EPA-650/2-74-070
July 1974
Environmental Protection Technology Series
-------
EPA-650/2-74-070
THE EFFECTS OF NOZZLE DESIGN
AND SAMPLING TECHNIQUES
ON AEROSOL MEASUREMENTS
by
F. H. Smith
ARO, Incorporated
Arnold Air Force Station , Tennessee 37369
Interagency Agreement No. EPA-IAG-0139(D)
ROAP No. 26AAM
Program Element No. 1AA010
EPA Project Officer: Dr. Kenneth T. Knapp
Chemistry and Physics Laboratory
National Environmental Research Center
Research Triangle Park, North Carolian 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON., D.C. 20460
July 1974
-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
CONTENTS
Page
1.0 INTRODUCTION 7
2.0 APPARATUS 8
3.0 DATA PROCESSING 13
4.0 PROCEDURE 15
5.0 RESULTS 18
6.0 CONCLUSIONS 21
7.0 RECOMMENDATIONS 21
REFERENCES 22
ILLUSTRATIONS
Figure
1. Velocity Profile along Vertical Centerline of
Wind Tunnel at Probe Mounting Location,
Nominal Flow Rate = 9.14 m/sec 25
2. Velocity Profile along Vertical Centerline
of Wind Tunnel at Probe Mounting Location,
Nominal Flow Rate = 15.24 m/sec 26
3. Velocity Profile along Vertical Centerline
of Wind Tunnel at Probe Mounting Location,
Nominal Flow Rate = 21.34 m/sec 27
4. Focal Volume Ellipsoid 28
5. Beam Interference Fringes 29
6. Oscilloscope Trace of Particle Crossing
Fringes in Ellipsoidal Volume 30
7. EPA Laser Velocimeter and Traverse
Installation 31
8. 1.27-cm Sampling Probes 32
9. EPA Side-Opening Probes 33
10. 1,91-cm Sampling Probes 34
11. 2.54-cm Sampling Probes 35
12. 5.08-cm Sampling Probe (Sharp Edge) 36
-------
Figure
13. 5.08-cm Sampling Probe (Square Edge) 37
14. EPA Sampling Probe Mounted in Wind Tunnel. ... 38
15. Schematic of Extraction and Sampling
System 39
16. Filter Housing 40
17. Theoretical Isokinetic Flow Requirements of
the EPA Sampling Probes 41
18. Schematic of Particulate Injection System. ... 42
19. Fiber Optics Particle Sizing of Diatomaceous
Silica 43
20. Frequency Response of Diatomaceous Silica. ... 44
21. Data Acquisition Instrumentation 45
22. Laser Velocimeter Data Acquisition and
Processing Instrumentation 46
23. Number of Measurements Required for a 90-
percent Probability (Z) of the Mean
Velocity 47
24. Number of Measurements Required for a 95-per-
cent Probability (Z) of the Mean Velocity. ... 48
25. Number of Measurements Required for a 99-per-
cent Probability (Z) of the Mean Velocity. ... 49
26. EPA Data Program Printout 50
27. Location of Probe Equal Area Measurement
Points 51
28. Location of Probe Velocity Measurements for
Zero Angle of Attack 52
29. Flow Streamlines Around a Probe 53
30. Constant Velocity Lines, 1.27-cm Square-
Edge Probe, AP=7.11 mm H2O, Pin > PQut 54
31. Constant Velocity Lines, 1.27-cm Square-
Edge Probe, AP = 25.4 mm H0O, P. <• P , 55
£ in ou "t
-------
Figure
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
Probe Performance at 9.14 m/sec and
Zero Angle of Attack
Probe Performance at 9.14 m/sec and
7.5 deg Angle of Attack
Probe Performance at 9.14 m/sec and
15 deg Angle of Attack
Probe Performance at 15.24 m/sec and
Zero Angle of Attack
Probe Performance at 15.24 m/sec and 7.5
deg Angle of Attack
Probe Performance at 15.24 m/sec and
Probe Performance at 21.34 m/sec and
Probe Performance at 21.34 m/sec and
7.5 deg Angle of Attack
Probe Performance at 21.34 m/sec and
15 deg Angle of Attack
1.27-cm Sharp-Edge Probe
1.59-cm Side-Opening Probe
1.91-cm Sharp- Edge Probe ,
2.54-cm Sharp- Edge Probe
5.08-cm Square-Edge Probe ,
1.27-cm Square-Edge Probe at 7.5 deg
1.27-cm Sharp- Edge Probe at 7.5 deg
Angle of Attack \ ,
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
-------
Figure Page
51. 1.59-cm Side-Opening Probe at 7.5 deg
Angle of Attack 75
52. 1.91-cm Sharp-Edge Probe at 7.5 deg
Angle of Attack 76
53. 2.54-cm Sharp-Edge Probe at 7.5 deg
Angle of Attack 77
54. 3.18-cm Side-Opening Probe at 7.5 deg
Angle of Attack 78
55. 5.08-cm Square-Edge Probe at 7.5 deg
Angle of Attack 79
56. 5.08-cm Sharp-Edge Probe at 7.5 deg
Angle of Attack 80
57. 1.27-cm Square-Edge Probe at 15 deg
Angle of Attack 81
58. 1.27-cm Sharp-Edge Probe at 15 deg
Angle of Attack 82
59. 1.59-cm Side-Opening Probe at 15 deg
Angle of Attack 83
60. 1.91-cm Sharp-Edge Probe at 15 deg
Angle of Attack 84
61. 2.54-cm Sharp-Edge Probe at 15 deg
Angle of Attack 85
62. 3.18-cm Side-Opening Probe at 15 deg
Angle of Attack 86
63. 5.08-cm Square-Edge Probe at 15 deg
Angle of Attack 87
64. 5.08-cm Sharp-Edge Probe at 15 deg
Angle of Attack 88
NOMENCLATURE 89
-------
1.0 INTRODUCTION
The sampling of moving gas streams for accurate deter-
mination of particulate content is at best a difficult task.
There are several variables existing in a flow of this
type which can affect the results of the sampling. Some
of these variables are
1. the mean gas velocity and local velocity
variations about the mean;
2. the temperature of the gas;
3. the particulate size distribution and
concentration in the gas, both of which
are time dependent and functions of the
duct design in which the gas is flowing;
4. the geometry of the sampling probe, the
angle of the probe with respect to the flow-
ing gas, and the extraction rate of the probe.
This investigation was concerned with the evaluation of
probe performance as a function of geometry, mean gas velocity,
probe angle of attack with respect to the flow, and extraction
rate. The performance of the probes was compared using the
sampling error determined from the flow parameters based on
a zero-error assumption at isokinetic sampling conditions.
Isokinetic, or equal velocity sampling, was accomplished
by using a variable speed wind tunnel to obtain the flow field
around the probes and then varying the probe extraction velocity
by connecting the probes to a controllable vacuum source.
Laser instrumentation was used in this study to confirm
the uniformity of the flow in the test section prior to testing
the probes. It was also used to measure the flow field adjacent
to the probe inlets to determine the isokinetic condition of
the probe flow. The use of the laser instrumentation facilitated
the effort because the accuracy of conventional aerodynamic
measuring instrumentation is susceptible to the same variables
as the extraction probes (for example, design, sensitivity to
angle of attack, and velocity of the gas streams). Conventional
instrumentation also has a perturbing effect on the flow which
it is measuring. Since laser instrumentation is perturbation-
less and insensitive to mechanical calibration errors, it was
used to make an accurate evaluation of the probes and also to
produce data that were heretofore unobtainable.
