SUPERVENTILATED FLOW PAST DELTA WINGS by Taras …
Transcript of SUPERVENTILATED FLOW PAST DELTA WINGS by Taras …
Department of the Navy Naval Ordnance Tes t Station Contract N123(60530)34767A
SUPERVENTILATED FLOW PAST DELTA WINGS
by
T a r a s Kiceniuk
Reproduction in whole o r in p a r t i s permi t ted for any purpose of the United States Government
Hydrodynamics Laboratory Karman Laboratory of Fluid Mechanics and J e t Propulsion
California Institute of Technology Pasadena, California
Report No. E-101. 5 July 1964
NOMENCLATURE
Symbols
z max
Definition Units
Length of model F t
Plan form a rea of model Ft2 in ~ t 2
Longitudinaldistance along F t model, upstream from trailing edge
Transverse distance on F t model, from model centerline
Half-breadth of model a t F t 1 ongitudinal position x
Submergence measured F t vertically from undisturbed f ree surface
Submergence measured to model centroid
Longitudinal distance to center of pressure
Half-breadth of model a t widest point
Model total apex angle
Model angle of attack
F ree stream static pressure
Cavity pressure
Free stream static head measured relative to atmosphere
Cavity pressure measured relative to atmosphere
Mass density of water
Degrees
Degrees
L ~ S / ~ t '
~ b s / ~ t '
F t of water
F t of water
Symbols
- Lift C~ - 2 t p V o A
- Drag C~ - 2 t~ V o A
- - Moment ( t . e . ) C~ z p I V:AL
NOMENCLATURE (continued)
Definition Units
Acceleration of gravity F t / ~ e c 2
F r e e s t r e a m velocity F t / ~ e c
Cavitation o r ventilation Dimensionless number
Lift coefficient Dimensionless
Drag coefficient Dimensionless
Moment coefficient about Dimensionless trail ing edge
Introduction
Although delta wings have been known for some t ime in aeronautics (1)(2);k
their introduction into a hydrodynamic context has been quite recent. As
in the flow of a i r , the delta wing provides a simple but useful configuration
for investigating three-dimensional problems in cavity flows. At the s t a r t
of the present work (1 960), only one theoretical study on this subject was
known(3). No information on flow pat terns , fo rce charac ter i s t ics o r
other propert ies were available for these shapes. It was accordingly
decided to embark on an experimental program with the a i m of providing the
basic charac ter i s t ics of the cavitating flow past delta wings, to observe and
outline any interesting features of these flows and, finally, to provide a
physical basis for any mathematical analysis of the flow that might be
undertaken. .Measurements of lift, drag and pitching moment and p res su re
distributions were made on a family of simple flat plate delta shapes of
varying apex angle; severa l configurations outside this family were a l so
tested. These included a diamond plan fo rm, r e v e r s e delta, and a delta
with a 90 degree bottom. All were without camber and were tested with
no yaw angle.
After completion of this work, the exhaustive t rea tment of Reichardt
and ~ a t t l e r ' ~ ) appeared which a lso deals with cavitating delta wings. It i s
believed, however, that the cur rent r epor t and that of Reichardt a r e
sufficiently different in scope and method to justify the presentation of the
present resul ts . =
Instruments and Apparatus
The experimental program described in this repor t was performed in
the Free-Surface Water Tunnel a t the California Institute of Technology. (5)
The working section of this tunnel, shown diagrammatically in F igure 1 ,
permi ts force and p r e s s u r e measurements to be made a t velocities up to
27 fps in the presence of a f r e e water surface. The boundary layer sk immer
located a t the upstream end of the working section, removes the slow-moving
top layer of water in the nozzle. The resulting velocity profile, labeled B
in Figure 1, shows no perceptible variation except a t the lower boundary
and s ide walls of the 20 inch by 20 inch working section.
.*, ' F
Numbers in parenthesis re fer to re ferences a t end of text.
