. Of the types of variable speed drives
available, the current source is typically the quietest, while a voltage source is
typically the noisiest.

• Avoid VAV fan operation at the point where variable inlet vanes must be nearly
closed. Inlet vanes tend to generate much low frequency sound as the vanes near
the closed position.

• Use a fan configuration that minimizes the system effect as much as possible.
Airflow into and out of the fan should avoid swirling and turbulence.

• Plenums can help attenuate considerable fan sound. This is a particularly effective
means of lowering the sound of return or exhaust systems. For supply systems,
plenums with one or two inches of sound absorbing material are especially effective
at attenuating fan sound.

96 Siemens Building Technologies, Inc.

 Duct Configurations

• When designing an HVAC system, never try to specify the resulting sound pressure
level of a HVAC system component. Instead, specify the required sound pressure
level (room NC or RC) to be attained in the space(s) served and where in the room
the measurement should be taken.

• When considering a specific fan for a particular application, always obtain the
manufacturer’s certified sound rating at the standard rating condition of 1 cfm and
1.00 in. WC.

Duct Configurations
• Size the duct system so airflow velocity does not need to exceed 1,200 fpm to meet
relatively low room sound pressure level requirements (such as, NC 35). Velocities
up to 1800 can be used for areas where NC 45 is acceptable.

• Avoid sharp duct bends, transition pieces with steep angles, sharp edge takeoffs and
junctions, and in general any element that causes air turbulence and higher pressure
drops. Use gradual angular transitions (15 degree maximum), radius type elbows
and takeoffs with turning vanes, and in general whatever type of configuration will
provide a more streamlined airflow in the entire HVAC system.

• Round sheet metal ducts are the most efficient at conveying airflow, but offer less
sound attenuation per given length than equivalent sized rectangular ducts.

• Ducts with internal linings of 1- or 2-inch thickness offer significant sound attenuation,
especially for higher frequency sound.

• Avoid direct duct runs between noisy rooms (duplicating machines, operating
equipment, etc.) and areas requiring low sound levels, to prevent room noise from
being transmitted by the ductwork (duct borne crosstalk).

• Wrapping the exterior of ducts with insulation will reduce the radiated sound, but will
not reduce the discharge sound level. It may sometimes even increase the discharge
sound since it prevents duct borne sound energy from radiating out from the
ductwork (in much the same way as insulation retards thermal losses).

Terminal Equipment
• Moderate airflow velocities are the key to minimizing the sound generated by air
terminal units. Consider going to a larger size air terminal to reduce the airflow
velocity and provide a more acceptable level of radiated and discharge sound in the
space served.

• Terminals serving individual office areas should be located in corridors rather than
above the office served to lessen the effect of the terminal produced discharge and
radiated sound.

• Whenever possible, terminals serving larger general offices should be located above
areas that will be less sensitive to HVAC sound, such as over copy machines,
printers, supplies storage, etc.

Siemens Building Technologies, Inc. 97

Chapter 6–Minimizing HVAC Sound

• Terminals should be mounted as high above the room’s ceiling as practical. At least
three feet above a dropped ceiling should be maintained whenever possible.

• Lined metal duct between a supply terminal and the air diffusers will best attenuate
the air terminal unit sound and allow the least sound breakout. Purposely locating
terminals to use longer runs to diffusers will enable more attenuation of terminal
sound.

• Minimize the use of flex duct as it is likely to generate sound at bends and wherever
sagging or compression occurs.

• Air diffuser sound is also directly dependent upon the type of diffuser and the airflow
velocity. For instance, a perforated diffuser has a higher sound level than a louver
type. Use an ample number of diffusers in a space to maintain a lower airflow velocity
and thus a lower sound level. Also, if a supply diffuser is used in a return/exhaust
application, airflow should be reduced to maintain the desired sound level.

• Use dampers to enable system balancing as far upstream as possible from the
spaces served by the system. Throttling dampers at diffusers should only be used for
small volume adjustments not requiring more than 0.10 in. WC pressure drop.

Table 29. Existing Ventilation System - Noise Troubleshooting and Potential Remedy.

