C. L.

MEADOR Exhaust Vent Silencer Design

NSF P r e d o c t o r a l Fellow,
Department o f M e c h a n i c a l Engineering,
Noise control problems exist where high pressure, high temperature gasses are vented to
Massachusetts Institute o f Technology,
C a m b r i d g e , Mass. the atmosphere. An example is used to illustrate an approach to the design of an
Assoc. M e m , ASME acoustically treated expansion chamber for a typical noise problem caused, by super-
heated steam vented to the atmosphere. An analysis is developed to aid in the design of
the acoustically treated expansion chamber type muffler for broadband frequency ap-
J. J. ALLAN I I I
plication. The algorithm as well as the supporting theory are sufficiently general so as
Assistant Professor,
Department o f Mechanical Engineering,
to be applicable to a number of high velocity gas-generated noise conditions. An Ap-
University o f Texas at Austin, pendix of terminology definitions is included.
Austin, Texas.
M e m . ASME

Introduction The ear is a sense organ specialized to receive and to respond to

airborne acoustic vibratory energy over a frequency range of
Ac
icorjSTiCAL noise may be defined as any unwanted about nine octaves and a dynamic range of more than 120 d b .
sound. The general noise control problem is complicated by the The ear's output consists of nerve impulses carried by the auditory
need to consider both airborne sound and solidborne vibration, nerve to the temporal lobes of the cerebral cortex. These im-
since each may create or be created by the other [ l ] . 1 Noise pulses convey auditory information to the brain. The series of
control, in its broadest sense, is the prevention or lessening of processes by which acoustic energy is transformed into nerve
noise before it is generated. Alternately, noise reduction is the impulses are in general very complicated. However, a complete
attenuation of noise after it has been produced. The obvious understanding of ear physiology is not essential to comprehending
control approach for an atmospheric steam vent is to reduce the the effects of damaging noise. For purposes of noise control, it
flow of steam. Since this is impractical or impossible in most in- will suffice to know that the ear is composed of three distinct re-
dustrial situations, the designer must use the noise reduction gions for sound transmission: the outer ear, the middle ear,
approach. Here, the pertinent design questions become: and the inner ear. The outer ear conducts acoustic energy to the
middle ear and the middle ear serves as an impedance matching
1 What is the definition of the existing noise problem in device to deliver the energy efficiently to the inner ear. Pressure
terms of emitted amplitudes and frequency spectra? waves in the inner ear are eventually converted to the nerve im-
2 What is an acceptable level of noise? pulses mentioned earlier. A series of tiny hair cells suspended in
3 What device or devices will attenuate the existing noise to fluid in the inner ear are attached to nerve cells and form an in-
the desired level? tegral part of the conversion process.
The Industrial Noise Control Problem. I n general, several deleteri- Exposure to sounds much louder than 90 or 100 db may pro-
ous effects occur as a result of high noise levels. Noise interrupts duce relatively long term (for more than a few minutes) changes
departmental communications between workmen and super- in the threshold of audibility for various sounds. The effect in-
visors, prevents coordinated operations which results in de- creases with prolonged exposure. Complete recovery takes
creased production, increases employee tension, and, in many in- place several hours or perhaps several days after the temporarily
stances, may contribute to higher accident rates. The lasting induced thresholds shift to lower regions. Temporary shifts of
effect of noise upon an individual depends upon the amplitude 20 to 30 db in the ears of young adults are the rule for many
and the frequency of the noise, the length of exposure, and the noisy industrial situations. This condition may be called "audi-
susceptibility of the person. Extremely high energy noise, even tory fatigue" and will vary greatly from one individual to the
of short duration, may permanently impair a person's hearing. next. I t is difficult to draw the line between physiological fa-
Conversely, long term exposure to noise above 110 db should be tigue and pathological injury followed by repair. I n any case,
avoided, although many persons may withstand subjection to the net effect seems to be about the same: a return to normal
high intensity noise up to 130 db for short durations without a auditory capability at some point after the exposure.
lasting effect [2]. Repeated exposure to intense noise over a period of months and
years will cause a gradual cumulative loss of hearing. The loss is
typically a high frequency loss, beginning a t about 4000 cps.
1
Numbers in brackets designate References at end of paper. The loss is apparently due to the degeneration of external hair
Contributed by the Safety Division for publication (without cells, and it is a "nerve" or "cochlear" hearing loss. No present
Presentation) in the JOURNAL OF ENGINEERING FOR INDUSTBY. Manu-
script received at ASME Headquarters, September 18, 1970. Paper test of hearing can reliably distinguish industrial hearing loss
No.71-Saf-A. from the normal hearing losses associated with the aging process

Journal of Engineering for Industry NOVEMBER 1 9 7 2 / 1007

Copyright © 1979 by ASME
Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use
 6 Octave band sound pressure levels may be converted to the
equivalent A-weighted sound level by plotting them on Fig. 1 and
noting the A-weighted sound level corresponding to the point of
highest penetration into the sound level contours. This equiva-
lent A-weighted sound level, which may differ from the actual
A-weighted sound level of the noise, is used to determine exposure
limits from Table 1.

