11

Noise

Wes Bussman, Jay Karan, Carl-Christian Hantschk, and Edwin Schorer

CONTENTS
11.1 Fundamentals of Sound ............................................................................................................................................... 239
11.1.1 Introduction ....................................................................................................................................................... 239
11.1.2 Basics of Sound .................................................................................................................................................. 240
11.1.2.1 Sound Pressure Level and Frequency ............................................................................................. 240
11.1.2.2 Decibel ................................................................................................................................................. 241
11.1.2.3 Sound Power Level ............................................................................................................................ 242
11.1.2.4 Threshold of Hearing ........................................................................................................................ 243
11.1.2.5 Threshold of Pain ............................................................................................................................... 243
11.1.2.6 Correction Scales ................................................................................................................................ 243
11.1.3 Measurements ................................................................................................................................................... 244
11.1.3.1 Overall Sound Level and How to Add dB Values ......................................................................... 246
11.1.3.2 Atmospheric Attenuation ..... , ........................................................................................................... 248
11.2 Industrial Noise Pollution ........................................................................................................................................... 250
11.2.1 OSHA Requirements ............................................................................................. :.......................................... 251
11.2.2 International Requirements ............................................................................................................................ 251
11.2.3 Noise Sources and Environment Interaction ................................................................................................ 251
11.3 Mechanisms of Industrial Combustion Equipment Noise ..................................................................................... 252
11.3.1 Combustion Roar .............................................................................................................................................. 252
11.3.2 Fan Noise ............................................................................................................................................................ 252
11.3.3 Gas Jet Noise ...................................................................................................................................................... 253
11.3.3.1 Gas Jet Mixing Noise ......................................................................................................................... 253
11.3.3.2 Shock-Associated Noise .................................................................................................................... 253
11.3.4 Valve and Piping Noise .................................................................................................................................... 254
11.4 Noise and Oxygen-Enhanced Combustion .............................................................................................................. 255
11.4.1 General ............................................................................................................................................................... 255
11.4.2 Example: Noise Spectra from an Oxy-Fuel Burner ...................................................................................... 256
Glossary ................................................................................................................................................................................... 257
References ................................................................................................................................................................................ 259
Bibliography ................................................................................................................................ ;.......................................... 259

Injuries range from short-term discomfort to perma-

nent hearing loss. According to recent statistics, more
11.1 Fundamentals of Sound
than 20 million Americans are exposed to hazardous
11.1.1 Introduction sound levels on a regular basis. There are approximately
28 million Americans who have some degree of hearing
Noise is referred to as unwanted sound; it is a common
loss: about one-third of these-more than 9 million-
by-product of our mechanized civilization and is an
have been affected, at least in part, by exposure to
insidious danger in industrial environments. Consi-
excessive noise.
dering the impact on people, noise is most often a source
The sense of hearing is a fragile and vital function of the
of annoyance, but it can also have more detrimental
human body. It resembles the sense of vision, more so than
effects, such as causing physical injury. Noise-related
the other senses, because permanent and complete damage

239
240 Oxygen-Enhanced Combustion

can occur quite easily in an industrial environment. So Nerve that sends

noise pollution has been recognized as a safety concern for
a long time, and has been appropriately regulated.
Within the industrial site itself, the immediate issue
with noise is one of employee safety. In addition to con-
cern with physical safety, it has been found that reduc-
tion in noise levels contributes to improved employee
work performance and morale.
Equipment is also affected by noise. In most cases, these
effects lie in the area of vibration control and are beyond
the scope of this chapter. Suffice it to say that noise and
mechanical vibration usually occur together. High levels
of mechanical vibration are accompanied by high lev-
els of noise and vice versa. Vibration leads to mechani-
cal fatigue, which reduces equipment life. The effects of
fatigue are frequently accepted as normal wear and tear Hair cells
if the equipment life cycle spans a reasonable duration.
FIGURE 11.1
This chapter is written as a practical guide, as well as (See color insert.) A cross-sectional view of the human ear.
a reference, for engineers involved in design, operation,
or maintenance of oxygen-enhanced combustion (OEC) one frequency. Any pure tone can be uniquely identi-
equipment. In addition, because this chapter provides a fied by two of its properties, namely, pressure level and
comprehensive coverage of the fundamentals of sound, frequency. A tuning fork is an example of a pure tone
the creative engineer will also be able to extend his or generator. Naturally occurring pure tone generators are
her knowledge to analyze other noise-producing indus- rare. Even musical instruments create notes that have
trial equipment. significant pressure levels at two or three multiples, or
harmonics, of the fundamental frequency of the note.

11.1.2 Basics of Sound

11.1.2.1 Sound Pressure Level and Frequency
Sound is the vibration of particles in a gas liquid
or solid. Pressure level defines the loudness of the sound,
while frequency defines the pitch or tone of the sound.
Peterson1 Pressure level is the amplitude of the compression, or
rarefaction, of the pressure wave. The common unit of
Sound is propagated through any medium in waves pressure level is decibel, abbreviated to "dB." Frequency
that take the form of pressure peaks (compressions) and is the number of pressure waves that pass by an arbi-
troughs (rarefactions). The pressure wave travels through trary point of reference, in a given unit of time. As such,
the medium at the speed of sound in that medium. The the measure of sound frequency can be cycles per sec-
auditory system in humans and most animals senses the ond (cps), and as with electricity, the commonly used
impingement of these pressure waves on a tissue mem- unit is Hertz (Hz); 1 Hz = 1 cps.
brane and converts them to electrical impulses that are Another important quantity in the description of
then sent to the brain where they are interpreted as the sound waves is the speed of sound, which is typically
information we receive from the sound. designated by the letter c. The speed of sound describes
Figure 11.1 shows a cross section of the human ear. the veloocity at which a sound wave propagates
Sound is collected and funneled into the ear canal by through the ambient air or other fluids (e.g., water). In
the outer ear. At the end of the ear canal, the sound other words, the speed of sound describes the distance
impinges on the ear drum. The bones of the middle ear a sound wave propagates per unit of time. The length
convey the ear drum's vibration to the inner ear. The of a sound wave is another important quantity in the
inner ear consists of a fluid-filled membrane called the description of sound. The wavelength, A, describes the
cochlea that has tiny hair cells on the inside. The hair distance between one wave crest and the next, that is,
cells sense the vibration conveyed to the cochlea and between one point of maximum compression in the
convert the vibrations into nerve pulses, which are then sound wave and the next. The frequency f, speed of
conveyed to the brain through the auditory nerve. sound c, and wavelength A are related as follows:
In reality, most naturally occurring sounds are com-
posites of different pressure levels at various frequen-
c=Axf (11.1)
cies. On the other hand, a pure tone is a sound at only
Noise 241

TABLE 11.1 less sensitive to sound at the extremes of low and high
Speed of Sound in Different Media and at Different frequencies and we will discuss this in more detail later
Temperatures in this chapter.
Temperature (OC) Speed of Sound (m/s) The wide range of frequencies in our hearing range
Fluid
may be conveniently handled by breaking it up into
Air, dry -40 307 octave bands. Each octave band represents a doubling
20 320 in frequency. Table 11.2 shows the 10 octave bands that
0 332 cover the hearing range and the center frequencies
20 344 that can be used to represent each octave band. Each
40 356 octave band extends over seven fundamental musical
Hydrogen 20 1316 notes.
Methane 20 448
Ethylene 20 331
Water 0 1403
11.1.2.2 Decibel
40 1529
80 1555 The unit of sound level, the decibel, is difficult to visual-
100 1543 ize and warrants some explanation. While it is possible
Steel 20 5180 to quantify sound in units of either power or pressure,
Soft rubber 20 54 neither unit is convenient to use because in practice one
has to deal with sounds that extend over a very large
range of power or pressure values. For example, the
The speed at which a sound wave propagates depends sound power of a whisper is 10-9 W, while the sound
mainly on the type of fluid and the temperature power of a jet plane is 103 W. The range of these two
of the fluid. Table 11.1 gives some examples for the sound sources spans 1012 W. The decibel, a dimension-
speed of sound in different media and at different less unit, was invented in order to represent these large
temperatures. ranges conveniently.
The typical range of human hearing extends from In the 1960s, Bell Laboratories coined the term deci-
20 Hz to 20 kHz. Young children can hear frequencies bel. The "deci" stands for the base 10 log scale on which
slightly higher than 20 kHz but this ability dimin- the decibel is based and the libel" was meant to repre-
ishes with age. This trend of reduced high-frequency sent Bell Labs. See Figure 11.2 for how decibel relates to
sensitivity continues with advancing age. Loss of watts.
hearing in humans in the later stages of life typically In Figure 11.2, the y-axis corresponds to the sound
manifests itself as diminished sensitivity to frequen- power (in watts) and follows a base 10 scale. The x-axis
cies from 10 to 20 kHz. Mechanically, this is due to corresponds to the sound power level in dB (PWL). The
the deterioration of the fine hair cells in the basilar line provides the relationship between the sound power
membrane. and the sound power level; for example, 120 dB is equal
It is important to note that the ear is not equally sen- to a sound power of 1 W. As an illustration of the 10glO
sitive over the entire range from 20 Hz to 20 kHz. This relationship, note that 110 dB is equal to one-tenth of
is vital to understanding how noise affects us and how a watt (0.1 W) and 100 dB is equal to a hundredth of a
noise control is implemented. The human ear is much watt (0.01 W).

