Académique Documents
Professionnel Documents
Culture Documents
Doctoral Dissertations
1972
This Dissertation - Open Access is brought to you for free and open access by Scholars' Mine. It has been accepted for inclusion in Doctoral
Dissertations by an authorized administrator of Scholars' Mine. This work is protected by U. S. Copyright Law. Unauthorized use including
reproduction for redistribution requires the permission of the copyright holder. For more information, please contact scholarsmine@mst.edu.
by
A DISSERTATION
UNIVERSITY OF MISSOURI-ROLLA
DOCTOR OF PHILOSOPHY
in
MECHANICAL ENGINEERING
1972
Adv1sor
T2800
222 pages
c .l
ii
ABSTRACT
A description
A region character-
ized by high input levels was found in which the resistance becomes
a linear function of the flow Mach number.
Empirical factors to
account for effects of finite amplitudes and steady flow are presented.
In addition to the impedance measurements conducted on the resonators,
response measurements were obtained and compared to those generated
using the empirical impedance expressions for both cold and heated
flow operations.
Measure-
iii
iv
ACKNOWLEDGEMENTS
My thanks to
Dr. W.S. Gatley for suggesting this project and the National Science
Foundation for the necessary funding.
I would like to especially thank Mr. Dick
Smith~
and Mr. Lee Anderson whose help in constructing the laboratory facility
and solving the day to day operational problems was sincerely appreciated.
I am especially
indebted to Dr. R.D. Rocke, Dr. L.G. Rhea, and Dr. L.R. Koval.
Three men who are not on the Rolla campus must also be mentioned.
They are Mr. Robert Brossemer, Mr. Pat Depew and Mr. Harold Shaw.
What instrumentation skill I may possess I owe to their tutelage and
only with this knowledge was the experimental work possible.
Finally, I wish to acknowledge and thank my wife Ruth Ann who
has persevered with me throughout this work.
TABLE OF CONTENTS
Page
ABSTRACT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
ii
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
iv
XV
I .
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
I I.
LITERATURE REVIEW. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A.
B.
C.
III.
11
E.
D.
14
19
F.
22
G.
Conclusions..........................................
25
EXPERIMENTAL APPARATUS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
27
A.
27
B.
32
C.
34
D.
37
vi
Page
IV.
V.
E.
42
F.
~................
44
EXPERIMENTAL PROCEDURES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
48
A.
48
B.
54
C.
59
61
A.
61
B.
67
C.
70
1.
Finite-Amplitude Variations.....................
70
2.
D.
VI.
78
84
1.
Finite-Amplitude Variations.....................
84
2.
95
E.
99
F.
130
132
A.
B.
132
140
vii
Page
C.
D.
1.
2.
E.
VII.
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
BIBLIOGRAPHY.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
VITA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
A.
B.
C.
D.
E.
F.
viii
LIST OF FIGURES
Figure
1.
Page
28
2.
30
3.
31
4.
36
5.
40
6.
41
7.
43
8.
45
9.
50
10.
53
11.
56
12.
Side-Branch Junction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
63
13.
71
14.
80
15.
16.
17.
18.
R-~
Neck
Diameter = 0. 375 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
19.
ix
Figure
20.
Page
21.
22.
97
98
25.
24.
93
23.
92
101
26.
27.
106
32.
3.96 inch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
d 0 = 1.50 inch, LB
31.
3. 96 inch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
30.
d 0 = 1.50 inch, LB
29.
28.
= 0.750 inch, LB
109
Figure
33.
Page
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
xi
Figure
45.
Page
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
xii
Figure
58.
Page
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
xiii
Figure
71.
Page
72.
73.
74.
75.
0.042 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
76.
0. 042. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
77.
A-1
B-1
C-1
D-1
E-1
xiv
LIST OF TABLES
Table
Page
I.
Resonator Configurations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
II.
72
77
V.
85
90
Resistance Coefficients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
94
XV
NOMENCLATURE
A+
Cf
d0
e
f
fc
fn
direction
xvi
max
min
p
max
min
pb
p.
p0
pr
pt
RB
acoustic resistance
Rbv
Rs
R
v
R .
VlS
Rey
Reynolds number
cross-sectional areas of tubing
xvii
max
min
location coordinate
location of standing wave maximum pressure locations relative
minimum
to tube termination
acoustic impedance
sl
Sr
o1
1
r
A
~
wavelength of sound
no-flow wave length of sound
area ratio
P0
mass density
angular velocity
I.
INTRODUCTION
increasing awareness
In
However, because of
One of
discrete pure tone through the filter and measuring its response.
By conducting a set of these tests, the filter response is obtained
over a desired frequency range.
Gatley (1)* employed this procedure to design a muffler system
for a small refrigeration unit.
It was decided
of system occurs at frequencies below 1000 Hz, this region was selected
for investigation.
In addition, it was decided to extend the investigation by heating the flow medium.
II.
LITERATURE REVIEW
One-dimensional linear theory has been used for some time for
the design of acoustic silencers and silencer elements.
However,
means has yet been devised to account for the deviations from linear
theory which occur in practice.
Nonlinear response of elements arise from a variety of effects,
namely; finite amplitudes of the acoustic waves, varying temperature,
a superimposed flowing medium, and sizing of elements so that some
frequencies have wave lengths violating the assumption of plane waves
(one-dimensional) used in linear theory.
Another factor which adds to the "black-art" concept of silencer
design is the techniques and terminology for describing the response
of these filters, e.g., the same term is used interchangeably by
different authors to describe different response parameters.
A.
This theory,
The acoustical
plane wave theory to describe the sound propagation into and out of
the filter elements.
The simplest form of wave motion propagated through a medium is
that of the plane waves of sound.
nal wave, i.e., the molecules transmitting the wave move back and forth
in the direction of propagation.
0.82D
( 2. l)
where:
A
The assumptions
necessary for the application of this theory are listed now in order
to show the limitations of this theory when applied to non-idealized
problems.
a)
b)
c)
The amplitude of the waves are small such that the acoustic
pressure is much less than the equilibrium pressure and the
incremental density is much less than the ambient densitv
of the fluid.
d)
The propagated waves cause adiabatic and reversible compressions and expansions.
h
8t
c2
82~
3t
2
3 u
2
3t
where:
~
dX 2
2 32~
2
3x
( 2. 2)
( 2. 3)
2
2 3 u
dX
(2.4)
acoustic pressure
particle displacement
particle velocity
p+
p+ e
~+
A+ e
ik(x-ct)
(2.6)
ik(x-ct)
u+ e
( 2. 7)
( 2. 5)
t~1be
when waves
p(x,t)
P+ e
ik(x-ct)
+ P
-ik(x+ct)
( 2. 8)
Davis and his colleagues (7) conducted a very extensive investigation of a large number of acoustic filters.
types of expansion chambers and resonators both as individual components and in combinations.
The con-
The
a change in cross-sectional area, some of the incident wave is reflected back toward the source as a plane wave and some is transmitted
through the discontinuity in the form of a plane wave.
obtained by this method are used to determine
11
Measurements
acoustic impedance'',
(3) presented the complete theory, accounting for both thermal and
This
CD
(2.9)
[L +(y-l)L ]
v
where:
Lv
em
em
With this constant, the value of the sound pressure of a plane acoustic
wave traveling in a waveguide is given for any location x bv:
-ax
Po e
where:
10
Many investigatlons have been conducted to experimentally confirm this constant (5,8,9).
10- 5
3.18
frequency in Hz
radius of tube in em
(2.10)
where:
He determined
that very little change occurred in this constant for loH Mach numbers
11
(MSO.l) and that for higher Mach numbers (M>O.l) the value was increased up to two to three times the no-flow value.
The waveguide
1)
2)
3)
They are:
This is
It is
The reactance
12
It is analogous to an electrical
Many investiga-
:~specific
acoustic impedance"
property of the medium, it is very useful Hhen discussing the transmission of sound waves from one medium to another.
Also, if by some
means one can determine the sound pressure of a free traveling wave at
a point in the medium, the particle velocity may also be determined
for that location from the following relationship:
13
( 2. 11)
The unit of
The magnitude of
This is
The "transmission
It
14
ATTN
( 2. 12)
Since, for the case of plane waves, the energy of a plane wave is
directly proportional to the square of the pressure, Equation (2.12)
can be written in terms of pressures as:
ATTN
20 log
10
Pincident
PTransmitted
(2.13)
15
Garrison
Region 2.
Region 3.
Region 4.
16
(2.14)
R
B
He determined that it
The correla-
tion was accomplished by establishing an analogy between flow resistance of an orifice and the acoustic resistance of the orifice.
Orifice flow resistance was defined by:
(2.15)
where:
P
P
urn
cf
z
u
17
where:
P
P
u
1
2
0
(2.16)
as the length of the column of air within the orifice and extending
beyond either side of it which contains the effective mass of air.
This mass oscillates with the acoustic wave.
which causes the orifice to have reactance.
18
effective length was the same as that used in the linear resistance
region.
(2.17)
where:
t
d0
If a simple rela-
tionship such as given for the orifice could be determined for sudden
changes in cross-sectional area, it would aid in simplifying their
design as components of acoustic filters.
Ronneberger (20), during his studies of the acoustic characteristics of ducts with sudden changes in cross-sectional area, obtained
the alternating pressure profile across expansions and also through
orifices.
a "loss coefficient 11 ,
19
E.
which could be applied to the series solution so that only the first
term of the series could be used.
across the area change were written and perturbations in the form of
an acoustic pulse were then applied.
the fact that both an incident and a reflected pulse were present
upstream from the change.
the downstream region which implied that the change in area was connected
to an infinitely long section or an anechoic termination.