Basic Noise Data. Noise almost defies scientific definition,

Noise as stated before is nothing more than unwanted sound.
More specifically, noise is any undesired disturbance and may h0
within certain discrete frequencies or may be across the entire
frequency spectrum. Noise may be intermittent, erratic, or con-
tinuous depending on the source. The term loudness is used in
conjunction with the term noise. However, noise level is mea-
sured and defined without reference to loudness. Loudness, not
presently definable by measurement, is to a large extent physio-
logical and psychological.
200 BOO 1000 Z000 4000 8000
BAND CENTER FREQUENCY- CPS Sound level measurement and analysis is difficult, to say the
least. The accepted unit of measurement is the dimensionless
Fig. I A - w e i g h t e d sound level ratio called the decibel or db. Plotting of noise levels or sound
pressure levels versus frequency is often required. The level for
a particular bandwidth is plotted along the center frequency of
the band. For example, if the SPL in the band 600 to 1200 Hertz
known as presbycusis. I t is suspected t h a t the two effects are is 84-db, this value is plotted on the center frequency, given by
additive, as though the exposure to noise merely accelerated the (600 X 1200)'/' = 850 Hz [2]. A narrow band does not permit
natural progress of presbycusis. The relations between hearing as much noise to pass as does a wide band. For example, the
loss and noise exposure involve frequency, sound level, temporal octave-band level of 84-db converted to half-octave is 81-db.
distribution, schedule of exposure and recovery, overall duration, This concept is graphically illustrated in Fig. 2. This correction
etc., and are too complex for a brief summary. should be applied when comparing, noise levels of different band-
Acoustic trauma is a term sometimes applied to cumulative widths.
permanent hearing loss from repeated exposures to noise, as in The apparent loudness of noise varies not only with the sound
industrial hearing loss. This term should probably be reserved pressure, but also with the frequency of the sound. Within the
for the effects of explosions, blasts, blows on the head, or a single
brief exposure to a very intense noise. T h e Walsh-Healy Public
Contracts Act recognizes this type of exposure by limiting expo-
Table 1 Permissible noise exposures 2
sure to impulsive or impact noise to 140 db peak sound pressure
level. This type of hearing loss may be very severe shortly after Duration per day, Sound level dbA,
the injury, but it generally tends to recover considerably before hours slow response
stabilizing [3], 90
In May, 1969, the Walsh-Healy Public Contracts Act was 6 92
4 95
amended to establish limits on noise exposure for employees of 3 97
companies with federal contracts of at least $10,000 annually. 2 100
Walsh-Healy amendments set a limit of 90 db (on the " A " scale) iVi 102
for continuous noise exposure during an eight hour working day 1 105
if ear protection is not used. Furthermore, the presumption of 72 110
l
/.i or less 115
the regulations is t h a t ear protection will be provided as a substi-
tute only when a noise reduction program is clearly not feasible.
2
The impact of these amendments has been significant in almost When the daily noise exposure is composed of two or more periods
every industry, since existing noise levels often exceed this limit, of noise exposure of different levels, their combined effect should be
considered, rather than the individual effect of each. If the sum of
in certain locations. The amended provisions of the Act are: the following fractions:
1 Protection against the effects of noise exposure shall be C„
provided when the sound levels exceed those shown in Table 1 Ti Tt" ' T,
when measured on the A scale of a standard sound level meter at
exceeds unity, then the mixed exposure should be considered to ex-
slow response. When noise levels are determined by octave ceed the limit value. C„ indicates the total time of exposure at n
band analysis, the equivalent A-weighted sound level may be de- specified noise level, and T„ indicates the total time of exposure per-
termined as follows: (see Fig. 1) mitted at that level.
As an example, consider an employee who works in three separate
2 When employees are subjected to sound levels exceeding high noise areas each day. Suppose he spends thirty minutes near a
those listed in Table 1, feasible administrative or engineering noisy exhaust vent putting out an average of 105 decibels, about
controls shall be utilized. If such controls fail to reduce sound forty-five minutes near process equipment where noise levels are 100
levels within the levels of the table, personal protective equipment decibels and an additional hour where ambient noise levels average
about 95 decibels. Assume that he spends the rest of his workday w
shall be provided and used to reduce sound levels within the relatively quiet areas. Since thirty min. = 0.5 hr and forty-five
levels of the Table. min. = 0.75 hr, we see from Table 1 that the combined effect of tins
3 If the variation in noise level involves maxima at intervals exposure pattern would be: ,-
of 1 sec or less, it is to be considered continuous.
(0.5/1.0) + (0.75/2.0) + (1.0/4.0) = 1.125
4 In all cases where the sound levels exceed the values shown,
a continuing, effective hearing conservation program shall be Since this factor exceeds unity, the mixed exposure exceeds the limit
administered. value as specified above. Notice that even though the employee WHS
exposed to relatively high noise levels for a total of only two hours
5 Exposure to impulsive or impact noise should not exceed and fifteen minutes of his workday, he was in violation of the new
140-db peak sound pressure-level. Walsh-Healy requirements!

1008 / N O V E M B E R 1972 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 \
\
\
\
k
l / « OCTAVE BAND
SB
1/3 OCTAVE BAND

e
CONVERSION IN «
DIFFERENCE IN DECIBELS
(TO BE SUBTRACTED FROM BAND BETWEEN TWO LEVELS
LEVEL TO OBTAIN SPECTRUM
-LEVEL) BEING ADDED?

Fig. 4 Continuation of Fig. 3

Fig. 2 Conversion in db

audible frequency range, 20 to 20,000 Hz, the ear is by no means

equally sensitive. If the sound pressure is held constant while
the frequency is changed, a sound t h a t is fairly loud at 4,000 Hz
may be entirely inaudible at 30 Hz. To the ear, a sound of 60-db
at 1,000-Hz may appear equally as loud as a sound of 70 db at
100 Hz. The loudness level of any sound is equal to the sound
pressure is decibels, of a sound of 1,000 Hz, which seems equally
as loud to the ear.
Often, various sound levels must be combined. Decibels are BACK0R0UND NOISE CORRECTION 8

combined on an energy basis, not added directly. Fig. 3 simpli-

fies the procedure by providing a graphic solution. As an ex-
ample: Using Fig. 3 to add the sound levels from Frequency
Band Numbers 7 and 8, first determine the difference in decibels
between the two levels being added. Band Number 7 = 70 db;
Band Number 8 = 70 db; subtracting 70 db - 70 db = 0 db. DB DIFFERENCE BETWEEN TOTAL NOISE AND BACKOI

From Fig. 4, add 3 db (for a zero db difference) to the higher Fig. 5 Background noise correction
level (70 db) to get a new SPL of 73 db.
When vent noises are measured, as in the case of a steam vent,
care should be taken to record only the direct airborne noise from
the source. However, complete isolation of the background

2 3 4 5 6 7 8 BAND NO.
95 IOO 93 82 75 7.0 70 SPL-DB

|S

PURE TONE
- 77.2
75 150 300 .600 IZOO 2400
150 300 600 1200 2400 4S00

OCTAVE BAND- CPS

; ) ) DAMAGE RISK CHART 2

Fig. 6 D a m a g e risk chart

noise is not usually possible. If both the source and the back-
>
ground noise are continuous, a correction is required as shown in
Fig. 5. The difference between the source noise level and the
background noise level determines the correction to be made. If
the difference is less t h a n 3 db, then the source noise level is less
DETERMINATION OF OVER-ALL SOUND than the background noise, and any correction should be con-
PRESSURE LEVELS FROM FREQUENCY sidered as approximate. In this case an attempt should be made
BAND L E V E L S '
to lower the background noise. If the difference is more than 10
fl db, the effect of background noise is nil and no correction need be
9 - 3 Determination of overall sound pressure levels from frequency
•""id levels made.