TABLE 11.2
The 10 Octave Bands
Octave Band (Hz) Center Frequency (Hz)
22-44 31.5
44-88 63
88-177 125
177-355 250
355-710 500
710-1,420 1,000
where
1,420-2,840 2,000
Lp = sound pressure level (dB)
2,840-5,680 4,000
Lw = sound power level (dB)
5,680-11,360 8,000
P = sound pressure (N/m 2)
11 ,360-22,720 16,000
W = sound power (W)
242 Oxygen-Enhanced Combustion

1 x 10 1
1 x 10°
1 x 10- 1
-- - - - - - - -- - -- - - -- -- ---
)f
1 x 10-2 V I
I

1 x 10-3 / I
I
I

~.... 1 x 10-4 / I
I
Q)
;?;
1 x 10-5 / I
I

'"CJ
0
0.. 1 x 10-6 / I

c::
::l
0 1 x 10-7 / I

~
VJ I
1 x 10-8
1 x 10-9 / I
I

1x 10- 10 / I
I
I
1 x 1O- 11 / I
I

1 x 10- 12 / I
I

o 10 20 30 40 50 60 70 80 90 100 110 120 130

Sound power level (dB)

FIGURE 11.2
Relationship of decibel to watts.

11.1.2.3 Sound Power Level

There is a subtle but important difference between the Sound pressure wave
at a distance r feet
terms sound power level (PWL) and sound pressure from the source
level (SPL). Sound power level is used to indicate the

~A ....
total energy-emitting ability of a sound source. In other
words, sound power is a characteristic of the sound
source itself. Sound pressure level, on the other hand, is
used to indicate the intensity of sound received at any \. Noise source with a
point of interest, from one or more sources. The illustra-
tion in Figure 11.3 shows the formula to calculate the
sound pressure level to be expected at a distance r from
r given sound power level
~ (Lwl mdiating outward.

The sound pressure level ~

a spherically radiating source of power level Lw: at this point can be
calculated as
Lp = Lw -10 IOglO ( 4nr2 ) + 10.5 (11.2)
Lp = Lw-lO loglO (4rr?) + 10.5
where r is in feet
where
Lp = sound pressure level (dB) FIGURE 11.3
Lw = sound power level (dB) Calculating sound pressure level at a distance r.
r = distance from source (ft)
If the earlier conditions are sufficiently fulfilled, then
Equation 11.2 is only valid exactly if the following con-
Equation 11.2 can not only be used to calculate the sound
ditions are fulfilled:
pressure level expected at a certain distance r from the
• Noise radiation from the source is uniform and source (knowing the sound power level Lw), but can also
equal in every direction. be used to back-calculate the sound power level of a
source from a measurement of the sound pressure level at
• The source is small compared to the distance r.
a known distance r from the source. The latter procedure
• There is no relevant noise contribution from is a technique that is standardized in DIN EN ISO 3744. 2
other sources at the point of interest. If the earlier conditions, however, are not fulfilled,
• There is no relevant influence on the noise at the then the sound pressure level predicted at a certain
point of interest from sound reflecting surfaces distance from a sound source could differ significantly
nearby. from the level actually measured, or the sound power
Noise 243

level of a source back-calculated from a measurement 130

of the sound pressure level can differ significantly 110
from the real sound power level of the source. In this P3' \ Threshold of pain
context, it is important to keep in mind that the sound ~ 90
Q)
;>
power level is an intrinsic property of the source, ~ 70
which is independent of the environment while the ....q;
:::s 50
sound pressure level not only depends on the distance ~
....q;
to the source but is also influenced by other sound 0... 30
'"d
c:
sources, the environment (reflecting surfaces etc.), and :::s 10
Vl0
the sound propagation conditions (temperature, atmo- -10
spheric pressure, wind, barriers, etc.).
-30
\0 lI'l Cf'l lI'l 0 0000000
r-I ,.....; \0 N lI'l 0000000
Cf'l r-I N lI'l 0 0 0 000
,.....i N~ ~ 00 v5 c5
N
11.1.2.4 Threshold of Hearing r-I

Frequency (Hz)
Figure 11.4 reveals a map of the threshold of hearing
FIGURE 11.5
in humans. The y-axis represents the sound pressure
Threshold of hearing and threshold of pain in humans.
level (SPL) in dB and the x-axis represents frequency.
Any SPL that falls below the curve at any given fre-
quency will be inaudible to humans. For example, a representing the threshold of pain is relatively flat. In
sound pressure level of 30 dB at 63 Hz will be inaudi- general, a sound pressure level of over 120 dB at any
ble, whereas a sound pressure level of 70 dB at the same frequency will cause pain. An important observation
63 Hz frequency will be audible. Humans are most that can be derived from the two curves is that if a
sensitive to sounds in the so called "mid-frequencies" sound is audible at very low or very high frequencies,
from 1 to about 5 kHz. This is generally the range of persons subject to this sound are very close to experi-
frequencies of the human voice and many other impor- encing pain.
tant common sounds. Additionally, at a constant level,
sound with a low or very high frequency will not have 11.1.2.6 Correction Scales
the same loudness sensation as that in the medium fre-
quency range. For example, a 100 Hz tone at a sound Sound meters are capable of measuring with equal sen-
pressure level 50 dB is perceived as the same loudness sitivity over the entire audible range. However, because
as a 1000 Hz tone at sound pressure level of 40 dB.3 humans do not hear with equal sensitivity at all fre-
quencies, the sound meter's measurement needs to be
modified to quantify what really affects humans. This
11.1.2.5 Threshold of Pain can be done using a correction curve. The most common
Figure 11.5 shows the threshold of pain superimposed correction is the A-scale correction curve, which resem-
on the threshold of hearing. Fortunately, the curve bles an idealized inverse of the threshold of hearing
curve (refer to Figure 11.6). An A-weighted sound level
130
80.--------------------------------------.
110
P3' 90 \\ I P3' 60~--------------------------------------1
~
Q)

~
;>
70 \ I ~
~ 40+---~----------------------------------1
q;
....
~ 50 ' " Human threshold of hearing I ~
q;
....
~, I
~ 20+---------~~---+----------------------1
trJ
....q;
0... 30 / trJ
....q;
'"d
c:
:::s 10
~~ ) ~
c:
O+------------------==-~~======~~~--I
:::s
---------------~~:--~-------
0
Vl o
Vl-20+-----~~--------~--------------------1
-10
-30 -40
o 0 0 0 000 o lI'l
,.....;
Cf'l
\0
lI'l
N
0
lI'l
0
0
0
0
0
0
0
0
0
0
0
0
lI'l 0 0 0 000 o r-I N lI'l 0 0 0 0 0
N lI'l 0 0 000 o Cf'l
r-I~ 01 '1'~ 00 v5 c5 ,.....i 01 ~ 00 v5
r-I N r-I
Frequency (Hz) Frequency (Hz)

FIGURE 11.4 FIGURE 11.6

Threshold of hearing in humans. A-weighted scale for human hearing threshold.
244 Oxygen-Enhanced Combustion