The reflec-
appropriate ratios.
Only
20
development.
The
b)
c)
d)
Po1~ell
Good agreement
21
Powell's theory.
ratio of the reflection factor with flow to that with no flow was independent of frequency and dependent only on the Mach number of the
incoming flow.
stage of this study, no information on the change in transmission factor with flow could be obtained for comparison to theory.
The steady state approach was applied by Ronneberger (20) to
sudden changes in cross-sectional areas.
momentum equations for the flow were written and small perturbations
representing the acoustic waves were applied.
An additional perturba-
tion term was introduced with the assumption that an entropy wave
was generated at the cross-section change and propagated with flow.
This perturbation appeared in the density term.
A general expression
22
A transition
region existed for each expansion which was independent of the Mach
number of the flow.
length for the perforated plate allowed a frequency range where the
wave-length of the acoustic input was greater than the length to be
investigated.
expression were in good agreement with the experimental data for the
perforated plate.
It was concluded that the developed expression was valid only if the
wavelength of the sound remained longer than the static pressure
transition length.
In his investigations of
23
made using the analogy described earlier for finite amplitude investigations.
The particle velocity within the orifice was determined to be
the governing parameter for the acoustic resistance for the case of
steady flow through the orifice.
(2.18)
R
v
where:
vt
high incident sound pressure levels, i.e., 160 db for his experiment,
the effective length of the orifice was the same as for the no-flow case.
24
Rv
~F (1+1.9MP)
(2.19)
(2.20)
where:
MP
The effectiveness of
For
25
impedance ... increases strongly and linearly with increasing flow velocity.11 He further stated that the critical parameter influencing the
change of the impedance with the flow was the particle velocity within
the orifice.
The orifice sizing employed by Garrison was such that the transition length of the static pressure change could be assumed small compared to the wavelengths of the acoustic inputs.
tion characteristics of test mufflers with and without steady flow were
obtained.
Conclusions
Linear acoustic filter theory is widely used in the design of
26
analysis were used for the high frequency work which helped to add
credance to the planned use of this method for the present investigations.
investigators was in substantial agreement for a wide variety of resonator neck diameters.
In addition, a flow-loss-coefficient
27
I I I.
EXPERIMENTAL APPARATUS
The standing wave tube method has been used by many investigators for response measurements of acoustic filters.
Simon (11)
designed and constructed the two-inch I.D. standing wave tube which
was used for the present investigation.
A slot, 52 inches
long by 1/8 inch wide, was then machined along the center of the flat
Teflon strips, 1/4
inch wide by 0.012 inch thick, were installed on the flat along both
sides of the slot to facilitate sealing.
attached to the steel band by two dowell pins, four small machine
screws and gasket cement.
was machined into the block; a probe fabricated from #12 hypodermic
tubing extended from the cavity into the tube.
passed through the steel band into the block, and a cigar shaped body
was silver-soldered to the other end.
the slot and into the interior of the tube until the center of the body
coincided with the center of the aluminum tube.
machined so that the inlet to the microphone cavity was located five
body diameters from the leading nose as shown in Figure 1.
This
+
Microphone Block
Extension
Tube
Original
Probe Tube
Probe
Body
D1
I~
Figure 1.
SD
00
29
A sketch of the
Figure 3.
Brass guide rails were used to maintain the microphone block
assembly and steel band in position over the slot.
extended the length of the flat area on either side of the slot and
were maintained in place by attachment to eight equally spaced cradle
supports.
tube and were bolted to a section of steel channel used as the base.
The sealing of the tube was accomplished by pressurizing two
lengths of surgical tubing placed immediately over the steel band.
The surgical tubing was constrained within brass angles attached to
the guide rails are shown in Figure 2.
helped to reduce the effort required to move the block when the tubing
was pressurized.
30
Probe
Body
Figure 2.
..
'
Cl)
='
.0
Cl)
f-'
;;
bO
3:
t::
ctl
t::
r-i
'"0
4-J
U)
4-1
..c::
p...
ctl
1-4
()()
4-J
..c::
p..,
~
r-i
::>
Cll
'"0
r-i
U)
C"')
Cl)
bO
='
1-4
r-i
31
32
ing pressures it was somewhat difficult to move the block; this provided impetus for the design of a mechanical drive system.
A roller positioned at each end of the flat maintained pressure
on the steel band and provided the seal at the end of the slot.
After
the steel band passed under the rollers it was directed away from the
tube and over an aluminum pulley mounted above the tube.
Thus each
end of the steel band ended above and parallel to the flat at which
point they were attached to each other by means of two small bolts.
The junction was located so as to prevent the bolts from coming into
contact with the guide pulleys and in this way provided a continuous
seal for all positions of the microphone block along the tube axis.
B.
The
air was supplied to the laboratory via 2-inch piping at 90 PSIG and
a nominal temperature of 75F.
At the exit of the supply line globe valve, the air line was
connected to a 150 cfm separator in an attempt to remove any oil
and water vapor present in the supplied air.
33
was installed within the plenum tank to enable the flow medium to be
heated to a maximum of 200F with the maximum flow rate tested (approximately 150 cfm).
A straight section of
pipe of ten diameters long was provided before the orifice and another
section of twenty diameters was provided downstream of the orifice.
This insured fully developed flow within the metering section.
and used to determine the volumetric flow rate and the mean flow
velocity.
analysis program.
By following the ASTM code and using an appropriate itteration
technique an accuracy of 2.% was obtained for the flow rate.
After the flow was metered the supply was passed through two
large expansion chambers.
34
entrance into the standing wave tube the temperature and static
pressure of the gas were again measured so that any changes in the
Mach number of the medium between the flow-metering section and
the tube entrance could be determined.
The flow-piping system between the plenum tank and the standing
wave tube as well as the termination tubing of the system was insulated using conventional magnesium pipe insulation.
The standing
tain a good seal between the steel band and the standing-wave tube
caused the microphone block to be difficult to move.
To alleviate
35
stalled over the acme thread nuts and constrained to move with the
nuts by using a pair of set screws.
The
belt drive for this pulley was driven from a set of speed-reducer
pulleys installed on an intermediate shaft and another belt drive to
a 1/3
Hp electric motor.
Hount
Bolt
-- -----
i'i
;'
IEf,
Set Screw
Acme Thread
Nut
Drive BJock
Screw
Hicrophone
Block Assembly
Figure 4.
(j\
37
obtainable with the pulley system; however, all operations were made
with a reduction of 15:1.
be the maximum speed of operation for which the sound pressure observed when the microphone was traversed by a location, and that observed at the same location with the microphone stationary, were
identical.
The position of the microphone block could be adjusted by hand
by turning the intermediate drive pulley shaft.
the discharge of the test elements was required to aid in the analysis
of experimental data.
match between the acoustic impedance of the standing wave tube and
the air of the room is obtained by using a horn having a catenoidal
38
profile.
where
fc
c
2Tih
(3.1)
cut-off frequency, Hz
Wood (27), on the other hand, states that the horn may be considered reflection free for the above-cut-off frequency only if the diameter
of the discharge of the horn approaches 4/TI wave lengths for this
frequency.
so that all the sound is transmitted into the air by providing a nearly
exact impedance match.
An expansion rate equivalent to a cut-off frequency of 45.5 Hz
was used in the design of the anechoic termination.
The criterion of
Wood could not be achieved because of the physical size which would
be involved; the horn would have to be approximately 21.2 feet long
with an exit diameter of 18.2 feet.
Myers (28) had determined that by partially filling the horn
with an insulating material he could achieve a satisfactory anechoic
termination without meeting the exit size requirement of Wood.
It
39
length was attached to the exit end of the caternoidal profile to give
a total length of 13 feet.
sealed using a
A collar was
installed at the throat of the horn in order to provide easy connection to the standing-wave tube system.
The interior of the horn was filled with glass fiber (approximately 2.5 lb/cubic feet density) wrapped around a two-inch I.D.
sleeve of 1/4-inch square wire mesh.
mately 18 inches from the throat of the horn and continued the entire
length to the exit.
Glass Fiber
Fill
r.
.tl
18in.
~1
NOT TO SCALE
2 in. diam.
_I
:X..:.
:L L
~ ~
_-L_ _-'-__
---
13 ft.
Figure 5.
Figure 6.
1-'
42
E.
These
In addition, the
A new
The cylinder was bored to provide a slip fit for the micro-
phone.
phone block so that the microphone face could not come in contact
43
Seal 0-Ring
Steel Collar
Steel Mount
Cylinder
Epoxy Filling
Figure 7.
Original
Mount Block
44
same O.D. as the cylinder and bored for a tight slip fit of the
microphone.
two assemblies.
Clay was also used to seal the set screw hole pre-
The micro-
phone was then inserted into the holding cylinder and locked in
place using four 10-32 screws.
Instrumentation System
The microphones used were of the piezoelectric-ceramic type.
diaphr~m
In
tal and when pressure is applied to the diaphragm, the stressing of the
crystal causes it to produce an electrical signal.
The
GR* 313A
Audio
Oscillator
~
Dynakit
40 Watt
Amplifier
GR 1560-PS
Microphones
J.B. Lansing
Acoustic Driver
,--
r-1
1._
B &K
Instrument
2406
VTVM
GR 1551
Sound
.......--+Level
Heter
GR 1521B
Graphic
Level
Recorder
AD-Yu
4066
L.--
Phase
Meter
r--
Figure 8.
Oscilloscope
Tektronic
Dual Beam
Selector
Box
L__
Out
In
GR 1564A
Wave
Analyzer
*GR:
General Radio
Vl
46
This is the so
Also, when
high amplitude sound waves were propagated in the tube there were
always harmonics of the primary frequency present.