-------
2.0 APPARATUS
2.1 WIND TUNNEL
The Environmental Protection Agency (EPA) probes were
tested in a small wind tunnel constructed at AEDC (Ref. 1).
The tunnel had a 35.6- x 50.8- x 121.19-cm test section with
solid walls made of 6.35-mm Plexiglas . The Plexiglas
walls allowed undisturbed optical investigation of the
flow fields in the vicinity of the probes. The test
section floor, or bottom wall, had a 254-mm-diam plate
set into it. The probes were mounted on this plate,
which was scribed in 5-deg increments to facilitate the
accurate alignment of the probes with respect to the tunnel.
The tunnel velocities were varied between 3.05 and 30.5 m/sec.
The velocity profile in the test section was surveyed with
the laser velocimeter (LV) at the selected test velocity be-
fore the probe was installed. The survey profiles, shown
in Figs. 1, 2, and 3, indicated that there were no large
velocity gradients across the test section. An aluminum
honeycomb with 9.53-mm hexagonal openings was installed
in the bellmouth of the tunnel to break up any gusts or large
turbulence formations that might enter.
2.2 LASER VELOCIMETER
This section will briefly cover the principal features
of the laser velocimeter so that those unacquainted with the
instrument can obtain a working knowledge of it. For a
complete development of the principles involved, see Ref. 2.
Basically, the instrument consists of the laser output
beam divided into two parallel, equal intensity, coherent,
TEM00 beams. These beams are passed through a lens which
brings them to a simultaneous cross and focus, forming an
ellipsoidal measurement volume of l/e2 relative beam intensity
(henceforth referred to as the probe volume) at the region of
intersection as shown in Fig. 4. The wavefronts of the two
beams interact in this volume and generate interference
fringes which are perpendicular to the plane of the beams,
parallel to the bisector of the beam angle, and sinusoidal
in intensity distribution (see Fig. 5). When a moving
particle passes through the probe volume, it scatters light
in proportion to the light power in the interference fringes.
This scattered light is collected by lenses and focused on
a photomultiplier tube, which converts it into an a-c signal
with some d-c shift, as shown in the scope trace in Fig. 6.
8
-------
Data conditioning and processing electronics then determine
the period, t, of the a-c component (Ref. 3). Since the
distance (6) between successive interference fringes can
be determined from the relationship
8 = A0/2sin (0/2) (1)
where Xo is the wavelength of the laser and 9 is the angle
formed by the two intersecting beams, the rate at which the
fringes are cut in the probe volume by a particle having a
velocity, V, is
' - f
V = (S (3)
Since i = J-, V = -£- (4)
or A
V - Ao
V ~
21 sin (8/2)
There are other considerations which arise in applying the
instrument to a particular problem. Some of these are the
relationship of the fringe spacing to the size of the par-
ticles present, the size of the optical probe volume formed
by the beams, the laser power density in this volume, and
the value of the velocity-period relationship, which must
be compatible with the ranges of the signal-processing
electronics. In this investigation a 4880& argon laser
line source was used, and the conversion was determined to
be 7.23 m/sec-jjsec. Interference fringe spacing, 6, was
7.23 Mm, and there were approximately 30 fringes across the
width of the ellipsoidal probe volume. The laser power 2
density at the geometric center cross section was 2.25 w/mm ,
which gave a data rate compatible with the acquisition rate
of the processing electronics. The LV was mounted on an
electrically driven three-component traverse which allowed
the accurate positioning of the probe volume throughout the
-------
test section. Each component of the traverse was equipped
with a position readout which had a resolution of 787
counts/cm. The traverse and LV installation are shown in
Fig. 7.
2.3 EXTRACTION PROBES
The Environmental Protection Agency requested that the
following three basic types of extraction probes be evaluated
(1) probes with a 90-deg bend and a square-edge inlet; (2)
probes with a 90-deg bend and a sharp-edge inlet of 15 cleg;
and (3) a vertical, cylindrical probe with a circular opening
in the side perpendicular to the axis of the probe. The
diameters for each type probe are listed below.
Square-Edge Probes Sharp-Edge Probes Side-Opening Probes
Diam, cm Diam, cm Diam, cm
1.27 1.27 1.59
5.08 1.91 3.18
2.54
5.08
The probes were all fabricated locally from 1.24-mm wall
thickness stainless steel tubing. The 1.27-cm tubing could
be smoothly bent through 90 deg, but the larger sizes buckled
in bending and were unsatisfactory. All probes larger than
1.27 cm having 90-deg bends were fabricated by cutting mitered
sections from the tubing and heliarc welding them together
to form the bend. This fabrication technique was approved
by the EPA prior to its use. The drawings of the probes, as
received from EPA, are shown in Figs. 8 through 13, together
with the modified fabrication drawings for the larger probes.
Each of the probes, regardless of its inlet or capture size,
initially terminated in a 1.59-cm-diam tube. This tube was
used for mounting the probes in the floor plate and for
attaching the vacuum line. This technique was found to be
satisfactory for all probes tested except the 5.08-cm-diam
probes. During testing of these probes it was found that
the flow rate was so large that the 1.59-cm tubing choked
or reached a sonic velocity, thereby limiting the flow.
This was corrected by removing the 1.59-cm tubing and re-
placing it with 3.18-cm tubing. After this modification
no further difficulty was experienced. All of the probes
were additionally modified by the installation of internal
and external static taps. These taps allowed test conditions
to be repeated by using the static pressure difference which
10
-------
was measured on a precision manometer. The taps were
installed on the horizontal centerline of the 90-deg
bend probes 1.91 cm behind the leading edge. The
taps were installed on the side-opening probes on the
vertical centerline at 90 deg to the axis of the opening.
They were located at 1.59 and 3.18 cm, respectively, below
the center of the opening on the 1.59- and 3.18-cm probes.
The taps were stainless steel capillary tubing which was
silver soldered to the probes and passed through the wind
tunnel floor plate for connection to the manometer. The
probes also had positioning rings installed on the 1.59-cm
tubing to position them at a constant height above the
floorplate. These rings were indexed to the probe center-
line and were also used to align the probes with the tunnel
floorplate. Figure 14 shows a typical installation of a
probe in the wind tunnel.
2.4 EXTRACTION AND SAMPLING SYSTEM
The extraction and particle sampling system used in
the test consisted of a filter housing, a vacuum source,
and connecting hoses and valves. A schematic drawing of
the system is shown in Fig. 15. The filter housing, shown
in Fig. 16, was designed to hold three 10.16-cm-diam filters,
together with backing screens and seals, and an inlet air
flow diffuser which directed the flow uniformly into the
three filters. The housing was designed with a 1.59-cm-diam
fitting to connect the probes directly. However, due to the
choking condition experienced with the 5.08-cm probes, the
housing was modified to a 3.18-cm fitting. An adapter was
used to connect the 1.59-cm hose to the housing for the
smaller probes. The filter housing outlet was connected to
the vacuum source with a 5.40-cm vacuum hose, and a 6.35-cm
guillotine valve isolated the filter housing from the vacuum
source. Coarse vacuum adjustments were made with the 6.35-cm
valve, and fine adjustments were made with a 1.91-cm globe
valve inbleeding atmospheric air. The vacuum source used on
the test consisted of two different vacuum pumps. A 1.42 m3/
min pump was used in testing all probes except the 5.08-cm-diam
probe. Figure 17 shows the theoretical isokinetic flow require-
ments for the probes at different velocities. Due to the ex-
tremely large flow rates of the 5.08-cm probes, a pump with
increased capacity was sought while testing was being conducted
using the smaller one. An 8.50 m3/min pump was subsequently
located and used for the 5.08-cm probes and also for the probe
angle-of-attack investigations and the sampling tests.