85 correlates reasonably well with hearing-damage risk in

;:0
80 industry and with subjective annoyance for a wide cat-
~ 75 egory of industrial and community noises. After apply-
Q)
;;- 70 ing the A-scale correction, the unit of sound pressure
..3:l
(l)
65 level becomes the dBA scale. Figure 11.7 shows a typical
~
~ 60 burner noise curve as measured by a noise meter (flat
~
Po< 55 scale) and the result after applying A-scale correction.
'"0
s:: 50 The other, less used correction scales are named, as
:::J
0
(/)
45 might be expected, B, C, and D. Referring to Figure 11.8,
40 one can see that the C-scale is essentially flat over the
l!) M l!) a a a a a a a
,....; \D C"l
...-l
l!)
C"l
a
l!)
a
a
a
a
a
a
a
a
a
a
range of interest and the B-scale lies somewhere between
M
,....;- 0S ..,;f< 00 0 the A- and C-scales. Given an understanding of the influ-
...-l

Frequency (Hz) ence of low-frequency sounds, one finds that the B- and
C-scales do not apply adequate correction in the lower
FIGURE 11.7
frequencies. Finally, the D-scale is different from the oth-
A-weighted burner noise curve.
ers in that it has a pronounced correction in the range of
2-5 kHz. The D-scale was devised for the aircraft indus-
try and is rarely used otherwise.
+20

+10 11.1.3 Measurements

0
;:0
A simple schematic of a noise meter is shown in
~
(l)
-10 Figure 11.9. The microphone is designed with a trans-
VJ
s::
0 -20
ducer that transforms pressure variations in air to a
Po<
VJ corresponding electrical signal. Since the electrical sig-
~ -30
(l) nal generated by the microphone is relatively small in
.?::
~ -40 magnitude, a preamplifier is needed to boost the sig-
Q)
p::; nal before it can be analyzed, measured, or displayed.
-50
Special weighting networks are used to shape the signal
-60 spectrum and apply the various correction scales dis-
-70
cussed earlier. The weighted signal then passes through
10 2 5 102 2 5 103 2 5 104 2 a second output amplifier into a meter. The meter and
Frequency (Hz) associated electronic circuits detect the approximate
root-mean-square (rms) value of the signal and display
FIGURE 11.8 it in units of dB.
Weighting curves A, B, C, and D.

Fast

Slow

FIGURE 11.9
Block diagram of a sound level meter.
Noise 245

Noise meters range from the simplest-microphone intervals and make a measurement at each interval.
and needle gauge-to sophisticated digital signal Typically, instruments are capable of
processing (DSP)-equipped analyzers. The more sophis-
ticated analyzers are equipped with fast Fourier 1. Octave band measurements
transform (FFT) capabilities that aid in accurate nar- 2. One-third octave band measurements
rowband analysis. In general spectrum analyzers 3. Narrowband measurements
allow the user to map the sound pressure level at dif-
ferent frequencies, or in other words, generate a curve Table 11.3 shows the usual octave and one-third octave
of the sound over different frequencies. However, bands. As the name suggests, a one-third octave band
there is a significant difference between instru- instrument makes three measurements in each octave
ments that make one measurement per octave band as opposed to the single measurement of the octave
and those that slice the octave band up into several band instrument. A narrowband instrument, on the

TABLE 11.3
Octave and One-Third Octave Bands
Octave One-Third Octave

Lower Band Upper Band Lower Band Upper Band

Band Limit Center Limit Limit Center Limit
12 11 16 22 14.1 16 17.8
13 17.8 20 22.4
14 22.4 25 28.2
15 22 31.5 44 28.2 31.5 35.5
16 35.5 40 44.7
17 44.7 50 56.2
18 44 63 88 56.2 63 70.8
19 70.8 80 89.1
20 89.1 100 112
21 88 125 177 112 125 141
22 141 160 178
23 178 200 224
24 177 250 355 224 250 282
25 282 315 355
26 355 400 447
27 355 500 710 447 500 562
28 562 630 708
29 708 800 891
30 710 1,000 1,420 891 1,000 1,122
31 1,122 1,250 1,413
32 1,413 1,600 1,778
33 1,420 2,000 2,840 1,778 2,000 2,239
34 2,239 2,500 2,818
35 2,818 3,150 3,548
36 2,840 4,000 5,680 3,548 4,000 4,467
37 4,467 5,000 5,623
38 5,623 6,300 7,079
39 5,680 8,000 11,360 7,079 8,000 8,913
40 8,913 10,000 11,220
41 11,220 12,500 14,130
42 11,360 16,000 22,720 14,130 16,000 17,780
43 17,780 20,000 22,390
The advantages of making broad band analyses of sound using octave or one-third octave band
filter sets are that less time is needed to obtain data and the instrumentation required to
measure the date is less expensive. The main disadvantage is the loss of detailed information
about the sound which is available from narrowband (FFT) analyzers.
246 Oxygen-Enhanced Combustion

other hand, uses digital signal processing (DSP) to or one-third octave band filter sets are that less time is
implement fast Fourier transform analysis (FFT), and needed to obtain data and the instrumentation required
in the current state of the art, FFT analysis allows the to measure the data is less expensive.
analyzed frequency range to be sliced up into a large When making sound measurements, several factors
number of smaller intervals, limited in number only regarding the nature of the source should be consid-
by the measured time interval's length and the avail- ered. For example, whether the source is a true point
able computer power. source in space (sound radiating spherically), whether
Figure 11.10 provides a comparison of the same the source is located close to a flat surface (sound radiat-
sound spectrum as analyzed using three different fre- ing hemispherical) or between two flat surfaces (sound
quency band intervals: octave band, one-third octave radiating a quarter of a sphere) will make a difference in
band, and narrow band. This comparison shows that how the measurement needs to be performed. However,
the additional resolution provided by narrower band a detailed discussion of measurement issues is beyond
methods can be of vital importance. In this example, the scope of this chapter and the reader may use some of
the level at 1 kHz, as recorded by the octave band the more comprehensive works in the list of references
instrument, is 90 dB; on the one-third octave instru- at the end of this chapter.
ment it is 85 dB and on the narrowband instrument it
is 70 dB. The lower resolution measurements produce
11.1.3.1 Overall Sound Level and How to Add dB Values
higher values due to the spillover influence of the
nearby peak at 1.8 kHz. In addition, in implementing As mentionedr most sounds are composites of several
noise control for this source, it is very valuable to know different levels at different frequencies. This is espe-
that it is the narrow peak at 1.8 kHz that is driving the cially true of industrial noise. Referring back to Figure
maximum noise. This knowledge helps to zero in on 11.7 (a typical burner noise curve), it can be seen there
the source. are significantly higher in two frequency zones, both of
However, as with many things, there is a cost asso- which will contribute to the apparent intensity experi-
ciated with high performance. For most applications, enced by a person working in the vicinity of the burner.
a one-third octave analysis is usually adequate. The It is difficult to describe this sound without using either
advantages of making broad band analyses using octave a diagram like the one shown or a table listing various

100r- 100

90 90

80 80

70 70

60 60

50
(a)

100

90

80

70

60

50
10 100 1k 10k
(c)

FIGURE 11.10
Same sound spectrum on three different intervals. (a) Octave-band spectruml (b) third-octave band spectruml and (c) narrow-band spectrum.
Noise 247

sound pressure levels occurring in the different octave To demonstrate how to use this equation, consider the
bands. The "overall sound level," a single number, has following example:
been devised to represent such composite sound curves
conveniently. If a single number is to be used to repre- Example
sent the whole curve, then it is not practical to use the
average of the various levels in the octave bands, since Add the following sound pressure levels together:
this number would be less than the levels at the peaks 88, 92, and 86 dB
and the peaks have the most influence on the listener.
Therefore, one must not confuse the average with the Ltotal( dB) = 10log lO (100.1X88 + 100.1X92 + 100.1X86) = 94.17 dB
overall sound level.
The overall sound level is calculated by adding the Some simple rules of thumb can be used to perform
individual levels in the various octave bands. In col- quick estimates. They are as follows:
umns 1 and 2 of Table 11.4, the burner sound curve
has been split up into its component levels in each 1. When adding dB values that are of equal mag-
octave band. In column 3, the A-weighted correction nitude or that differ by I, the sum is 3 dB added
has similarly been split up and listed. Column 4 gives to the greater number.
the A-corrected values for the sound curve by simply 2. When the two values are different by 2 or
subtracting column 3 from column 2. Now, the values 3 dB, then the sum is 2 dB added to the greater
in column 4 must be added to obtain the A-weighted number.
overall sound level.
3. When adding two values that differ by 4-9 dB,
Since the decibel is based on a loglo scale, simple addi-
then the sum is 1 dB added to the greater
tion cannot be used. For example, if two values of equal
number.
magnitude are added, say 100 dB and 100 dB, the result
is 103 dB. The formula used to add sound pressure levels 4. For values that differ by 10 dB or more, the sum
is as follows: is just the larger number.
5. Always start with the smallest number in the
list and add it to the next larger number.