The input to the wave analyzer was also connected to one channel
of a dual-trace oscilloscope to provide a visual monitor of the data.
The filtered output of the analyzer was connected to the second
channel of the oscilloscope to provide a monitor to insure that the
analyzer was performing properly.
47
The sound-
level meter was used only as an amplifier so that the input signal
to the phase meter channel A could be maintained at a level of 800
mv 200 mv, for improved accuracy in phase measurements.
The oscillator signal was also connected to a Dynakit 40 watt
amplifier which in turn was connected to the acoustic driver which was
used to generate pure tone signals in the standing wave tube.
maximum amplitude of approximately 153 db re 0.0002
~bar
at the
entrance to the test elements was obtained for the frequency range
200 to 600 Hz.
48
IV.
EXPERIMENTAL PROCEDURES
One
test procedures.
A.
If the
response of the microphone probe did not change with variation of the
incident sound pressure amplitude, no error was encountered.
This
lt was felt
Also, a neces-
49
This
along the longitudinal axis such that a slip fit would occur between
the plug and an installed microphone.
was flush with the face of the plug and perpendicular to the axis of
the tube.
The details of
was anticipated, meaning that the incident wave was totally reflected.
In the development of the standing wave tube analysis detailed
in Appendix A, it is shown that the sound pressure amplitude at a
standing wave maximum is described by the following expression:
max
ax
le
max + Ble
-ax .
mln
( 4. 1)
Where A, and B, are the magnitudes of the incident and reflected waves
respectively.
max
A [e
1
max
+ e
-ax
max]
(4.2)
Locking Sleeve
(Threaded into Piston)
Compression and
Spacer Sleeve
Microphone
Seal 0-Ring
Figure 9.
Piston Plug
lJI
0
51
-1
max
This
analysis was used as the basis for the microphone-probe response calibration.
In the experiments, the oscillator was set at a desired frequency and the output level increased until a sound pressure level of
140 db was observed at the end-plug microphone.
level which could be measured with the microphones that were used.
A standard reflection-factor analysis was made of the standing
wave.
The
magnitude of the sound pressure at the maximum nearest the termination was obtained and compared to the level at the end-plug.
This
gave the response of the microphone probe for the given input frequency.
By adjusting the oscillator output level, changes in the probe
response with varying acoustic wave amplitudes were determined for
the preset frequency.
52
With
at the nearest one-half wave length distance from the fixed position
of the transmission microphone.
The amplitude of the sound pressure was determined at both
microphone locations.
The difference in
1.2
l.l
:::>
r- 1.0
z(9 0.9
~ 0.8
0.7
~
I-
0.6
~ 0.5
u..
0.4
Heasured
Theory
0.3
r- 02
w
_J
0.1
LL.
()
~ o.o 0.0
0.1
Figure 10.
0.2
0.3
0.4
0.7
0.5
0.6
FREQUENCY (KHZ)
0.8
0.9
1.0
Vl
54
the amplitudes gave the relative difference between the two probe
configuration responses.
The dif-
ference between the measured phase angles are also compared to the
theoretical value.
(4.3)
where:
not change with varying incident pressure; i.e., the response was
independent of amplitude.
The calibrations were used in the data-analysis computer program
to correct the test information for the probe responses.
Using this
55
Figure 11.
time.
the precision counter which was also used for continuous monitoring
throughout the test operation.
The position of
the minimum was then designated as the mean of the two measurements.
The microphone block was positioned at the minimum using the
hand adjustment of the drive system.
Flow Metering
Section
Plenum
Containing
6 Strip
Heaters
Acoustic Driver
Test Element
Transmission
Tube
Figure 11.
Anechoic Termination
ln
(J'\
57
min 2
where:
(4.4)
min 1
Thus, if the Mach number is known, the wavelength of the sound source
can be found.
(4.5)
Similarly, if only one minimum is located within the measurement
section, the maximum can be located when the wavelength of the sound
is known.
c0
C[l -
0 579
] em/sec
D(Tif)l/Z
(4.6)
58
( 4. 7)
The values
and locations of the minimum nearest the tube termination and its
adjacent maximum were used as inputs to the computer program for
calculation of the reflection factor and the acoustic impedance of
the test element.
tested~
the fre-
quency range investigated was from approximately one-half the resonant frequency to twice the resonant frequency.
For an expansion
chamber, the range of interest was based on the length of the chamber,
Lc; specifically, the frequency range for which the sound wavelength
is one-half Lc to twice Lc was used.
When the reflection factor was computed, it was assumed that
the attenuation constant was given by Beranek's expression,
Equation (2.10), for both static and steady flow conditions.
59
The
maximum flow rate of the present system was below this value.
Also,
tube and a transmission tube with a fixed microphone station was employed
in the present work.
test set-ups and helped reduce the influence of flow parameter variations which could have caused significant errors.
As the final step of the reflection factor measurements, the
standing wave microphone was positioned at the maximum pressure location.
between the filtered signal and the amplified oscillator output was
recorded from the phase meter.
amplified for this measurement so that the phase meter had two inputs
of similar voltage.
their slopes were also nearly equal thus increasing the accuracy of
the data.
The switch box was used to connect the fixed transmission tube
microphone to the input of the analyzer.
of the filtered signal were recorded.
60
The stand-
ing wave maximum was selected to improve the accuracy of this calculation.
Also,
61
V.
Side branch resonators have been used for many years as components
in acoustic filtering systems.
The rrir;arv
At
P0 c
; which produces a reflected
s
wave.
In addition, part of the incident acoustic energy may be
transmitted into the branch and dissirated there.
transmitted beyond the side branch and dm,Jn the \vaveguide mav be
reduced by one or both of these mechanisms.
62
If a sinusoidally excited
plane wave incident on the left hand side of the side branch is given
by
P.
-i (wt-kx)
( 5 .1)
( 5. 2)
-i(wt-kx)
( 5. 3)
pi
Al
Pr
B] e
pt
- iult
-iwt
-iwt
( 5. 5)
(lf
..0
p..
1"4
0..
.j...J
cu
p..
U)
l-i
0..
-..
63
64
p.
l
(5.6)
From the definition of acoustic impedance the volume velocities associated with each of the waves may also be obtained:
u.
l
pi/poc/s
ur
pr/p 0 c/s
ut
pt/poc/s
ub
pb/Zb
zb
( 5. 7)
where
Equations (5.6) and (5.8) can now be combined to give the more useful
continuity condition of continuity of impedance.
(5.9)
The left hand side of Equation (5.9) is the inverse of the impedance
looking into the junction from the upstream side.
Bv using Equations
65
z=
poe
(5.10)
(5 . 11)
where
(5 . 12)
The impedance of the side branch can be written in its two basic
components , resistance and reactance .
( 5 . !3)
66
p c
(5 . 14)
(--o- + R )2 + X 2
2s
(5 .15)
area , called
varying wave then yields the average energy p r esent in the wave .
servation
Con-
following expression :
(5 . 16)
ju~cture .
It is obtain ed as a function of the impendance of the branch by combining Eq uations (5.14), (5 . 15), and (5.16) , and is given by
p c
<+)
~
(5.17)
It is desirable to
67
HoJ~1oltz
r esonator
syste~
response.
are assumed .
w~tv es
in
( 2]JpW)
(5.18)
68
R .
VlS
4R ( t+d)
s s d
0
where:
t
s0
(5.19)
69
wM
(5.20)
0
where:
wM
mass reactance
density of medium in-neck
This
is due to the fact that the oscillating mass is contained not onlv
in the neck but actually extends slightly beyond both ends of the
neck.
~1
is given by:
+ .85d
(5.21)
It is defined by
we
(5.22)
70
where:
we
compliant reactance
p 1
w o eff
s
C.
p c
--wv
( 5. 2 3)
First,
By varying
response testing; however, a mechanical failure occurred in its construction so that no conclusive data were obtained.
71
Backing Piston
Microphone
Holder Assembly
Sealing 0-Ring
5 in. I. D.
1/4 in. Wall
Aluminum
Cylinder
Resonator
Base
Connecting
Tube
Figure 13.
72
Table I .
Resonator
Number
Neck
Length
(in)
Resonator Configurations
Neck
Diameter
(in)
Cavity
Diameter
(in)
Back i ng
Depth
(in)
R-1
1.00
o. 750
5.047
1.67-6 . 55
R-2 *
1.00
1 . 250
5 . 047
1.67-6 . 55
R-3
1. 0 0
1 . 500
5.047
1. 95-5 . 36
R-4
0 . 953
0.375
2.000
1. 94-5.4 1
Using the expression for the reactance of the resonato r p reviously developed and the definition of resonance as the frequency for
which the reactance is zero , the well-known expression for the
resonant frequency of a Helmholtz resonator is obtai n ed name ly ,
(5.24)
between the standing \vave tube and the anechoic tenninatin , a standard reflection factor t est was conducted over the desired frequency
range.
Thus, by in ter?olating
On ce t h e
73
The
the oscillator supplying the acoustic signal and the backing depth
of the resonator volume was varied.
the
Since the
standing wave tube was not designed for precision impedance measurements, the accuracy of the phase angle and/or the microphone location
was subject to possible error. Fbr
~accurate
impedance calculation,
Thus,
74
two microphones and the phase difference of the two signals , the
acoustic impedance of the resonator was determined using the expression
developed in Appendix E .
There are t\vO criteria \vhich may be used for determining the
resonance conditions .
The backing de pt h
was then measured and the effective length calculated from Equation
(5 . 24).
c~).
