THE IMPROVEMENT OF WIND TUNNEL
DIFFUSER Cr~PACTERISTICS
By
HENRY G. WEBB JR.
B.Ae.E.,
Rensselaer
Polytechnic
1942
Institute
and
ZUP~~ICK
JOSEPH E.
B.S. University
of Pittsburgh
1943
SUBMITTED IN PARTIAL FULFILLMENT OF THE
REQUIfiEMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
IN
AERONAUTICAL ENGINEERING
From the
MASSACHUSETTS
INSTITUTE OF TECHNOLOGY
1949
Signature
of Authors:
HenryjG.
Webb Jr.
{Vose~h E. uubadick
Certified
by:
De#artment
of Aeronautical
Engineering,
Chairman,
Sept. 2, 1949
Department
~vraduate Students
,:"
"
.~
I
l
Committee
on
Sept. 2, 1949
Prof. Joseph S. Newell
Secretary
of the Faculty
Massachusetts
Cambridge
39,
Dear Professor
Institute
of Technology
Massachusetts.
Newell:
In accordance
with the requirements
of the faculty, we
hereby submit a thesis entitled,
THE IMPROVEMENT OF \~fJND TUNNEL
DIFFUSEH
ln partial
fulfillment
Master of Science
CHARACTERISTICS
of the requirements
in Aer~nautical
for the Degree of
Engineering.
/fHenry G. Webb,
Joseph E.~nick
j?
.
ACKNOWLEDGMENT
The authors
sincere
thanks
the preparation
to express
to the following
their
persons
tunity of working
Joseph
Bicknell
on the project
1n
for g1v1ng
us the opporduring
the
of the work.
Wright
for his aid in the construction
obtaining
necessary
the Aeronautical
machine
work
test equipment.
Machine
involved
Shop
during
To the sponsors
making
and
who were of assistance
and his suggestions
To Mr. Hatry Hilberg'of.the
Staff
appreciation
of this thesis.
To Professor
course
wish
the funds available
Also
to Mr.
for his splendid
J.R.
Wind Tunnel
and 1n
Maddox
cooperation
of
on the
the project.
of the Diffuser
Research
for the construction
To the typist Eleanor
the Vari-Typer.
Brothers
of the equipment
Mugnai
Project
for
of the equipment.
for her work performed
on
TABLE OF CONTENTS
Page
I
Purpose
II
Summary
III
IV
V
Introduction
Description
1
.
3
of Apparatus
A.
The Tunnel and Diffuser
3
B.
Nacelles
4
C.
The Windmills
D.
Damping Screen
5
E.
Instrumentation.
5
.
4
1.
Pressure Measurements
5
2.
Direction
5
3.
Tuft Studies
4.
Miscellaneous
Indicator
6
.
6
7
Test Procedure
A. Pressure Measurements
B. Wind Direction
and Windmill
7
Speed
8
Measurements
9
C. Tuft Studies
VI
Test Results
A. Velocity
and Discussion
.
. .
10
Surveys
11
B. Pressure Recovery
C. Turbulence
D. Tuft Studies
10
.
.
. . .
12
13
Page,
VII
VIII
IX
Conclusions
14
. . . • .
A.
General
B.
Recommendation
14
Remarks
for Further
Study
14
References
16
Appendix
17
A.
Nomenclature
B.
Pitot Tube Calibration
19
C.
Direction
20
D.
Pressure
Recovery
E.
Velocity
Surveys
F.
Run Index
G.
Figures
17
.
Indicator
(Drawings
Characteristics
and Data
and Data
22
26
30
and Photographs)
32
I
This experimental
to determine
investigation
was conducted
the effect of freely rotating windmills
and time variations
Wind Tunnel
PURPOSE
of velocity
diffuser.
optimum position
on the space
in a model of the Wright Brothers
It is required to determine
and blade configuration
the lowest turbulence
primarily
both the
of the windmill
level and most uniform velocity
to give
distribution
at the diffuser exit.
A further characteristic
to be considered
loss associated with each. diffuser windmill.
windmills
is the energy
Thus a comparison
of
with damping screens or other devices capable of produc-
ing the improvements
mentioned
above may be made.
II SUMlv1ARY
windmills
The installation
of several
in two positions
~n a model of the Wright Brothers
Wind Tunnel diffuser was tested.
distribution,
turbulence,
by the proper selection
single configuration
types of freely rotating
It was found that the velocity
and energy losses could all be reduced
of windmill
was optimum
location
and blade form.
for the simultaneous
No
achievement
of all three effects.
A description
diffuser
characteristics
velocity distributions
q (Y~U2) variation
variation
behavior
included.
is given.
employed
Data are presented
for the
at the tailpipe exit, and also as a weighted
Turbulence
results, but attention
in their determination.
measurements
The interpretation
are given as
is called to the inaccuracies
Diagrams
of tufts placed on the diffuser
used to estimate
to study the
in the form of contour plots of the
with duct radius.
quantitative
involved
of the methods
illustrating
the
and tailpipe walls are
of a modified
diffuser
efficiency,
energy losses, is explained.
Specific
c~nclu~ions "to be drawn from these tests are
made and recommendations
for further investigation
are given.
III
INTRODUCTION
The work contained
herein
tion of the effects of a windmill,
velocity
distribution
is an experimental
placed
and turbulence
The existing
problem
investiga-
in a diffuser,
level of the stream.
that brought
the work in this
thesis about was that the flow in the Wright Brothers
Tunnel
is of such a nature
as to produce
is these eddies that make it impossible
steady stream required
for the testing
the flow is satisfactory
by R.A. Summers,
a honeycomb
the maximum
of flutter models,
restrained.
type of testing
This problem
~ethod of attacking
or a series of damping
section.
The damping
screens
of fluid passing
through it, since the resistance
the velocity
variations
to the square of the local wind velocity,
although
in which
at or near
pressure
of the stream
is proportional
the high speed areas
losing more total head than the low speed regions.
flow is thereby
improved
velocity
the section
across
turbulent
motions
scale.
in two respects,
is reduced
The effect of damping
has been investigated
Schubauer,
ref. 6.
Nevertheless,
transf~s
screens
speed regions
as required.
flow in this manner
speeded-up
flow is subjected
is relatively
by the screen is reduced
In addition
to delay and possibly
to a positive
prevent
pressure
- 1 -
to
to adopt a device
to 'smoothing'
in the boundary
diffuser.
of
and G.B.
from high speed regions
the velocity
the
on. wind tunnel
section
it is more efficient
excess energy
in
them to motions
by H.L. Dryden
low, the amount of energy dissipated
The
along with decreasing
Since the speed in the maximum
a minimum.
the variation
of large scale by reducing
turbulence
which
and
is by
screen with a uniform
reduces
smaller
It
is discussed
the problem
drop coefficient
therefore
eddies.
ref. 5.
The standard
placing
undesirable
Wind
to have the uniform
for the present
a rigid model is completely
on the
out the
layer can be
seperation
gradient,
to the low
where
the
such as in a
A windmill
of appropriate
design, placed in a diffuser
and allowed to rotate freely is one such device capable of
producing
the same effects as the damping screen.
The windmill
is driven by the high speed regions and drives the low speed
regions producing
turbulence
an even velocity distribution.
in a manner analogous
It acts on the
to a rotating cutter.
If a
piece of wire fed into the cutter were such that the blade came
around at the same time the cut in the wire did the resulting
pieces would be the same as the ones fed in.
If the same rate of
feed was kept and the cutter rotated faster the resulting
pieces
would be smaller.
the
resulting
Similarly,
turbulence
in the case of the windmill
is a function of what goes into the windmill
and the R.P.M. of the windmill
There occurs two classes of turbulence,
termed 'coarse grained' and 'fine grained'.
the 'coarse grained' turbulence
turbulence
windmill
differentiating
The wind mill reduces
but allows the 'fine grained'
to pass through unaffected.
is analogous
generally
The blade pitch of the
to the mesh of the damping screen in
between the scale of turbulence which will be
reduced and that which will not.
The use of a windmill
employed
to any great extent.
subject is by A.R. Collar,
cal and assumes a small
velocity
distribution.
to improve the flow has not been
The only report available
ref. 3.
on the
Most of the work is theoreti-
steady variation
from a uniform
Very little test data is presented.
- 2 -
IV
A.
The Tunnel
DESCRIPTION
and Diffuser.
The tunnel
of and assembled
and diffuser,
in following
section,
the setting
section,
the diffuser,
pipe,
figures
TD 1n notation,
order:
chamber,
the blower,
the bell mouth,
and the 22.3
is comprised
the expansion
15 inch constant
the
inch constant
section
tail
66, 56, 57 and 58.
The power
It is an American
cubic
OF APPARATUS
source
High
for the tunnel
Speed Blower,
feet per minute.
is a centrifugal
No. 245 and rated
blower.
11,000
at
The tunnel was run at a constant
R.P.M.
1780, since the driving motor is a 3 phase induction motor.
The outlet of the blower is a rectangular section of 19 x 26~
of
inches with
the centrifugal
impeller
e
to the right viewing
offset
into the stream.
To prevent
blower
gap of ~ inch exists
The flexible
seal,
between
figure
vibrations
from being
the blower
57, between
strip of ~ inch by 4 inch rubber
transmitted,
and the expansion
section.
these two is made with
bolted
to the flange
a
a
of the
8
blower
and to the flange
the irregular
series
of
velocity
distribution
section.
produced
To improve
by the blower
a
6 screens, 16 mesh and .009 inch dia, are evenly spaced
1n the expansion
section,figure
to the last screen
expansion
of the expansion
section
66.
in this section.