To better understand why these rules work, refer to the

chart in Figure 11.2. From the chart it can be seen that
1 W is equal to 120 dB.
= 10Ioglo(10o.lLl + 10 o.lL2 + 10 o.lL3 + ...)
1 W=120dB
1 W=120dB
where
2W=123dB
Ltotalis the total sound pressure level (dB)
Li is each individual level (dB) On the chart, 2 W registers 123 dB on the line. Similarly,
n is the number of sound pressure levels to be added the reason that numbers 10 dB or more in difference are
neglected is because
TABLE 11.4
1.0 W = 120 dB
A-Weighting of the Burner Sound Curve
0.1 W=110dB
from Figure 11.7
Frequency SPL A-scale CF SPL 1.1 W = 120 dB
Hz dB dB dBA
Since the 110 dB contributes only a 0.1 W, it is neglected
31.5 72 -39 33 in the approximation. The example becomes more vivid
63 75 -26 49 when adding two numbers that differ by 20 dB or more.
125 79 -16 63
250 79 -9 70
1.00 W = 120 dB
500 72 -3 69
1,000 69 0 69 0.01 W = 100 dB
2,000 68 1 69 1.01 W = 120 dB
4,000 78 1 79
8,000 83 -1 82 Rule number 5 is especially necessary when adding a
16,000 80 -7 73 list that contains several numbers that are almost equal
248 Oxygen-Enhanced Combustion

TABLE 11.5
Addition Rules
What Is the Overall dBA Level?
A-Scale
Frequency (Hz) SPLdB CFdB SPLdBA

~
31.5 72 -39 33

63 75 -26 49 ~49

125 79 -16 63 ~~ 63
250 79 -9 70 ~ ~ 71
500 72 -3 69
~ ~73

1,000 69 0 69 ~ ~75
2,000 68 1 69 ~ ~ 76
4,000 78 1 79 ~ ~ 80
8,000 83 -1 82
~ ~ 84
16,000 80 -7 73
~ ~ 85

Overall sound level 85dBA

Caution: overall SPL does not = average SPL

in value and one or more that are 10 dB greater, such as two levels can be found in the upper part of the nomo-
in a list that contains six values of 90 dB and one value gram. For example, add the following pressure levels:
of 100 dB. If we begin to add from the 100 dB value we 90 + 90 dB. The difference between these two values is
will arrive at a wrong result. It should be noted that the zero. Locating zero on the lower scale of the nomogram,
rules provided are approximations. For exact calcula- one finds a value of 3 directly above it; therefore, 90 dB +
tions, the formulas should be used. 90 dB = 93 dB.
Table 11.5 shows the effect of applying the addi-
tion rules to the values generated by breaking up the
burner noise curve. At the end of the addition list, 1 dB 11.1.3.2 Atmospheric Attenuation
has been added to compensate for any errors due to When a sound wave travels through still air, it is absorbed
approximation. or attenuated by the atmosphere. Over a couple of hun-
As an alternative to using the previous rules of thumb dred feet, the atmosphere does not significantly attenuate
for level addition, the nomogram in Figure 11.11 can be the sound, however; over a few thousand feet, the sound
used to determine the sum of the two levels Ll and L 2: level can be substantially reduced. The amount of sound
For a certain difference between the two levels Ll and that is attenuated in still air largely depends on the atmo-
L2 in the lower scale of the nomogram, the correspond- spheric temperature and relative humidity. Figure 11.12
ing level increase I1L to be added to the higher one of the depicts the atmospheric attenuation for aircraft-to-
ground propagation in sound pressure level per 1000 ft
f..L----+ (300 m) distance for center frequencies of 500, 1000,2000,
4000, and 8000 Hz. Notice that the atmospheric attenua-
o 0.5 1 1.5 2 2.5 3
tion is larger at higher frequencies than at lower frequen-
cies. For example, suppose that we are 1000 ft (300 m)
away from a noise source and that the atmospheric tem-
perature and relative humidity is 80°F (27°C) and 10%,
20 109 8 7 6 5 4 3 2 1 o respectively. The plots in Figure 11.12 show that the atmo-
spheric attenuation for 500 Hz is approximately 2 dB
+ - - - - L1-L2
whereas for 8000 Hz the attenuation is 55 dB.
FIGURE 11.11 Atmospheric attenuation, outdoors, can also be
Nomogram for level addition. affected by turbulence, fog, rain, and snow. Typically, the
Noise 249

Temperature (0C)
-10 o 10 20 30 40
- 40
10
I I I I I
I I I I I I
355/710 Hz, fourth octave band, GMF 500 Hz S
- 20 0
0
,...,
0
o ,.,.,~ ....
("".*--.
;:q
.0;.,
0 "d

<:t: Relative _
o
40
humidity, %
g
,..., 10

---
;:q 710/1400 Hz, fifth octave band, GMF 1000 Hz --,-- 10

-- '-.-----'*
,.-....................-I_'. . . 'M#'
. .....,.-...,
'"0 I 1"--,,- --20 20
S
~
ci ?~
o ..........
~
., ~~m 30 0
.~ ~ ". .;';. 0
,...,
0

c:: o ~
50 0 ;:q
<l)
'"0
-~~-**~ 70
~
.~ 20 90
..... 60
<l)
...c::: 1400/2800 Hz, sixth octave band, GMF 2000 Hz
~
o
---•. -.,,,---~,
S / - 40 S
~ 10 ·/7
l/
7 ~ I"~"-' •..."
0
0
-~
~ ~ r-....._,,-..- '--- "'.'''.." ,...,
0

~ ~
i' ..........
-----r- - ;:q
20 "d
: / ~ ~." ["
....•..
V I---
:~ _/--~-

o 0
o 10 20 30 40 50 60 70 80 90 100
(a) Temperature (OF)

Temperature ("C)
-10 0 10 20 30 40

2800/5600 Hz
<:t: 30 seventh octave band 100
0
0
,...,
0 GMF 4000 Hz Relative
;:q humidity, (%) 80
"d
ci 10
.g 20 --20 S
ro 0
::l 30 60 0
c::
<l) --50 ,...,
0

~ 70 ;:q
'"0
.~ --90 40

20

0 0
0 10 20 30 40 50 60 70 80 90 100
(b) Temperature (OF)

FIGURE 11.12
Atmospheric attenuation (a) 500, 1000, and 2000 Hz, (b) 4000 Hz. (From Beranek, L.L., Noise and Vibration Control, McGraw-Hill, Cambridge,
MA,1988.)
(continued)
250 Oxygen-Enhanced Combustion

Temperature (0C)
-10 o 10 20 30 40

III I I I /"Nl.
,LJ\ II II
_
50
~ 5,600/11,200 Hz
eighth octave band -t
I 1\\ _ 160

~
t----r-- GMF 8,000 Hz /" I'- / \ Relative- 140
g 40 L __ L/ 'A \ -
o 1.------+---r~I- humidity-
(%) _

I~ IFm~mm-. 10I 120

""'"
P3;
'"0
2'
j I

~~
o / / '\ '--20
'.::lg 100 S
30
ilK"\ i/l\ '"
i2~ ·'l\"'~~' ;~
0
s:: 0

~
<J)
r- " _ 0
""'"
P3;
80 '"0

//iJ lll~ . . "'j-9~

.~

~0.. 20
'"o 1/// ~~~~ ~ 60

~ 'f- / // '1'--- 'I~~>--_ _

S
~
40
10
if
/ ~L
V
-~-+-rj-I
_ 20
?~}./--/ 1 1 1 1 J
o 0
o 10 20 30 40 50 60 70 80 90 100
(c) Temperature (OF)

FIGURE 11.12 (continued)

Atmospheric attenuation (c) 8000 Hz. (From Beranek, L.L., Noise and Vibration Control, McGraw-Hill, Cambridge, MA, 1988.)