Noise Problem Potential Remedy

Excessive Fan Sound • Ensure vibration isolation mounts are present and
properly installed so the fan floats on its mounts.
This is one of the most common noise problems and
many times is due to a multitude of factors. Often there is • Determine the fan’s operating point on the fan curve.
a poor inlet and outlet duct arrangement resulting in Determine if the system static pressure drop can be
excessive air turbulence and pronounced system effect. reduced and thus enable the fan operation to be
There can also be higher pressure drops in the duct moved to a more efficient point (higher) on the fan
system than originally estimated. Sometimes there will be curve.
a need for more airflow than the design intended. This all
leads to attempting to handle these conditions by • Determine if the system airflow and fan RPM can be
“speeding up the fan” and results in a significant increase reduced. Seal up all sources of system air leakage.
in the generated sound. • Improve fan inlet and outlet airflow if the present
arrangement causes excessive turbulence and
pressure loss.
• Install a sound attenuator in the fan outlet duct (see
Noise Attenuation Devices)
Excessive Air Supply Terminal Discharge Sound • Install lined flexible duct between the terminal and the
diffusers.
This is usually caused by a relatively high airflow through
the terminal and insufficient attenuation after the • Install more diffusers with longer lined flexible duct
terminal. after the terminal to deliver the same air change rate
for the space. (This lessens the airflow per diffuser
and increases the attenuation by utilizing more
terminal discharge paths.)
• Consider installing a larger air supply terminal.

98 Siemens Building Technologies, Inc.

 Terminal Equipment

Noise Problem Potential Remedy

Excessive Room Supply Air Diffuser Sound • Ensure that the flexible duct at the connection to the
diffusers is in relatively good vertical alignment. The
This is usually a higher pitched whistling sound caused
connecting duct should be relatively straight for 1 to 2
by a high pressure drop across air diffusers. (Ensure the
duct diameters immediately prior to the diffuser collar.
problem is the diffuser by removing it and noting if a
substantial reduction in the sound level occurs.) • Where multiple diffusers serve a space, ensure that
the airflow volume is equally divided among the
diffusers.
• Fully open the diffuser throttling damper. Add any
necessary throttling damper(s) farther upstream,
nearer to the air terminal unit.
• Replace the diffusers with larger ones or those with a
significantly lower sound rating.
Excessive Room Exhaust Air Sound • When an exhaust plenum is used above the room
ceiling, determine if the sound is actually due to a
This may be due to a sound source located close to the
piece of equipment (such as, terminal unit) located
exhaust grille in a ceiling plenum arrangement, or to the
close to the exhaust grille. If so, relocate the exhaust
exhaust system itself. Exhaust system noise is usually
grille away from the noise source.
higher pitched sound often caused by a high pressure
drop across the air grille. (Ensure the problem is the grille • In ducted exhaust systems, ensure the duct
by removing it and noting if a substantial reduction in the centerline is relatively straight for 1 to 2 duct
sound level occurs.) diameters immediately prior to the grille assembly
collar.
• Replace the grille assembly with a larger one or
multiple grilles if the problem is due to high airflow
through the grille.
• Lower the static pressure in the exhaust system to
prevent excessive pressure drops.
• Seal the exhaust system to minimize any excess
airflow that is caused by leaks into the system.
Noise Transmission From Another Area • Use lined ductwork between the noisy area and the
affected areas.
• Locate the duct termination point in the source room
to a location less susceptible to picking up the sound.
(Avoid locations directly above or near sound
producing equipment.)
• If possible, add a lined S curve in the duct connecting
to the noisy area.
• Lower the sound at the source by enclosing the
equipment within a sound absorbing enclosure.
• Reduce the overall sound level in the source room by
using wall and ceiling sound absorption (acoustical)
linings.

Siemens Building Technologies, Inc. 99

Chapter 6–Minimizing HVAC Sound

Sound Attenuation Devices

It is often necessary to add sound attenuation devices when it is determined that a sound
level cannot be adequately attenuated by the duct system itself or the HVAC system
configuration. Noise attenuation devices are divided into two categories; active and passive.
Active attenuation involves using state of the art technology to generate opposing waves of
sound energy that are intended to cancel the offensive sound wave energy and thus
significantly reduce or eliminate the resulting offensive sound. Passive attenuation is based
upon the absorption of sound energy (and its conversion into heat) by acoustical material,
typically applied as liners, baffles, or insulators.