The effective
75
The high
Therefore,
length value.
an infinite plane wall and is, according to Kinsler and Frey (6),
given by
+ 0.85 d 0
(5.25)
where:
t
d
neck length
neck diameter
The latter term of the right-hand side of (5.25) is termed the end
correction".
76
+ o.
Tid 2
8(--0 ) 1/2
+ 0. 71 d 0
(5.26)
The third expression used was that found by Garrison in his resonator tests to define the effective length for a high frequency,
small physical size resonator used in an array:
t + d
(5.27)
The deviations
of the mean experimental values from the above expressions were obIn addition, the deviation of all experimental values from
tained.
S1
eff
( 5. 2 8)
Table II.
1
Resonator
Number
Proportionality
eff
mean
Constant
(Experimental)
B1
t+O. 71d
t+. 85d
0
t+d
Avg.
Max.
Avg.
Max.
Avg.
Max.
R-1
1.535
.788
5.0
-12.9
3.1
- 6.87
7.1
-18.5
R-3
1.834
.628
6.2
-24.2
4.7
-11.00
10.6
-26.5
R-4
1. 399
1.202
3.4
7.8
4.9
12.50
- 4. 7
3.4
-....J
-....J
78
2.
section
were possible.
After a desired flow rate was set using the flow-metering orifice,
a series of resonance tests were performed using the frequencies employed
for the static finite amplitude tests.
100 to 350 Hz.
from that used for the no-flow case was due to the fact that the introduction of a flowing medium changes the acoustic propagation velocity.
The propagation velocity of the waves traveling with and against the
flow is obtained by multiplying the static value by the quantities
(l+M) and (1-M) respectively.
The resonator was tuned for resonance conditions by adjusting
the backing depth until a minimum sound pressure level occurred at the
microphone.
79
A total
of three flow rates were used with Mach numbers of 0.042, 0.083, and
0.107.
The effective length of the resonator neck was observed to decrease with increasing flow past Mach number.
gations have been performed using resonator arrays with the most
extensive work being done by Mechel, Mertens, and Schilz (31) and
Garrison (19).
For this
(5.29)
or,
1
eff
[1 - ~ ] M
(5.30)
effNF
The effective length used for the no-flow case was obtained using
Equation (5.28).
80
occurred
for Resonator R-3 (1.5 inch neck diameter) with values of 1.13 at
200 Hz and M=0.042; and 3.92 at 275 Hz and M=0.083.
determined using all the data points obtained was s
= 2.19.
This is
of the same order of magnitude as found by Garrison for his high frequency resonator arrays.
Assuming s
number exceeds the value such that El M exceeds the static mass end
correction length, the oscillating mass would not occupy the entire
resonator neck.
that s
There is no
5.0
4.0
...
1I
\U
f-
0 R-1
---
..
r::l
R-3
R-4
3.0
w
-l)
f- lL
lL
2.0
(.!)
z 0w
w
.....J
()
1.0
g
LL.
0.0
0.0
.04
.08
.12
.16
.20
Co
I-'
82
The second correlation of effective length with flow was accomplished assuming that the change could be accounted for by a reduction
in the mass of gas extending beyond the resonator neck.
This is re-
presented by
effNF
- 6L
(5.31)
or
+ S1 .
+ ol[l - 6 LJ
6L
(5.32)
Thus, if
(5.33)
01
Inspection of
these data indicated that for the two smaller diameter resonator necks
tested, namely 0.375 in. and 0.750 in., the value of the ratio increased
more rapidly as the Mach number was increased.
tested did not have the same trend but appeared to be sloping more
toward a constant value.
1.0
0.9
0.8
0.7
0 R-1
8 R-3
0.6
_J
G()
'<l
_J
R-4
0.5
0.4
0.3
0.2
0.1
0.0
0.00
0.04
0.08
FLOW PAST
Figure 15.
0.12
0.16
0.20
MACH NO.
84
Ingard and
Labate (17) have shown that the changes in resonator impedance are
directly related to the flow circulation through the neck.
No finite amplitude effects were observed in the effective
lengths obtained during the flow past testing.
In addition, operation
Again, as with
All
of these tests were performed with the resonator tuned to the test
85
As a
Table III.
Frequency
% Deviation
150
Avg.
+2.1
Max.
+ 4.54
175
+9.6
+11. 70
200
+7 .1
+ 9.30
The value of
86
Analysis
This
Yr
( 5. 34)
where:
[
R
--b
exp
t+d
-
( 5. 35)
1] (-d--)
0
approx~ma
t e 1y +
- 9 . 0%o
,_..
10
LLJ
(/)
"VJ
'
1-
0
8
LL
.........
100 Hz
125 Hz
150 Hz
175 Hz
200 Hz
0!
::>
10
'
...
0
0 w
<! 0
pu
R-B-s-
<{
~
(f)
Nonlinear
Region
-Cf)
w
a::
10 0.1
10
100
():)
-.....!
,.......
10
LLJ
G
0
(/)
.,_
l.J...
.......
'o/
::E
Hz
Hz
Hz
Hz
Hz
sr
pu
rf
:::>
0
200
225
250
275
325
102
...
u w
<( u
\-J t;(A
z
<(
r-
'
(f)
(/)
w
I
a: 10 0.1
\1
W
<!>~ '
v-
VI
.
w~
~ Nonlinear
Region
10
100
Cf:)
Cf:)
(04
.........
(.)
w
U)
lL
'
15() Rz
EJ
8.
175 Hz
200 Hz
250 Hz
~ ~,.
0
0
10
() w
<l: ()
<!
r--
(/)
(/)
w
0:
o/Lb '"~\'w~~
10 0.1
puR=B
b
r s0
0
~Nonlinear
Region
'
10
tOO
co
\.0
90
sr .
, from which
(5.36)
(5.37)
As an aid in defining the nonlinear region of each of the resonators, the specific acoustic resistance was plotted versus particle
velocity; samples are shown in Figures 19, 20, and 21.
As the neck
Table IV
Resonator
Number
u 0 -Critical
(fps)
Reynolds
Number
R-1
6.0
1950
R-3
4.5
2930
R-4
12.0
1950
-wu
10
8 R-1
8 R-4
(/)
Nl
0
-
....
l1..
t- . . . . .
(/)
::::> #
0-
en 100
0:::
...
u w
LL U
u z
w <I:
Cl.. ~
(f)
El
El
(/)
-(/)
w
et:
t6'
01
PARTICLE
Figure 19.
[!]
/
L__
RB = puo
10
VELOCITY, uo
100
(FT/SEC)
1..0
f-l
........
10 1
(/)
0
-
Nf
R-3
\o1
R-4
.._
lJ...
r- '~
(j)
::J
0-
<(
en IOo
rr_
...
LL
(.)
VI
u z
w <{
a_ ~
en cn
'W
0
'W
'o/
'\1
R = pu
B
o
(/)
w
0::
16
01
PARTICLE
Figure 20.
100
10
VELOCITY, uo
(FT/SEC)
\0
N
,_..
10\
(.)
(/)
Nl
.....
LL
:::J
#
.._,
8 R-1
0 R-3
/1
R-4
r- '~
(/)
0
~ ~ 10
,...
0
lJ_
w
u
u z
(/)
t-
<!
(l.t(/)
(f)
w 101
0::
A
8
01
PARTlCLE
Figure 21.
[!]
EJ
G
0
8.
/t----
RB = pu o
10
100
VELOCITY, uo (FT/SEC)
1.0
94
Although
A Reynolds number
A different coefficient
The results of this
Table V.
Resonator
Number
Resistance Coefficients
Linear
Resistance
End Correction
Finite Amplitude
Nonlinear
Coefficient (Sr)
(y )
R-1
0.715
.941
R-3
1.113
. 871
R-4
1.173
. 834
95
critical particle velocity and the coefficient was refined using only
the data for particle velocities equal to or greater than this value.
Since acoustic resistance and flow resistance are analogous, as the
acoustic resistance increases so does the flow resistance, and thus,
the orifice coefficient of the neck would increase.
If the expression
(2.16)
2
then the coefficient Br corresponds to 0.36/cf s o .
It is assumed that
96
this chapter.
(5. 38)
bv
where:
sr
was observed
When flmv
was introduced into the waveguide and the neck pressure, which was a
function of the driver output and the impedance load on the driver,
was relatively low, a large increase in the resistance occurred.
As the neck pressure was increased the resistance decreased in value.
At a sufficiently high neck pressure, the ratio of the resistance with
flow to the no-flow value became independent of any further increase
in neck pressure.
1.95.
This was the same value obtained by Garrison (19) in his investigation
of resonator arrays.
97
3.8
3.4
M = 0.083
<:>
N = 0.095,
T = 666R
0.042
= 0.107
2.6
~ 2.2
a::
'>
.Q
a:
1.8
1.4
0.2
0.4
0.6
0.8
1.0
1.2
( PSF)
1.4
98
4.2r----,-----r----~--~:---~----~----
3.8
3.4
= 0.042
(>
M = 0.107
0. 083
2.6
~ 2.2
0:
..........
>
..0
a::
I .8
1.4
0.2
0.4
0.6
0.8
1.0
1.2
( PSF)
1.4
99
Of the
two resonators, the larger diameter neck tests exhibited a much more
rapid transition to the high amplitude region.
Similar testing of
the resonator having the largest diameter neck was not accomplished
because of svste.m limitations.
air flow temperature of 206F and are also shown in Figure 23.
E.
Therefore, the
response for each of the test elements was obtained in the form of
reflection and transmission factors.
100
(5.39)
0
Thus, if the Reynolds number was greater than 2100, the acoustic
obtained for each configuration with both no-flow and cold flow
operations.
values were in good agreement, see Figures 24 through 35, pages 101
through 112.