more turbulence present in the air, the more the sound TABLE 11.6
is attenuated. There appears to be conflicting evidence Sound Levels of Various Sources
as to whether or not fog attenuates sound. It is recom-
Threshold of hearing 0 dBA
mended that no excess attenuation be assigned to fog or
Rustle of leaves 10 dBA
light precipitation.
Normal conversation (at 1 m) 30 dBA
Minimum level in Chicago at night 40dBA
City street, very busy traffic 70dBA
Noisiest spot at Niagara Falls 85dBA
Threshold of pain 120 dBA
11.2 Industrial Noise Pollution Jet engine (at a distance of 50 m) 130 dBA
Rocket (at a distance of 50 m) 200 dBA
Thus far, sound has been discussed. So what is noise? An all-
encompassing definition would be that noise is any unde-
sirable sound. By saying this, the concept is introduced that 87-88 dBA in the aromatic and paraffin facilities and
what is considered to be noise is somewhat relative, and 89 dBA in alkylation facilities. In comparison, work-
depends on several temporal and circUmstantial factors. ers in the warehouse, health clinics, laboratories, and
For example, it is not unusual for a person to encoun- offices were generally found to be exposed to much
ter sound pressure levels of 100-110 dB at a sporting lower levels.
event, in a stadium full of cheering fans, and yet not be Noise can damage hearing, and can cause physi-
perturbed by it. On the contrary, the barely 45 dB sound calor mental stress (increased pulse rate, high blood
of a dripping faucet may cause considerable annoyance pressure, nervousness, sleep disorders, lack of concen-
in the quiet of the night. Table 11.6 gives some examples tration, and irritability). Irreparable damage can be
of noise levels. caused by single transient sound events with peak lev-
Industrial noise pollution is a major concern for soci- els exceeding 140 dBA (e.g., shots or explosions). Long-
ety as a whole. In a recent survey, the effects of exposure duration exposure to noise exceeding 85 dBA can lead
to noise in refinery workers were studied extensively. to short-term reversible hearing impairment and long-
A cross section of workers in different divisions/units term exposure to levels higher than 85 dBA can cause
was chosen. It was found that noise levels averaged permanent hearing loss.
Noise 251

TABLE 11.7
OSHA Permissible Noise Exposure
Sound Pressure Level, dBA
Duration per Day (h) (Slow Response)
8.0 90
6.0 92
4.0 95
3.0 97
2.0 100
1.5 102
1.0 105
0.5 110
0.25 or less 115
FIGURE 11.13 Note: Exposure to impulsive or impact noise should
Typical ear plugs and muffs. not exceed 140 dBA.

The following is a mathematical model based on

empirical data (ISO 1999) used to calculate the maxi- where
mum permissible continuous noise level at the work en is the total time of exposure at a specific noise level
place that will not lead to permanent hearing loss: Tn is the total time of exposure permitted at that level
and shown in Table 11.7
LA,m (dBA) < 85+ 101oglO( ~:) (11.3)
If the exposure factor exceeds 1.0, the employee's expo-
sure is above OSHA limits. If OSHA identifies such a
where Tn is the daily noise exposure time in hours. situation, a citation may be issued and a grace period
Wearing ear protection devices at continuous noise defined in which the employer must correct the viola-
levels greater than 85 dBA can prevent or reduce the tion or face penalties as high as $10,000 per day.
danger of permanent hearing damage. There are
two levels of protection commonly used by indus-
trial workers to reduce noise levels: ear plugs and ear 11.2.2 International Requirements
muffs as shown in Figure 11.13. Ear plugs can lower
the noise level by 5-45 dB, depending on the type of Regulations aimed at protecting individuals from
plug, sound frequency, and how well the user inserts industrial noise pollution have been enforced in
them into the ear. Ear muffs are designed to cover the almost all industrialized countries. The noise caused
entire ear and typically reduce noise levels by 5-50 dB in industries and the work place is generally taken as
depending on the type of ear muff and the frequency a serious issue.
of the sound. Most countries have adopted 85 dBA as the limit for
the permissible noise. At any work place with sound
levels exceeding 85 dBA, ear protection devices must
11.2.1 OSHA Requirements be worn, and workers exposed to this level should have
Title 29 CFR, Section 1910.95 of the Occupational Safety their hearing level checked periodically.
and Health Act (OSHA) pertains to the protection of
workers from potentially hazardous noise. Table 11.7
shows OSHA permissible noise exposure levels. 11.2.3 Noise Sources and Environment Interaction
OSHA requires that the employer must provide protec-
Although noise pollution caused by industrial sec-
tion against the effects of noise exposure when the sound
tors is minor compared to that caused by road and
levels exceed those shown in Table 11.7. When the daily
rail traffic, industrial noise receives more attention
noise exposure consists of two or more periods of noise
due to public representation. ISO 1996 provides a
exposure at different levels, their combined effect should
set of international regulations for noise protection
be considered rather than the individual effects of each.
in residential neighborhoods located near industrial
According to OSHA the exposure factor (EF) is defined as4:
areas.
National or local authorities must enforce noise limits
(11.4) that should not be exceeded in the neighborhood. The
magnitude of limiting values, additional charges for
252 Oxygen-Enhanced Combustion

tonality and impulsive noise, and the legalities change

not only from country to country, but sometimes within
different states and regions in the same country. In gen-
eral, nighttime noise limits are 10-15 dB lower than that
for the daytime. co
~
~
2:l
~
'0
Z

11.3 Mechanisms of Industrial

Combustion Equipment Noise
There are four major mechanisms of noise production Frequency (Hz)
in combustion equipment. They can be categorized as
either predominantly high-frequency or low-frequency FIGURE 11.14
Typical noise signature emitted by an oxy-fuel burner.
sources. They are
Combustion roar emitted is not highly directional
1. Low-frequency noise sources
and is considered to be a monopole source. That is, it
a. Combustion roar
is analogous to a spherical balloon whose surface is
b. Fan noise
expanding and shrinking at various frequencies and
2. High-frequency noise sources emitting uniform spherical sound waves.
a. Gas jet noise The noise emitted from oxy-fuel burners typically has
b. Piping and valve noise two mechanisms at work, namely, combustion roar and
gas jet noise. Combustion roar typically resides in the
lower frequency region of the audible frequency spec-
11.3.1 Combustion Roar
trum, while gas jet noise occurs in the higher frequen-
To better understand combustion roar, the mixing pro- cies, as illustrated in Figure 11.14. Oxy-fuel burner noise
cess taking place between the fuel and the oxidant on can have a spectrum shape and amplitude that can vary
a very minute scale is considered. A well-blended mix- with many factors. A few of these factors include the (1)
ture of fuel and oxygen will burn very rapidly if the fired duty, (2) fuel-oxygen ratio and mixing, (3) design
mixture is within the flammability limits for that fuel. of the burner, (4) internal shape of the furnace, and (5)
However, a raw fuel stream that depends on turbulence acoustic properties of the furnace lining.
and momentum to mix in the oxygen, and achieve a
flammable mixture, tends to create a slower combus-
tion process. In either case, when regions in the mix-
11.3.2 Fan Noise
ing process form a flammable mixture and encounter
a source of ignition, combustion takes place. Typically, The noise emitted from industrial fans typically consists
the closer an oxygen-fuel mixture is to stoichiometry of two noise components: broadband and discrete tones.
prior to encountering an ignition source, the more rapid Vortex shedding of the moving blades and the interaction
the combustion. of the turbulence with the solid constructed parts of the
The noise emitted from each small region of rap- fan create the broadband noise. This broadband noise is
idly combusting mixture adds up to create what is of the dipole type, meaning that the noise is directional.
called combustion roar. Therefore, combustion roar On the other hand, the discrete tones are created by the
is largely a function of how rapidly the fuel is being periodic interactions of the rotating blades and nearby
burned. In addition, in the context of combustion upstream and downstream surfaces. Discrete tonal noise
equipment like oxy-fuel burners, usually the larger from fans is usually the loudest at the frequency at which
the fuel flow rate, the more the turbulence in the a blade passes a given point. The tonal frequency is eas-
combustion process. Since turbulence directly influ- ily calculated by multiplying the number of blades times
ences the mixing rate, high turbulence processes also the impeller rotation speed in revolutions per second.
produce more combustion roar. Thus, it is more accu- The broadband and discrete tonal noise emitted from
rate to state that the level of combustion roar gen- fans can radiate from both the suction and pressure side
erated by a combustion process is a function of the of a fan and through the fan casing. The noise can radi-
amount of fuel burned and how rapidly one arranges ate downstream through the ducting and discharge into
to burn it. the environment at an outlet. Installation of mufflers
Noise 253

and silencers on the suction and the discharge sides of Region of

the fan, as well as wrapping of the casing and the ducts maximum jet

are common methods for reducing fan noise.