Passive Sound Attenuation Devices

Passive sound attenuation is the most prevalent method of sound attenuation since it
involves relatively low cost materials and common design and installation methods.

Linings
Internal duct lining and plenum lining is perhaps the most effective means of attenuating
sound. Note that the lining must be on the inside of the duct and must be at least one-inch
thick to be effective. Crosstalk, which occurs when room sound travels to another room by
means of a common duct connection, can mostly be eliminated if the interconnecting duct is
lined.

Note that it is important that duct lining not be allowed to become wet. Otherwise, it must
invariably be removed and replaced since once it becomes moist, it is virtually impossible to
sanitize it against the bacteria growth that inevitably occurs.

Although fiberglass has been the most prevalent lining composition, alternate materials such
as fibrous metal is also available. Fibrous metal lining differs from fiberglass in that it is not
applied as a soft thick material, but rather it is compressed into a thin sheet. Further, fibrous
metal can be tuned to closely match the frequency needing attenuation by altering its
permeability, texture, thickness and the space between the duct wall and the material.

Duct Silencers and Attenuators

These devices, also known as sound traps, are passive sound attenuation components
designed for inserting into duct systems. Configurations are available for standard sized
round and rectangular ducts.

They offer only limited attenuation in the low frequency (125 Hz and below) octave bands,
and moderate attenuation at the high frequency bands. Their maximum attenuation is in the
mid frequency (1,000 Hz) octave bands. Apart from the cost of silencers, they require a
certain amount of physical space and will create an additional duct system pressure drop.

Be sure to follow the manufacturer’s instructions regarding the proper installation and location
for a duct attenuator or silencer. Typically, there should be a certain number of duct
diameters (1.5 in. to 2 in.) between the discharge of a fan and the silencer as well as
between the silencer and other duct elements (elbows, etc.) for proper functionality.

100 Siemens Building Technologies, Inc.

 Sound Attenuation Devices

A sound trap is typically larger in external dimensions than the duct it is intended to be used
with, since it must incorporate internal sound absorbing material and baffles. Therefore, it can
require an angular duct transition piece at the inlet and outlet of the silencer. The transition
element should not utilize angles greater than 15 degrees when connecting to fan outlet
ducts.

Duct silencers come in many configurations and lengths and their pressure drops vary.
Typically, the smaller the silencer, the greater will be the resulting pressure drop.

Ceiling and Wall Absorbers

In some applications, such as mechanical equipment rooms and large plenum areas above
ceilings, ambient sound may be attenuated by adding absorbers to the wall and ceiling (if
sufficient open wall space exists) or hanging sound absorbers from the ceiling. When applied
to hard wall areas (such as, masonry walls) they may provide up to 10 dB of attenuation to
the ambient sound pressure level in the room.

Enclosures
When it is not practical to reduce the sound level of certain equipment (pumps, compressors,
etc.), the next best approach is to consider building a sound isolating enclosure around the
sound source or adding an acoustical type of barrier between the sound source and the area
affected. Normally, conventional drywall construction on studs with insulating material
between the two inner surfaces of the drywall, will handle the majority of equipment noise
problems. In some of the more severe instances, it may be necessary to use two layers of
drywall on each side of the studs.

Aside from the obvious access considerations when enclosing equipment in this manner, the
enclosure will have maximum effectiveness if ample space exists between the inner wall of
the enclosure and the equipment. This obviously results in a larger enclosure, but enables
the enclosure to be more effective at absorbing the sound energy.

Active Sound Attenuation Devices

Active sound attenuation is a more recent technical innovation for attenuating sound. One of
the more common methods consists of sensing the sound profile and generating opposing
sound waves that are nearly equal to that of the original sound source, but are 180 degrees
out of phase with the source.

This results in a cancellation type of effect that can dramatically decrease or almost eliminate
the offensive sound that location. Figure 21 shows an active sound attenuation application.