A series
of tufts were
The construction
is of ~ inch plywood
and tapers
tied
of the
from a
19 x 26~
4
inch rectangular
blower
flats at the settling
The settling
8
outlet
to a regular
octagon,
4 foot across
chamber.
chamber
is also of ~ inch plywood
construc-
4
tion with
circular
holes
a tuft observation
port of 8 inches
port on the top for the light
are located
the regular
on the vertical
octagon
at station
The bell mouth
by
source.
and horizontal
12
The
inches
and a
four static
centerline
of
1, given by figure 66.
and all of the tunnel
of .042 inch gage steel with welded
- 3 -
joints
aft 1S constructed
on each particular
section.
The contraction
satisfactory
cone surfaces
are not completely
because of warping and generally
poor fabrication.
The 15 inch constant section contains
equally spaced on the periphery
4 static holes
at station 2, given by figure 66.
This section and the 4~o
half angle diffuser
section
are a scaled down model of the Wright Brothers Wind Tunnel.
to the lack of rigidity of the gage of the metal a plywood
was used to bring the diffuser
Due
frame
section into round at the wind-
mill.
B.
Nacelles.
The forward nacelle,
figures 59 and 67, is a scale
model of the Wright Brothers Wind Tunnel
and supports.
fan drive motor fairing
A scaled down model of the propeller
in the full
size tunnel is not present in this set up, since the blower
produces
nacelle
the stream in this case.
is removable
to take the
A cone at the end of the
%
{nch shaft of the windmill
hub,
figure 66 and 71.
The aft, nacelle,figure
68,does not exist in the Wright
Brothers Tunnel but is used here as a mount for the large aft
windmill.
c.
Windmills.
The forward and aft windmill
chord and diameter
are made of ~
16
are very similar except ~n
of the blade figures 69 and 70.
inch 17ST dural having a constant
chord, with the
leading edge rounded and the trailing edge tapered.
taper being about 15% of the chord.
The blades
The width of
The blades are riveted to a
threaded steel shank which in turn mounts
into the wind mill hub,
figure 71.
This one hub is used for both the forward and aft
windmills.
The twist was put into the blades by the device.
shown in figure 65.
This allows for a given increment
per inch of the blade.
the protractor
of twist
The total twist was then checked with
arrangement
of figure 60, which was also used to
set the blade angle for the various configurations.
- 4-
The windmills
D.
are tested with their respective
Damping Screen.
An 18 mesh .010 inch diameter
the
nacelles.
3
4
screen S was tacked to
inch plywood frame at the end of the tailpipe.
was made with this configuration
to determine
One run
the effects of
screen on turbulence.
E.
Instrumentation.
1.
Pressure
measurements.
The velocity
nacelle were obtained
distributions
ahead of the forward
by a pitot static tube mounted on the
survey rake, pitot tube and wind direction
figure 66.
The diffuser
readings.
obtained
indicator
and tail pipe were removed
The data for the velocity
distributions
suppor~
for these
was
from a 31 tube survey rake, figure 63 and 72.
total head tubes and 2 static tubes were connected
vertical c;llcohol
manometer
board, all pressures
relative
pressure.
to
atmospheric
in inches of water.
The 29
to a
being measured
The scale is calibrated
The average velocity was obtained
readings of the {our static holes in the settling
from
chamber,
figure 66 and the four static holes in the 15 inch constant
section, figure 66 and 58, connected
to the same mpnometer
board as the survey rake.
2.
Wind Direction
Indicator
By following
vane direction
scale
indicator
d.isturhance
in flow direction,
can be used to measure
in the flow.
shall, for the purposes
including
the variations
The definition
of turbulence
as
these large scale disturbances.
to build a single vane
indicator with an oil damper proved unsuccessful
the large effect of temperature
upon.
the large
of this report, be interpreted
After several attempts
indicator
a
employing
on the oil viscosity,
only aerodynamic
of air speed.
A disadvantage
ing ratio must be used if the natural
- 5 -
an
damping was decided
This would have the added advantage
independent
because of
of a damping ratio
is that a low damp-
frequency
is to be high
enough to respond
to a reasonable
range of disturbance
frequencies.
The final design,
Sec. IX-C had stability
fig. 73, described
up to a deflection
above which the indicator
was slightly
more fully in
of about 40°,
unstable.
The three vanes are made of .005 inch brass shim
stock while the supports
tubing
are of hollow stainless
.034 inch outside diameter.
The vertical
.020 inch dia. piano wire with conical
is assembled
by soldering
two adjustable
pivot,
3.
screws
tips.
and is mounted
steel
pivot 1S
The indicator
in a fork with
that hold, act as bearings
for the
fig. 64 and 73.
Tuft Studies
A series of 2 inch tufts were attached
scotch tape to the surface of the diffuser
fig. 62.
With the large windmills
necessary
to use the Strobotac-Strobolux
the tufts 1n the diffuser,
and tailpipe,
installed
especially
with
on N2
it was
apparatus
to see
for the high windmill
speeds.
4.
Miscellaneous
Ihe windmill
speeds were measured
Radio Co. Strobotac-Strobolux
the same reason the ambient
pressure
were recorded
the Wright Brothers
for references
air temperature
for each run made.
together with all pressure
with a General
measurements
Wind Tunnel.
- 6 -
purposes.
For
and atmospheric
These
data,
etc. are on file at
V
A.
TEST PROCEDURE
Pressure Measurements
and Windmill
Speed
Before any runs were made all connections,
tunnel static orifices,
from the
pitot tube, and rake, to the manometer
board were checked for leaks.
With the single pitot tube installed on the support
traverses
in four directions
were made for a complete
velocity at the end of the 15 inch dia. constant
survey of
area section.
Dynamic pressure heads were recorded and corrected
for instru-
ment error.
The variation of static pressure
over the cross section
of the tailpipe exit was checked with the single pitot tube.
Since the variation was found to be small all subsequent
were made with the survey rake incorporating
tubes.
two static pressure
The position of the static holes was adjusted
error caused by proximity
to the adjacent
runs
till the
total head tubes was
eliminated.
After allowing the motor to accelerate
ning speed on 220 V. the windmill
rotational
to normal run-
speed was taken
with a Strobotac-Strobolux.
The hub and one blade were marked
to allow quick determination
of the true speed.
ings were then taken.
approximately
read-
The total head orifices on the rake were
1 inch inside the tailpipe
the large downstream
Pressure
windmill
installed
for most runs, but with
some pressures
were
taken with the rake pushed inside the tailpipe up almost to the
windmill.
No trouble was experienced
with vibration
of the rake
in either position due to the rigidity of the mount.
level in the manometer
atmospheric
nearest
pressure
the horizontal,
was adjusted till the tube connected
read zero.
.01 inch water.
All heads were measured
pressure was measured.
to
to the
The rake was set at four angles from
_45°, 0°, + 45°, and
were read and recorded
The alcohol
~o,
individually.
and pressure
Thus a time average
If the flow was very unsteady
might take as long as 10 seconds, but usually
7
heads
each reading
all heads were
read for each rake position
in ahout35
seconds.
line total head was read four times during
variation
of the average
time interval.
The variation
pressure
heads
alcohol
density,
increased
B.
flow velocity
since
accuracy
ment
taken.
instrument
although
were estimated
most
oscillations
deflection
the following
oscillated
sinusoidally
oscillations
fairly
the averaging
elsewhere
simple
Readings
finding
amplitude
-
of many
detected.
a time average
If the vane
of
~ + 6°.
i
10°, the
If the
-
superposed
and varying
but with
can be made.
time interval
The
10 seconds.
the deflections
The
frequency,
is not as simple,
different
were
can he read
case may be taken.
7T
the instru-
were readily
behind
for
deflections
at its natural
as true flow direction
ful only in comparing
angular
the instrument.
he (~)(~100)
was approximately
in this report
was vertical
the movement
a constant
is composed
estimate
to he interpreted
of
and the
By viewing
into
the reasoning
would
the process
accurate
built
frequently
with
deflection
movement
visually.
of lower frequency
To illustrate
indicator
The
variation
be small,
and maximum
lighting,
to the protractor
oscillated
time average
would
of the indicator
The average
from above, w~th proper
by reference
to be negligible.
Measurements
The axis of rotation
of the indicator
over this
is not required.
Wind Direction
all readings
each run the time
for temperature
the correction
the center-
could be checked
was found
are not corrected
Since
practice
As pointed
presented
used
for
out
herein
variations,
a
are not
hut are use-
configurations.
taken with
the instrument
at the following
positions,
--
- 8 _
1----
HOR.
t.
c.
Tuft Studies
By means
of tufts placed
the type of flow along
following
in the diffuser
the wall was placed
and tailpipe
in one of the three
categories,
1.
Relatively
2.
Rough,
smooth
unseparated
but with
scale turbulence
3.
Separated
- 9-
considerable
large
VI
A.
Velocity
TEST RESULTS AND DISCUSSION
Surveys
Fig. I shows a nearly uniform
the diffuser
inlet.
The velocity
velocity
decreases
wall on the left side, facing upstream.
the slightly
eccentric
irregular
outlet
distribution
at
more rapidly near the
Whether
this is due to
fairing of the bell mouth or to the
from the centrifugal
fan is uncertain.
result is that the flow tends to separate
more readily
The
on this
side of the diffuser
The data obtained
the tailpipe
Appendix.
has been plotted
q or q20
or average,
is a quantity
from the duct centerline,
the horizontal
Appendix.