In summary, fan noise can usually be addressed as
follows:

1. Silencers can be installed at the suction and

pressure sides of the fan particularly for fans
communicating with the atmosphere on either
the suction or the pressure side and thereby cut
down on noise coming out of these portals.
2. Acoustically enclose the fan casing to address
noise radiated from or transmitted through the
casing surface.
maximum jet
3. Acoustically isolate the ductwork leading to
mixing noise
and from a fan.
FIGURE 11.16
At the design stage, one may consider the use of low- Illustration showing the region of maximum jet mixing noise.
noise motors and the use of impellers with more blades,
reduced tip speed, etc. The source of gas jet mixing noise begins near the
nozzle exit and extends several nozzle diameters down-
stream. Near the nozzle exit, the scale of the turbulent
11.3.3 Gas Jet Noise eddies is small and predominantly responsible for the
Gas jet noise is very common in the combustion indus- high-frequency component of the jet mixing noise. The
try and in many instances it can be the dominant noise lower frequencies are generated further downstream of
source within a combustion system. The noise created the nozzle exit where the large-scale orderly pattern of the
when a high-speed gas jet exits into an ambient gas usu- gas jet exists.
ally consists of two principal components: gas jet mix- Gas jet mixing noise consists of a broadband fre-
ing noise and shock-associated noise. 5 quency spectrum. The frequency at which the spec-
trum peaks depends on several factors, including the
(1) diameter of the nozzle, (2) Mach number of the gas
11.3.3.1 Gas Jet Mixing Noise jet, and (3) temperature ratio of the fully expanded jet
Studies have shown that a high-speed gas jet exiting to the ambient gas.
a nozzle will develop a large-scale orderly pattern as The overall sound pressure level created by gas jet
shown in Figure 11.15. This orderly structure is known mixing depends on several variables including the (1)
as the "global instability" or "preferred mode" of the jet. distance from the gas jet, (2) angle of the observer rela-
The presence of both the small-scale turbulent eddies tive to the gas jet centerline, (3) Mach number, (4) fully
within the jet and the large-scale structure is respon- expanded gas jet area, and (5) density ratio of the fully
sible for the gas jet mixing noise. expanded jet to the ambient gas.
The maximum overall SPL of gas jet mixing noise
occurs at an angle between approximately 15° and 30°
relative to the centerline of the gas jet as illustrated in
Figure 11.16.5 As one moves in either direction from this
angle, the noise level typically drops off significantly.
For example, the overall SPL created by gas jet mixing
can be reduced as much as 25 dB when one moves from
an angle of maximum noise (15°-30°) level to an angle
directly behind the nozzle (180°).

11.3.3.2 Shock-Associated Noise

When a gas nozzle operates above a certain pressure,
Locations where orderly patterns are developing
a marked change occurs in the structure of the gas jet.
FIGURE 11.15 Above a certain pressure called the critical pressure, the
Development of orderly wave patterns within a high-speed gas jet. gas jet develops a structure of shock waves downstream
254 Oxygen-Enhanced Combustion

Shock waves

Nozzle exit

FIGURE 11.17
Photograph showing shock waves downstream of an air jet. (Photograph by Wes Bussman.)

of the nozzle as shown in Figure 11.17. The critical 11.3.4 Valve and Piping Noise
pressure of a gas jet typically occurs at a pressure of
When a gas flowing steadily in a pipe encounters a
12-15 psig (0.8-1 barg), depending on the gas composi-
valve, a change in the flow pattern and pressure will
tion and temperature. These shock cells consist of com-
occur that can create turbulence and shock waves down-
pression and expansion waves that repeatedly compress
stream of the valve. Typically, when valves are partially
and expand the gas as it moves downstream. Using
closed, creating a reduction in flow area, the small flow
Schlieren photography, several investigators have seen
passage behaves much like an orifice and produces jet
as many as seven shock cells downstream of a nozzle.
noise. As discussed previously, turbulence and shock
These shock cells are responsible for creating two addi-
waves create mixing noise and shock-associated noise.
tional components of gas jet noise: screech tones and
This noise can radiate downstream through the pipe
broadband shock-associated noise.
and exhaust into the environment at an outlet and/or
Screech tones are distinct narrowband frequency
radiate through the pipe wall not only into the space
sound that can be described as a "whistle" or "screech."
near the valve itself but also at relatively large distances
The literature reports that these tones are emitted from
from the valve. Usually, butterfly valves and ball valves
the fourth and fifth shock cells downstream of the noz-
are noisier than globe valves. Butterfly valves and ball
zle exit as shown in Figure 11.18.6
valves typically have a smaller vena contracta than a
The sound waves from these shock cells propagate
globe valve operating at the same pressure drop, which
upstream, where they interact with the shear layer at
results in higher levels of mixing and shock-associated
the nozzle exit. This interaction then creates oscillat-
noise. As a general guideline when the pressure ratio
ing instability waves within the gas jet. When these
across a valve is less than approximately three, the mix-
instability waves propagate downstream they interfere
ing noise and shock-associated noise are within about
with the fourth and fifth shock cells, causing them to
the same order of magnitude. However, for pressure
emit the screech tones. Screech tone noise is not highly
ratios greater than three, shock noise usually dominates
directional (monopole noise source), unlike gas jet mix-
mixing noise?
ing noise.
Valve and piping noise abatements include sound-
Broadband shock-associated noise occurs when
absorptive wrapping of the pipes and valve cas-
the turbulent eddies within the gas jet pass through
ings, installation of silencers between the valves and
shock waves. The shock waves appear to suddenly
the connecting pipes, and the use of low-noise valve
distort the turbulent eddies, which create a noise
designs with multiple pressure stages. Acoustical pipe
that can range over several octave bands. The broad-
lagging is similar to thermal pipe insulation. However,
band, shock-associated peak frequency noise typically
whereas acoustical pipe lagging also provides excel-
occurs at a higher frequency than the screech tone
lent thermal insulation, many thermal insulations
peak frequency.
Noise 255

Nozzle exit

Shock associated
~ noise spectrum

Mixing noise
spectrum

Noise frequency

FIGURE 11.18
Screech tone emissions.

provide only poor noise control. Rigid insulations for noisy (e.g., in steel mills) so that emissions from the
cold service (such as foam glass installed on smaller- burners are not among the primary contributions to
diameter pipes) can actually aggravate the noise situ- the overall noise level.
ation by conducting the noise to the outer surface. The noise emissions from GEe systems strongly
Although acoustical energy radiated per unit area of depend on the burner design in question and the
insulated and jacketed pipe is less than for the same combustion characteristics. As the range of possible
noninsulated pipe, the surface area of an insulated and operational modes of GEe burners is far wider than
jacketed pipe is greater. The product of these two fac- for normal fuel-air burners, so is the range of noise
tors can cause larger-diameter jacketed pipes to radiate emissions that can be observed with GEe burners. For
more noise than bare pipes. 8 example, a variation in the generated noise of more
than 60 dB has been observed for the same GEe burner,
depending on fired duty, fuel-oxygen ratio and mixing
rate. Some exemplary values for the A-weighted sound
power level radiated into the firebox from different
GEe burners are as follows:
11.4 Noise and Oxygen-Enhanced Combustion
• Burner design duty 6 MW: 63 dBA (fired duty
11.4.1 General 0.5 MW), >115dBA (fired duty 3.0 MW)
To date, the noise generation mechanisms specific to • Burner design duty 1.3 MW: 61 dBA (mini-
oxygen-enhanced combustion (GEe) are still not well mum load), 98-108 dBA (nominal load), 118 dBA
understood and there seems to be only little research (overfired)
dedicated to characterization and prediction of their • Burner design duty 3.5 MW: 110 MW (nominal
noise emissions. A possible reason is that GEe burner load), > 125dBA (maximum load)
systems are usually tightly enclosed, which provides
some reduction of the sound transmission from inside For the reasons outlined earlier, it is not possible to pre-
the burner/firebox to the outside. Another reason may dict the noise emissions of GEe burners by simple cor-
be that, in many practical applications, GEe burners relations. Following is a summary of the most important
are operated in environments that are generally very qualitative conclusions that can be drawn from reports
256 Oxygen-Enhanced Combustion

made in the literature or from observations made by

GEe system suppliers and operators:

• Noise generation strongly depends on the

degree of premixing of fuel and oxygen: better
premixing with hot and compact flames close
to the injector nozzles leads to higher noise
generation.
• Separate injection of fuel and oxygen in par-
allel streams-which takes longer to mix and
results in long reaction zones-generates less
noise.
• An extreme example are some applications in
glass or metal reheating where fuel and oxy-
gen are introduced in a way as to distribute
the combustion reaction over the whole firebox
volume, resulting in a very uniform and steady
volumetric heat release. In this case, typically,
very little noise is generated.
• In systems with high injection velocities of fuel
and/or oxygen, more noise is generated.
FIGURE 11.19
• For most systems, the increase in the noise gen- Oxy-fuel burner.
erated becomes less with higher burner loads.