1.1
w
0
::::>
1.0
Theory
0.9
z 0.8
(!)
Measured
0.7
a: 0.6
g(.)
~
-~
0.5
0.4
0.3
() 0.2
w
_J
LL
0.1
w
0::
0.00.0
0.1
0.3
0.2
FREQUENCY
Figure 24.
0.4
0.5
0.6
(KHZ)
f-'
0
f-'
1.1
w l.O
0
~ 0.9
~
0.8
<(
:;E
0.7
0::
0
0.6
I-
()
0.5
<(
u... 0.4
0.3
Theory
(/)
(/)
- 0.2
Measured
(/) 01
z
~
r-
00
o.0
0.1
Figure 25.
0.2
0.3
0.4
FREQUENCY (KHZ)
0.5
0.6
t--"
0
N
I. I
w
Theory
1.0
0.9
Measured
z(!) 0.8
<(
~
0.7
0::
0.6
(.)
0.5
~ 04
5t;
w
0.3
_~
0.2
O.l
0.00.0
LL
0.1
Figure 26.
0.2
0.3
0.4
FREQUENCY (KHZ)
0.5
0.6
f-'
0
l..U
1.1
1.0
~ 0.9
~
0.8
<{
~
0.7
~ 0.6
1-
u
0.5
<(
LJ... 0.4
0.3
Theory
(/)
CJ)
0.2
Heasured
(/) 01
z
00
o.o
0.1
Figure 27.
0.4
0.3
0.2
FREQUENCY (KHZ)
0.5
0.6
1--'
0
~
1.1
1.0
0.9
0.8
0.7
(.9
<(
a: 0.6
~(.)
0.5
~ 0.4
6-
0.3
Theory
t;
0.2
_j
Ql
0.00.0
w
LL
0.1
Figure 28.
Measured
0.4
0.2
0.3
FREQUENCY (KHZ)
0.5
0.6
1-'
0
V1
l.l
w 1.0
Theory
~ 0.9
Heasured
a o.a
<(
::E 0.7
0::
0
..__
0.6
u
0.5
<{
LL. 0.4
z
0
0.3
(f)
(f)
0.2
U)
01
00
o.o
0.1
Figure 29.
0.2
0.3
0.4
FREQUENCY (KHZ)
0.5
0.6
I-'
0
(j\
1.2
Cl
1.1
I
::)
--
r- 1.0 I
z- 0.9
(!)
<!
~
p.AJ'"'~
Theory
Heasured
0.8
0.7
0:::
0.6
0.5
t<{
lL
0.4
03
(.)
02
....J
01
r-
lL.
((
0.0
I-
0.0
0.1
0.2
Figure 30.
0.3
0.4
0.7
0.5
0.6
FREQUENCY (KHZ)
0.8
0.9
1.0
t--'
0
"-J
1.2
0
:::::>
1.1
-z
1.0
._
(.9
<!
0.9
~ 0.8
n: 0.7
0
t;
0.6
lL
0.5
0.4
<{
Theory
Measured
(/) 0.3
(/)
(/)
0.2
z 01.
<(
g:
0.0 0.0
0.1
0.2
Figure 31.
0.4
0.7
0.6
0.5
FREQUENCY {KHZ)
0.3
0.8
0.9
1.0
co
1.1
l.O
0.9
Theory
z(.9 0.8
<( 0.7
Measured
0:
0.6
0.5
()
~ 0.4
5
t;
w
-
0.3
0.2
_J
0.1
0.00.0
LL
0.1
Figure 32.
0.2
0.3
0.4
FREQUENCY (KHZ}
0.5
0.6
,__.
0
1.0
1.1
w 1.0
0
~ 0.9
0.8
0.7
<(
a::
0 0.6
Theory
.....
(.)
0.5
<{
lL 0.4
z
0 0.3
en
en 0.2
Measured
en Ql
z
00
o.o
0.1
Figure 33.
0.2
0.3
0.4
FREQUENCY (KHZ)
0.5
0.6
I-'
I-'
l.l
w 1.0
0.9
Theory
t-
z(!) 0.8
<( 0.7
Heasured
0:
~(.)
lt
0.6
0.5
0.4
~ 0.3
t;
0.2
_J
0.\
0.00.0
w
LL.
0.1
Figure 34.
0.2
0.3
0.4
FREQUENCY (KHZ)
0.5
0.6
I-'
f-1-'
1.1
w 1.0
0
~ 0.9
~ 0.8
<t
0::
0.7
-~
0.6
(.) 0.5
<t
LL. 0.4
z
0 0.3
Theory
Measured
(f)
en 0.2
~
(/) OJ
z
g o.oo.o
0.1
Figure 35.
0.4
0.2
0.3
FREQUENCY (KHZ)
0.5
0.6
j-J
j-J
1 13
The response
ments were obtained over the frequency range while maintaining the
neck pressure as measured by the piston microphone at 90 db.
Similar
measurements were made over a narrow frequency range centered about the
resonant frequency with a constant 110 db neck pressure.
The trans-
mission factor was computed from the impedance measurements made using
the pressure-phase differential method.
A nominal variation of
length with flow were determined from the empirical relations develcYDCd
from the impedance measurements.
measuring the neck pressure with the present test equipment it was not
possible to maintain the desired neck pressure.
Therefore, at the
conclusion of the response testing, an average value of the neck pressure over a narrow frequency range centered at resonance was obtained
from the test data.
l.l
w \.0
0.9
z 0.8
<( 0.7
(!)
Q P neek=90 dB
cr 0.6
g
(.)
neck
=110 dB
05
.
~ 0.4
0.3
cw 0.2
_) Ql
u...
0.00.0
0.1
Figure 36.
0.2
0.4
0.3
FREQUENCY (KHZ)
0.5
0.6
f--1
f--1
~
1.1
w 1.0
::>
l-
0.9
~
0.8
<(
~
0.7
nee k=90 dB
nee k=llO dB
a:
0 0.6
I-
u
0.5
<(
u.. 0.4
z
0 0.3
(f)
(f)
0.2
(/) 0.1
I-
00oo
.
0.1
0.2
0.3
FREQUENCY
Figure 37.
0.4
0.5
0.6
(KHZ)
f--'
f--'
Vl
1.1
w 1.0
0
0.9
Linear Theory
0.8
Flow Theory
<(
~
0.7
0:
0.6
::::>
~
(!)
~
(.) 0.5
~ 04
-5
t-
Heasured
0.3
u 0.2
w
_J
LL O.l
w
a:: 0.00.0
Figure 38.
---0.1
0.2
0.3
0.4
FREQUENCY (KHZ}
0.5
0.6
1-'
1-'
0\
1.1
w
0
1.0
~ 0.9
0.8
0.7
0::
0
0.6
<{
(.) 0.5
<t
l.J... 0.4
Linear Theory
0.3
Flow Theory
(/)
en
-
0.2
Measured
(/) 01
1--
00
oo
.
Figure 39.
0.1
0.4
0.3
0.2
FREQUENCY (KHZ)
0.5
0.6
I-'
,.......
'-.I
l.l
w 1.0
0.9
Linear Theory
0.8
0.7
Flow Theory
(..!)
<(
~
0:
g
(.)
Measured
0.6
0.5
~ 0.4
5
t;w
0.3
0.2
_, O.l
lL
0.00.0
O.l
Figure 40.
0.3
0.4
0.2
FREQUENCY (KHZ)
0.5
0.6
I-'
I-'
00
1.1
1.0
i=
0.9
~
0.8
<(
~
0.7
~ 0.6
.....
(.)
0.5
<(
I.J... 0.4
z
0 0.3
Linear Theory
Flow Theory
(f)
en
0.2
Measured
en
z 01
00
o.o
Figure 41.
0.1
0.4
0.3
0.2
FREQUENCY {KHZ)
0.5
0.6
f-i
f-i
1..0
l.l
w 1.0
0.9
<(
0.8
0.7
0::
0.6
f20
05
.
0.4
Linear Theory
o.3
Flow Theory
(!)
5
t;w
0.2
_J
Ql
00
LL.
o.o
Figure 42.
0.1
Measured
0.4
0.3
0.2
FREQUENCY (KHZ)
0.5
0.6
t-'
N
1.1
w 1.0
- - Linear Theory
~ 0.9
Flow Theory
~ 0.8
<(
Measured
0.7
~ 0.6
......
(.)
0.5
<(
LL. 0.4
z
0 0.3
CJ)
(/) 0.2
:?!
(/) 01
00
o.o
Figure 43.
0.1
0.4
0.3
0.2
FREQUENCY (KHZ)
0.5
0.6
r-'
N
r-'
1.2
w l.l
0
::)
I-
z
(!)
1.0
0.9
~ 0.8
0.7
I-
0.6
~ 0.5
LL
0.4
Linear Theory
~ 03
t;
0.2
_j
01
000.0
u..
Flow Theory
0.1
0.2
Figure 44.
0.3
Measured
0.7
0.4
0.5
0.6
FREQUENCY (KHZ)
0.8
0.9
1.0
1'--'
N
1.2
w
0
::J
f-
\.\
-z 1.0
<..9
<t
~
0::
0.9
0.8
--
I
I
Flow Theory
"'--
Linear Theory
Measured
0.7
~
u 0.6
<{
LL
0.5
0.4
0
Cf)
(f)
0.3
~ 0.2
en
z 0.1
<{
a::
r- 0.0
0.0
O.l
0.2
Figure 45.