Journal of Engineering for Industry NOVEMBER 1 9 7 2 / 1009

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 wr»mnn CONSERVATION CRITERIA TOR BROAD BAND NOISE WORKMAN EXPOSURE
LEVELS AT WH!CH HEARING CONSERVATION PROCEDURES
ARE RECOMMENDED FOR PERSONS EXPOSED APPROXIMATELY
DAILY FOR ABOUT 25 YEARS.
UPPER LIMIT
DESIREABLE LIMIT

FOR THE CONSERVATION OF THE

HFABINIi OF SPEECH2

DAILY EXPOSURE
/ / r/TT J I 1111 h
TIME, MINUTES in

/ / C'ONTI'NUQUS / / / / / / / jj

1— 1 1 I 1 1
20 75 150 300 600 1200 2400
75 ISO 300 600 1200 2400 4800

FREQUENCY BAND- CPS

MAXIMUM NOISE LEVEL

FOR WIDE BAND NOISE 2

Fig. 8 M a x i m u m noise levels for w i d e band noise

300 600 1200 2400

600 1200 2400 4800

FREQUENCY BAND-CPS

Fig. 7 Hearing conservation criteria

Noise Criteria and Standards. There are many standards relating

to the effects of noise on man. However, the most widely ac-
cepted standard to date is the Damage Risk Criterion [2] il-
lustrated in Fig. 6, which applies to long term exposure to con-
tinuous noise. The more the noise exceeds 85 db, the more urgent
the need for noise reduction. Intermittent exposure is con-
sidered less damaging than exposure to continuous noise. Fur-
ther, it is generally conceded that hearing loss is directly related
to noise intensity and length of exposure according to the Hearing
Conservation Criteria, Fig. 7. This criteria sets the maximum
noise level (at specific frequencies) that can be experienced for a
given period without apparent ear damage.
The maximum noise level for wide-band noise is related to
workman exposure to noise in Fig. 8. Following the reasoning
that intermittent noise is less damaging than continuous noise of
the same intensity, the upper portion of 85-95 db was designated
continuous. Each category was sub-divided to show both a de-
sirable, as well as a maximum, level. The maximum noise limits
set by Fig. 8 whole, accepted by industry. It is ap-
parent, however, that no single noise standard will meet the re-
quirements of all industry. Industrial requirements, local con-
ditions, individual annoyance factors, and the like must all be
considered. This situation is emphasized by Fig. 9, where a
random plotting of various noise standards indicates little or no 2*00
4S0O
agreement in certain areas. This plot also shows the overall
FREQUENCY BAND- CPS
noise spectrum that specific standards must cover to be effective.
A tentative noise criteria for wide-band noise has been developed Fig. 9 Various m a x i m u m noise criteria for w i d e band noise
in Fig. 10. This criteria is recommended as a guide where a
specific standard or limit of noise is not specified [2].
Typical Silencer Classification. Theoretically, silencers are di- relatively broad-band performance is obtained, the category is bi -'
vided into two distinct categories. These are (a) reactive or re- described as being low reactive.
flective and (b) absorptive or dissipative. All practical designs Optimum broad-band silencer designs, then, are grouped ir1"
obtain noise reduction, to some extent, by both reflection or three basic categories:
energy absorption and dissipation: thus, such a distinction is
somewhat arbitrary. 1 Low-reactive
The purely reactive silencer is seldom applied in broad-band 2 Absorptive (dissipative)
noise applications due to its narrow-band performance. Where 3 Combinations

1010 / N O V E M B E R 1 9 7 2 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 When combined, the general performance spectrum of each of
the three types covers the entire audible frequency range.
Various silencer configurations t h a t are used are illustrated in
Figures 11,12,13, and 14.
Dimensions and weight of the various types of silencers are ex-
tremely important: especially in tight installations where space
and support come at a premium. This is one of the reasons t h a t
it is generally a necessity to have more than one choice to
select from. I n many applications, the success of the project
depends greatly on this choice.

Preliminary Analysis
Exhaust Vent Silencing Criteria. A common fallacy in vent and
blowdown silencing design is to attempt to comply with all
possible requirements, irrespective of the service or application
involved. Parameters t h a t are most often to be optimized are:

1 Maximum noise reduction

2 Minimum pressure drop
3 Minimum size and weight
4 Durability, long life, and freedom from maintenance
5 Lowest possible cost
It is obvious t h a t seldom, if ever, will all of these be met.

lio 300 600 1200 24" 00

300 600 1200 2400 4800
FREQUENCY BAND - CPS

Fig. 10 Tentative noise criteria for wide band noise

JPE I
g
flCAL LOW REACTIVE Sll ^NCFR

TYPE 2
I I
TYPICAL ABSORPTIVE SILENCER2

ACOUSTICAL PACK

INTERNAL
d/I/ll/L/U/J/M/JLl UlK
BAFFLES

PERFORATED OR

!i i
SLOTTED TUBES
\f77Tmrr/m r/rr/niJ
a) ANNULAR CORE

i i
i i

a) STANDARD 2 CHAMBER 6) OPTIMUM 3 CHAMBER ACOUSTICAL PACK

\JM/lIllUtUl/l/W± UUl/l/Xi
YXjfT/r/T/rj/irr/r/ r/jnnnnn.
b) STRAIGHT-THROUGH

I I
PERFORATED OR ' I I
SLOTTED TUBES

ACOUSTICAL PACK
INTERNAL BAFFLES

I 1 <mnmiuzn>
mn/minn>
c
) INLINE LOW-LOSS 2 CHAMBER d) INLINE LOW-LOSS 3 CHAMBER (nmni rrn TTTT^
c) SPLITTER MODULE

Fig. 11 Typical low-reactive silencers Fig. 12 Typical absorptive silencers

Journal of Engineering for Industry NOVEMBER 1972 / 1011

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 TYPE 3
TYPE 3
TYPICAL COMBINATIONS'
TYPICAL COMBINATIONS

MULTIPLE TUBE FT7]