For an optimal combustion, various parameters of the at a constant heat release rate of 10 mmBtu/h (2930 kW)
burner need to be adapted and set correctly, which is with varying concentrations of oxygen.9- 12 The low-
not always possible for combustion with air and with frequency noise is attributed to combustion noise and
oxygen alike in the same burner. Therefore, the noise resides in the frequency range less than approximately
emissions of an GEe system burning the same rate of 1000 Hz. The high-frequency noise is attributed to the
fuel with air or with oxygen are not always comparable gas jet noise and resides in the frequency range greater
as the combustion characteristics may not be optimal in than approximately 1000 Hz. Although the burner heat
both modes of operation. release rate is constant for all cases, notice that as the
An advantage of GEe as compared to combustion in oxygen flow rate increases, the combustion noise also
air is that it is easier to obtain a stable flame in a wider increases; this increase is attributed to higher turbulent
range of operating conditions: This can lead to reduced intensity of the oxidant-fuel mixture. That is, more tur-
noise generation and provides more flexibility when bulence causes the oxidant-fuel mixture to burn more
the system parameters need to be varied in order to rapidly, resulting in more combustion noise. The data
avoid self-excited combustion oscillations, which usu- also show that increasing the flow rate of oxygen also
ally occur at very specific parameter settings only. Last increases the gas jet noise. Notice that as the flow rate
but not least, in terms of acoustics, care also needs to of oxygen increases, the peak value of the gas jet noise
be taken in the design of the fuel and oxygen supply shifts slightly to higher frequencies; this peak frequency
systems of GEe burners: High flow· speeds and control is primarily a function of the gas jet velocity and gas
valves with high-pressure drops may create consider- port diameter. An increase in the gas jet velocity or a
able noise outside of the actual burner and, thus, make decrease in the port diameter results in a shift to higher
a noticeable contribution to the sound pressure levels frequencies.
received in the vicinity. Figure 11.21 is a plot showing the noise spectra from
a diffusion style, oxy-fuel burner firing natural gas at
various heat releases with a constant flow rate of oxy-
11.4.2 Example: Noise Spectra from
gen (2 volumes of oxygen per volume of fuel). Again,
an Oxy-Fuel Burner
the low-frequency noise is attributed to combustion
The noise emitted from an oxy-fuel burner (Figure 11.19) noise and resides in the frequency range less than
is predominantly gas jet noise and combustion roar. For approximately 1000 Hz and the high-frequency noise
example, Figure 11.20 is a plot showing the noise spectra is attributed to the gas jet noise and resides in the fre-
from a diffusion style, oxy-fuel burner firing natural gas quency range greater than approximately 1000 Hz.
Noise 257

Region of Region of
combustion noise gas jet noise
110
~ __~A ~ ____~~~______~
r V ,
Volume O 2
=3
VolumeNG
100
~
~
a:l 2
>-
~
90
~
'0
Z
-----=1
Volume NG
80 Burner heat release = 10 mmBtu/h
O 2 =Oxygen
NG = Natural gas

70~--------~--------~--------~--------~~~

10 100 1,000 10,000 100,000

Frequency (Hz)

FIGURE 11.20
Noise spectra from an oxy-fuel burner firing at a constant burner heat release. (Adapted from Song, X.-R., Experimental study of combustion
noise generated by oxygen-fuel burners, 12th National Conference on Noise Engineering, Williamsburg, VA, pp. 97-102, 1993.)

130 Region of Region of

combustion noise gas jet noise
~__~A.~__~ ~_ _~A~_ _ _~
r V \
110

~ 15 mmBtu/h
~
90
~
~
\
<l.l \
rf)
......... "'"
'0
Z 70 \
\~
4mmBtu/h

Volume02 = 2
50
VolumeNG
1 mmBtu/h
Natural gas
30
10 100 1,000 10,000 100,000
Frequency (Hz)

FIGURE 11.21
Noise spectra from an oxy-fuel burner firing at various burner heat release. (Adapted from Song, x.-R., Experimental study of combustion
noise generated by oxygen-fuel burners, 12th National Conference on Noise Engineering, Williamsburg, VA, pp. 97-102, 1993.)

The plot shows that increasing the burner heat

release increases both the combustion noise and the
Glossary
gas jet noise. Notice that for the low heat release case
(1 mmBtu/h)f the noise is dominated by the combus- Absorption: Conversion of sound energy into another
tion noise. Howeverf as the heat release increasesf the form of energYf usually heatf when passing
gas jet noise becomes more prominent. At a burner through an acoustical medium.
heat release of 15 mmBtu/hf the gas jet noise domi- Absorption coefficient: Ratio of sound absorbing effec-
nates over the combustion noise. tivenessf at a specific frequencYf of a unit area of
258 Oxygen-Enhanced Combustion

acoustical absorbent to a unit area of perfectly N ear field: That part of a sound field, usually within
absorptive material. about two wavelengths of a noise source, where
Acoustics: Science of the production, controt transmis- there is no simple relationship between sound
sion, reception, and effects of sound and of the level and distance.
phenomenon of hearing. Noise emission level: dBA level measured at a specified
Ambient noise: All-pervasive noise associated with a distance and direction from a noise source, in
given environment. an open environment, above a specified type of
Anechoic room: A room whose boundaries effectively surface; generally follows the recommendation
absorb all incident sound over the frequency of a national or industry standard.
range of interest, thereby creating essentially Noise reduction coefficient (NRC): Arithmetic average
free field conditions. of the sound absorption coefficients of a mate-
Audibility threshold: Sound pressure levet for a speci- rial at 250, 500, 1000, and 2000 Hz.
fied frequency, at which humans with normal Phon: Loudness level of a sound, numerically equal
hearing begin to respond. to the sound pressure level of a 1 kHz free
Background noise: Ambient noise level above which progressive wave, which is judged by reliable
signals must be presented or noise sources listeners to be as loud as the unknown sound.
measured. Pink noise: Broadband noise whose energy content is
Decibel scale: Linear numbering scale used to define a inversely proportional to frequency (-3 dB per
logarithmic amplitude scale, thereby compress- octave or -10 dB per decade).
ing a wide range of amplitude values to a small Power spectrum level: Level of the power in a band 1 Hz
set of numbers. wide referred to a given reference power.
Diffraction: Scattering of radiation at an object smaller Reverberation: Persistence of sound in an enclosure after
than one wavelength and the subsequent inter- a sound source has been stopped. Reverberation
ference of the scattered wavefronts. time is the time (in seconds) required for sound
Diffuse field: Sound field in which the sound pressure pressure at a specific frequency to decay 60 dB
level is the same everywhere, and the flow of after a sound source is stopped.
energy is equally probable in all directions. Root mean square (RMS): The square root of the arith-
Diffuse sound: Sound that is completely random in metic average of a set of squared instantaneous
phase; sound that appears to have no single values.
source. Sabine: Measure of sound absorption of a surface. One
Directivity factor: Ratio of the mean-square pressure metric sabine is equivalent to 1 m 2 of perfectly
(or intensity) on the axis of a transducer at a absorptive surface.
certain distance to the mean-square pressure Sound: Energy transmitted by pressure waves in air
(or intensity) that a spherical source radiating or other materials, which is the objective cause
the same power would produce at that point. of the sensation of hearing: commonly called
Far field: Distribution of acoustic energy at a much greater noise if it is unwanted.
distance from a source than the linear dimensions Sound intensity: Rate of sound energy transmission
of the source itself. See also diffraction. per unit area in a specified direction.
Free field: An environment in which there are no reflec- Sound level: Level of sound measured with a sound level
tive surfaces within the frequency region of meter and one of its weighting networks. When
interest. A-weighting is used, the sound level is given in dBA.
Hearing loss: An increase in the threshold of audibil- Sound level meter: An electronic instrument for mea-
ity due to disease, injury, age, or exposure to suring the RMS of sound in accordance with an
intense noise. accepted national or international standard.
Hertz (Hz): Unit of frequency measurement, represent- Sound power: Total sound energy radiated by a source
ing cycles per second. per unit time.
Infrasound: Sound at frequencies below the audible Sound power level: Fundamental measure of sound
range, that is, below about 16 Hz. power, defined as:
Isolation: Resistance to the transmission of sound by P
materials and structures. Lw = 10log-dB
Pa
Loudness: Subjective impression of the intensity of a
sound. where
Masking: Process by which the threshold of audibil- P is the RMS value of sound power in units
ity of one sound is raised by the presence of of watts
another (masking) sound. Pais 1 pW
Noise 259