11
-------
2.5 PARTICLE INJECTION SYSTEM
2.5.1 Design and Operation
The particulate injection system, shown schematically
in Fig. 18, was designed and built using the fluidized bed
principle. Air at 0.14 kg/cm2 was injected into the bottom
of the container, thus floating the particulate in the top
of the container. A 3.18-mm ejector nozzle in the injection
tube, supplied with 3.52-kg/cm" air, provided enough suction
to draw off the dispersed particulates from the container and
eject them at the tunnel bellmouth. An electrically driven
stirrer was mounted on the container and stirred the particu-
lates at 10 rpm to eliminate any channeling by the fluidizing
air supply. The particulate container was also vibrated to
keep the particulate at a uniform level in the container.
The solenoid valves, the stirrer, and the vibrator were all
connected electrically to an interval timer switch for
simultaneous operation.
The additional flow in the wind tunnel resulting from
the ejector and the induced air flow with the suspended
particulates was calculated using compressible flow theory.
The resulting velocity increase at a nominal tunnel velocity
of 9.14 m/sec, the worst case, was 0.6 percent, or 0.05 m/sec.
This was considered negligible since the uncertainty in the
unperturbed velocity profile in the wind tunnel was determined
to be greater than this amount.
2.5.2 Particles
The selection of the particles to be used in the sampling
portion of the test was based on three considerations: (1) mean
particle size, (2) size distribution, and (3) material density.
It was desired to select some material that generally had physi-
cal characteristics similar to those combustion byproducts found
in stack gases. It was also desirable to have a material that
was commercially available, nontoxic, and not overly expensive.
An inspection of the sizes of air-borne particles (Ref. 4)
indicated that the size range from 1 to 50 ^ would include a
large portion of fly ash, cement dust, pulverized coal, foundry
dust, and smelter dust. The apparent densities of this type
of material were found to range from 160.2 kg/m3 for lamp
black to 240.3 kg/m3 for charcoal, with slag and coke included
in this range (Ref. 5). After surveying many materials whose
characteristics fell into these categories, diatomaceous
silica was selected. It is inexpensive and readily available
and has an apparent density of 224.3 kg/m3 (Ref. 6). Samples
were sized on a fiber optics particle sizer (Ref. 7), and a
12
-------
representative of the size distribution is shown in
Fig. 19. Statistically, the average linear dimension
was 5.53urn, and the average diameter on a volume basis
was 10.03pm. These parameters (size and density) were
input into an existing computer program to determine the
flow-following capabilities of the material. Figure 20
shows the plotted results, which indicate that the diato-
maceous silica is capable of following 95 percent of turbu-
lent velocity fluctuations up to a frequency of 42,500 Hertz.
This was felt to be a much higher frequency level than would
be encountered in the low-speed wind tunnel.
2.6 DATA ACQUISTION INSTRUMENTATION
The data acquisition instrumentation used for the
tests consisted of a relatively standard instrument
package which is routinely used at AEDC with the laser
velocimeter. A block diagram of the system is shown in
Fig. 21. The output signal from the 931A photomultiplier
(PM) tube was sent to a Tektroniks Model 7623 oscilloscope,
where the signal could be displayed and amplified if desired.
The signal from the oscilloscope was used to select inputs
to the Doppler Data Processor (DDP) (Ref. 3), which subjected
the signal to certain logic considerations such as signal-
to-noise ratio and periodicity and, upon successful passage
of the signal, determined its period. A narrow band filter
was used on the signal input to the DDP to reject any noise
signals outside of the frequency range of interest. The
binary coded decimal (BCD) output of the DDP went to a
Hewlett-Packard (HP) 2547A coupler and an HP 2515A scanner,
which sequenced inputs of position, date, run number, and
period-to-velocity conversion constant, and then transmitted
them to a Kennedy Model 1600 Incremental Tape Recorder. The
recorder stored the data on magnetic tape for processing by
a computer. An HP 5050A line printer was paralleled with
the tape recorder so the data could be monitored directly
if desired. This instrumentation was rack mounted as a unit
and is shown in Fig. 22.
3.0 DATA PROCESSING
The data obtained with the LV consist of a set of values
varying about some mean. Each value is obtained in a random
time sequence; therefore, it must be treated statistically.
In working with such data, it is important to determine when
a sufficiently large data sample has been obtained to consti-
tute a statistically valid observation. This determination
13
-------
allows the acquisition of the data to be discontinued
when the data obtained satisfy the restraints that have
been placed upon it, thereby saving run time and data
processing time. These restraints are the desired accuracy
of the mean value and the confidence level or probability
of obtaining this accuracy. These factors are related by
the expression
V[TOL] _ V[TOL] ( ^
aV
where
Z = number of standard deviations away from the
mean velocity
V = mean velocity
TOL = tolerance or desired accuracy
N = number of samples
crV = standard deviation of the sample averages =* S/VN~
1/2
i (Vi - V)^
standard deviation
N-l
The above expression (6) can be rewritten for N as
N -=^_ (7)
l_V(TOL)J
but turbulent intensity (TI) = S/V; therefore,
r i2
N =
LTOLJ
This expression was solved for several values of Z, TI, and
TOL and is shown in Figs. 23 through 25.
14
-------
When values of the wind tunnel turbulent intensity
with the probe installed were determined from data, Figs.
23 through 25 were used to obtain a value of N commensurate
with the tolerance and probability function Z. A 90-percent
probability of obtaining an average velocity within 1 percent
of the actual average velocity was used for the experimental
work. This gave values of N between 4 and 60. Since it
was necessary to have a constant value of N for compatibility
with the computer program, N = 50 was used for all data. A
computer program was used to discard any measurements that
were more than two standard deviations, 2a, away from the
mean. The remaining data were then recalculated to obtain
a new mean and standard deviation. This mean and standard
deviation, calculated using the data period, were converted
into velocity units using the LV conversion constant. An
example of the computer program printout is shown in Fig. 26.
The Doppler Data Processor was operated in the 1.5—percent
window mode. This caused it to reject all LV input data
whose eight-cycle period varied more than 1.5 percent from
the five-cycle period. This implies that the data from
particles that were accelerating or decelerating by more
than 1.5 percent, as they traversed the probe volume, were
rejected. It also increased the validity of the data by
eliminating any aperiodic electronic noise which might occur
in the frequency range of the data.
4.0 PROCEDURE
The sampling probes were installed in the wind tunnel,
and velocity surveys were made of the flow field at different
conditions of tunnel velocity and extraction rate to determine
the extent of the flow disturbances resulting from their
presence. These preliminary tests indicated that the dis-
turbances extended transversely approximately 1.5 times the
probe inside diameter (I.D.), and axially upstream approxi-
mately twice the probe I.D. These values were used as the
bounds of the survey area with transverse traverses made
at 1/4, 1/2, 1, and 2 probe I^D.'s axially upstream of the
probe inlet plane. The axial location of the probe inlet
plane was determined with the LV under probe operating condi-
tions of tunnel velocity and extraction prior to the test.
This was necessary because the probes deflected slightly
under the aerodynamic loads imposed on them during flow.