0.3
0.4
0.7
0.6
0.5
FREQUENCY (KHZ)
0.8
0.9
1.0
;--<
1.1
1.0
Linear Theory
:::> 0.9
~
Flow Theory
0.8
<(
~
0.7
(.9
Measured
a:: 0.6
g
(.)
0.5
~ 0.4
5
t;
w
_J
LL.
0.3
0.2
01
~ o.oo.o
Figure 46.
0.1
0.2
0.3
0.4
FREQUENCY (KHZ)
0.5
0.6
f-.'
.c-.
1.1
0
w 1.0
0
:::>
r- 0,9
z
t!)
c:::(
0.8
::E 0.7
Linear Theory
0::
0
0.6
Flow Theory
0.5
....-
0.4
0.3
(/)
(f)
(/)
Measured
0.2
OJ
<{
0::
r- 0.00.0
Figure 47.
O.l
0.4
0.2
0.3
FREQUENCY (KHZ}
0.5
0.6
1--'
N
Vl
1.1
w 1.0
0
::J
Linear Theory
0.9
Flow Theory
t-
0.8
<(
~
0.7
(.!)
:1easured
a:: 0.6
~0
0.5
~ 0.4
5
t;w
-
0.3
0.2
_J
0.1
0.00.0
LL
Figure 48.
0.1
0.3
0.4
0.2
FREQUENCY (KHZ)
0.5
0.6
1""-'
N
(J\
1.1
00
w 1.0
::::>
..._
0.9
z
(!)
<(
0.8
0.7
0::
0
0.6
Linear Theory
t;
0.5
Flow Theory
lL
0.4
0.3
<(
Measured
(/)
en 0.2
~
(/) 01
~
l-
0.00 0
.
Figure 49.
0.1
0.3
0.4
0.2
FREQUENCY (KHZ)
0.5
0.6
f-'
N
.......;
128
First, there
This
occurred for this resonator and only where the transmission factor
approached unity.
the energy is transmitted beyond the test element and toward the
termination whete it can be reflected.
1.2
w 1.1
0
::>
~
1.0
z<.9 0.9
~ 0.8
0.7
r-
0.6
~ 0.5
LJ_
0.4
5-
0.3
t;w
_J
LL
Flow Theory P
k
nee
4t
0.2
120 dB
0.1
~ o.o 0.0
OJ
0.2
Figure SO.
0.3
0.5
0.6
0.7
FREQUENCY (KHZ)
0.4
0.8
0.9
1.0
.......
N
\0
130
F.
Conclusions
Finite-amplitude effects encountered in acoustic side branch
A region
131
heated flow operations did not reveal any variation in the results
not accounted for by effect of the temperature on the wave length.
Because of the complex variation of the resistance with flow
and acoustic pressure,
1so
db).
In
side branch element response, the acoustic impedance must be investigated since its' variation is the cause of the response change.
Thus,
132
VI.
EXPANSION CHAMBERS
Little is known
method does not account for any flow effects which may be present, a
linearized perturbation approach was decided on for the present work.
The phenomena occurring at an expansion or contraction must be delt
with separately since the mechanisms available for acoustic losses
are considered to be quite different.
For the case of the contraction, it was assumed that no flow
losses occurred, i.e., the flow was considered isentropic.
a reasonable assumption for the flow Mach numbers used.
This is
In addition,
the work of Ronneberger (20) for Mach numbers of the order of magnitude similar to those of the present investigation indicated minimal
flow losses.
for the flow can be written for the contraction shown in Figure 51,
as follows:
133
M
a
pi
...
l'~
-,.
sb
pt
pr
sa
Figure 51.
P.
sa
-,.
Pr
'I
...
- pt
-p
...
sb
,,
Figure 52.
v~
( 6. 1)
u
a
2
ub
2
_l_
y-1
_ L Pb
y-1 p
( 6. 2)
where:
area ratio
density
static pressure
c /c
p
area is assumed to be terminated anechoically so that onlv a transmitted wave is present beyond the change.
isentropic relationship (6), namely, p
pa~pa
+ p.l + p r
p.+p
pa-+pa +
pb ~pb + pt
pt
pb~pb
-c
( 6. 3)
135
n[p
l
- a
( 6. 4)
ca
(6.5)
pt
pi
2(1+M )n
a
( 6. 7)
~ cb
ca
(l+M ) (1+- - - ) (-)
b
Ma
2 cb
ca
1.0
0.9
a:: 0.8
Iu 0.7
pi--+-
<t
LL
---+
p ....__
r
0.6
---....pt.~
--
~ 0.5
t;
0.4
_J
0.3
u..
~ 0.2
0.1
0.0
0.0
0.1
0.2
0.3
EXIT
Figure 53.
0.7
0.6
MACH NO., Mb
0.4
0.5
0.8
0.9
1.0
f-'
0'.
1.3
1.2
0::
l.l
~ 1.0
<t
LL
0.9
z0 0.8
U5
0.7
(/)
~ 0.6
en
~ 0.5
0.4
Ma
Q3
P.--L
pr-J
____._~ ~
0.2
000.0
0.1
0.2
0.3
0.4
0.6
0.5
EXIT MACH NO. , Mb
Figure 54.
0.7
0.8
0.9
1.0
1--'
-......!
138
Generation of
ring vortices which interact with the sound field (20,32) cause losses
to occur at the expansion thus negating the use of the isentropic
conditions.
In addition entropy
u -+u
a
a
+ pr
p -+p + p.
~
a a
pa-+pa
pi+pr
c
where:
pb-+pb + pt
p -+p
a a
P t +o
+ -2c
( 6. 8)
139
pt(l+Mb)+~
(6.9)
p (l+M ) - --1-- 0
t
b
y-1
(6.10)
Because of the inclusion of the entropy wave there are now three
unknowns involved in Equations (6.9) and (6.10); therefore, a third
equation is necessary to produce a unique solution.
This is provided
First the
For the
(6.11)
2
2
P i CnMa +2n(Ma +l)+p r CnM a -2nMa +1)
(6.12)
140
Equations (6.9), (6.10), and (6.12) are now solved to obtain the
reflection and transmission factors for the adiabatic expansion.
details of this development are given in Appendix E.
The
The variation
that the sound wave produced by the shed ring vortices discussed for
the expansion are instigated at the edge of the expansion by the initial
sound wave (20).
discussed.
Thus,
0.0
,---,----:.,-----y--.---::n--.-----,----,-----,.----
-0.1
-0.2
Isentropic
0: -0.3
0
~ -0.4
Adiabatic
~ -0.5
z -0.6
0
i=
-0.7
()
p. __...
w
-0.8
_j
w-09
0::
Pr
-+---
L1..
'\
-\.0
-I.'
-1.2
0.0
0.1
0.2
Figure 55.
0.4
0.6
0.5
EXIT MACH NO., Mb
0.3
0.7
0.8
0.9
1.0
.......
l:'~
1.3
1.2
a:: 1.1
~ 1.0
~ 0.9
z0 0.8
o.1
en
C/)
~
en 0.6
~ 0.5
pi ___.
Isentropic
Adiabatic
pt
Ma ~
0::
1- 0.4
pr-
03
0.2
000.0
0.1
0.2
Figure 56.
0.3
0.6
0.4
0.5
EXIT MACH NO. , Mb
0.7
0.8
0.9
1.0
I-'
~
1.0
0.9
0::
I(.)
<(
lJ..
0.8
0.7
0.6
t\pi:_
~pt-e
pr
z 0.5
0
I() 0.4
w
_J
LL
0.3
w 0.2
a::
0.1
0.0
0.0
0.1
0.2
Figure 57.
0.7
0.6
0.4
0.5
0.3
EXIT MACH NO., Mb
0.8
0.9
l.O
1--'
~
6.0
0:::
0
5.0
4.0
z
en
(/)
3.0
Pi-+
Pr...,_
t-
(.)
~~-
....,.pt~
(/)
<(
2.0
ct:
t-
1.0
0.00.0
0.1
0.2
Figure 58.
0.4
0.6
0.3
0.5
0.7
EXIT MACH NO. , Mb
0.8
0.9
1.0
f--1
~
~
145
A complete description
i.e.,
Hhen
The
tion propagates out the discharge and the reflected portion traverses
back across the chamber.
146
+ z2 + . . .
Iz I <1.
Thus,
R T T e -ikl
2 1 3
( 6. 13)
1-R R e-i2kl
2 3
and
(6.14)
The expressions
5 in. l.D.
Seal 0-Ring
Discharge
Extension
2 in. l.D. x 2 l/4 in.
O.D.
Inlet Coupling
Figure 59.
Adjustable
End Plate
f--1
+='
'.!
148
of 30 inches.
In
incident wave was increased in approximately 3db steps and the tests
repeated until the maximum level obtainable with the amplifier and
acoustic driver system
(~
varied over the range for which the sound wavelength is one-half to
two times the length of the expansion chamber.
As pointed out earlier in Section B, the linear response normally
used for the design of expansion chambers is obtained by letting
M
a
This
1.2
w
l.l
r-
1.0
0
:J
z
<!)
0.9
~ 0.8
0.7
r-
0.6
~ 0.5
u..
-~
0.4
0.3
t;
0.2
_J
0.1
LL
~ o.o 0.0
0.1
0.2
0.3
0.4
0.5
FREQUENCY
Figure 60.
0.6
0.7
0.8
0.9
1.0
(KHZ)
f--1
\.0
0
:J
1.2
1.1
..... 1.0
z
<D
<!
0.9
0::
0.7
0
<!
0.6
~ 0.8
0
LL
0.5
0.4
(J)
(/)
C/)
0.3
0.2
~ 0.1
g:
0.0 0.0
0.1
0.2
0.3
Figure 61.