With reference to Figure 21, note that the sound source can be anything, although
attenuating fan sound, particularly because of the low frequency components, is a most
common HVAC application.

Siemens Building Technologies, Inc. 101

Chapter 6–Minimizing HVAC Sound

The principle of the attenuation system is rather straightforward. The source input
microphone is located in the area where the offensive sound is present. The control unit
creates a near mirror image output signal of the sound sensed by the input microphone, but
180 degrees out of phase with the sound waves of the offensive sound. The control unit
amplifies the output signal to nearly match the strength of the offensive sound, and the output
speaker produces a sound that cancels much of the audible source sound. An error input
microphone picks up the remaining audible sound and enables the control unit to further
modify the output signal as may be required, to more nearly cancel the offensive sound at the
location where it is desired to have the maximum attenuation.

Figure 21. Active Sound Attenuation System.

The advantage of this type of approach is that no significant change in the existing HVAC
ductwork is normally needed, and the resulting sound is almost free of the most annoying
components such as the low frequency tones.

Active sound attenuation systems are best at attenuating the difficult low frequency sounds
from 250 Hz and lower, but are limited at higher frequencies. Thus, if the sound is generally
loud over the entire octave range, full range (broadband) attenuation may need to include
some passive attenuators to reduce the higher frequency sounds (500 Hz and up).

Using active sound attenuation is advantageous to fan sound on large existing HVAC
systems where it may not be possible or practical to change fans or make major changes to
the existing ductwork. However, if air terminal unit sound is the major problem, it may be
more cost effective to replace the terminals or add passive attenuation at those specific
locations.

102 Siemens Building Technologies, Inc.

 Sound Measurement Instrumentation

Sound Measurement Instrumentation

Whenever the sound level becomes an issue, it must be measured and quantified before it
can be resolved. Handheld battery operated sound level meters are available. These typically
have a tube-shaped microphone that protrudes out of the top of the unit and a meter or
display of the sound level at the selected octave band. Meters vary as to features with the
more sophisticated ones offering a digital display of sound dB levels. Some models even
provide a display in graphical format showing each separate octave band’s dB value as
separate bars or integrated into a dB curve.

A sound level meter should offer separate dB readings (octave band filters) for each of the
eight8 octave bands (63, 125, 250, 500, 1,000, 2,000, 4,000, and 8,000 Hz). It is even more
desirable if the meter can also provide a 31 Hz octave band reading.

Sound Measurement Procedure

Holding the meter in one’s hand and walking around a room is OK for initial general
examination to determine if there is a specific noise problem. However, when making actual
measurements for recording data, and when making before and after type measurements,
the meter should be affixed to a tripod and fixed so the microphone is at the 5 foot (average
ear height) level. Follow the specific manufacturer’s instruction manual to position the meter
with respect to the primary source of the sound. This may require the microphone tube to
form a certain angle (such as, 45 degrees) with the direct line to the sound source.

Record the dB level at all octave bands when taking data. (Use the Sound Measurement
form provided in the Appendix of this document to record sound level data.).

Use masking tape of other suitable means to mark or record the exact location of the meter
and tripod if they must be removed from the room, so that measurements taken at another
time will be consistent with the original measurement conditions.

Siemens Building Technologies, Inc. 103

Appendix
This Appendix contains blank copies of certain graphs and forms that have appeared in this
document. They are intended to be copied and used for sound measurement and analysis. It
includes the following topics:

• NC and RC Curves, Tabular Listing

• NC Curve

• RC Curve

• Sound Analysis Worksheet

• Sound Measurement Worksheet

NC and RC Curves, Tabular Listing

NC Curves vs. Sound Pressure Level Decibels Tabular Listing

NC 63 125 250 500 1,000 2,000 4,000 8,000

CURVE Hz Hz Hz Hz Hz Hz Hz Hz
15 47 36 29 22 17 14 12 11
20 51 40 33 26 22 20 17 16
25 54 45 38 31 27 24 22 21
30 57 48 42 35 31 30 28 27
35 60 53 46 40 36 34 33 32
40 64 57 51 45 41 39 38 37
45 67 60 54 49 46 44 43 42
50 71 64 59 54 51 49 48 47
55 74 67 62 58 56 54 53 52
60 77 71 67 63 61 59 58 57
65 80 75 71 68 66 64 63 62