Measurements were made on each of the four transverse
traverses at ±1.5 I.D., ±1.25 I.D., ±1 I.D., ±3/4 I.D., ± the
center position on the probe lip, and at five selected
15
-------
locations across the probe opening. The internal flow
area of the probe was divided into three equal areas,
and the five measurements were made on the centers of
these areas. As shown in Fig. 27, one measurement
was made on the probe center, which is the center of
area 1, and measurements were made on the centers of
areas 2 and 3. The velocities for the three areas were
averaged to obtain the mean velocity in the probe.
Vi + —-—
V - -..-2
prohp
(9)
OV-, + Vo H- V* + Vn -4- Va
/. v ^ T V2j_ ' ^2 ^ 1 ^2
The velocities used in this average were from the I.D./4
axial position only, which was the closest approach to the
probe inlet that could be made because of the beam angle.
All measurements were made in a horizontal plane on the
probe centerline. Figure 28 shows the 60 locations at
which velocity measurements were made for each condition
of extraction rate and velocity. The locations for these
measurements were calculated for each probe using the
probe I.D. and then converted into count readings for the
digital-to-analog position readout panel meters, whose
resolution was 787 counts/cm. The velocity profiles in
Figs. 30 and 31 are shown as a function of the probe
I.D. for convenience in plotting and comparison. The pre-
ceding procedure for determining the measurement locations
applied only to the 90-deg bend probes at zero angle of
attack. The measurement points for the side-opening probes
were determined in the same manner, with two exceptions.
Instead of the probe inside diameter, the diameter of the
probe inlet opening was used as the normalizing parameter,
and the probe velocity was calculated using two equal
areas instead of three. Two equal areas were also used
for the probes at angles of attack of 7.5 and 15 deg
because as the probes are yawed the projected frontal
area becomes elliptical, with the smaller dimension
occurring in the line of measurement. The use of the two
areas facilitated the measurements across the smaller
distance. The offset distance from the center of the
projected ellipsoid to the tunnel centerline was calculated
for each probe at 7.5 and 15 deg and was used as the
reference around which the transverse measurements were made.
16
-------
The procedure for setting the extraction flow rate
consisted of establishing a flow through the probe and
then adjusting it to give a selected static pressure dif-
ferential across the probe. Since the differential for
isokinetic conditions was unknown, differentials above
and below the null or zero differential, as well as the
zero differential, were set. These points were -12.7mm
H2O differential (static pressure inside probe > static
pressure outside probe), 0 differential, and +12.7mm I^O
differential (static pressure inside probe < static
pressure outside). Experience showed that because of the
steep slope of the curve defined by these points, an addi-
tional set point was desirable in the positive differential
sector. Therefore, a +25.4-mm H20 differential point was
added to the schedule. This allowed good definition of the
curves and more accurate determination of the static pres-
sure differential for isokinetic conditions. The pressure
differentials were set on a Meriam Model 34FB2 micromanometer
which was directly connected to the static pressure leads
from the probes. The effect of the isokinetic condition
on the probe flow is shown in Fig. 29. Figure 29a illustrates
a less than isokinetic condition in which the flow entering
the probe is decelerated; by continuity, a portion of the
flow in the projected area of the probe must be rejected to
the outside. Figure 29b shows a condition greater than
isokinetic in which the probe is capturing flow from out-
side the projected probe area. It is readily apparent
why sampling under either of these conditions should be
greatly in error. A true or 100-percent isokinetic condi-
tion has the flow streamlines in the projected probe area
entering the sampling probe and those outside the area
passing the probe with a minimum disturbance of the flow.
It is not possible to have a flat probe velocity profile
whose magnitude is the same as that of the free stream since
there is a boundary layer developing as the flow enters the
probe. Because of this, the velocity in the center portion
of the probe must be slightly greater than the mean to account
for the boundary velocities which are less than the mean
velocity.
The procedures used in sampling involved the extraction
and sampling system and the particulate injection system.
Before a sampling test was conducted, groups of three filters
were weighed to the nearest lO"4 grams. The filters were
identified by group number and placed in plastic bags for
protection. Before the filters were-installed, the vacuum
system was turned on to remove any stray or residual particles
that might be in the probe, filter housing, or hoses. The
17
-------
filters were then installed and the vacuum adjusted to
give the desired differential pressure between the probe
inside and outside static orifices. As soon as conditions
were stable, the particle injection system was energized
and operated for sixty seconds. Any change in the probe
AP due to particle buildup on the filters was corrected by
adjusting the probe vacuum during the test interval. When
the interval timer turned the injection system off, the
vacuum line was closed and the filters were removed from
the housing and replaced in their plastic bags. These
filters were returned to the chemical lab for reweighing.
The weight differential was identified with the probe and
test conditions for data reduction.
The standardization and calibration procedures for
this test were conducted as follows:
1. The laser velocimeter period-to-velocity constant
was determined initially and reverified monthly.
2. The wind tunnel manometer and the probe static
pressure differential manometer were disconnected
and zeroed prior to each test.
3. The traverse position indication instrumentation
was checked before each run and rezeroed and
spanned if required. A mechanical scale and
pointer on each axis of the traverse was used
as a standard for the position indicators.
4. The laser velocimeter alignment with the wind
tunnel was verified before each test.
5. The Doppler Data Processor was returned to the
laboratory monthly, and the operation of its
counter circuits was checked and verified.
5.0 RESULTS
The results of this test are presented in Figs. 30
through 64. These figures represent a condensation of 282
plots of velocity profiles obtained with the LV on the
different probes. These plots were not formally published
because of their quantity and also because their most perti-
nent information has been extracted and is presented herein.
They are of interest in examining the flow disturbances of
specific probes and can be used to develop constant potential
fields for specific probes. Figures 30 and 31 are examples
18
-------
of this for the 1.27-cm square-edge probe. Figure 30
is for a AP of 7.11 mm H20, P^n > Pout» or less than
isokinetic, and Fig. 31 is for a AP of 25.4 mm H2O,
Pip < POut> or greater than isokinetic. These figures
illustrate how the flow is drastically accelerated or
decelerated, depending on the kinetic condition of the
sampling.
The condensed plots used only the traverse data
from the I.D./4 axial position, the closest position to
the probe. The data are plotted against the arguments
of probe AP and the percent-isokinetic condition.
These plots have grouped the probes by velocity and
angle of attack, allowing comparison of the different
probe designs at the same conditions. Figure 32, at 9.14
m/sec and zero angle of attack, shows the 90-deg bend
probes grouped together and crossing the isokinetic line
at AP's of 0.76 to 2.54 mm H2O, Pin < POut. The 1.27-mm
square-edge probe requires the greatest AP of the 90-deg
bend probes to reach isokinetic flow. However, it is
considerably more efficient than the side-opening probes,
which have not reached isokinetic flow conditions at a
AP of 25.4 mm H2O, P^n < POut- Figure 33, with the flow
also at 9.14 m/sec but with the probes at 7.5 deg angle
of attack, shows the probe curves somewhat spread but not
greatly deteriorated. Figure 34, for 9.14 m/sec and
15 deg angle of attack, shows .the 5.08-cm, 90-deg bend
probes operating more efficiently than the other similar
probes, whose performance is between 90 and 95 percent
isokinetic at a zero AP.
Figure 35, zero angle of attack at 15.24 m/sec, shows
much steeper curves for the probes although the zero-AP
crossing of the curves is about the same as that at 9.14 m/sec.