OUTLET
ri ACOUSTICAL PACK

PERFORATED OR
SLOTTED TUBES

INTERNAL BAFFLES
u vit
ACOUSTICAL PACK f,\ \i
t) ANNULAR CORE /DIFFUSER f) SPLITTER /DOUBLE DIFFUSER

a) STANDARD b) /DIFFUSER

ACOUSTICAL PACK

TU^VZZZZ^ZZZZZZ^^ZTZZX^

VZZ77Z7ZZZZ7/Z, YZZZZZt
z^nj
PERFORATED OR
SLOTTED TUBES g) INLINE LOW-LOSS 2 CHAMBER

Fig. 14 Continuation of Fig. 13

INTERNAL BAFFLES

ACOUSTICAL PACK

c) OPTIMUM d) /DIFFUSER

Fig. 13 Typical combinations

The selection of silencer type and actual silencer sizing is based

mainly on the following considerations.
1 Type of application and service (intermittent or continuous)
2 Intensity of unsilenced noise
3 Peak frequency of unsilenced noise
4 Specified noise criteria (required attenuation)
5 Flow rate and type of gas being vented
6 Allowable pressure drop
7 Upstream pressure and temperature
8 Cost, weight, and size restrictions
The first consideration in vent silencing should be to determine
whether the noise source is continuous or intermittent. Con-
tinuous service is venting under near stable flow conditions. In-
termittent service by definition, is other than continuous. In- NON C R I T I C A L FLOW
termittent noise is evaluated in terms of (a) amplitude and fre- AIR VENT TO A T M O S P H E R E 2
llNSSLENCEO NOISE L E V E L S
quency content, (6) duration of blowdown, and (c) frequency of
occurrence. Any noise that is periodic is considered to be in-
termittent. Most accepted noise standards permit intermittent 130 140 150
i
levels of 5-10 db above those established for continuous service.
UNSILENCED SPL DO AT 10 FT. FROM SOURCE
Maximum or optimum noise reduction can only be obtained if
the noise to be attenuated is there in the first place. This is one Fig. 15 Air vent to atmosphere—unsilenced noise levels

1012 / N O V E M B E R 1972 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 VENTS TO ATMOSPHERE
CORRECTION FOR VALVE
SIZE2 VENTS TO ATMOSPHERE
CORRECTION FOR DISTANCE 2
(EXTENDED FROM 10 FT.)

SPL CORRECTION - DB

Fig. 16 A i r v e n t to atmosphere—correction for v a l v e size

-20 -'30

CORRECTION FOR HYDROCARBON GASES SPL CORRECTION- DB

Fig. 18 Vents to atmosphere—-correction for distance

Low frequency noise is produced from low pressure, large valve,

and high mass flow vents to atmosphere. High frequency noise
is generally associated with high pressure, medium to small valve,
high velocity vents. Fig. 19 gives the peak frequency of the un-
silenced noise utilizing a formula developed b y Strouhol [2].
The peak frequency refers to the fundamental frequency at which
maximum occurs. Multiples of the fundamental may be of equal
or even greater amplitude. This depends in part on the valve
type and size, and the piping downstream of the valve. I t is
next to impossible to predict whether or not harmonic frequencies
will occur or when they do occur and a t what multiple the roll-off
will start.
Peak frequency of the unsilenced noise determines the silencer
type to some extent. However, in the past the selection of type
has been somewhat arbitrary, about t h e only agreement being
that volume is needed for the low frequencies and some sort of
acoustical treatment for the high frequencies. Without a proven
and useable theory, experience has been the deciding factor.
Thus, silencing in general is as much of a n art as it is a science.
I t is important t h a t silencer sizing be based on flow data.
CORRECTION- DB Actual sizing is normally based upon velocity and pressure drop,
both of which are a function of silencer type, flow rate, molecular
Fig. 17 Correction for gas specific g r a v i t y
weight, and temperature of the gas.

144 X A C F M
Velocity fpm (1)
reason why performance data, to be practical, must be based on Area
the specific input intensity of t h e unsilenced noise. Otherwise,
attenuation curves are meaningless. The known formulas for A C F M = Flow rate, actual cfm
calculation of unsilenced noise levels are generally exceedingly (lb/hr X Specific Volume)
complicated and for the most part are empirical rather than ACFM = (2)
theoretical. A solution to t h e generalized equation for sound 60
Power [2] can be realized b y consecutively using Figs. 15, 16, 17, Area = Effective Silencer Largest Cross-sectional Open Area,
an
d 18. An example is given a t the end of this section. sq in.

Journal of Engineering for Industry NOVEMBER 1 9 7 2 / 1013

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 VALVE THflOAT DIAMETER, INCHES OCTAVE BANDS, H i , USAS Z24.I0, 1933

IOS

too
/ ///TEMPERATURE I

DECIBELS
93

90

as

LEVEL IN
80
FREQUENCY CHART
F.qn
VENT TO ATMOSPHERE 2 75

STEAM FLOW- a»,«OQpph

PEAK FREQUENCY CP3

Fig. 2 0 Octave band analysis

Fig. 19 Peak frequency chart for vent to atmosphere

Pressure drop across a silencer may be calculated using the con-

ventional formula as applied in both atmospheric and pressure OCTAVE BANDS, H i , USAS Z 2 4 . I 0 , 1953
600- 1200- 2400- 4800- 9600
(closed system) applications [2], eoo (200 2400 4900 9600 19200

Pressure Drop = (MWt X 530 X * X c)/(28.9 X Ti) (3)

MWt = Mol. wt. of steam
Ti = Temperature deg F
h = Velocity Pressure, " H 2 0 . "
c = Silencer Constant, valued from 0.3 to 4.0
HEARINS 'CONVERSATION

Determining Peak Frequency. Suppose t h a t the existing condi-

tions at an industrial facility are not satisfactory in terms of
satisfying noise control criteria as specified by the Walsh-Healy
Public Contracts Act. The problem could be an inadequately
g 70-f
sized muffler on top of a steam exhaust vent. Recent octave
band analyses in Figs. 20 and 21 indicate sound pressure levels 1,
are far in excess of hearing conservation ciiteria for this system.
Since unsilenced noise levels are unknown, it is advisable to READINGS TAKEN 4 0 FT. .BELOW THE TOP OF THE MUFFLER

approximate these levels by analytical tools. STEAM FLOW - S 9 , 6 0 0 pph,

Peak frequency is actually the lowest harmonic at which peaks Fig. 21 Octave band analysis
should occur in the unsilenced octave band analysis. I t may be
approximated if variables such as letdown valve throat diameter,
molecular weight and temperature of the vapor are known.
The conditions of a typical industrial steam plant are currently
as follows:

Superheated steam is vented to the atmosphei'e Table 2 Estimation of unsilenced noise levels (correction for various
Letdown valve—8 in. nominal dia gases)
Downstream pressure—175 psig Gas Correction
Downstream temperature—480 deg F Carbon Monoxide CO + 0.4db
Pipe size—10 in. dia, schedule 30 Hydrogen H2 +23.0
Maximum flow—186,000 pph Helium He + 19.0
Molecular weight—18 Methane CH, + 5.0
Nitrogen No + 0.2
StrouhaPs algorithm for determining peak frequency was pre- Oxygen 02 -0.9
Carbon dioxide -2.2
sented graphically in Fig. 19. Based on the above information, Superheated steam co220
H + 3.7
the peak frequency seems to be about 600 cps as illustrated in Saturated steam H20 - 5 . 0 (Avg)
Fig. 19.
Therefore, at a distance of ten ft from the unmuffled pipe, we
Determination of Unsilenced Noise Levels. The unsilenced noise
would expect an unsilenced sound pressure level of 145.0 db.
level is based on such variables as upstream pressure, upstream
Assuming a distance of 25 ft between the exhaust vent and the
temperature, valve size, and gas specific gravity. This level may
nearest person, Fig. 18 yields a distance correction factor of —8
be approximated using Figs. 15, 16, 17, 18, and Table 2. Based
db, to get a net of 137.0 db overall sound pressure level (SPL)'
on our data, these criteria yield the following results:
This figure should not be construed as the SPL a t all frequencies,
Fig. 15 + 122.5 db but, in fact, is the net SPL of all frequencies combined. (See the
Fig. 16 addition of SPL's at different frequencies in Figs. 3 and 4.) A
+ 18.8 db
Table 2 very conservative approach would be to assume t h a t no in-
+ 3 . 7 db
dividual frequency would yield more than 140 db sound pressure
+ 145.0 db level, as can be seen from Figures 3 and 4.

1014 / NOVEMBER 1 972 Transactions of the ASME;

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 reduction at different frequencies. Since the exhaust steam vent
with the high pressure, high temperature superheated vapor being
emitted will undoubtedly produce broad-band noise, this reactive
design without the dissipative effects of the lining is not satisfac-
tory.
I t would be well to digress for a moment to discuss some of the
important assumptions generally found in the literature t h a t are
inherent in describing the acoustic performance of the reactive
EXPANSION CHAMBER muffler. I t is assumed:

Fig. 22 Expansion chamber 1 T h a t the flow through the system does not affect the sound
propagation.
2 T h a t temperature variations within the system do not
affect the sound propagation. The effect of the average tem-
Designing the Expansion Chamber perature on the velocity of sound and wave length has been taken
The Acoustically Treated Expansion Chamber Transmission Loss into account in equation (6).
Equation. The general muffler design problem is poorly supported 3 T h a t sound pressures are small compared to absolute pres-
theoretically. Most information concerning the subject is em- sures, so that nonlinear effects are negligible.
pirical. The primary reason for this unhappy situation is simply 4 T h a t muffler wall surfaces do not conduct or transmit
that sound pressure wave transmission is a n exceedingly compli- sound.
cated process even under ideal conditions. When flow paths are 5 T h a t sound is propagated in plane waves, unattenuated by
complicated by twists, turns, expansions, constrictions, and viscosity or heat conduction.
acoustical treatments, the theoretical difficulties in modeling
Theoretical explanations of flow and sound interactions are not
the silencer system become quite immense. A sad by-product of
available at present: therefore, assumption number one is the
this situation is that, with only empirical information to work
most questionable. Even with the other qualifications, the fore-
with, it is often impossible for the designer to estimate whether a
going material can serve as a usef ul guide when estimating muffler
given muffler design that worked under one set of flow circum-
performance [4].
stances will work under a slightly different set of flow conditions.
The effects of adding dissipation to an expansion chamber are
If faced with higher gas flow rates, the designer must decide how
derived from Davis' analysis by changing the phase propagation
much he must increase the size of an existing design to satisfy new
exponent from kle to (kle + jale). One discovers that, by making
criteria. In many cases, if he cannot find a nearly identical
the substitution and carrying out some algebra, the transmission
silencer for a similar gas flow situation whose success is supported
loss is described by a reasonably formidable looking equation
empirically, he may only be able to make an educated guess as to
called here the hyperbolic form of the transmission loss equation:
his new silencer dimensions.
There is clearly a very real need for a great deal of experimental T.L. = 10 log10 {[cosh (al.) +
and theoretical work in this area. No doubt new government
1/2(TO + l / m ) sinh (<rle)V cos2 (M„) + [sinh (<rl,) +
regulations will continue to stimulate research and progress
toward optimal silencer design. 1/2(TO + l/m) cosh (o-ZJP sin 2 (W,)}db (8)
Fortunately, some analytical tools do exist for some types of
silencers, and they have been incorporated here to generate an <j = attenuation per unit length for lined duct, nepers.
algorithm intended to take a lot of drudgery out of the design of TO = ratio of area of expanded or lined sections to area of
one type of broad-band silencer. Specifically, the silencer initial (and final) sections of duct.
modeled here is an acoustically lined expansion chamber. This
is basically a combination reactive and dissipative muffler in k = wave number.
which several mechanisms interact favorably to reduce sound
levels. l„ = muffler length.
The analysis of this system begins b y considering the reactive Even the very patient designer would probably not find this
elements of noise reduction without including the dissipative equation susceptible to a large number of slide rule calculations.
effects of the lining. A simplified diagram, Fig. 22, illustrates the The obvious next step is to automate the transmission loss calcu-
salient aspects of the reactive element. The behavior of this lations via the digital computer.
unit is described in terms of two dimensionless parameters TO Because many digital computers have exponential and trigono-
and kl„ metric functions, b u t do not have hyperbolic functions, the follow-
ing form of equation (8) is given:
muffler cross-sectional area <S2
(4)
duct cross sectional area Si Exponential form of the Transmission Loss Equation (5)
1 le
2ir 2irf , _ e" ' + e~*
k = wave number = -r- = ft ' (5) T.L. = 10 :
A c ((
le = muffler length, ft + l/2(m + l/m) cos2 kl„