Siemens Building Technologies, Inc. 105

Appendix

RC Curves vs. Sound Pressure Level Decibels Tabular Listing

RC 31.5 63 125 250 500 1,000 2,000 4,000

CURVE Hz Hz Hz Hz Hz Hz Hz Hz
25 – 45 40 35 30 25 20 15
30 55 50 45 40 35 30 25 20
35 60 55 50 45 40 35 30 25
40 65 60 55 50 45 40 35 30
45 70 65 60 55 50 45 40 35
50 75 70 65 60 55 50 45 40

NC Curve

Noise Criterion (NC) Curve

106 Siemens Building Technologies, Inc.

 RC Curve

RC Curve

Room Criterion (RC) Curve

Siemens Building Technologies, Inc. 107

Appendix

Sound Analysis Worksheet

HVAC system -sound analysis Sheet of
Analysis By:
Date:
System/Room Identity:
Notes: ______________________________________________________________

Example HVAC System Sound Analysis Worksheet.

108 Siemens Building Technologies, Inc.

 Sound Measurement Worksheet

Sound Measurement Worksheet

Example HVAC System Sound Measurement Worksheet.

Siemens Building Technologies, Inc. 109

Appendix

110 Siemens Building Technologies, Inc.

Glossary
This glossary describes various terms and acronyms used in this application guide. For a
comprehensive listing of building control terminology, see the Technical Glossary of Building
Controls Terminology and Acronyms (125-2185).

A-Weighted Sound Level

A single number in dB that represents the effect of all frequencies of a given sound on the
human ear. Since human hearing is less sensitive to very low and very high frequencies, the
contribution of these is reduced in terms of the resulting sound level to provide an A-weighted
value. Sound level meters typically also provide an A-Weighing sound measurement feature.

This approach is typically used to measure ambient sound levels in conjunction with
compliance with allowable noise exposure limit regulations. However, using the A-Weighted
approach is not recommended for analyzing lower sound levels for the purpose of achieving
a proper balanced room ambient sound level.

Also see C-Weighted Sound Level.

anechoic termination
A device used in acoustical laboratories at the end of a test duct in conjunction with
determining certified sound power ratings for HVAC equipment. The anechoic termination
prevents excessive end reflection of the sound waves back into the test duct where they
would interfere with the waves generated by the piece of equipment being tested. Also see
End Reflection.

acoustics
The science of sound, its measurement and its control.

acoustical louver
A specific type of louver used in air transfer openings between rooms to reduce sound
transmission through the louver.

airborne sound
Sound waves that travel through the atmosphere as opposed to sound that travel through
elements of a building structure (such as, pipes, beams, walls, floors, etc.).

Siemens Building Technologies, Inc. 111

Glossary

attenuate, attenuation
To decrease the sound power level resulting in a lower sound pressure level at the point of
concern. Also see insertion loss.

background sound
Sound from sources other than the one being measured or analyzed. In a room being
analyzed for HVAC sound, background sound may typically include cooling fans on electronic
equipment, copy machines, conversation, and even outside sources such as traffic, etc.)

broadband sound
Sound that is composed of frequencies covering much of the entire audible range as
opposed to tonal sounds that are composed of a narrow frequency range.

C-Weighted Sound Level

A method of ambient sound measurement that provides uniform weighing of frequencies
between 70 and 4,000 Hz, but reduces the effect of frequencies above and below this range.
C-Weighted measurements are much closer to true sound pressure levels than A-Weighted
sound level measurements.

decibel (dB)
When used in conjunction with acoustics, it’s an expression of the relative strength or
intensity of the sound power level or sound pressure level. The unit is based upon the
logarithmic scale so that a 50 dB sound power level actually represents an energy level that
is 100 times greater than that of a 30 dB sound power level.

dosimeter
An instrument for registering the occurrence and cumulative duration of sound that exceeds a
predetermined level at a specific location. A Dosimeter is most typically used when analyzing
an area with regard to ensuring compliance with allowable noise exposure limit regulations.

end reflection
The return of sound power energy back into a duct system when the duct ends abruptly or
undergoes an abrupt change in area. When end reflection occurs, the sound pressure level is
reduced in the area that the duct serves. Thus, when a duct terminates at a room diffuser,
end reflection will cause a reduction of the sound power that would otherwise enter the room.

fan sound power

The sound power that a fan radiates into a standard test duct. Fans are tested for and
certified as to their sound power level by the Air Movement and Control Association (AMCA).