This indicates that the kinetic condition, or percent isokinetic,
is not as sensitive to changes in AP as it was at 9.14 m/sec.
The 5.08-cm sharp-edge probe gives the best performance in
terms of highest percent isokine.tic condition at zero AP.
At 7.5 deg angle of attack and 15.24 m/sec (Fig. 36), the
performance showed an improvement for all the probes except
the 1.59-cm side-opening probe, which decreased. In Fig. 37,
at 15 deg angle of attack, the performance decreased uniformly
for all the probes except for the 5.08-cm ones, which remained
relatively the same, and the 3.18-cm side-opening probe, which
improved.
The performance curves at 21.34 m/sec, Fig. 38, were
much steeper that at 15.24 m/sec, but the AP required for
isokinetic flow conditions changed very little. At 7.5 deg
angle of attack, Fig. 39, there was very little change in
the zero AP performance with the exception of the 1.27-cm
square-edge probe, the 3.18-cm side-opening probe, and the
19
-------
5.08-cm sharp-edge probe, all of which improved a small
amount. Figure 40, 15 deg angle of attack at 21.34 m/sec,
showed a general decrease in performance except for the
3.18-cm side-opening probe and the 5.08-cm sharp-edge
probe, both of which improved slightly.
These results may be summarized as follows:
1. The probe static pressure differential (AP) required
to perform isokinetic sampling varies directly as the stream
velocity.
2. The static pressure differential required for iso-
kinetic sampling varies inversely as the probe diameter.
3. The square-edge lip design on a probe requires a
much greater static pressure differential. This differential
varies from a factor of 2 for the 1.27-cm probe to a factor
of 5 for the 5.08-cm probe.
4. The side-opening probes require a much greater
static pressure differential than the 90-deg bend probes
to achieve isokinetic sampling. Since these probes were
not designed with the same side opening-to-I.D. ratio, it
is not possible to isolate the observed differences in their
performance.
The performance of the probes was also presented as a
function of percent sampling error plotted versus the probe
pressure drop, AP, in mm of water. These plots were developed
with the premise that at isokinetic sampling conditions there
is a zero sampling error. Each plot shows the performance
of a single probe at the three test velocities. The plots
are arranged by ascending size and lip design; that is, the
data on the 1.27-cm-diam square-lip probe are first, and the
5.08-cm-diam sharp-lip data are last. Figures 41 through 48
are the zero angle-of-attack plots, 49 through 56 are the
75-deg angle-of-attack data, and 57 through 64 are the data
for an angle of attack of 15 deg. Typically the lower
9.14 m/sec velocity data have sampling errors which are
larger per unit change in probe AP. As the stream velocity
increases, the data become less sensitive to error for a
unit change in the probe AP. The side-opening probes (Figs. 43
and 46) were so inefficient that isokinetic conditions were not
met over the normal range of testing used for the other probes.
However, an extension of the curves to the horizontal "zero
error" line will give an indication of the probe AP required
to reach isokinetic conditions for these probes. This is
typically true at the three different angles of attack for
these probes. The three velocity plots for any probe and
20
-------
angle-of-attack situation have nearly the same sampling
error at a probe "zero" AP or null condition. This is
typical for all probes except the 5.08-cm square-edge
probe at an angle of attack of 7.5 deg, shown in Fig. 55.
The 9.14 m/sec data were considerably lower in value than
the two other velocities for this case. These data were
checked, and nothing was found to invalidate them. This
effect did not occur at 15 deg and was not apparent to
the extent noted in any of the other data.
6.0 CONCLUSIONS
There are significant errors in extraction probe
sampling that are caused by sampling at nonisokinetic
conditions. The magnitude of the errors is a function
of several variables, the major ones of which were
evaluated in this investigation. The effects of the
variables on the sampling error are as follows:
1. Velocity: the sampling error is greater for low
velocities (9.14 m/sec) than for high
velocities (21.3 m/sec) for the same
static pressure differential (inside
to outside) across the probe.
2. Probe Diameter: the larger probes appeared to
be more efficient at all velocities than
the smaller probes. The larger (5.08-cm-
diam) probes closely approached isokinetic
flows at a zero static pressure differential.
3. Probe Shape: the sharp-edge probes in all cases were
more efficient than the square-edge probes.
The two side-opening probes were much less
efficient than the 90-deg bend probes.
4. Angle of Attack: in general the efficiency of the
probes decreased with increasing angle of
attack. There were some exceptions where
the efficiency apparently increased. These
cases were few, however, and generally were
limited to the 7.5-deg angle-of-attack cases.
7.0 RECOMMENDATIONS
This study has demonstrated techniques that are applicable
to a field in which there is limited information available.
21
-------
These techniques can also be applied to the related problem
of where the samples are taken in the duct. The concentra-
tion of airborne particulates is directly affected by flow
disturbances such as bends, T's, and Y's in the ducting and
by the location of the sampling probe. This information
can be obtained by electro-optical techniques, and the
effect of the variables can be shown in nondimensional form
for general sampling application.
The location and size of the static taps on the probes
should be standardized. This would allow the use of probes
based on the calibration of one sample probe, and would
thereby eliminate the need for individual probe calibration.
The sampling probes should be evaluated to determine
an optimum, or minimum total pressure loss design. A
review of the literature indicates that even simple shapes
such as 90-deg bends can be optimized. These techniques
are applicable to sampling probes and would result in more
efficient probes that would require less suction pressure
for isokinetic sampling.
REFERENCES
1. Kroeger, Richard A. "Wind Tunnel Design for Testing
V/STOL Aircraft in Transition Flight." AEDC-TR-
72-119 (AD 749154), September 1972.
2. Lennert, A. E., Brayton, D. B., Crosswy, F. L., Goethert,
W. H., and Kalb, H. T. "Laser Metrology." AGARD
Lecture Series No. 49 on Laser Technology in
Aerodynamic Measurements, presented from June 14-18,
1971 at the von Karman Institute. Published
March 1972.
3. Kalb, H. T., Brayton, D. B., and McClure, J. A. "Laser
Velocimetry Data Processing." AEDC-TR-73-116
(AD 766418), September 1973.
4. Sheeny, James P., Achinger, William C., and Simon, Regina A,
"Handbook of Air Pollution." U. S. Department of
Health, Eduction and Welfare, Robert A. Taft
Engineering Center, Cincinnati, Ohio, 05226.
5. McAdams, William H. Heat Transmission. McGraw-Hill Book
Company, Inc., New York, 1954. (Third Edition).
22
-------
6. Kreith, Frank. Principles of Heat Transfer. Inter-
national Textbook Company, Scranton, PA, 1958.
7. Bentley, H. T. "Fiber Optics Particle-Sizing System."
AEDC-TR-73-111 (AD 766647), September 1973.
23
-------
e
o
tance .
cc
•H
Q
1— 1
d
o
-H
0)
18
16
14
12
10
8
6
4
2
0
-2
_4
-6
-8
-10
-12
-14
-16
18
•~
-
-
-
h-CH
h-OH
-
1 0— 1
r-^H
I-^>H
KH
hCH
K*
rCH
~
-
1 1 i
8.5 9.0 9.5 10.0
Average Velocity, m/sec
Figure 1. Velocity profile along vertical centerline of wind tunnel
at probe mounting location, nominal flow rate = 9.14 m/sec.