LP
11
~)12
satisfies
The quantity~,
distribution
duct to regions
energy
the energy
from
in the
q2D at any radius r
lS,
at the radius
of mass, but introduces
the equation
for conservation
the great change
is transferred
near the wall.
Fl.
of
from Eqn. 14.
in the velocity
At the low blade
from the center of the
Unfortunately
the section
at these high rotational
of the
speeds
losses.
To reduce the energy
blades were twisted
position
or the mean q is calculated
blade near the hub is stalled
thus increasing
on the radius
heads obtained
caused by the small windmill
angles considerable
This
2
the continuity
Fig. 2 shows clearly
fig. 6
(I)
a slight error by not satisfying
energy.
J
8, the number of pressure
This equation
dependent
used to calculate
(h :
r.
q to the mean q.
line as is the data presented
rt~
=
in the
on fig. 2 through
but not on an angular
reference
The equation
on the curves
the effect of the fans on the q
the data has been plotted
as a ratio of a weighted,
wheren
directly
To show more readily
distribution
weighted
from the rake survey at the end of
loss caused
by blade stalling
to give the small windmills
- 10-
F3 and F5,
the
fig. 8.
With a twisted blade the effect of a change
is much less than that for an untwisted
clearly
in fig. 4.
uniform
twist, as for Fs, and a helical
obtained
blade.
This is shown
The twist of Fs is intermediate
twist.
between
a
The results
from Fs and Fs show that when the blade twist is close
to that of a helix a change
in pitch distribution
effect on the blade loading distribution
blade.
in blade angle setting
For this reason an analysis
of the blade element propeller
mounted
a fairly uniform
The blade angle setting
of the windmill
velocity
or approaches
and so the actual distribution
nacelle
distribution.
for the best average
is such that the flow still separates
the diffuser
q distribution
on the upstream
av~age
required
by any form
lead to large errors.
of the weighted
either of the three windmills
N1 will produc~
than for an untwisted
theory will
From a consideration
has a smaller
distribution
separation
in
is still not too
satisfactory.
The action of the windmills
end of the. diffuser
operates
uniform
is quite different.
in a flow having reached
velocity,
the windmill
mounted
Here the windmill
its greatest
and the tailpipe
helps further
following
to give a uniform
F2 has much the same effect
tion.
Windmill
behind
the hub and stalled
diviation
immediately
velocity
With F4 this loss is eliminated
distribution
is obtained
from a
behind
distribu-
as F1 but the wake
section of the blades
noticeable.
actual distribution
at the downstream
1S
more
and a good velocity
both as a weighted
average
and as an
for a blade angle of about 20~ as shown by
fig. 6 and fig. 53.
B.
Pressure
Recovery
The pressure
conversion
reflects
of kinetic
recovery
energy
both the relative
the diffusion
process,
a measure
to potential
magnitude
of the efficiency
energy.
As such it
of the total head loss in
velocity
distribution
at the
exit.
From fig. 7 it is seen that the tailpipe
pressur~
of
as well as the loss of static pressure
head rise caused by a non-uniform
tailpipe
1S
recovery
by about 6%.
increases
This is due to the improvement
- 11 -
the
in
velocity
distribution.
recovery
if the velocity
3%.
The possible
distribution
Thus the values of pressure
reflect primarily
pipe.
further
were uniform
recovery
in pressure
is only about
at the tailpipe
the total head loss in the diffuser
This loss is caused by skin friction
eddylos~es,
increase
and simil~
losses associated
exit
and tail-
on the duct wall,
with the nacelle-fan
group.
Windmill
settings
decreases
Fl
by reducing
the losses at large blade angle
the eddy losses,but
at low blade angles due to stalling
of the fan.
F2 has a limited effect on the diffuser
effect at all blade angle settings
and blade root.
this windmill
wall.
C.
near the hub
loss, and reduce the other losses considerably
less effect
The windmill
separation
at
than either F3 or F6•
recovery
This is caused by separation
and by earlier
directly
flow it has an adverse
due to stalling
The large drop in pressure
nacelle
Since windmill
All twisted blades on the other hand eliminate
low blade angles, F4 having
apparent.
loses its effectiveness
due to Nl
is
on the aft portion
of the
of the flow on the diffuser
has little effect on the losses associated
with the nacelle.
Turbulence
The direction
indicator,
fig. 73 which is described
Sect. "IV-E-2 and IX-C gives a visual
turbulence.
eliminating
promoting
The intensity
separation
dissipation
in smoothing
By preventing
begins
Furthermore
the turbulence
by smoothing
to increase
it again.
on the duct centerline,
ref. fig. 10.
the blade
windmill
Fl
stalling
is more apparent
Fig. 11 shows that the
obtained
chiefly
to the absence of blade stalling.
of turbulence
to that
torque.
This latter effect
results
added to the reduction
out large scale
in the diffuser,
from F6 are slightly
its scale by
might be compared
up to a point where
by
as well as by
by reducing
out an engine's
separation
of low frequency
can be reduced
of the turbulence
the action of the windmill
of a flywheel
reduces
of turbu~mce
in the diffuser" wall,
means of the windmill.
turbulence
indication
in
better
than with Fl, due
The damping
screen
for the few configurations
- 12 -
tested.
As shown in fig. 12, windmill
diffuser
has little net effect on the turbulence
since the
decrease
due to improved
offset by the
increaseq
turbulence
being twisted
turbulence,
the diffuser
D.
flow in the diffuser
generated
to a regular
although
F2 at the end of the
1S
by the blades.
Windmill
F4,
helix, gives a marked decrease
in
not as much as that given by F1 or Fs, since
separation
is not completely
eliminated.
Tuft Studies
The flow along the diffuser
and tailpipe
wall can be
studied by tufts.
The motion of the tuft indicates
whether
flow has separated
from the surface or is about to separate.
The flow is quite good with the large nacelle
With N1 in the flow separates
with the small windmill
being seen.
badly.
This is improved
N1 out.
greatly
F1 at low blade angles, no separation
The flow is almost equally
angles with windmill
the
F5•
The improvement
is not as great, considerable
separation
smooth at low blade
in flow with F2 and F4
occurring
on the left
side of the diffuser.
From a consid~ation
and pressure
recovery,
it is apparent
N1 is the largest component
of windmills
by reducing
of the tuft studies,
turbulence,
that the drag of nacelle
of the diffuser
head loss.
The use
does not reduce this loss, but does decrease
separation
in the diffuser.
by N1 and that associated
can be reduced
by
thp.
The turbulence
with flow separation
nroper choice of windmill.
- 13 -
losses
created
in the diffuser
ti
-J
2
et.
UJ
(f)
:J
U-
10
_.
..
lL t0
Z
<{ 0
}-
:l
~
Z ~et
0
r:JlL
2
CO
_0
Cf.v
U')
~
0
00
uJ
cf)
ell
W
uJ
d.
~
o ~
~;
UJ
d
UJ
:J
oCf.
V
O~H'Nl-t
m
o
~ri
Ci.
UJ
>-
~z
:J
°0 et
rZ
(/)
Gl
tU
~
uJ
~tO
g2
:::>
If)Ci.
o
~
I
-q-
C'J
0
I
~
I
~
,
\0
,
<0
FIG. I
(f)
UJ
u
1
I
2
en
0
:>
4
fi.
0
Z
:J
l-
..J
Z
uJ
c.J
UJ
U
W
£)
Ii
Z
uJ
et
w
W
LL
II
tL
aJ
rt oooooouJ
II
oooooocr
-en CO " \0 lO'V
Z2ZZZZZZ
m,...cocnQ=~-..r
211 1/" "
en ~cC..~~<C..~~
0
d)
~
>-
(()
C!C1aetetet
::>~:J:J:J::>:J::J
Cl
0<10t>0006
"b'/V\
N
0
lC
ro
-
guJ
(f)Q.
LL
Z
0
u
C)
:l
t<{
ct
0
z
CO
r-~"
ULL
00
--l
UjO
uJ
>2
~~
~
a
w
>
«
o
- - 2aQ,a
>-~r
o
FIG 2
cJ)
l1.J
<J
I
I
Z
en
:J
0
«
et
0
2
t::>
ro
~
J-LU
--
(f)
..J
0
cO
~
>en
1ft
I-
(/)0- u.
00. 2
..J 0
>-
t:~
z
0
ULL
00
~
.J
wo
u..J
OOV
000
"
II I'
1O~1f'\
c!Le.~
~lnu)
--ZZZ
£YO:
::l:J::J
(1
0<]0
"""'M
N
-
0
FIG 3
en
UJ
-zI
I
0
C)
<Il
~
:::>
U.
u
z
B
o
0
)2 :>a:
~~
<{
a
UJ
~
o
Z
0
J=
::>
-lC
aUJ
~a.
-00.
>~
r-~
Uu..
00
..J
UJO
)2
uJ
~t:r
<3:
ct.
UJ
~
f/)
.J
(0
0
>-
~
CIl
~
Z
Q
..
tz
0
-
t1
~
~
~
II
Z
0
\)
o 000
<'
00000\.0
CO,....d)lt}~(1)
..
0
.' " .' "
<!L~C!l..ec...~\!L
I'--COIDO-N
___
NNN
:J:::>::>::>::>:J
2ZZZZ2
ct.etCf.Cf.Cf.et.