In English units kle can be approximated by grfe _L e~<Ae

/e"u
1/2 (m + 1/TO) 1 sin 2 kle J (9)
kle = ( 2 T T / V 1 130) ( 5 3 0 / ( 7 + 460) )V' (6) +
The equation describing the transmission loss (T.L.) for this One interesting aspect of this T.L. equation is the relatively
system is: flat response of the system across a broad frequency spectrum
when compared with a purely reactive system. This added
T.L. = 10 log 10 [l + 1/4(TO - l / m ) 2 sin 2 (kl,)] (7) feature makes the new silencing system a definite candidate for
The T.L. for this system will be a periodic function in kl„ re- the broad-band attenuation requirements of the steam exhaust
peating every ir radians (180 deg). The obvious drawback is vent or other high temperature, high pressure gas vents.
that the T.L. characteristics will vary widely across a broad fre- There are too many salient variables involved in the common
quency spectrum, having substantial dips in the effective noise acoustically-lined expansion chamber design problem to make a

Journal of Engineering for Industry NOVEMBER 1972/ 1015

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 D1SSIPAT1VE EXPANSION CHAMBER TRANSMISSION LOSS ANALYSIS"* single optimization algorithm feasible. Some of the parameters
that could be optimized or treated as constraints are:
MUFFLER LENGTH = 6, FEET
1 muffler length
MUFFLER I.D.. = 55. INCHES
2 muffler width
AREA RATIO =30, 3 muffler volume
4 type of acoustical treatment
FREQUENCY SIGMA TRANSMISSION LOSS 5 gas velocity through the muffler
CPS NEPERS DECIBELS 6 depth of acoustical treatment, and
7 higher attenuation of one frequency versus another.
75 0,080 50.1
I t appears t h a t the most general approach would be to provide an
150 0,080 56,1
algorithm t h a t would analyze a predetermined set of silencers,
300 0,090 52,0 and then provide broad-band performance spectra to the designer
600 0,090 57.3 so t h a t a decision could be made from a large group of potential
1200 0,100 50.6 designs. A program has been written based on this algorithm
and implemented in the design of a silencer for a typical industrial
2400 0.100 55.3
exhaust steam vent. The program provides transmission loss
1800 0.120 59.5 analysis across several frequencies for each muffler design con-
9600 0.120 53,5 sidered. The details of the muffler selected follow.
12000 0,120 59,5 Expansion Chamber Dimensions (an example). Although availabil-
0.120 57.2 ity and other factors may influence the particular acoustical
15000
lining utilized, we have assumed the use of an unbonded glass
fiber that can operate under temperatures as high as 1,000 to
Fig. 2 3 Computer output from Transmission Loss program
1,100 deg F . Attenuation data for frequencies above 4000 cps is
relatively scarce. A set of values for sigma has been assumed
and these are illustrated in Fig. 23, as a function of frequency.
Once the physical parameters of the acoustical lining have been
determined, it becomes a relatively simple procedure to establish
the relevant physical dimensions of the silencer using the trans-
mission loss equation.
After these values have been run in the transmission loss pro-
REFLECTOR gram, t h e designer can simply scan the output until he arrives at
a distribution of transmission losses across the frequency range
t h a t satisfies his design criteria. A constant sound pressure level
of 140 db was assumed here at all frequencies tested, and the task
is to bring the SPL down to 90 db. A transmission loss of 50 db
at each frequency must be generated. Fig. 23 illustrates the
computer output for the smallest volume dissipative expansion
chamber t h a t the designer felt met the design criteria. The
actual muffler is illustrated in Fig. 24. Dimension (h refers to
-ACOUSTICAL
muffler I D in the T.L. equation which here is 55-in. Dimension
BLANKET le refers to muffler inside length which here is 6.0 ft. Drain plugs
are supplied to release any condensate t h a t might form in
acoustical blanket.

•EXTERNAL
The high pressure, high temperature turbulent steam is
SHELL tremely corrosive and a special lining must be supplied to pro
the acoustical blanket. Fig. 25 may be used to determine
type of protective facing necessary, given the average gas velc
through the muffler [4]. Velocity is given by.
-PERFORATED
STAINLESS
STEEL
144 X ACFM
Velocity fps
INTERNAL (Area X 60)
SHELL
Area = Effective silencer open area, sq in.

(lb/hr) X (Specific Volume, cu ft/lb)

ACFM =
60

(186,000 X 37.96)
117,500 A C F M
60
DRAIN PLUG

(144 X 117,500)
Velocity = 120:
(3.14 X 27.5 X 27.5 X 60)

FLANGE The facing that is recommended for this application is

lustrated with the acoustical blanket (glass fiber) in Fig.
The steel sections are 14 gauge stainless. This type of fa<
should be quite durable over a long period of time. The o
ACOUSTICALLY LINED EXPANSION CHAMBER - NO SCALE area of the holes in the facing illustrated in Figs. 24 and 26 she
be no less than 20 percent of total inside muffler area (here, I
Fig. 2 4 Acoustically lined expansion chamber ir X cfc2/4 = 171,360 sq in.). T h e theoretical transmission

1016 / NOVEMBER 1 972 Transactions of the ASME

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 PERFORATED METAL
(AT LEAST 2 0 % OPEN)
MATERIALS

SLASS FIBER
MAX. VELOCITY
FPS

m ** "^
-PERFORATED
STEEU
STAINLESS

CLOTH

BLANKET -2 CORRUGATED STEEL

SHEETS

•a a &• •#

PERFORATED METAL -WIRE SCREEN

( 2 0 % OPEN) SLASS FIBER
CLOTH
WIRE SCREEN
BLANKET 75-150

;*r«i
"GLASS FIBER CLOTH
PERFORATED METAL
(20% OPEN)
2 CORRUGATED STEEL
SHEETS (30% OPEN) -ACOUSTICAL BLANKET
GLASS FIBER
WIRE SCREEN CLOTH
Fig. 2 6 Protective f a c i n g for acoustical blanket
150-225

PERFORATED METAL WIRE SCREEN

(20% OPEN)
2 CORRUGATED STEEL
SHEETS ( 3 0 % OPEN)
PERFORATED METAL GLASS FIBER
(40% OPEN) CLOTH
BLANKET 225-250

PROTECTIVE FACINGS FOR ACOUSTICAL LININGS SUBJECTED

TO HIGH-VELOCITY GAS STREAMS 4

0I5SIPATIVE EXPANSION CHAH6CR TRhttBUISSIQH LOftft AHALYSl..