112 Siemens Building Technologies, Inc.

 Sound Measurement Worksheet

frequency
When applied to sound, it’s the number of complete pressure wave fluctuations per second.
The measurement unit is cycles per second called Hertz (Hz). Human hearing allows us the
hear sound within the range of about 20 Hz to 20,000 Hz.

insertion loss
The reduction in sound power level due to the physical ability of something to absorb or
dissipate sound power. Also see attenuation.

octave band
A limited range of sound frequencies for the purpose of acoustical measurement and
analysis. To reduce the amount of individual frequencies that need to be measured when
analyzing a sound source, the 20,HZ to 20,000,Hz frequency range of human hearing is
divided up into what’s referred to as standard octave bands. These bands are used by
virtually everyone in the industry for consistency in specifying sound power levels and
measuring sound pressure levels. The standard bands are identified by their mid frequency:
16, 31.5, 63, 125, 250, 500, 2,000, 4,000, and 8,000 Hz. The bands have been selected so
that each band covers the frequency range where its lowest frequency is one half of its
uppermost frequency. Thus, the 125 Hz band covers from 84 to 167 Hz, while the 250 Hz
octave band covers from 167 to 334 Hz.

one third octave bands

For very precise laboratory sound measurement and analysis, the standard octave bands are
each further divided into three narrower bands. (This is used by the AMCA sound
measurement laboratory when determining certified fan sound power level ratings.)

sone
A linear unit of loudness as experienced by the human ear at a frequency of 1,000 Hz.

sound masking
Sometimes also called white noise. It is the intentional addition of a background sound that
tends to provide just enough broadband intensity to cover background sounds. In office
environments, a well designed HVAC system provides sufficient sound masking due to its
sound. The background sound used for sound masking should ideally not be noticeable
when present, and must not be at an objectionable level (too loud) or have identifiable tones.

sound power level

The power (dB) rating of a source of sound energy that is an indication of its potential for
loudness.

Siemens Building Technologies, Inc. 113

Glossary

sound pressure level

The pressure (dB) on a unit area caused by a sound source some distance away. Sound
pressure is therefore only the effect of a portion of the total sound power output of a sound
source, much like the force of the air on a given area of tissue paper is only a small portion of
the total output power of a fan. Although both sound power and sound pressure are
expressed in dB, they are not referring to the same effect since decibels have no dimension
and only represent ratios within the parameter in that they are applied.

room effect
The reduction in the effect of the sound power level emitted to a room. Room effect has the
potential for lowering sound pressure level measured at a given point as the total size of the
room increases, and with the room’s ability to absorb sound due to furnishings, wall
coverings, sound treatment, etc.