0<30[><>0
.
(0
,'VN\.
o
(\J
0
o
FIG.4
lU
en
I
(j
,
2
et)
0
:>
<:!
Cf.
Z
0
-
::>
I-
oJ
Cf.
t-W
(/)(1
On
>-~
r-~
.J
cf)
u:
u.J
Ci.
Ioooooo
•
.. "
II
"
..,°0000000
ZO)CO,,~lOv(t')
iu
() <!l.c..e..\C..~~
rtlrr)rOct)
O<lO(><>oOA
II
"~N\
:J:J:>:::>:>:>:>:>
dCf.dCf.arta.et.
ZZ2ZZZZZ
NN~
~tOl'~O-NtO
O tto
~
>-
(f)
I'll
LC
2
2
Q
t-
z
0
-~IL
Z
0
u
-
N
0
co
FIG.5
en
~
0
:>
f/)
,
u
J:
UJ
UJ
::>
UlL
~~
<{
>
UJ
«
a.
>Z
z
o
°0
-.J
WQ ~d
o
o
o
o
FIG.G
U')
I
lU
u
I
2
(()
Q
:>
~
IU
.J
~
Z
<t
uJ
a
~
.J
dl
Ul
>1
&
uJ
a.
J
a:
~
LL
0
0
2
UJ
I-
~
':I.t~
lJJ
,., r4
-iLlafZ2_.2,..
2z-i"izZ2-
>
0
o<JOl><>oOA
U oooo~aDO
ill I-t-t-t- t-rt-
a
a
uJ
:J
(/)
en
lU
CJ.
a.
d.
llJ
tf)
:>
u..
lL
a
u.
0
0
2
IJJ
~
"Lb<
~~~
iz"i
222
Illl
... r4 ,..
n
Q
&-0----0-\0
-0---..co
~
o
Ql..
J
0
l1J
t1
~
lU
ILl
Ul
FIG. ?
o
~
~
0
tn
0
\P
,...0
0
co
o
m
(/)
:J
-
lJ
<{
CY
(()
>
uJ
..J
l!)
z
«
uJ
0
<l
..J
..J
..J
~
0
Z
~
l3
uJ
rJ.
~
£:)
I
OOH.9't~
,
......
OCO('lO<O\n(Q
Od1I'-Nr-~N
CJ\CX)r-,....<..9
It)
CO--dl
'-D
~cri~lti~"":'cri<riQ
/
o
(,()
III
It'
:J
~
N
~
2
0
ll..
a
~
~
'P
en
u.: o~lOm~m\SicOci
mcol"\O
lI)lnltlltl
~
jf
//
:> N
-oooooooq
(f)
C!l..l[
lftrll
II
II
JI
o
co
I
/
I
I
>tNb'HS
o
~
0
o
FIG.8
g]
u
I
z
J
I
(/)
0
<{
et
UJ
(!)
...J
2
<1:
W
0
~
..J
t-
<J 0 [>
I- J-
Q 0 £)00
r
0
I-
<>
0
(0
0
~
It')
0
0
I'
0
\D
FIG.9
(fl
lU
C!l...
J
d.
£!)
z
3
a
_ ~
~
r4
~
ll.
Z 2
Cl iiz-z-z-
~
UJ
o
V)
'~'d''tj
>
-
~
UJ
0
iClJ!
0001
en
r-
Z
uJ
~
lL
-
Z
0
o.
cJ. r
0
Z
~ 0
U
z
II
(!)
to
.J
en
0
10
~
~
_ N rn - ~
(t\
-
~~~~~~K~~
~u.-lCa
~iiZ[t~~oo
~g~r-rto<J[Ju~'tJ.o..<J)J
.ijJ
rJ tn
~t0
~
0
et)
I1J
uJ
Cl
l!)
~o.
o
0
0)
~
~
In
0
ca..
J
uJ
(!)
~ en
uJ
..J
~
Z
<!
>
uJ
a:
q
t-'
lLJ
«
Z
(!)
..J
:J
«
Cl
0
llJ
uJ
.J
U.
(,)
l-
0
£il
.J
«
0
uJ
z
~
lu
.~
... f\
o
...J
~.
~
D
l:l
,f:)J3A)
C'I
--~--- o
o~
«
vd
a -~"OUle'\l
'0
""'0-
~.
o
~~
-j(\
~
.a.oco
s '3 =n:j~da - e..Y
z
<r
a
uJ
<{
-.J
en
~
z
0
F
u
l1J
J
11.
uJ
0
a:
2
~
:J
~
«
\!)
<{
:J .J
c:J. :J
- F« 2
'0
a
-2 :>
2
0
0
V
<(
x
a
~
CJ
ILl
0
o
\0
-&.3 3l:1~~
o
6
N
0
(t)
~
0
0
It')
0
U)
0
"
0
co
o
m
FIG.IO
rn
UJ
111
d
t!)
tU
0
I
~
lJ..tIl
z-
0
Z
..
c:J. t-
0
~
l1J
J
l?
z
<t
uJ
a
j
CO
~
Z
0
t-
o
W
J
en u.
0
r" uJ
oJ
~
_ rl
r{\
tr
Q
-
..uJ
N
d\
.JJI
-NIfl
Q. ~
(fl.
.
ell.
en..
tl' ~ u.=
D
d)
D 0 000
d)
~~~t ~~ ~.-:
0.
~ ~ If»
0
~i2-2-iiZZ2Z
t
If'
~~~~~r r r r
> 0
~
o<lCJo~bcf<fc(
~~~
Yo-
0
)(
<t
~
0
~
a
",
~
III
0
0
U)
0
,.....
g
en
IlJ
l1J
a
1LI
~
I
a
~
UJ
..J
2
~
cq:
CJ
W
~
J
£0
~
2
(J
0
t-
IlJ
..J
lL
UJ
Q
d
..J
<t
:>
Z
"
<{
UJ
<{
<i
C1
g-
~
f:f-<f
N
X))UI
ev
:>
V
a-
a.
uI
>
I
<t
~
2
:J
U 0 2 a
:J <i
0 ~ d. .J
~
j
Z a
Z
0
u
0 II
z
t-
(J
UJ
a
0
------000
'"
S 3"3 ~~"3
Yo-Y'a.
~ '.
c:a..
I
0
W
~
rt.
uJ
w
CI)
FIG. It
~
1"\
0
0
~
0
In
0
\SI
0
,.....
0
()
(/)
J-
z
I~
~ 2
W Z
J
ILl
0.
2
<[
~
«
~
t-
0
z
./
a ~ ci5
::> ...,.
~ z
Z
<3: ~
UJ ~
~
uJ
u
o
a. t~
rf)
0..
0
t; z v ~
~
5
cJ - Z ,.,
:> ~
Q ~
r:t.
~
z
~
l!)
Z et.
-:l
u.
Z
o
::>
~
u
~
....
o
)(
~
C!.
uJ
~
«
o
o
<0
o
o
v
I/)
u.J
lLJ
Cl.
t!)
Co.
0
,
o w
lI)
o
'-D
Q
I()
o
(J)
IlJ
.J
~
2
<:{
u.J
o
«
..J
co
~
z
o
F
(J
uJ
.J
1L
IlJ
o
r:i.
<
..J
::>
C)
Z
«
«
uJ
~
C!.
ILl
>
<t
o
N
o
o
o
m
I()
o
'"
o
\D
tn
~
o
et)
o
N
o
F(G
o
. 12.
fI)
1LJ
W
Ci
C)
UJ
J
o
ca..
'"
Z
lU
2
(t
lJ.J
:J
Cfl
<l
lL"
c4
IU 2
Z
2
0
0: t-
0
IU
..J
l!)
2
IL.I
~
£)
..J
<t
lO
~
Z
0
F
w
U
.J
lL
Ul
z ~o
CJ
i[
~
:J
<!
Z
~
U 0 "4' d
2 <l
Q
:) .J
Z ~C! d ::>
~
::>
Z
0
z
0
u
<t
b
l1J
CJ.
~
~
><
0
o
U>
~~
(f)
A-
~
r:
[j
Q.
N d\
t-=
II
o <J
~~
o
N
"""'",
t~
GJ
o
0
N
0
ct)
0
V
(I)
d
UJ
I1J
<!l...
I
~
0 0
10
0
\D
0
"
o
(1)
UJ
..J
C)
z
IlJ
«
«
a
.J
CO
~
Z
0
u
~
uJ
.J
lL
lU
Q
tJ.
..J
4
:l
u
z
<{
III
~
\J
rJ.
IlJ
>
<{
o
_.... rfI
~[t
o
~
£)
,
UJ
C)
uJ
W
d
en
FIG.I'3
0
N
0
ct)
0
V
~
0
\0
0
""
0
CO
(J)
o
Explanation
The diagrams
observer
would
The
the tailpipe
inside
area between
circles
represented
on the following
see looking
wall,
of Tuft Study
upstream
the diffuser
flow, unseparated
represent
the tailpipe
and middle
between
wall.
as follows,
flow
Separated
into
the outside
by cross hatching
Rough
pages
and the area enclosed
represents
Smooth
Diagram
The
what
the
thus,
circle
represents
the middle
and
type of flow is
FfC.14
TUFT
STUDIES
CONFIGURATION:
NOTED
FIG.IS
TUFT
STUDIES
CONFIGURATION:
TDN,F.