25
-------
18
16
14
12
10
8
6
a
0 4
Distance,
o to
rH 2
o
• H
£ -4
0)
-6
-8
-10
-12
-14
-16
-18
••
-
-
1 O 1
- 1 U I
I o I
I \J 1
1 0 1
0 1
h-OH
1— 0— 1
h-o^-t
I—OH
-
-
iii]
15.0 15.5 16.0 16.5 17.0
Average Velocity, m/sec
Figure 2. Velocity profile along vertical centerline of wind tunnel
at probe mounting location, nominal flow rate = 15.24 m/sec.
26
-------
u
bance
w
•H
Q
ertical
18
16
14
12
10
8
6
4
2
0
-2
-4
-6
-8
10
-12
-14
-16
18
—
-
-
-
, Q ,
1 C\ I
1 U 1
1 0 1
t— 0— 1
h- 0 — 1
1 i 1 1
20.0 20.5 21.0 21.5 22.0
Average Velocity, m/sec
Figure 3. Velocity profile along vertical centerline of wind tunnel
at probe mounting location, nominal flow rate = 21.34 m/sec.
27
-------
1/e Relative Signal
Amplitude Ellipsoid
ts)
oo
1/e Relative Beam
Intensity
Figure 4. Focal volume ellipsoid.
-------
End View
Interference Fringes
Gaussian
Intensity
Distribution
Fringe Spacing
2 sin (y/2)
1/e Relative
Beam Intensity
Figure 5. Beam interference fringes.
-------
Figure 6. Oscilloscope trace of particle crossing fringes
in ellipsoidal volume.
30
-------
Figure 7. EPA laser velocimeter and traverse installation.
-------
5.08
7
1.27R
\
7.6
15.24
.02 Diam
1.27-cm Square-Lip Sampling Probe
08
1
^
I
/S.27R
\
/**- 15 deg
7.6
15.2
-I
.02 Diam
~T
1.59
Diam
*
1,59
Diam
1.27-cm Sharp-Lip Sampling Probe
All Dimensions in Centimeters
Figure 8. 1.27-cm sampling probes.
32
-------
16.51
13.97
1.59
Diara
0.95 Diam
1.59-cm Side-Opening Probe
2.54 Diam
3.18-cm Side-Opening Probe
All Dimensions in Centimeters
Figure 9. EPA side-opening probes.
33
-------
5.72
75 dee' TyP
15 deg
1.91 Diam
7.62
15.24
Modified Design Fabricated
at AEDC
1.59
Diam
5.72
7
1.27R
\
•15 deg
1.91 Diam
7.62
15.24
EPA Design
1.59
Diam
All Dimensions in Centimeters
Figure 10. 1.91-cm sampling probes.
34
-------
7.62
Modified Design Fabricated
at AEDC
2.54 Diam
62
/
J
^-w
/ 1.91R
\
/*x— 15 deg
m—O. Rd TUnm
7.62
15.24
EPA Design
1.59
Diam
All Dimensions in Centimeters
Figure 11. 2.54-cm sampling probes.
35
-------
1.59
Diam
8.26
r
8,26
5.08 Diam
Modified Design Fabricated
at AEDC
15 deg
5.08 Diam
1.75R
\
15 deg
15.88
EPA Design
All Dimensions in Centimeters
Figure 12. 5.08-cm sampling probe (sharp edge).
36
-------
8,26
-5.08 Dlam
1.59
Diam
Modified Design Fabricated
at AEDC
8.26
All Dimensions in Centimeters
Figure 13. 5.08-cm sampling probe (square-edge).
37
-------
U)
00
Figure 14. EPA sampling probe mounted in wind tunnel.
-------
EPA
Sampling
Probe
1.91-cm
Valve
•{&•
i
Filter
Housing
6.35-cm
Valve
•Exhaust to
Atmosphere
Vacuum
Pump
Figure 15. Schematic of extraction and sampling system.
-------
-
A E D C
1459-73
Figure 16. Filter housing.
-------
3.0 r
0
o
9 12 15
Velocity, m/sec
18
21
24
Figure 17. Theoretical isokinetic flow requirements of the EPA sampling probes.
-------
Interval Timer Switch
0.14 kg/cm Regulated
Air Supply-
3.52 kg/cm Regulated
Air Supply-
Electric
Vibrator
Solenoid
Valve
Stirrer Drive Motor
•Particulate Stirrer
Ejector Nozzle
Solenoid
Valve
-Particulate
Injection
Tube
•Particulate
Container
Figure 18. Schematic of paniculate injection system.
-------
Ell
E10
E9
E7
E6
E5
\
OE-4 5E-4 10E-4 15E-4 20E-4 25E-4 30E-4 35E-4 40E-4 45E-4
Diameter, cm
Figure 19. Fiber optics particle sizing of diatomaceous silica.
-------
1.0
*
%
>
a
06
o
o
0.8
0.6
c
a
(V
s
o
c
o
•rt
•P
Gt
•H
0.4
0.2
•H
u
o
1-1
0)
10"
10'
Frequency, Hz
Figure 20. Frequency response of diatomaceous silica.
-------
Oscilloscope
PM Tube
Scope "A"
Gate
Positional
Inputs:
x,y,z
Manual Inputs:
Date, Run Number,
Conversion Constant
Bandpass Filter
mm
I I
Scanner
Doppler Data
Processor
Coupler
Incremental
Tape
Recorder
Line
Printer
Figure 21. Data acquistion instrumentation.
-------
~
Figure 22. Laser velocimeter data acquistion and processing instrumentation.
-------
10,000
1,000 -
c
o>
E
O)
ISI
0}
d>
0.05 0.10 0.15 0.20 0.25
Turbulent Intensity, S/V
0.30
0.35
Figure 23. Number of measurements required for a 90-percent
probability (Z) of the mean velocity.
47
-------
10,000
1,000
o>
OJ
s
0>
-O
100
10
0.05 0.10 0.15 0.20 0.25
Turbulent Intensity, S/V
Tolerance
0.01
0.02
0.30
0.35
Figure 24. Number of measurements required for a 95-percent
probability (Z) of the mean velocity.
48
-------
10,000
Tolerance
1,000 -
1/1
0>
1_
13
S
S
"o
o>
0
0.10 0.15 0,20 0.25
Turbulent Intensity, S/V
0.30
0.35
Figure 25. Number of measurements required for a 99-percent
probability (Z) of the mean velocity.