114 Siemens Building Technologies, Inc.

Index
Example Elbow Sound Power Level Calculation,
A
36
Active Sound Attenuation Devices, 103 Example Fan Sound Power Level Calculation,
Air Delivery Device Sound, 43 28
A-Weighted Sound Level, 16 Example HVAC System Sound Analysis, 67
Example Plenum Attenuation Calculation, 49
B Example RC Analysis, 21
Background Sound, 2 Example Rectangular Duct Attenuation
Basic System Design Criteria, 97 Calculation, 52
Blade Frequency Increment, 24 Example Rectangular Duct Elbow Attenuation
Calculation, 61
C
F
Calculate Damper Blockage Factor BF, 31
Calculate Pressure Loss Coefficient C, 31 Fan Aerodynamic Sound, 24
Calculate the Velocity Factor U, 31 Fan Efficiency, 25
Ceiling and Wall Absorbers, 103 Fan Sound Components, 24
Commentary on HVAC System Sound, 88 Fan Sound Power Level Calculation, 26
Computer Program Sound Analysis, 3 Fan Sound Power Level Data, 25
Fans, 98
D Flexible Duct Connection to Diffusers, 44
Damper Airflow Noise, 30 Fume Hood Sound, 89
Decibels, 9 G
Determining an RC Rating, 20
Discharge Sound, 93 Getting Help, III
Discharge Sound and Radiated Sound, 44 H
Duct Attenuation, 51
Duct Configurations, 99 HVAC Sound Transmission, 2
Duct Elbow B, 71
I
Duct Elbow J, 80
Duct Elbows, 59 Introduction to HVAC Sound Attenuation, 47
Duct Section A, 70 Introduction to HVAC System Sound Analysis,
Duct Section C, 73 67
Duct Section E, 75 Introduction to Minimizing HVAC Sound, 97
Duct Section G, 76
J
Duct Section I, 80
Duct Sections L, 82 JC Factor
Duct Silencers, 63 Junction and Takeoff Airflow Noise, 39
Duct Silencers and Attenuators, 102 Junction and Takeoff Airflow Noise, 38
Duct Takeoff/Junction H, 77 Junction D, 73
Duct Takeoffs and Divisions, 61 Junction F, 76
E K
Elbow Airflow Noise, 34 K Factor
Enclosures, 103 Damper Airflow Noise, 32
End Reflection, 63, 83 Elbo Airflow Noise, 35
Environment Adjustment Factor, 64 Junction and Takeoff Airflow Noise, 38
Example Damper Sound Power Level
Calculation, 33 L
Laboratory Applicability, 2

Siemens Building Technologies, Inc. 115

Index

Laboratory Elements, 45 Scope of This Guide, 1

Laboratory Room Ambient Sound, 89 Send Comments, III
Laboratory Room Sound Analysis, 88 Sound Analysis Worksheet, 110
Linings, 102 Sound Attenuation Devices, 102
Sound Breakout and Break-in, 45
N
Sound Measurement Instrumentation, 105
NC and RC Curves, Tabular Listing, 107 Sound Measurement Parameters, 8
NC Curve, 108 Sound Measurement Procedure, 105
NC Curves, 17 Sound Measurement Worksheet, 111
Sound Power Level, 8
O Sound Pressure Level, 11
Octave Bands, 13 Sound Wave Parameters, 6
Organization of Guide, I Sound Wave Propagation, 5
Sources of Sound in HAVC Systems, 23
P Space Effect, 65, 83
Passive Sound Attenuation Devices, 102 Step 1. Actual Operating Conditions Increase,
Perforated Diffuser, 83 26, 68
Plenums, 48 Step 1. Measure Existing Sound Pressure, 20
Purpose of this Guide, I Step 2. Blade Frequency Increment (BFI), 27,
29, 69
R Step 2. Mark Average Sound Pressure, 21
Radiated Sound, 91 Step 3. Efficiency Correction, 27, 29, 69
Radiated Sound Attenuation, 66 Step 3. Plot Curve of Octave Band, 21
RC Curve, 109 Symbols, III
RC Curves, 18 T
Rectangular Acoustically Lined Sheet Metal
Ducts, 54 Terminal Equipment, 99
Rectangular Unlined Sheet Metal Ducts, 51 Terminal Radiated Sound -Example Analysis, 90
Rectangular Unlined, Externally Insulated, Sheet Terminal Radiated Sound -Example Analysis 2,
Metal Ducts, 53 94
Reference Materials, II U
Reheat Terminal, 82
Round Acoustically Lined Sheet Metal Ducts, 58 U (Velocity Factor)
Round Unlined Sheet Metal Ducts, 58 Damper Airflow Noise, 31
S

116 Siemens Building Technologies, Inc.

Siemens Building Technologies Inc.
1000 Deerfield Parkway
Buffalo Grove, IL. 60089-4513
1-847-215-1000

125-1929
Copyright © 2009 Siemens Building Technologies, Inc.
Country of Origin: US

www.sbt.siemens.com