~"400
FIC.,..lt;
TUFT
STUDIE.S
CONFIGURATION:
TDN.Fs
~-700
~esoO
Fl Coy.l1
TUFT
STU DIES
CONFIGURATION:
TDN,N2F2.
FIC.. lB
o
TUFT
STUDIES
FIG. 19
TUFT
STUDIES
(3 = 60°
(3
~
I:
30°
~c
I:
40°
2.0°
VII CONCLUSIONS
A.
General Remarks
Detailed
conclusions
to be drawn from this investiga-
tion are as follows,
(1) The presence
of a large nacelle
inlet increases
turbulence
induces
flow separation
earlier
near the diffuser
and energy losses greatly,
on the diffuser
and
wall.
(2) The velocity
distribution
at the tailpipe
poor both with and without
the upstream
nacelle
(3) The use of a windmill
upstream
nacelle
does not greatly
tion at the tailpipe
energy
improve
losses, and decreases
turbulence
mounted
reduces energy
appreciably
at the tailpipe
blade stalling
if tip stalling
only
but reduces
a good velocity
distribution
Recommendations
at either
root or tip, although
wili normally
occur.
reduced
turbulence
tested.
for Further Study
The windmill
mounted
results.
velocity
should have sufficient
screen appreciably
for the few conditions
uniform
outlet
in the diffuser
losses only slightly,
and promotes
(6) A damping
most promising
reduces
exit.
twist to prevent
B.
distribu-
turbulence,
near the diffuser
(5) The blade of the windmill
it is doubtful
the velocity
the area of flow separation.
the amount of flow separation
slightly,
installed.
on the re~. of the
exit, but does decrease
(4) A windmill
reduces
mounted
exit is
at the diffuser
It is capable
distribution,
exit showed the
of giving a fairly
while at the same time reducing
the level of turbulence.
To reduce the turbulence
damping
screens
is suggested.
screen is considerable,
still further the use of
Since the pressure
any reduction
of energy
- 14 -
drop across a
losses elsewhere
1S
desireable.
The best way to do thi~
nacelle, but since considerations
into this decision
is merited here.
1S
to remove the forw~d
other than aerodynamic
enter
WB~T no further discussion
in the case of the
A windmill mounted on the forw~d
nacelle
could be used, since its favorable effect on energy
losses has
been demonstrated.
The next step should be to design. and test an improved
wind direction
incorporate
indicator.
The indicator
should preferably
the following characteristics,
1.
Single vane type
2.
Damping
ratio approximately
0.7, independent
of
air velocity
3.
High natural frequency
4.
Responsive
at all ai~ velocities
only to change in direction
of air
velocity.
The single vane type will have no undesireable
large deflections
accurate
and a high natural frequency
response characteristics.
these requirements
aerodynamic
are needed for good frequency
For the instrument
should be independent
the damping
ratio and
of air velocity.
All
non-
springs and dampers, but by doing this the instrument
to changes in air speed, an undesireable
Some compromise
will have to be made, but further
study of this problem is needed.
A method of recording
indicator movement which has no influence
instrument,
to have the same
can be achieved by incorporating
becomes more sensitive
feature.
A damping ratio of 0.7
'under all flow conditions
natural frequency
at
and can be made small enough to give an
local value of wind direction.
characteristics
instability
the
on the behavior
such as a type of optical measurement,
of the
would be most
satisfactory.
With an accurate wind direction
of turbulence with such combinations
screens as has been suggested
indicator
of windmills
the reduction
and damping
above should be investigated.
- 15 -
VIII
1.
P£FERENCES
Peters, He: Conversion of Energy in Cross-Sectional Diffusers
under Different Conditions of Inflow. T.M. No. 737, N.A.C.A.,
1934.
2. Patterson, G.N.: Modern Diffuser Design.
Aircraft Engineering, vol. X, No. 115, Sept. 1938, pp. 267-273.
3.
Collar, A.R.: The Use of a Freely Rotating Windmill to Improve
the Flow in a Wind Tunnel. R. & M. No. 1866 British A.R.C. ,
1939.
4.
Den Hartog, J.P.: Mechanical Vibrations. Me Graw-Hill Book Co.,
Inc., 1940, pp. 221-228.
5.
Summers, R.A.: An Investigation'of the Effect of Honeycomb on
Large-Scale Disturbances in Wind Tunnels. Massachusetts
Institute of Technology. Thesis for masters degree in aero.
engr., 1946.
6 .. Dryden, H.L. and Schubauer, G.B.: The Use of Damping Screens
for the Reduction of Wind Tunnel Turbulence. Jour. Aero. Sc.,
vol. 14, No.4, April 1947, pp. 221-228.
7.
Schwartz, I.A.: Investigation of an Annular Diffuser-Fan
Combination Handling Rotating Flow. R.M. No. L9B28, N.A.C.A.
25 Apri I 1949.
- 16 -
IX APPENDIX
A.
Nomenclature
A
cross sectional
CL
lift coefficient
c
damping constant
D
duct diameter
h
total pressure
I
moment of inertia
k
spring constant
1
centerline
area of duct
1
ft lb sec radft
head
inches of water
2
slug ft
ft lb rad-1
distance
rotation
to quarter
undamped
natural
from aX1S of
cho~d of surface
static pressure
q
local dynamic pressure
weighted
cycles sec.1
frequency
p
head
dynamic
inches of water
head (Y~U2)
pressure
inches of water
mean dynamic pressure. head at any
particular
r
inches of water
head at
any given radius
q
ft
station
radial distance
inches of water
from duct center-
line
S
projected
T
torque
u
local free stream velocity
surface
area
ft lb
angle of attack
damping
ratio
angular
POSt
ft sec.1
rad
ition
rad
time mean angular deflection
rad
maximum
rad
angular deflection
coefficient
of viscosity
2
Ib sec ft-
slug ft -3
density
- 17 -
Subscripts
1,2,3
i
Configuration
T
station
(see fig. 66)
location
uncorrected
for instrument
error
Hef. Fig.
symbols
Tunnel
only, up to but not including
66
diffuser
D
diffuser
N1
nacelle
66
and tailpipe
(mock up of WBWT
prop. and motor
67
fairing)
N2
aft windmill
F1
small windmill,
untwisted
blades
8,
small windmill,
linearly
twisted
8,
small windmill,
variably
twisted
large windmill,
untwisted
blades
F4
large windmill,
constant
H
windmill
S
damping
lR mesh,
nacelle
68
pitch
69, 71
8, 69, 71
8,
70, 71
8,
70, 71
71
hub only
screen
blades
69, 71
at end of tailpipe
.010. inch dia.
- 18 -
B.
Pitot Tube Calibration
The pitot tube used in Runs 1-3'was calibrated in
Run 43 in the M.l.T. Student Wind Tunnel ag~mst the 'standard'
pitot tube used by the Wright Brothers Wind Tunnel. The procedure was to make two runs over the same speed range; first with
the standard pitot tube mounted in the test section, then with
the uncalibrated pitot tube and mount in the test section with
the static pressure orifices in the same location as .for the
standard pitot tube. Total and static pressure heads were read
from a vertical manometer when the tunnel speed had been
stabilized at the correct value. The correction is computed as
(g-qi) and plotted vs qi in fig. 20.
below.
RUN 43
The data is tabulated
7/1/49
heads - inches H2O
PlTOT TUBE RIG
STt~~ARD PITOT TUBE
E.A.S.
M.P.H.
20
30
40
50
60
70
80
90
100
h
.16
.46
.82
1.28
1.81
2.48
3.21
4.09
4.98
p
q
h.
1
Pi
qi
(q-qi)
.00
.00
.00
.00
.00
.00
.00
.02
.02
.16
.46
.49
.02
.83
1.28
1.81
2.48
3.21
4.08
.04
.05
.06
.09
.12
.16
.47
.79
1.23
1.76
2.39
'3.09
4. g;
.19
-.01
.03
.05
.05
.09
.12
.15
.16
.82
1.28
1.81
2.48
3.21
4.07
4.96
_ 19 -
3.92
4.80
FtG.20
Z
J
o- ~
Z
C1.
~~
In
ct. ~
Q
lO z
-Uj
..J 0
~
~
«F
rn
U rn
'V
UJ. r
UJ
I
0
2
41-=
F~
I
etl
f- ~
gz
-
to
j
Q.a.
E)
0
~
\0
l:J'3.J.. '9/\/\
N
S3H:::>NI
CO
q
~t- t)
~
0
0
0
~
."
c.
Direction
Indicator Characteristics
The dynamic response
characteristics
tion indicator can be best described
1. Damping
of the wind direc-
by the parameters
ratio, ,
2. Undamped natural frequency, nn
Considering
the indicator
degree of freedom the equation
to be a body with a single
of motion
for the condition
of a
uniform air stream is,
Ie
where
e
+
cB
0
(2)
the angular deflection
1S
and all derivatives
ref. 4,
=
+ ke
from the equilibrium
are taken with respect
the expressions
for'
to time.
position,
As shown in
and nn are,
, = 2JkI
c
and
1
=
nn
(4)
21TJ~
the coefficient
c is calculated
for each component
from the
expression
(5)
and the coefficient
k
=
dT
de
=
k
1S
given by
qS dCt..1
da
(6)
where T is torque in ft. lb. and I is the distance
the centerline
of the indicator
measured
along
from the axis a-a to the quarter
chord line of the surface considered.