Fig. 25 Protective f a c i n g s for acoustical linings subjected t o high
velocity gas streams

FREQUENCY IN CP3

Fig. 27 T r a n s m i s s i o n loss a n a l y s i s

characteristics for this example system are illustrated in Fig. 27.

All frequencies have T.L.'s greater than 50 db as required.

Special thanks are due to Mr. Bob McCorkle of Delta P In-

Summary corporated, Subsidiary of Burgess-Manning Incorporated for per-
The recently amended Walsh-Healy Public Contracts Act mission to quote material from their Silencing Handbook; 1968
focused on one aspect of the much publicized pollution problem, Edition.
specifically t h a t of noise or sound pollution control. This Act
provides limitations on noise exposure for employees of com-
panies with federal contracts of at least $10,000 annually.
References
The intent of this article is to bring together some of the more Cited
salient aspects of the noise control problems that exist where high 1 Allen, Clayton H., "Guidelines for Quieter Equipment," Me-
pressure, high temperature gasses are vented to the atmosphere. chanical Engineering, Jan. 1970, p. 29.
The analysis here is supplemented with an example noise control 2 Golden, B. G., Silencing Handbook: 1968 Edition, Burgess-
Manning, Dallas, 1966.
problem for a typical industrial steam facility. The problem is
3 Harris, Cyril M., Handbook of AToise Control, McGraw-Hill,
resolved by the application of a theoretical model for an acousti- New York, 1957.
cally treated expansion chamber type muffler. This silencer siz- 4 Beranek, L. L., Noise Reduction, McGraw-Hill, New York,
ln 1960.
g algorithm as well as the supporting theory are sufficiently
general so as to be applicable to a number of high velocity gas- 5 McEver, M. D., Meador, Larry, and Phair, M. W., "Exhaust
Vent Muffler Design," Unpublished Technical Report No. Dup-
generated noise conditions other than the specific one men- 3-0270, Mechanical Engineering Department, University of Texas at
tioned here. Austin, May 1970.

O f Related Interest
1 "Acoustical Terminology," American Standards Association,
Acknowledgments 1st ed. (1960).
2 Beranek, L. L., "Design for Quiet, "Machine Design, Vol. 39,
The authors wish to credit the work of Mr. M. W. Phair, and Sept. 14, 1967, pp. 174,-184.
Mr. M. D . McEver, who as design team members with Mr. 3 "Computation of Loudness of Noise," American Standards
feador conducted the original research as reported in [5]. This Association, 1st edition, 1960.
4 Davis, Hallowell, M. D., "The Hearing Mechanism," Hand-
*ork was performed in the Mechanical Engineering Department, book of Aroise Control, McGraw-Hill, New York, 1957.
tne University of Texas at Austin, under the guidance of Professor 5 Driskell, L. R., "Piping of Pressure Relieving Devices,"
J
°hnJ.Allan. Petroleum Refiner, July 1960, pp. 28-32.