49
-------
9ATA
KEDUCTIfiN FHR FCCH STAFF LUV UATA
RAh DATA PRINT JUT
1101
700
700
700
70U
700
70373
1313
342*
3355
1470
3312
700
700
700
700
700
2200
3433
3359
344 3
3463
3326
S04
700
700
700
700
700
3300
3347
3452
3302
3355
3433
3502
700
7UO
700
700
700
4401
3392
3310
3331
3370
4dfi8
23733
TAPE RECORD NUMBER
700
700
TO\J
700
700
5540
3429
3331
33b7
3417
3342
8660
700
700
TOO
700
700
3316
337.*
3363
3290
3473
700
700
~76u
700
700
1
3395
3385
3378
3335
3417
700
700
700
TOO
700
3391
3263
3339
3390
3361
700
700
700
700
700
3336
3308
3379
3191
J309
700
700
700
700
700
3358
3524
3329
32«3
3415
RUN NUMBER 1
POSITION - X "
ID - 0.50U
DA It 7- 3-73 CONVERSION CONSTANT
0.45 Y » -0.75 i - 4.33
23.733 (FT/SEU/HHZ
in
O
VFUCIMFTEK DATA - PFKIQU
0.3433
0.3359
0.3443
0.3493
0.3347
0.3452
0.3302
0.3355
UF DOPPLbR DATA IN MICROSECONDS
0.3392
0.3310
0.3331
0.3370
0.3429
0.3331
0.3367
0.3417
0.3318
0.3374
0.3363
0.32*0
0.3395
0.3385
0.3378
0.3335
0.3391
0.3263
0.3339
0.3390
0.3336
0.1306
0.3379
0.3191
0.3358
0.3524
0.3329
0.3283
0.3318
0.3429
0.3355
0.3470
0.3326
0.3433
0.4886
0.3342
0.3473
0.3417
0.3361
0.3309
0.3415
0.3312
NlWtJER UF MEASUREMENTS * 50
CEKKJD- AVEKAGE = 0.3398HICKOSLCUNDS
STANUAHO DEVIATION -
0.0224MICROSECONDS
VELOCITY -
AVERAGE = 69.845 FEtT/SECONO
STANDARD DEVIATION - 4.598 FT/SEC
6.58 PER CENT
VElDTTtfETtft UATA - PERIOO OF OOPW.ER DATA IN MICRCSECONCS
CORRECTED TO AVtRAGE 4-/-2.00 STANDARD OtVIATIUNS
0.^433
0.3359
0.3443
" OV3481
0.3326
0.3347
0.3452
0.3302
0.3355
0.3433
NUMBER UF MEASUREMENTS
PERIOD^ AVERAGE » 0.
O.J392
0.3310
0.3331
0.3370
0.0
* 49
3368H1CKQSECUNOS
STANDARD DEVIATION »
VtLUCITY •-
AVERAGE -
0.3429
0.3331
0.3367
0.3417
0.3342
0.3318
0.3374
0.3363
0.3290
0.3473
0
0
0
" 0
0
.3395
.3305
.3378
".3335
.3417
0.3391
0.3263
0.3339
0.3390
0.3361
0
"0
0
0
0
.3336
.3308
.3379
.3191
.3309
0.3358
0.3524
0.3329
0.3283
0.3415
0.3318
\j.34~29~~
0.3355
0.3470"
0.3312
O.O042MICRGSECONCS
70.476 FEET/SECOND
STANDARD DcVIAlIQN
1.305 FT/SEC
1.85 PER CENT
Figure 26. EPA data program printout.
-------
Probe ID
Lip
Lip
Probe OD
Figure 27. Location of probe equal area measurement points.
51
-------
en
to
-1.5 ID
-1.25 ID
-1 ID
-0.75 ID
-Lip
-Center Area 3
-Center Area 2
Center Area 1
+Center Area 2
+Center Area 3
+Lip
+0.75 ID
+ 1 ID
+1.25 ID
+1.5 ID
± Area 3 = ± ±R
± Area 2 = ±
± Lip
1
- I ID + 0.124 I
\
+y
Flow
Figure 28. Location of probe velocity measurements for zero
angle of attack.
-------
a. Less than isokinetic conditions.
b. Greater than isokinetic conditions
Figure 29. Flow streamlines around a probe.
53
-------
-1.4 r-
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
Y/ID
0'
0.2
0.4
0.6
0.8
1.0
1.2
1.4 -
9.14
8.53
7.92
9.14
-4.42-
4.88
9.14
8.53
6.71
6.71
8.53
I
0,25 0.50
9.14
8.53
8.53
9.14
1
X/ID
9.14
9.14
Figure 30. Constant velocity tines, 1.27-cm square-edge probe,
AP
mm H2O, P
in
out.
54
-------
1.2
1.4
9.75
9.75
X/ID
Figure 31. Constant velocity lines, 1.27-cm square-edge probe,
25.4mmH20, Pin < Pout.
55
-------
s
A
C
o
04
V
e
en Square-Edge Probe
en Sharp-Edge Probe
en Side-Opening Probe
en Sharp-Edge Probe
en Sharp-Edge Probe
en Side-Opening Probe
en Square-Edge Probe
en Sharp-Edge Probe
-5 -
-10 -
-15 -
-20
-25
60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 32. Probe performance at 9.14 m/sec and zero angle of attack.
-------
A 1.27
_O 1.27
Ol.59
g 1.91
Q 2.54
03.18
N 5.08
V 5.08
cm Square-Edge Probe
cm Sharp-Edge Probe
Side-Opening Probe
cm Sharp-Edge Probe
cm Sharp-Edge Probe
cm Side-Opening Probe
cm Square-Edge Probe
cm Sharp-Edge Probe
-25
20 30 40 50 60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 33. Probe performance at 9.14 m/sec and 7.5 deg angle of attack.
-------
A
C
25
20
15
10
V/l
00
S
V
C
-5
-10
-15
-20
-25
A 1.27-ca Square-Edge Probe
1.27-cm Sharp-Edge Probe
1.59-cm Side-Opening Probe
1.91-cm Sharp-Edge Probe
2.54-cm Sharp-Edge Probe
_O3.18-cm Side-Opening Probe
N 5.08-cm Square-Edge Probe
V 5.08-cm Sharp-Edge Probe
20 30 40 50 60 70 80 90 100 110
laokinetic Condition, percent
120
130
140
150
Figure 34. Probe performance at 9.14 m/sec and 15 deg angle of attack.
-------
5
A
B
o
N
s
V
fi
A 1.27-cm Square-Edge Probe
_Ol.27-cm Sharp-Edge Probe
Ol.59-cm Side-Opening Probe
1.91-cm Sharp-Edge Probe
2.54-cm Sharp-Edge Probe
3.18-cm Side-Opening Probe
N 5.OS-cm Square-Edge Probe
V 5.OS-cm Sharp-Edge Probe
-10 -
-15
-20
-25
20
30 40 50 60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 35. Probe performance at 15.24 m/sec and zero angle of attack.
-------
8
A
C
O
a"
a
<3
v
c
A 1.27
Q 1.27
O 1.59
D 1.91
0 2.54
03.18
N 5.08
V 5.08
cm Square-Edge Probe
cm Sharp-Edge Probe
cm Side-Opening Probe
cm Sharp-Edge Probe
cm Sharp-Edge Probe
cm Side-Opening Probe
cm Square-Edge Probe
cm Sharp-Edge Probe
-25
20 30 40 50 60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 36. Probe performance at 15.24 m/sec and 7.5 deg angle of attack.
-------
A
B
25
20
15
10
o
N
-5
-10
v -15
c
Hi
-20
-25
Al.27-cm Square-Edge Probe
Ol.27-cm Sharp-Edge Probe
Ol.59-cm Side-Opening Probe
D 1.91-cm Sharp-Edge Probe
V 2.54-cm Sharp-Edge Probe
-<>3.18-cm Side-Opening Probe
N 5.08-cm Square-Edge Probe
V 5.08-cm Sharp-Edge Probe
20 30 40 50 60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 37. Probe performance at 15.24 m/sec and 15 deg angle of attack.
-------
§
A
C
25
20
15
1O
o
PI
V
C
-5
-10
-15
-2O
-25
A 1.27-en Square-Edge Probe
_Ol.27-cm Sharp-Edge Probe
y 1.59-cm Side-Opening Probe
3 1.91-cm Sharp-Edge Probe
0 2.54-cm Sharp-Edge Probe
:>3.I8-cm Side-Opening Probe
N 5.08-cm Square-Edge Probe
V 5.08-cB Sharp-Edge Probe
20 30 40 50 60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 38. Probe performance at 21.34 m/sec and zero angle of attack.