- 20 -
The moment of inertia I is
based on a density
of .310 lb/cu-inch
giving some allowance
for all components,
for the soldered
connections.
sions are taken from fig. 73 and the computed
coefficients
are listed in the following
thus
The dimen-
values of all
table for an air speed
of 100 M.P.H.
I
1
2.208 x 10 -8
2
2.360
X
10-8
3
.566
X
10-8
8
1.595 x 10-
4
I
c
k
=
=
=
,=
1.084
X
10-6
4.045
X
10-3
1.156
X
10-6
-2.961
X
10-3
10-3
ft lb ra d -1
cycles per second
forces on components
effects between
in this simple analysis.
2 and 3, as well
all components
The low damping
of the input signal near the natural
that quantitative
However
results obtained
the variation
is altered allows a qualitative
intensity
X
0.15
as mutual interference
accurate.
1.448
1
2.604 x 10 -6 ft 1b see ra d-
The aerodynamic
amplification
k
c
2
6.729 x 10-8 slug ft.
nn .= 20
neglected
1C4
,a,
Component
has been
ratio causes
frequency
from the indicator
so
are not
of output as the configuration
estimate
of turbulence.
-- 21 -
of the change in
D.
Pressure Recovery
H. Peters,
efficiency
ref. 1, gives an analysis of diffuser
in which it is shown that the commonly
used expression
(7)
1S
in error when the velocity distribution
are not uniform.
The correct equation
P2
T/TOT
where subscripts
-
a~ the inlet and exit
is shown to be,
Pi
(8)
1 and 2 refer respectively
to the diffuser
inlet
and outlet, and
The constants X and Yare
the deviation
equation
necessarily
qreater than 1 and reflect
from a uniform distribution.
By applying
to the diffuser - tailpipe combination
efficiency
of energy conversion
=
P4
-
this
to find the
between station 2 and 4 we have
P2
From an inspection
(9)
of fig. 1 it appears
that the constant
X will not differ greatly from 1.00 due to the nearly uniform
velocity
distribution.
introduced
by taking y=
investigation
Since (A2)2
A4
is approximately
1.00 will not be ~eat.
the assumption
that X = Y.=
- 22 -
~ , the error
For a preliminary
1.00 is considered
justifiable
in V1ew of the simplification
lengthy computation.
modified
efficiency
Pressure Recovery
To avoid misinterpretation
we define this
as,
= __
P-:=4=-_-_P_2_ ::: P4-
- ~
A2
q2 ~-(A4)
/
of an otherwise
P2
(10)
~
J
I~
)
-
-
V
STA.I
STA.2
To calculate
q2 and q4.
5TA.3
pressure
recovery
Since we have assumed X=
it is necessary
1.00, Bernoulli's
to find
equation
gives,
(11)
from continuity
(12)
from which we find
q2 =
P1
-
P2
(13)
~(A2)j
A1
But,
A2 =
177 sq in
A1 = 1910 sq 1n
A4 = 391 sq in
- 23 -
(14)
so that,
Pressure
Recovery
=
1.249 (P4 -P2 )
(Pi -P2)
- 24 -
(15)
Pressure
At Station
Pressure
25
29
38
39.
41
Recovery
-%
84.2
46.5
48.0
51. 9
54.5
55.0
54.8
52.2
55.0
56.8
57.6
48.4
52.1
53.1
54.8
4
6
7
8
9
10
11
12
14
15
16
17
18
19
20
Run
Data
4
Run
At Station
Recovery
Run
21
22
24
26
27
28
30
31
32
33
35
36
~7
38
39
40
3
Pressure
Recovery
37;4
39.4
40.8
43.1
43.3
- 25 -
- %
Pressure
Recovery
56.4
56.6
45.7
45.6
44.9
44.2
45.1
44.4
43.1
42.0
44.0
45.5
46.2
47.7
.48.5
49 ..
3
- %
E. Velocity
00
~
\0
M
~
~
~
~
Surveys
C'1
~
r-; ~ ~
~
M
0
C'J
0
N
C'tMooO~-.::r~
l.f')~~~MC'10
o
o
o
~
~
~
C'J
,....j
00
00
C'J
00
~
~
M
~
~
M
M
l.f')
M
M
M
M
r-r,
~
N
00
,....j
l.f') 0
C'J
0
C'1
N
C'J
\0
0
00
00
\0
,
~
~
N
0
0
~
I
M
C'1
~
0
l.f')
l.f')
M
~
\0
-.::r ~
,....j~
~
I
00
trJ
~
0
~
~
0
~
\0
~
l.f') N
l.f') \0
\0
~
~
~
l.f')
00
~
l.f')
00
00
~
0
0
M
~
N
l'-
M
0
00
0
0
V")
~
,....j
l.f') 0
J
C'1
000
\0 l.f') l.f') l.f') M
C'1C'1,....jM~O
,....j,....j,....jM
\0
0
~
C'1
l.f') ~
0
~
00
00
~
~
,....j~
~
~
00
~
0
,....j~
~
~
0
,....j~
~
~
0
\0
~
00
N
l.f') \0
,....j~
~
,....j,....j,....j~
0\
00
0
~
\0
~
00
,....jr-l
0
,....j~
\0 ~
00
~
~.
0
~
~.~
~
~
and Data
\0
~
~
00
l.f') 00
00
N
0 ~
~
M
~
C'1
~
C'J
<:0
C'1
00
0'1
~I
ro
00
~
~
l.f')
0
\..'") M
~
~
0\
0
l.f')
\0
~
0\
00
0
~
~
0
l.f')
0
l.f')
0
00
r-l
\0
~
l.f')
N
I
~
00
,....jM
I
M
I
r-l
O~O'\~Ol'-~oo
~
Moo~\OO
N ,....j,....j~
Lf') ~
M
C'l
C'J
~
0
l.f')
l.f')
~
0
0\
00
\0
0
l.f')
00
~
0
~
~
~
00
~
~
0
M
0
0
If')
~
0
M
0
M
C'J
~
0
l.f') \0
MOl'~
~
0
N
~
l.f') 0
~
~
\0 \0
~
,....j~
~MooO\oorol.f')OO\N
MMr-l~l.f')~~M~M
~
~
\0
~
l.f')
~
0
0
l.f')
0
M
0
0
~
l.f')
0\
C'1
C'1
0
~
o
l.f')
00
~
00
~
~
C'1
~
-26-
C"\I
O
~
0\
0\
0\
0
~
0\
0
~
M
Ln
~
0\
0
eo
\0
0
Ln
~
C"\I
~
~
~
0
C"\I
ro
~
t-
0
~
~
~
~
C"\I
C"\I
~
~
1.1")
~
~
0
~
0
~
~
~
0
~
~
1.1")
~
tM
0
t-
~
C"\I
0
~
~
\0
0
~
CO C"\I
0
1.1")
CO
0
t-
1.1")
C"\I
0\
0\
0
0
CO
0
0
~
0\
1.1")
t-
0
0
0\
~
~
0
t.r.l
0\
eo
~
t-
tt.r.l
~
0\
C"\I
M
CO
~
t.r.l
~
~,
CO ~
1.1")
0
~
0
0
~
0
0
~
~
0\
C"\I
I
I
M
0
....-j<
(Vi
0
M
0\
t-
CO
C"\I
0
M
0
I
~
C"J
~
0
M
0
~
0\
CO
\0
M
M
C"J
....-l
M
C'J
C"~
~
C"\I
CO
0
~
CO ~
0\
0\
eo
t.r.l
0\
0\
\0
~
0\
C'1
~
0\
CO
.'
~
Ln
0\
0
.-I
~
C"l
0
eo
~
U")
O\o\COMt-O
0
0\
0
.0
0\
0
~
t.r.l
~
CO
CO
0\
C"\I
eo
0
U")
0
o
0\
0
eo
t.r.l
teo
0\
~
1.1")
~
0
~
M
..
~
~
~
~t-
M
0
~
\0
0
~
0
Ln
~
0\
~
t-
LnoLnOtnOtr.>
tr.>
M
1.I")tr.>OC'lMLn
C"\ICClMM~O
OO~~~~
0
0\
~
~
o
tnLn~O\~..q.OOC'JO\OJ\OOttt0\
0
0"\
t....-l~~C"\I~~~~OOO\O\roroMMO
t-M\OCO\OO1.l")\O~COt-~~tt0
1.1")
tCO
~
~~~~~~~~O~O\O\COCOM~O
~
CO
M
~
CO 0
0\
0
eo
M
eo
OJ
~
1.1")
~
~
0\
C"\I: \.f)
M
t.r.l
0
~
t-CO~O\~M~"'~~O\~O\~U")~t-~
ooeoO\O\O....-l~~~OOO\O\OOMt-O
("f)
~
t.r.l
CO
0\
~
0\
~
C"\I
M
o
~
\0
0
o ..q'
o ~
0\
o
~
0
~
o
o
o
M
If)
roM
~
M
~
~
ro
lO
M
M
~
~
M
ro
\00
~
~
~
0\
M
«:f'oo:::T'
C'1
I
C'J
I
C'1
C'J
0
ro
0
M
0
M
~
0
C'J
\.(')
ro C'1
ro '" lO
~
\0
C'J
"'.
.....