0.7
0.4
0.6
0.5
FREQUENCY (KHZ)
0.8
0.9
1.0
......
\Jl
1.2
'
w
0
::l
~
l.l
LO
~ 0.9
<{
0.8
0.7
0:::
0
0.6
~ 0.5
LL
0
1-
0.4
03
(.) 0.2
w
~
0.1
w
(t:
0.0
0.0
0.1
Figure 62.
0.2
0.3
0.7
0.6
0.5
FREQUENCY (KHZ)
0.4
0.8
0.9
1.0
f--1
\Jt
f--1
1.2
w
1.1
r-
l.O
(!)
0.9
::::>
<l:
0::
t;
0.8
0.7
0.6
<{
lL
0.5
0.4
(f)
(/)
0.3
0.2
z
<!
01.
g:
0.0 0.0
(/)
0.1
0.2
0.3
Figure 63.
0.7
0.6
0.5
FREQUENCY (KHZ)
0.4
0.8
0.9
l.O
f-1
1...11
N
1.1
LLJ
0
1.0
:::> 0.9
~
(9
0.8
~ 0.7
a::: 0.6
(.) 0.5
~ 0.4
-~
0.3
(.) 0.2
w
~
0.1
Ct:
0.0
0.0
0.1
Figure 64.
0.2
0.3
0.4
FREQUENCY (KHZ)
0.5
0.6
I-'
Vl
1.1
w
0
1.0
0.9
:::>
z
(!)
<{
0.8
0.7
~ 0.6
~
~ 0.5
lL
0.4
0.3
(f)
(f)
0.2
::E
01
0.00.0
0.1
0.2
0.3
FREQUENCY
Figure 65.
0.4
0.5
0.6
(KHZ)
f-"
Vl
J:'-
155
For
perimental data to the predicted values did not yield totally conelusive results.
struct an element that exhibits three-dimensional effects in the useable frequency range of the test equipment, its transmission factor
was extremely low.
Two flow rates, M=0.084 and M=0.108, were used for the evaluation of the response of the expansion chamber elements.
Reflection
156
range and the results compared to the values calculated using the
mathematical model previously developed.
The experimental results were in very good agreement with the
predicted response for the 5 inch I.D.
chambers as shown in
tion factors were in very good agreement with the predicted values;
however, the measured transmission factors tended to be lower in the
high attenuation frequency range, see Figures 70 and 71.
a region of high reflection factor.
This is
indicates the dependency of the total reflection factor of the expansion chamber on the reflection factor of the expansion.
The agreement
However,
examination of the bore of the PVC tubing used to construct the 2.90inch I.D. chamber revealed a rough surface texture which would cause
the viscous effects along the wall to be greater than for the s~ooth
surface finish assumed.
1.2
w 1.1
:::>
..... 1.0
0.9
~ 0.8
0:
t;
<l:
l.J..
z
0
0.7
0.6
0.5
0.4
--- '\
--Linear
Theory
Measured
03
0.2
~ 01
w
0:::
0.0
0.0
0.1
0.2
0.3
0.4
0.5
FREQUENCY
Figure 66.
0.6
0.7
0.8
0.9
1.0
(KHZ)
1--'
Vl
-....,J
1.2
w
::::>
1.1
....
-z 1.0
(.9
<!
0.9
- - Linear
Theory
~ 0.8
(L
0.7
t;
0.6
LL
0.5
<!
---
Flow Theory
Measured
z 0.4
en 0.3
en
~
en 0.2
<!
0::
I-
0.1
0.0 0.0
O.l
0.2
0.3
0.4
0.5
FREQUENCY
Figure 67.
0.6
0.7
0.8
0.9
l.O
(KHZ)
f--'
\.Jl
co
1.2
w l.l
0
::)
1.0
I-
f//-
~ 0.9
<!
~
0.8
0.7
0::
0
r-
Linear Theory
0.6
~ 0.5
LL
Flow Theory
Measured
0.4
~ 0.3
t;
0.2
_J
0.1
lL
~ o.o 0.0
0.1
0.2
0.3
0.4
0.5
FREQUENCY
Figure 68.
0.6
0.7
0.8
0.9
1.0
(KHZ)
1--'
ll1
1.0
1.2
w
0
:::>
I-
1.1
-z 1.0
<.9
0.9
I - - Linear Theory
0.8
o:: 0 7 I
o I
~
(.) 0.6
Flow Theory
Measured
<(
lL
0.5
0.4
(/) 0.3
(f)
(/)
0.2
~ 0.1
0:::
t-
0.0
0.0
O.l
0.2
0.3
0.4
0.5
FREQUENCY
Figure 69.
0.6
0.7
0.8
0.9
l.O
(KHZ)
f--'
0\
1.1
w l.O
0
::J
~
(.9
0.9
M = 0.084
M = 0.088, T
596R
0.8
~ 0.7
a:: 0.6
~ 0.5
(.)
~ 0.4
-5r-
0.3
u 0.2
w
_j
LL
0.1
cr: 0.00.0
Figure 70.
0.1
0.4
0.3
0.2
FREQUENCY (KHZ)
0.5
0.6
f-'
(j\
f-1
1.1
w 1.0
0
:::>
I-
0.9
(.!)
0.8
<(
0.7
0::
0
t-
0.6
(.)
0.5
~ 0.4 1z
1
0.3 -
0.2
(f)
(/)
M = 0.084
M = 0.088, T = 596R
I -
en 01
<(
0:
00
o.o
Figure 71.
0.1
0.4
0.2
0.3
FREQUENCY (KHZ}
0.5
0.6
.........
0'N
163
Two test operations '\vere performed with heated flow and using
the 2.90-inch chamber at the test element.
ture attained was 180F.
ments were accomplished and the results compared to the cold flow
operations.
va
1 (p
ua
+ L ) ( u +u) 2
2
a
c
where:
(6.15)
164
ua
(6.16)
(6.17)
Similarly for sound propagating against the flow
Pr
r
p c (l-M)
a
(6.18)
Similarly for a
when the medium is flowing through the standing-wave tube and element,
i.e., M>O.
165
D~
basic types of elements, the side branch resonator and the expansion
chamber.
If more than
two elements are employed, the first two are combined using Equations
(6.13) and (6.14).
lumped model for analysis with the next element in the system.
The
166
This was
not cause any variance in the total system response according to the
linear theory, it was felt that this might not be true when flow
superposition is involved.
The results of these tests are given in the form of reflection
and transmission factor curves in Figures 72 through 77.
the data agrees reasonably well with the predicted values.
In general,
The
Conclusions
A linearized perturbation approach as used in this investigation
gation, the assumption of an adiabatic expansion and isentropic contraction for expansion chambers with low flow Mach numbers is deemed
valid.
tion velocity and, thus, the acoustic wave length for the temperature
range investigated.
1.1
l.O
0
:J
0.9
0.8
.....
(!)
~ 0.7
0.6
u..:
o.4
-5f-
0.3
0::
~
(.) 0.5
~
(.) 0.2
w
0.1
ii
w
0:::
0.0
\0
\I
-
0.0
Figure 72.
\9
Theory
0.1
0.2
0.4
0.3
FREQUENCY (KHZ)
0.5
0.6
f-.'
()',
-...,J
l.l
:::>
1.0
0.9
(.!)
0.8
<(
0.7
cr::
0
0.6
0.5
0.4
0.3
....
Theory
(/)
(/) 0.2
~
(/) 01
z
~ 0.000
1-
Figure 73.
0.1
0.4
0.2
0.3
FREQUENCY (KHZ)
0.5
0.6
0 -
D -
I-'
0\
00
l.l
0
:J
~
(.!)
l.O
0.9
0.8
~ 0.7
0:
0.6
~ 0.5
(.)
ti:
0.4
0.3
-~
(.) 0.2
u:w
et:
Flow Theory
01
0.0
0.0
0.1
0.2
0.3
FREQUENCY
Figure 74.
0.4
0.5
0.6
{KHZ)
0 -
D -
1--'
0\
I.D
1.1
w 1.0
::>
1-
0,9
(.!)
0.8
0.7
<!
Flow Theory
~ 0.6
i-
~ 0.5
lJ... 0.4
z
0.3
C/)
C/)
0.2
Ql
<{
00
o.o
0.1
0.2
0.3
0.4
0.5
0.6
FREQUENCY (KHZ)
Figure 75.
1--'
-.....!
0-
0 -
1.1
w 1.0
0
::J
I-
(.9
0.9
0.8
~ 0.7
0::
0.6
~
u 0.5
~ 0.4
5
-
r()
0.3
0.2
Flow Theory
~ 0.1
w
Cl:
Figure 76.
0.0
0.0
0.1
0.4
0.3
0.2
FREQUENCY (KHZ)
0.5
0.6
0-
0-
-....,J
1--'
1.1
\.0
Theory
~ 0.9
a o.s
<(
0.7
~ 0.6
..._
u
0.5
<{
LL
z
0
0.4
0.3
(f)
en 0.2
~
(/) 01
z
00
o.o
0.1
0.2
0.3
0.4
0.5
0.6
FREQUENCY (KHZ)
Figure 77.
1--'
-.....)
0-
D -
173
ent approach than used in the present work, it is felt by the author
that the results herein
174
VII.
work was to provide the designer with the capability to account for
the variation of the response from linear theory caused by these
phenomena.
Two basic types of elements were examined; the side branch resonator and the simple expansion chamber.
attenuation were present for the element types, each requiring appropriate analysis techniques and analytical modeling.
Finite amplitude effects were observed for no-flow operations.
At the conclusion of these tests the elements were examined under
conditions of flow past and flow through the element for the resonators
and expansion chambers respectively.