-------
A
C
25
20
15
10
o
X
I o
0.
-5
-10
V
-15
-20
-25
Al.27-cm Square-Edge Probe
1.27-cm Sharp-Edge Probe
1.59-cm Side-Opening Probe
1.91-cm Sharp-Edge Probe
2.54-cm Sharp-Edge Probe
3.18-cm Side-Opening Probe
5.08-cm Square-Edge Probe
V 5.08-cm Sharp-Edge Probe
I
8
20 30 40 50 60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 39. Probe performance at 21.34 m/sec and 7.5 deg angle of attack.
-------
8
A
B
o
N
E
o\
25
20
15
1O
-5
-10
-15
-20
-25
A 1.27-cm Square-Edge Probe
~ 1.27-cm Sharp-Edge Probe
1.59-cm Side-Opening Probe
1.91-cm Sharp-Edge Probe
2.54-cm Sharp-Edge Probe
_O3.18-cm Side-Opening Probe
N 5.08-cm Square-Edge Probe
V 5.08-cm Sharp-Edge Probe
8
20 30 40 50 60 70 80 90 100 110
Isokinetic Condition, percent
120
130
140
150
Figure 40. Probe performance at 21.34 m/sec and 15 deg angle of attack.
-------
40
0)
u
t-,
0)
0.
o
t*
rt
w
30
20
10
-10
-20
-30
-40
-50
-30
O 9.14 m/sec
A 15.24 m/sec
a 21.34 m/sec
J I
-20
-10
0
AP, mm
10
20
30
40
Figure 41. 1.27-cm square-edge probe.
65
-------
40
30
20
10
o>
u
h
o
a
o
be -10
c
a,
rt
OT
-20
-30
-40
-50
-30
O 9.14 m/sec
A 15.24 m/sec
O 21.34 m/sec
-20
-10
0
AP, mm
10
20
30
40
Figure 42. 1.27-cm sharp-edge probe.
66
-------
40
30
20
10
c
0)
u
o 0
t,
o
IN
be -10
c
•rt
rH
a
§
w
-20
-30
-40
-50
O 9.14 m/eec
A 15.24 n/sec
O 21.34 m/eec
I I I
.30 -20 -10 0 10 20
AP, mm H20
Figure 43. 1.59-cm side-opening probe.
30
40
67
-------
O 9.14 a/sec
A 15.24 m/sec
21.34 m/sec
Figure 44. 1.91 -cm sharp-edge probe.
68
-------
40
30
20
10
e
0)
u
IH
0)
Q.
I,
O
t* -10
•H
i— I
Q.
§
03
-20
-30
-40
-50
-30
O 9.14 m/sec
A 15.24 n/sec
O 21.34 m/sec
I
I I
_20 -10 0 10 20
AP, nun H20
Figure 45. 2.54-cm sharp-edge probe.
30
40
69
-------
40
30
20
10
0)
o
(H
0)
a
(H
O
u> -10
c
Q.
rt
-20
-30
-40
-50
-30
I
I
O 9.14 in/sec
A 15.24 m/sec
D 21.34 m/sec
-20
-10 0
AP, mm
10
20
30
40
Figure 46. 3.18-cm side-opening probe.
70
-------
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
Figure 47. 5.08-cm square-edge probe.
-------
O 9.14 in/sec
A 15.24 in/sec
D 21.34 in/sec
Figure 48. 5.08-cm sharp-edge probe.
72
-------
40
30
20
10
c
0)
OJ
a
o
(K -10
C
a.
-20
-30
-40
-50
-30
O 9.14 m/sec
A 15.24 m/sec
HI 21.34 m/sec
I I
-20
-10
20
30
0 10
AP, mm HgO
Figure 49. 1.27-cm square-edge probe at 7.5 deg angle of attack.
40
73
-------
40
30
20
10
0)
o
o
£
a
a
-10
-20
-30
-40
-50
-30
I I
I
O 9.14 ra/sec
A 15,24 m/sec
D 21.34 m/sec
I I I I
0
AP, mm
10
20
30
Figure 50. 1,27-cm sharp-edge probe at 7.5 deg angle of attack.
40
74
-------
-------
40
30
20
10
u
* 0
a
o
-10
a,
a
-20
-30
-40
-50
-30
I 1 1 1
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
J I
-20 -10 0
AP, nun
10
20
30
40
Figure 52. 1.91-cm sharp sharp-edge probe at 7.5 deg angle of attack.
76
-------
40
30
20
10
4)
O
0
Ll
t* -10
a
rt
CO
-20
-30
-40
-50
O 9.14 ra/sec
A 15.24 m/sec
a 21 .34 in/sec
I I I I I
-30 -20 -10 0
AP, not
10
20
30
40
Figure 53. 2.54-cm sharp-edge probe at 7.5 deg angle of attack.
77
-------
30
20
10
01
o
o>
Q.
be -10
Q,
rt
w
-20
-30
-40
-50
-30
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
-20
-10 0
AP, mm
10
20
30
40
Figure 54. 3.18-cm side-opening probe at 7.5 deg angle of attack.
78
-------
40
30
20
10
m
u
O
be -10
c
0,
«
w
-20
-30
-40
-50
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
I I I I I
-30 -20 -10 0
AP, mi
10
20
30
Figure 55. 5.08-cm square-edge probe at 7.5 deg angle of attack.
40
79
-------
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
-30
Figure 56. 5.08-cm sharp-edge probe at 7.5 deg angle of attack.
80
-------
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
-50
-30
-20
Figure 57. 1.27-cm square-edge probe at 15 deg angle of attack.
81
-------
O 9.14 m/sec
A 15.24 m/sec
21.34 m/sec
Figure 58. 1.27-cm sharp-edge probe at 15 deg angle of attack.
82
-------
40
30
20
10
0)
u
0
£
a
I
-10
-20
-30
-40
-50
-30
-20
O 9.14 m/sec
A 15.24 m/sec
a 21.34 in/sec
I I
-10 0
AP, mm
10
20
30
40
Figure 59. 1.59-cm side-opening probe at 15 deg angle of attack.
83
-------
40
30
20
10
HI
o
IH
V
a
t* -10
c
•H
i—I
a
I
-20
-30
-40
-50
-30
J I
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
J I
-20
-10 0
AP, mm
10
20
30
40
Figure 60. 1.91-cm sharp-edge probe at 15 deg angle of attack.
84
-------
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
I I I I I
Figure 61. 2.54-cm sharp-edge probe at 15 deg angle of attack.
85
-------
40
30
20
10
0)
y
U
O
c
•rt
i—I
Q,
«
-10
-20
-30
-40
-50
-30
-20
O 9.14 m/sec
A 15.24 m/sec
D 21.34 m/sec
I I I I
-10
0
AP, am
10
20
30
Figure 62. 3.18-cm side-opening probe at 15 deg angle of attack.
40
86
-------
m/sec
D 21 .34 m/sec
-50
-30
Figure 63. 5.08-cm square-edge probe at 15 deg angle of attack.
87
-------
O 9.14 m/sec
A 15.24 m/sec
O 21.34 m/sec
Figure 64. 5.08-cm sharp-edge probe at 15 deg angle of attack.
-------
NOMENCLATURE
b Laser beam radius
f Rate at which fringes are cut in
probe volume
p Probe static pressure
t Period of a-c component of PM tube
signal
V Velocity
6 Interference fringe spacing
B Angle formed by intersecting beams
^o Wavelength of laser
89
------- |