sound pressure: Dynamic variation in atmospheric 3. W. Daiminger, KR Fritz, E. Schorer, and B. Stuber,
pressure. The pressure at a point in space minus Ullmann's Encyclopedia of Industrial Chemistry, Vol. B7,
the static pressure at that point. VCH, Weinheim, Germany, pp. 384-401, 1995.
sound pressure level: Fundamental measure of sound 4. A. Thumann and RK Miller, Secrets of Noise Control,
Fairmont Press, Atlanta, GA, 1974.
pressure defined as
5. L.L. Beranek and LL. Ve'r, Noise and Vibration Control
Engineering, John Wiley & Sons, Inc, New York, 1992.
P 6. H. Shen and C.KW. Tam, Numerical simulation of the
Lp = 20log - dB
Po generation of axisymmetric mode jet screech tones, AIAA
J., 36(10), 1801, October 1998.
7. L.L. Beranek, Noise and Vibration Control, McGraw Hill
where Book Co., New York, 1971.
P is the RMS value (unless otherwise stated) 8. L.D. Frank. and D.R, Dembicki, Lower plant noise with
of sound pressure in units of Pascals lagging, Hydrocarb. Process., 71(8), 83-85, 1992.
9. X.-R Song, Experimental study of combustion noise
Po is 1 pPa
generated by oxygen-fuel burners, 12th National
Conference on Noise Engineering, Williamsburg, VA, pp.
Sound transmission loss: Ratio of the sound energy 97-102, 1993.
emitted by an acoustical material or structure 10. S. Laux, Private communication, R&D Combustion,
to the energy incident upon the opposite side. Praxair, Inc., Danbury, CT, 2011.
Standing wave: A periodic wave having a fixed distri- 11. T. Niehoff, Private communication, Linde Gas, Non
bution in space that is the result of interference Ferrous and Mining, Germany, 2011.
of progressive waves of the same frequency and 12. J. von Scheele, Private communication, BOC India
kind; characterized by the existence of maximum Limited-A Member of The Linde Group, India, 2011.
and minimum amplitudes that are fixed in space.
Thermoacoustic efficiency: A value used to character-
ize the amount of combustion noise emitted
from a flame; defined as the ratio of the acousti-
cal power emitted from the flame to the total
heat release rate of the flame. Bibliography
Ultrasound: Sound at frequencies above the audible American Petroleum Institute, 50,125-146,1972.
range, that is, above about 20 kHz. RS. Brief and R G. Confer, Interpreting noise dosimeter results
Wavelength: Distance measured perpendicular to based on different noise standards, Am. Indust. Hygiene
the wavefront in the direction of propagation J., 36(9),677-682, 1975.
between two successive points in the wave, S.c. Crow and EH. Champagne, orderly structure in jet turbu-
which are separated by one period; equal to the lence, J. Fluid Mech., 48(3), 547-591, 1971.
ratio of the speed of sound in the medium to the A.H. Diserens, Personal noise dosimetry in refinery and chem-
ical plants, J. Occup. Med., 16(4),255-257,1974.
fundamental frequency.
A. Gharabegian and J.E. Peat, Saudi petrochemical plant noise
Weighting network: An electronic filter in a sound level
control, J. Environ. Eng., 112(6), 1026-1040, 1986.
meter used to approximate, under defined con- HFP Acoustical Consultants, Effect of flow parameters on flare
ditions, the frequency response of the human stack generator noise, Proceedings of Spring Environmental
ear. The A-weighting network is most com- Noise Conference: Innovations in Noise Control for the Energy
monly used. Industry, Banff, Alberta, Canada, April 19-22, 1998.
White noise: Broadband noise having constant energy IEC 61672-1, Electroacoustics-Sound level meters-Part 1:
per unit of frequency. Specifications, 2002.
ISO 1683, Acoustics-Preferred reference values for acoustical
and vibratory levels, 2008.
ISO 532, Acoustics-Method for calculating loudness level, 1975.
ISO 1996-1, Acoustics-Description, measurement and assess-
ment of environmental noise-Part 1: Basic quantities
and assessment procedures, 2003.
References
ISO 3744, Acoustics-Determination of sound power levels
1. A.P'G. Peterson, Handbook of Noise Measurement, 9th edn., and sound energy levels of noise sources using sound
GenRad, MA, 1980. pressure-Engineering methods for an essentially free
2. ISO 3744, Acoustics-Determination of sound power field over a reflecting plane, 2010.
levels and sound energy levels of noise sources using ISO 8297, Acoustics-Determination of sound power levels of
sound pressure-Engineering methods for an essentially multisource industrial plants for evaluation of sound pres-
free field over a reflecting plane, 2010. sure levels in the environment-Engineering method, 1994.
260 Oxygen-Enhanced Combustion

ISO 9613-1, Acoustics-Attenuation of sound during propaga- R. Reed, Furnace Operations, Gulf Publishing, Houston, TX, 1981.
tion outdoors-Part 1: Calculation of the absorption of HS. Ribner, Perspectives on jet noise, AIAA J., 19(12), 1513,
sound by the atmosphere, 1993. 1981.
ISO 9613-2, Acoustics-Attenuation of sound during propaga- J.P. Roberts, PhD thesis, London University, London, U.K.,
tion outdoors-Part 2: General method of calculation, 1996. 1971.
ISO 9614-1, Acoustics-Determination of sound power lev- G. Seebold and AS. Hersh, Control flare steam noise,
els of noise sources using sound intensity-Part 1: Hydrocarb. Process., 51, 140, 1971.
Measurement at discrete points, 1993. G.K. Selle, Steam-assisted flare eliminates environmental con-
N. Krishnamurthy et al., Studies on low-intensity oxy-fuel cerns of smoke and noise, Hydrocarb. Process., 73(12),
burner, Proc. Combust. Inst. 32,3139-3146,2009. 77-78,1994.
W.W. Lang, Ed., A commentary on noise dosimetry and stan- B.N. Shivashankara, W.e. Strahle, and J.e. Henkley,
dards, Proceedings of Noise Congress-1S, Gaithersburg, Combustion noise radiation by open turbulent flames,
MD, September 15-17, 1975. Paper 73-1025, AIAA Aero-Acoustics Conference, Seattle,
J.e. Maling, Jr., Ed., Start-up silencers for a petrochemical com- WA,1973.
plex, Proceedings of the International Conference on Noise J.E Straitz, Improved flare design, Hydrocarb. Process., 73(10), 61---66,
Control Engineering, December 3-5,1984. 1994.
A. Powell, On the noise emanating from a two dimensional jet A. Thomas and G.T. Williams, Flame noise: Sound emission
above the critical pressure, The Aeronaut. Quart., 4, 103, 1953. from spark-ignited bubbles of combustion gas, Proc. Roy.
A.A. Putnam, Combustion noise in the hand glass industry, Soc., A294, 449, 1966.
Tenth Annual Symposium on the Reduction Cost in the Hand E. Zwicker and H. Fastl, Psychoacoustics-Facts and Models,
Operated Glass Plants, WV, 1979. Springer Verlag, Berlin, Germany, 1990.