\0
I
M
C'1
~
0
M
~
0 M
~
ro
~
~
~
M
0
~
C'1 C'1 C'1 <:'1 C'1 M
~
~o
0
~
~
~
~
~
0
~
0
.-I
0\
~
'"
If)
ro
I
If)
~
00
0
~
~
~
~
M
M
M
C'1
0
M
~
~
0
~
.-I
M
M
lO
0
~
~
M
'"
M
M
ro
'"
ro
0
ro ~
000
0
0
0
M
\.(')
0
.-I
~
~
ro
~
~
~
M
0
.-I
M
M
0\
M
~
~
M
ro
0
MM
M
.-I
~
\.Olf)
0
M
ro
~
0
0
0
ro
C'1
0
~
\0
'"
~«:f'~lOM~MlOO\
0\
~
lO
ro ~
.-I C'1 .-I M
~
If)
0
M
0\
C'1 0
\00\
Mro~.-I\OO\
lO \0 ro
If)
ro
V
.-I .-I
C'1
~
~
C'1
C'1
~
C'1 ro
lO-O
If)
.-I
.-I
'"
M
M
0
lO \.(')
~
lO
M
lO .-I ~
OO\"'COrolOOO
0
0
~
0
lO
0
.-I
0
0
ro
C'1
0
.-I
~
~
0
ro
co
ro
.-I
M
C'1
M
ro ~
ro .-I C'J
CO CO \.(')
M
0\
ro
00- 0\
ro M
~
'"
~
~
'"
M
0
~
0
0
ro
0
'"
0
0\
If)
.-I
0
.-I
\0
0'\
~
~
C'1
«:f'M
.-100
~
0\
M
0
M
ro
M
M
~
.-I
.-I
~
~Olf)Olf)OlO
0\
<:'1 .-I
\0
0
0
M
ro
ro
lO
C'1
~
If)
~
\0
~
M
~
~
0
~
0
'" ~
«:f'M
00 '"
'"
M
~
0
.-I
~
~
0
C'1 ~
ro ~
C'1 M
'"
~
ro
M
~
C'1
\0
If)
.-I
-a)
MOOM\OroM
MM~OOO"l::f<
MMM~'MM
-.:T
~
ro
~
C'1
CO '"
~
0\
0\
0'\
0
~
CO
oo:::T'MMMMC'1
o
o
~
lO
ro
-28-
E-t
<:
<
Q
~
H
S
CO
H
f:S
I-f
en
Q
0"
c::"
I:IJ
0
...q<
Z
~
~
0\
M
~
~
eo
Z
('I")
~
0
0\
....-4
....-4
0
....-4
~
0
eo
0
,......j
0\
0\
~
...-i
...q<
If)
0
,......j
eo
....-4
....-4
('I")
eo
0\
0
CO
\0
\0
~
C'l
....-4
0\
0'
Ln
0\
0'\
Ln
M
0\
CO
0\
00
0
('I")
CO
t-
-.:f'
~
~
If)
0\
t--
eo
\0
eo
I
I
I
I
I
I
...-t
t'I
0..
~
~
0
r-
0
M
0
0
0
...-t
0
M
.0
C'\I
0
C'\I
~
M
0
eo
M
....-4
M
....-4
t-
0\
t-
\0
...-t
If)
eo
CO
CO
r-i
0
M
\0
eo
eo
If)
.....
C'\I
C'1
t--
M
...-i
00
.....
0
If)
,.....;
If)
0
0
0\
..-l
0
0\
eo
....-4
If)
0\
If)
\0
M
0
0\
0\
tn
M
0\
t-
\0
.....
CO
C'\I
~;
..
...-t
....-4
C"I?
~
.-;
0
0
('I")
....-4
0
C'\I
M
\0
,......j
I:"'-
.-;
....-4
0
...-i
0\
C'\I
...-i
.-;
.-;
....-4
\0
0\
....-4
C'\I
If)
0
0
C'\I
C'\I
l./j
0
...-\
\0
C'I
If)
C'\I
.....
0
...-\
....-4
0\
CO
M
....-4
l./j
'<:f
...-i
....-4
0:::>
C'I
0
r-t
If)
M
C'\I
...-i
C'\I
M
0
O
....-4
~
....-4
C'l
M
....-4
...-i
0\
C'\I
t-
,......j
....-4
M
If)
....-4
C'\I
CO
M
0\
0'.
C'\I
CO
t--
..-l
~
....-4
M
..-l
r-t
....-4
If)
..-l
0
C\
('I")
~
."dl
....-4
\0
g;
...-i . ...-i
...-t
r-i
....-4
CC
0
C
....-4
....-4
....-4
..-l
0\
C'J
\0
..-l
M
M
If)
\0
.0
M
C"'?
If)
C'\I
M
\0
0\
\0
C'\I
....-4
('I")
C'\I
0
('I")
;Z;
\0
C"I?
...-t
r-t
0\
0\
C'f')
0
....-4
0
M
0
0
C'.-,
0
0
CO
0
~
C'\I
CI:l
....-4
....-4
If)
C"I?
\0
0\
\0
....-4
..-l
.-;
0\
....-4
~
...-t
\0
....-4
If)
C"I?
..-l
0\
..-l
0
eo
0\
P"'"'\
C'I
M
M
..-l
M
C'1
..-l
...-t
0\
C'l
0
0
t-C'\I
C'I
t0
..-l
\0
0\
If)
CO
0
..... .....
t-
I
~
....-4
C'\I
..-l
....-4
C"I?
0,
0\
....-4
0
0
'<:f'
....-4
..-l
0\
If)
...-i
0\
«:ft
C'\I
If)
....-4
If)
0\
C".J
CO
C'\I
C'\I
0::>
0
0
...-t
CO
If)
r-t
O
0
(;".1
M
If)
\0
0\
L'')
0\
....-4
0
t-
..-l
C"'l
0\
0
.....
t0
M
...-i
....-4
0
If)
....-4
....-4
..-l
t-M
....-4
Ln
C'l
....-4
..-l
C'\I
t-
..-l
..-l
...-i
~
0
..-l
C'\I
CO
....-4
....-4
..-l
t-
P.
.-l
C"I?
If)
C'\I
..-l
\0
,......j
-.:i'
O
If)
0
0
0
~
eo
r-t
tr)
t-
M
If)
t-
M
0
eo
t-
If)
C'\I
M
..-l
\0
If)
0
0\
~
t-t--
...-i
N
('I")
If)
Z
C"I?
CO
"<::fl
~
0
Itr'
t'I
'-....
A
tr'
C'')
..-l
-.::f'
~<..;;;
~rr'
::J
H
W
a::
C"I?
C"I?
~
CC;
Z
I
H
en
~
H
Q
~
~
E-t
H
~
~
0'
-29-
40"
320"
60"
300"
70°
290'
80"
280'
100"
260"
210"
1500
320"
40')
100'
40"
320"
50')
310"
60"
300"
70"
290"
80"
280"
90'0
2 70"
160"
200"
170"
190"
180<>
1800
II-T:::r:.~~L~1=~::::
I ~+-+--~+- \~::.t:.::t.'\:M--l'''C1=]~--+-1--=t=~H\=r:...\-I:7'\A-;:<::'PIn:~
100"
260') --
150"
210"
1900
1700
200"
1600
2100
1500
110'
2300
130')
2200
140"
40"
320"
50')
310"
60"
300~
70°
290'
80.0
280)
150"
210"
160"
200"
170"
190"
1800
1800
1700
1900
1600
2000
210"
1500
100'
110'
250'"
130"
2300
140"
2200
40°
320"
50"
310°
60°
300~
70°
2900
80'"
280')
loon
260"
190"
1700
2000
1600
210°
1500
250"
110')
230"
130"
2200
140"
40"
320"
50')
310"
60"
300"
80"
280'
100"
260"
30"
330"
20"
340"
o
350"
10"
340"
20"
330"
30"
40"
320"
50')
310"
60"
300~
70"
290'0
80')
280')
0
100
260"
30"
330"
20"
340"
150"
210"
40°
320"
50"
310"
60°
300"
2900
0
80
280'
30"
3300
20"
3400
10"
350')
3300
30"
210"
150"
40"
320"
50')
310"
60"
300"
70°
290"
80°
280'
100"
260"
350"
10"
40"
320"
70"
290'
80"
280')
100"
260"
30"
3300
20"
340"
10"
3500
330"
30"
210"
150"
40"
320"
50')
310"
60"
300"
70"
290'
~
210"
1500
2200
140"
2300
130')
240"
120"
250-'
110')
280"
80'-'
2000
1600
80°
280')
1900
1700
270"
90'
180<>
ISO')
90"
270'
170"
190"
260'
100"
160"
200"
100"
260"
150"
210')
40"
320"
50')
280':'
310"
80"
2BO"
270"
90')
220"
140"
2300
130"
240"
120"
110"
-', 250"
BO"
90"
270~
100"
260"
210"
150"
320"
40"
350"
10"
40"
320"
310"
50"
o
50"
310'°
60"
300~
100"
260"
110"
2500
1200
240"
210"
150"
40"
3200
50"
310"
60"
300~
BO"
280')
100"
260')
40°
3200
50"
310°
60°
300"
700
290'
80'"
280.0
-
170"
190"
1800
1800
1900
1700
210"
150"
40"
3200
50')
310"
60"
300"
70°
290"
80"
280'
100"
260"
110"
2500
340"
210"
150"
40"
320"
50"
310"
60"
300"
80"
280')
350"
10"
2100
1500
50°
240"
120"
230"
130')
2200
140"
40"
320"
50"
310"
60"
300"
70"
290'
80"
280'
100"
260" -
1900
1700
200"
1600
210"
150"
2200
140"
400
320"
50"
3100
60°
300"
70°
290'
80°
280')
100°
260"
110"
2500
90''>
40°
320'~
60°
300~
70°
290"
80"
280"
210°
1500
40"
320"
50"
310"
60"
300"
80"
280'
350"
10"
320"
40°
50"
- 280"
80°
250"
110')
230"
130"
2200
140<'
50"
310"
60"
300"
70°
290'
80°
280'
o
350"
10"
210"
150"
40"
320"
50')
310"
60"
300"
70"
290"
I~
2000
1600
210"
1500
0
400
320
0
50"
3\0
600
300~
70°
290'
80~
2600
0
100
260')
-
30"
330"
20"
3400
10°
350')
350"
100
210"
150"
50"
3100
600
300"
70°
2900
80°
280')
2100
1!S00
100"
40"
320"
50<)
310"
60"
300"
80"
280"
100"
260"
110"
2500
2000
1600
210"
1500
40"
320"
50"
310"
60"
300"
70°
290"
3500
10"
200"
1600
210"
1500
320'"
40"
40°
3200
50"
310"
60"
300"
70°
2900
80"
280')
I
210°
1500
320'"
40°
40"
320"
50"
310"
60"
300"
70°
290"
80"
280"
2000
1600
2100
1500
250"
110"
2200
140"
40"
320"
50')
310°
60°
300"
80°
280'
o
3500
10"
,/
2500
110"
2300
1300
2200
140"
F.