H'lMal of Engineering for Industry NOVEMBER 1972 / 1017

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 6 "Engineering Can Control Process Plant Noise," American 8 Decibel (do). The decibel is one-tenth of a bel. Thus,
Society of Safety Engineers, Mar. 1969, pp. 7-9. the decibel is a unit of level when the base of the logarithm is the
7 Floyd, J. K., "Controlling Industrial Noise," Mechanical Engi- tenth root of ten, and the quantities concerned are proportional to
neering, Vol. 90, Oct. 1968, pp. 32-36.
8 Geiger, P. H., Noise Reduction Manual, University of Michi- power.
gan, 1956. 9 Effective sound pressure (root-mean-square sound pressure).
9 "Harnessed Sound Gets Industry Learning," Chemical Engi- T h e effective sound pressure at a point is the root-mean-
neering, May 20,1968, pp. 75-82. square value of the instantaneous sound pressure, over a
10 Hopkins, J. M., "Noise Pollution: A New Problem," Hy- time interval a t the point under consideration. In the case
draulic Power, Vol. 47, May 1968, pp. 124-126. of non-periodic sound pressures, the interval should be long
11 Industrial Noise Manual, American Industrial Hygiene Asso-
ciation, 1966. enough to make the value obtained essentially independent of
12 Ingard, IT., "Nonlinear Attenuation of Sound in a Duct," small changes in the length of the interval. T h e term "effective
Acoustical Society of America, Vol. 43, Jan. 1968, pp. 167-168. sound pressure" is frequently shortened to "sound pressure."
13 King, A. J., The Measurement and Suppression of Noise, 10 Free Field. A free field is a field (wave or potential) in a
Chapman and Hall, London, 1965. homogeneous, isotropic medium free from boundaries. In prac-
14 Koyamai, S., and Saito, Y., "Noises of a Thermal Power
Plant," Technical Review, May 1968, pp. 86-90. tice it is a field in which the effects of the boundaries are negligible
15 "Noise," Product Engineering, Vol. 39, April 22, 1968, p. 65. over the region of interest. The actual pressure impinging on an
16 "Plastic Foam Absorbs Sound," Material in Design Engineer- object (e.g., electro-acoustic transducer) placed in an otherwise
ing, Vol. 65, April 1967, p. 29. free sound field will differ from the pressure which would exist at
17 Sanders, O. J., "Identification and Diagnosis of Noise Prob- that point with the object removed, unless the acoustic impedance
lems," Noise Control, Mar. 1958, p. 15. of the object matches the acoustic impedance of the medium.
18 "Simple Method of Identifying Acceptable Noise Exposures,"
11 Free-field room (cinechoic room). A free-field room is a
Acoustical Society of America, Vol. 42, Oct. 1967, pp. 810-819.
19 "Sound and Psyche," Mechanical Engineering, Vol. 90, Aug. room in which essentially free-field conditions exist.
1968, pp. 40-41. 12 Infrasonic frequency (subsonic frequency). An infrasonic fre-
20 Sutton, P., "Design of Noise Specification for Process Plants," quency is a frequency lying below the audio-frequency range.
Sound and Vibration, Vol. 8, July 1968, pp. 33-43. 13 Loudness. Loudness is not a physical measurement of
21 Wiener, F. M., Sound Propagation Outdoors, McGraw-Hill, noise, but rather a psycho-acoustic response. It is "the intensive
New York, 1958.
attribute of an auditory sensation, in terms of which sounds may
be ordered on a scale extending from soft to loud. Loudness de-
APPENDIX pends primarily upon the sound frequency and wave form of the
For convenience, the definitions of several terms used in this stimulus."
paper and in the noise control field in general, have been as- 14 Loudness contours. Loudness contours are curves which
sembled here. References [2] and [3] can be consulted for a more show the related values of sound pressure level and frequency
comprehensive list. required to produce a given loudness sensation for the typical
listener. A single contour gives the sound pressures at different
1 Absorption coefficient (a). Absorption coefficient for a surface frequencies which are necessary to produce equal loudness sensa-
is the ratio of the sound energy absorbed by a surface of a medium tions.
(or material) exposed to a sound field (or to sound radiation) to 15 Loudness level. The loudness level, in phons, of a sound is
the sound energy incident on the surface. The stated values of numerically equal to the sound pressure level in decibels, relative
this ratio are to hold for an infinite area of the surface. The to 0.0002 microbar, of a simple tone of frequency 1000 cps which
conditions under which measurements of absorption coefficients is judged by the listeners to be equivalent in loudness.
are made are to be stated explicitly. 16 Masking.
2 Ambient noise. Ambient noise is the all-encompassing noise (A) Masking is the process by which the threshold of audibility
associated with a given environment, being usually a composite for one sound is raised by the presence of another (masking)
of sounds from many sources near and far. sound.
3 "A-Scale." The "A-scale" refers to a sound pressure level (B) Masking is the amount by which the threshold of audi-
measurement in decibels when this measurement is referenced to a bility of a sound is raised by the presence of another (masking)
base pressure of 0.0002 microbars. sound. The unit customarily used is the decibel.
4 Attenuation. Attenuation is the reduction, expressed in 17 Microbar, dyne per square centimeter. A microbar is a unit
decibels, of the sound intensity a t a designated first location as of pressure commonly used in acoustics. One microbar is equal
compared to sound intensity at a second location which is to 1 dyne per square centimeter. The term " b a r " properly de-
acoustically farther from the source. For an acoustic signal notes a pressure of 10° dynes per centimeter. Unfortunately, in
which travels in open air, the observation points must lie on the acoustics the bar was used to mean 1 dyne per square centimeter.
same radial line from the source. For a signal confined to a chan- I t is recommended, therefore, in respect to sound pressure that
nel or duct, the intensity values are averages over the cross-sec- the less ambiguous "microbar" or "dyne per square centimeter'
tion. Reflected signals are disregarded in computing the at- be used.
tenuation. 18 Narrow band. Applies to a narrow band of transmitted
5 Band width. When applied to a band pass filter, band width waves, with neither of the critical or cutoff frequencies of the filler
is determined by the interval of transmitted waves between the being zero or infinite.
low and high cutoff frequencies. 19 Noise. Unwanted sound.
6 Cutoff frequencies. The frequency response bandwidth of 20 Noise level. Noise level is the value of noise integrated over
various sound measuring devices is defined by lower and upper a specified frequency range. I t is expressed in decibels relative
bounds known as t h e cutoff frequencies or half power frequencies. to a specified reference. In air the acoustical noise level is
At the half power frequencies, t h e response is 70.7 percent of the usually measured with a sound-level meter, and is hence the same
midfrequency gain. as the sound level of the noise. For. special purposes other
7 Damage risk criteria. A damage risk criterion specifies the measuring techniques are used and must be specified.
maximum allowable exposure to which people may be exposed if 21 Octave. An octave is the interval between two sounds hav-
risk of hearing loss is to be avoided. A damage risk might in- ing a basic frequency ratio of two. By extension, the octave 15
clude in its statement a specification of such factors as time of ex- the interval between any two frequencies having the ratio 2:1-
posure, noise intensity and frequency, amount of hearing loss The interval, in octaves, between any two frequencies is the
t h a t is considered significant, percentage of the population to be logarithm to the base two (or 3.322 times the logarithm to the base .
protected, and method of measuring the noise. ten) of the frequency ratio. 1

1018 / N O V E M B E R 1972 Transactions of the AS ME J

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 22 Peak sound pressure. The peak sound pressure for any speci- 27 Sound pressure level (SPL). The sound pressure level, in
fied time interval is the maximum absolute value of the in- decibels, of a sound is 20 times the logarithm to the base 10 of the
stantaneous sound pressure in that interval. In the case of a ratio of the pressure of this sound to the reference pressure. The
periodic wave, if the time interval considered is a complete period, reference pressure shall be explicitly stated. The reference pres-
the peak sound pressure becomes identical with the maximum
sure of 2 X 1 0 - 4 microbar is most commonly used.
sound pressure.
23 Phon. The phon is the unit of loudness level. 28 Transmission loss (TL). Transmission loss is the reduction in
24 Pure lone. A pure tone is a sound wave, the instantaneous the magnitude of some characteristic of a signal, between two
sound pressure of which is a simple sinusoidal function of the stated points in a transmission system. The characteristic is
time. often some kind of level, such as power level or voltage level; in
25 Resonance. Resonance of a system in forced oscillation acoustics, the characteristic t h a t is commonly measured is sound
exists when any change, however small, in the frequency of excita- pressure level. If the levels are expressed in decibels, then the
tion causes a decrease in the response of the system. transmission loss is likewise in decibels.
26 Sound. Sound is an alteration in pressure, stress, particle
displacement, particle velocity, which is propagated in an elastic 29 Wavelength (X). The wavelength of a periodic wave in an
material, or the super-position of such propagated alteration. isotropic medium is the perpendicular distance between two
Sound is also auditory sensation which is usually evoked b y the wavefronts in which the displacements have a difference in phase
alterations described above. of one complete period.

Journal of Engineering for Industry NOVEMBER 1 9 7 2 / 1019

Downloaded From: http://manufacturingscience.asmedigitalcollection.asme.org/ on 01/29/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use