Only flow-induced
175
The coefficient
Sr
That is,
0.36
cf
(7.2)
so that, as the geometry of the resonator neck (and the flow coefficient)
change, the value of
will change.
176
This would be
As the
acoustic pressure is increased the ratio of the flow to no-flow resistance decreases and approaches a constant value for sufficiently high
pressures.
amplitude region the analogy of Garrison (19) applies; i.e., the ratio
of the flow to no-flow resistance is a linear function of the flow
177
~v
~NF
(1
+ 1. 95
M)
( 7. 3)
Thus the
maximum reduction results in an effective length equal to the thickness of the neck; i.e., the end correction goes to zero.
This is in
qualitative agreement with the results of Phillips (30) for his single
resonator tests.
number with the effective length asymptotically approaching the thickness of the neck at a relatively low value of the steady flow Mach
number
(~
0.12-0.15).
2.
3.
178
4.
5.
For Reynolds numbers greater than 2100, assume that the resonator will be operated in the finite amplitude region and compute the particle velocity assuming
Sr =
1.0.
6.
Using u
7.
effv
for M>O.l5,
- 61[1-.15M].
(1+1.95M).
NF
If the assumed pressure is less than 130 db, select a reason-
resistance is given by
Rb
10.
Flow
effects were observed and the response generated using the developed
analytical models '\vere in good agreement with the measured values.
A mathematical model for a simple expansion chamber was developed
by using the steady state approaches of Powell (21) and Ronneberger (20)
to obtain the responses of the individual discontinuities.
The
179
individual response parameters were then combined using a lumped parameter approach to obtain the total element response.
It is recommended that
dimensions~
cl~racteris
v,raves.
180
BIBLIOGRAPHY
1.
2.
3.
4.
5.
6.
7.
DAVIS, D.D., JR.; STOKES, G.M.; MOORE, DEWEY; and STEVENS, G.L.,
JR.; (1954) "Theoretical and Experimental Investigation
of Mufflers with Comments on Engine-Exhaust Muffler
Design", Nat'l Advisory Committee for Aeronautics
Report No. 1192.
8.
9.
10.
11.
SIMON, V.H., III (1969), "Apparatus and Techniques for the Plane
Wave Analysis of Acoustic Filters," Thesis, University
of Missouri-Rolla, p. 196.
12.
13.
BERANEK, L.L.,
181
14.
15.
16.
17.
18.
19.
20.
21.
22.
POWELL, ALAN, "Theory of Sound Propagation through Ducts Carrying High-Speed Flows", Jour. Acous. Soc. Amer., v. 32,
no. 12, December 1960, pp. 1640-1646.
23.
24.
25.
26.
lA2
27.
28.
29.
30.
31.
MECHEL, F. : MERTENS, P. , and SCHILZ, W. , "Research on SoundAbsorbent Ducts with Superimposed Air Streams"
Final Report (AMRL-TDR-62-140, v. TT and Jll),
Physikalisches Inst., Univ. Gottingen, West Germany,
December 1962.
32.
33.
DAVIES, P.O.A.L. and DWYER, M.J., "A Simple Theory for Pressure
Pulses in Exhaust Systems", Proc. Instn. Mech. Engrs.,
1964-65, 179(1)' 365.
34.
183
VITA
Additional educational
184
APPENDICES
185
Appendix A
REFLECTION FACTORS IN STANDING WAVE
TUBE MEASUREMENTS
The
follows:
-i6
B1 relative to A1 .
Termination
Figure A-1
186
(A-2)
p(x,t)
or
p (x)
{"\)
A1 e~xe
ik x
+B 1 e
-ax
i(8-k 2 x)
(A-3)
where
kl
k2
~
0
ko
T-rn
___w__
ko
c 0 -u
r-M
diP~~)
2
1
0.
187
(A-5)
By order of magnitude arguments, the first term of (A-5) can be
shown to be negligible compared to the second term.
or
nn; n=0,1,2,3 ...
X
(A-7)
mln
8-(k +k 2 )x
1
max
(A-8)
8-(k +k )x .
1 2 mln
(A-9)
(A-9):
min 2
83TI
kl+k2
Thus,
min 2
min 1
2
2n(l-M )
2 w/c 0
188
or
(A-10)
(A-ll)
where
Pr
2ax .
m1n
2
-2R+R e
-2ax .
m1n
(A-12)
Equations (A-ll) and (A-12 are now solved simultaneously for the
reflection factor (R):
+ (--r--)2 e
2ax .
2axr
m1n_e
=O
(A-13)
Pmin
where
8-(k +k )x
1 2 r
Since (A-13) is quadratic in R it can be solved using the fa~iliar
expression
189
-bl b 2 -4ac
2a
(A-14)
where
a
Pr
2
c-->
p .
m1n
-e
-2ax r
Pr
2
-2 [ ( - - ) +cos]
pmin
2ax
p
c
-2ax
min
= c-r-)2 e
min
2ax
-e
(A-15)
Pmin
For the
ations that the reflection factor must have magnitude less than or
equal to unity.
= Pmax and cp
0 or
(A-16)
For the case of the propagation medium in motion (M>O), the reflection factor can have a magnitude greater than unity and is obtained
using the positive root of the radical in (A-14)
ax .
m1n
Pmaxe
R
ax
-ax
-axmin
Pmaxe
max
+pmine
-prnine
max
(A-17)
190
191
Appendix B
DERIVATION OF TRANSMISSION FACTORS
In addition, a por-
..
xl
Al
Bl
..
Figure B-1
-1
I
I
1..
x2
A2
B =0
i iwt
e
(B-1)
192
(B-2)
(B-3)
(B-4)
(B-5)
i1jJ2
p2 = p2m e
(B--6)
i( 1~2-l)Jl)
P2m
plm
(B-7)
Thus,
i(1jJ
-w1 )
ax
[e
ik x
1 1
-ax
+Re
-i(8-k x )
2 1
(B-8)
193
194
Appendix C
P2
and
as shown in Figure
C-1.
s0
Figure C-1
zneck
s u
0
R
neck
+ iXneck
(C-1)
zvol
_ 1pc
wv
(C-2)
s u
0
s u
0
195
zneck
tot
s u
0
zvol
s u
0
(C-3)
0
(C-4)
o neck
( C-5)
zneck
Since p
and p
2
i~
(UV
(C-6)
(C-7)
where
196
+ i
__l_ cos
P2
(C-8)
zneck
-i
~ r-1 cos +
wv
P2
pl
P2
sin - 1]
(C-9)
The impedance of the neck can now be written in its resistive and
reactive components
2
~sin cp
wv
neck
neck
= p~ [_l cos - 1]
wv
(C-10)
sin
(JJV
~pl
wv
cos
(C-11)
relationshi~
19 7
Appendix D
EVALUATION OF THE ANECHOIC TERMINATION
This is
The evalua-
A reflection factor
In addition,
though the horn might have a relatively large reflection factor, the
energy which is reflected back to, and through, the test element is
quite small in comparison to the primary reflected wave.
\.2
~
l.l
1.0
Static
Max Flow
~ 0.9
<l:
~
0.8
0.7
0::
0
t-
0.6
~ 0.5
u..
0.4
~ 0.3
~ 0.2
w
.....J
0.\
000.0
LL
0.\
0.2
0.3
0.4
0.5
FREQUENCY
Figure D-l
0.6
0.7
0.8
0.9
LO
(KHZ)
I-'
\0
00
199
Appendix E
PERTURBATION ANALYSIS FOR RESPONSE WITH FLOW
pb
pi
t>
Pa
u
pt
sb
pr
Figure E-1
can be written:
Mass:
(E-1)
nuapa
Energy:
a
2
Pa
--L-y-1 pa
~
== -2-
+_L
y-1
(E-2)
200
where
Y
n = s ;s
a
area ratio
With the added assumption that the pressure acting on the area
(Sb-Sa) is equal to the upstream pressure, p ; the conservation of
a
momentum can be written
p u 2n
a a
(E-3)
direction of flow.
u -+u
p a -+p a
+ Pl + Pr
p+o
Pb -+o' b
c
t
+ --2-
where:
201
flow terms, the continuity equation for the acoustic waves in the
presence of steady flow is given by
c
(E-4)
n[pi(l+Ma)-pr(l-Ma)] cb
a
p_-p
1
p c
a a
a]
+ _x_ [
y-1
p +p.+p
a 1
r ]
p.+p
1
r
pa +
2
c
(E-5)
(E-6)
202
1
- - ---
Y-1
{~
Mb
[p.(l+M
)-pr(l-M )]
1
a
(l+M )
- Pt
(E-8)
[(l+Ma)(Mb + __1__)]
y-1
(E-9)
Similarly, substituting the expression for ~o from (E-4) into (E-7)
and simplifying yields
Pr
2
[n(M -2M
pi
a
a
+ -)
n + Mb(l-Ma)
ch.
-c nJ
a
cb
2
Mbn(l+Ma) - - - n(M +2M + l)
c
a
a
n
a
(E-10)
c;
Pr
(l+Ma)
pi
c
b
(1-M )
a c
a
(E-ll)
203
(l+M_) { (1-Ma)~[-1-(l+Mb)+MlJ
a y- 1
- -b
(E-12)
As the Mach number of the incoming flow reduces to zero the expressions
(E-ll) and (E-12) reduce to the linear theory form; i.e.,
n-1
n+l
_2J]_
n+l
(E-13)
(E-14)
204
Appendix F
LIST OF EQUIPMENT
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Bruel and Kjaer Co., Electric Voltmeter 2-200000 c/s, Type 2416,
Serial Number 186526.
12.
205
13.
14.
15.
16.
17.