Index of Runs
Run No.
Da ta Qbt lained
Configuration
1
T
Pressures
2
T
Pressures
3
T
Pressures
4
TD
Pressures
5
TD
Wind
6
TDNi
Directions
Pressures
7
TDNiFi
(3=90°
Pressures
8
TDNi Fi
/3=80°
Pressures
9
TDNiFi
(3=70°
Pressures
10
TDNi Fi (3~60°
Pressures
11
TDNi Fi
,6=50°
Pressures
12
13
TDNi Fi
/3=40°
Pressures
14
TDN1Fs
/3=50°
Pressures
15
1DN1Fs /3=40°
Pressures
16
1DN, Fs
/3=34°
Pressures
17
18
19
TDN1 F5
Pressures
TDNiF5
TDN1F5
(3=80°
/3=70°
(3=60°
20
IDNi F5
(3=50°
Pressures
21
TDNiF5
/3=40°
Pressures
22
TDN1 F5
(3=36°
Pressures
23
TDNiS,&TDNiF5S,
TDNi
Wind Directions
& TDN1 F 1 Allj3
1
Pressures
Pressures
TDN1 F5AII
f3
Wind Directions
,8=50°
2°4
TDN1N2H
Pressures
25
TDN1N2H
Pressures
26
TDNiN2F2
(3=90°
Pressures
27
TDN1N2F2
(3=70°
Pressures
28
TDNiN2F2
(3=70°
Pressures
29
TDNiN2F2
f3=70°
Pressures
30
TDNiN2F2
/3=60°
Pressures
- 30
-
31
TDN1N2F2,6=50°
Pressures
32
TDN1N2F2,6=40°
Pressures
33
TDN1N2F2 ,6=30°
Pressures
34
TDN1N2F2All
Wind Directions
35
TDN1N2F4 /3=70°
Pressures
36
TDN1N2F4 /3=60°
Pressures
37
TDN1N2F4 /3=50°
Pressures
38
TDN1N2F4 ,6=40 °
Pressures
39
TDN1N2 F4 ,6=40 °
Pressures
40
TDN1N2 F4 {3=200
Pressures
'41
TDN1N2F4 /3=20°
Pressures
42
IDN1N2F4
43
Pitot
',6;tDN1N2H
All
/3
tube calibrations,
- 31 -
Wind Directions
Student
Tunnel
G.
gu res
C ~8
n, sad
Forw
i.ure
D__0
tor a ~_
acelle
,;;»
N1 installe
6
iff ser and t il 1pe - 32 -
1n diffu er.
Fig re 57
10 er ana ex
ns.o
section
'i ure 58
Settling
c .am
r,
ell mouth
se tion.
- 33 -
nd cons ant area
Fi ure 59
Tacelle
1
and
indmill F1
Ei,ure 60
Blade angle c ec' Jlg and
- 34 -
indmill F1
Figure 61
Nace Ie N2 and windmill
4
'igure 62
'acelle Ni,
windmill Fi diffuser an
- 35 -
tufts
.. re 6
Rake ads
pport
Fi ure 64
ind direction
indicator
- 36 -
and support
Figure 65
Blade twister and blade from windmill F4
- 37 -
FIGURE-66.
Amerrcan H S Fan No.245
cap.
11.000
C FM
SIX
No.
screens4 foot spacing
16 mesh x .009 WI re
/l[Z6'.19 ..
I blower
outlet
to man""e!er
i'o
tube No. 32
reet.J
rnano:netler
tube
No.33
diffuser
section
direction IndIcator
tube
support
.J
r~qu,!ar octagon sect Ion Lsettlrnq chamber
-14-0
across f1ats-lnSldel
observohon
port
I
\
\.
II-+----.-t---
13i"
5- ~,,_/../_/_/_/_/_/_/_/_/_/_/..,..._/'_/_'-'-/~L~.'-
4:~6" _/ __/ _/ ---+0---
---+---/_/2:~~"
~
-'-' '_/
23'-10-1"
'%
INCH:
1 FOOT
/
./
GENERAL
ARRANGEMENT
DIFFUSE R RESEARCH
PROJECT
FIGURE-67
diffuser
section
i
-4----8
t-
--+--~
,
/
,
1
-1C'
t\l") l
~
-----#-~
-
,
-+
._+-
,
_
.. E--__
I-
-E
a-L~
---+--~
-f- -------=-~--+--_+__f_-------~
oJ
~--------18.J.
.~--
'bros. wind
tunnel
\
----4-
,- ------1--
----et-
\,
I
----)
-
---~~~
.z/~ ............
---...1
QUARTER
dwg. No. D-3 -10-6
,
\
....----2 S
wright
--~
\
\~--+-----
.r e f:
r-,7~
I
I
~
,
I
hub
\
I
(
~
\
\
=-_.~_t?
for wlnd-mdJ
I
I
I
tori cone
\!
I
\
removable
nacelle - N 1
forward
SIZE
FORWARD NACELLE
DIFFUSER
RESEARCH
PROJECT
FIGURE-58
/
~C1ft
noceJJe-N2.
sup port for wind-mill
-tall
pIpe
--7\
,
,
\
--
,
'
,
,I
/
I
\
I
\
I
I
\I
\
\
hub
\
I
/
""
,,I
I
I
\
\
--<t-
I
\
/
r
;Jo
0
\
\
I
,
I
)
'"
~
---~~
QUARTER
ref:
wnght bros. wind
dwg. No. D-3-20-"7
tunne I
S.'Z E
---.....---
AFT NACELLE
DIFFUSER
RESEARCH
PROJECT
w
a:
:J
l)
u..
(J)
o.
to t'-
<f)
4" ~
g
(\J
....
.... ~....
~
W <0
CD
<{ ~
ft)
(\)
~
0
u
C
OJ
C
.....-
-0
-0
0
E
a
~
L-
~
L-
"-~ ~
OJ '-- 0
-0
-0
W
, fry''}
r<!
0
0
,
~I~
U
CD
0
0
I
(1)
• iI
-to-
~L:~
.z
.
• • t:::r' CI'"
..c.
-00
L-
oeD
~I
CN
-00
-+-
:::leX)
C
C
OJ
o
u....
u-
::)
~'(f)
~a::
-w
I
~a:
-w
.-J(j')
...JW
«
(Der
.-JU
<{I
o
wn.
(J)cC
o
-,
liJ
b
F IG URE-.6~t70
-
W
N
(j)
W
u
-
5
.....
.
U
I
z-
I
c..o
~
~~-O
'ia:>0ld 9-IO:>ld'<~\
o
~
""'
-E.-':
r
'Ico
.............
..........
w
N
U1
:::>
..J
..J
l.l..
Cl..
a:
o
J
W
U
t-
FIGU RE- 7"1
CD
:)I
Il)
..Jer
--.1«
_W
.w
2U1
0((
Z
w
~a:
U1
:::>
La.....
l.L
o
1
.
-00
eN
-I
•
~rt)
&flCD
o
\-
...00
.....
2
...c. •
~
• • c:r" C'"
~~-o
4-L:
(l)
..0
:J
-+-
:J
-+-
o
...0
V')
(l)
.L)
::J
-+-
-0
0
~
..r.
0
-+-
0
-+-
~ -8' ---E
~
cJ)
W
N
(/)
o
a:
,
.-I
Z
W
o
FIGURE -72
~
U
W
J
o
a:
0...
I
U
WeJ:
~«
«~
a:w
a:
a:
(/)
w
::>
l.J..
u..
o
FIGURE-73
--
-_---t--- ................
.......-- --+---
+~i
.oos thickness
VIEW
FULL
A-A
SIZE
WIND DIREC'TION INDICA TO R
DI FFUSER RESEARCH
PROJECT
t::r"
C
U
a..
o
\Il
I
l/)
o
~
..c.
u
N
W
(J)
.F IGURE-74
~
J
&3
o
a:
CL
wI
cnU
:::>0:
w
J-<{
Ow
J-(f)
0:
J-a:
CL
(J)
w
:::>
u..
u....
o