The fo rces and moment acting on the model were measured on the
three-component force balance. This balance, descr ibed in grea ter detail
in Reference 6, i s an electrical-mechanical type to which the model i s
attached by means of a shielded adjustable s t ru t assembly. Simultaneous
readings of forward lift, r e a r l if t , and drag a r e combined by a simple
computational procedure into lift, d rag , and pitching moment coefficients,
the la t ter being taken about the trail ing edge of the model. The elevating
mechanism to which this balance i s attached can be r a i sed o r lowered s o
that the model can be completely submerged, o r i t can be r a i sed through
the f r e e surface and out of the moving water .
The support-s t rut assembly consis ts of an inner support s t ru t and an
outer s t reamlined fairing. Within the fairing a r e contained the tubes which
supply a i r to the cavity and which permi t the cavity p r e s s u r e to be
measured. No mechanical sea l i s provided between the cavity bubble and
the outside a i r above the water surface, a condition which might have
resul ted in the maximum cavity p res su re being l imited by the leakage of
a i r f r o m within the cavity. To obviate this possibility, prel iminary
experiments were made using a water sea l , in which a generous external
supply of water a t low p ressu re was permitted to fill the space between
the fairing and the support s t rut . Although this sea l worked very well, i t
was found to be unnecessary since the r a t e of a i r l o s s past the s t ru t was - found to be small compared to the a i r entrainment a t the downstream end
of the cavity.
The angle of attack of the model could be changed during the course of
the run by means of the adjusting knob located a t the upper end of the s t ru t ,
but was l imited to an excursion of plus o r minus 10 degrees f r o m the p re - se t
midpoint. The dial indicator, reading in thousandths of a n inch, had
previously been calibrated to indicate angle of attack. This dial indicator
can be seen in F igure 2 which shows a s lender delta wing and the lower end
of the force balance during a typical run at moderate submergence. The
lower end of the support s t ru t fairing and the point of attachment a t the
model were not fa i red since they did not contact moving water inside the
a i r supported cavity.
The supplementary a i r supply probe can be seen in these same photographs,
extending upstream f r o m i t s surface-piercing support s t ru t to i t s terminating
point within the cavity. This supplementary supply was used only to check
whether o r not the upper value of cavity p res su re was being limited by the
a i r supply ra te obtainable through the piping within the s t ru t fairing. When
it was conclusively proved that this was not the case , the auxiliary a i r supply
was removed and the a i r supply l ines within the model support s t rut fairing
were used alone.
The p res su re in the cavity was measured on a water manometer since
the range of cavity p res su res usually encountered in a i r supported cavities
of this type a r e within a foot o r two of atmospheric p res su re for velocities
l e s s than 25 fps. In order to make su re that the p res su re tap line was f ree
of water a t all t imes , some a i r was continually bled into the cavity space
through the p res su re measuring tap. The p res su re drop associated with
this small a i r flow ra te through the p res su re line causes the cavity p res su re
manometer to read about . 01 feet too high. This e r r o r was easi ly compen-
sated for by adjusting the scale a t the s t a r t of the run by f i r s t raising the
model out of the water and using the ambient a i r p res su re a s the known zero
p res su re reference.
Description of Models --- The design requirement of each model was that i t be hydrodynamically -
equivalent to a flat plate, wetted only on the p res su re side and not contacting
the cavity wall on i t s upper surface. This goal could be met easi ly and
inexpensively by machining the model f r o m flat aluminum stock beveled
45 degrees along the sides of the hydrofoil. If handled carefully the edges
remain reasonably sharp, and prel iminary tes t s proved that the model did
not deform under the influence of the hydrodynamic loads. The plates were
made in thicknesses that var ied f rom 1 / 8 inch to 1 / 4 inch, depending upon
the strength and deflection requirements imposed by the part icular geometry
and point of support. The models tested, a s well a s cr i t ical dimensions, a r e
shown in Figure 3. Provisions for making p res su re distribution surveys of
the 15 degree model consisted of 1/16 inch diameter piezometer openings
distributed a t key points along the bottom of the model, communicating to
flexible polyethylene 'tubing by means of the short lengths of bent b r a s s
tubing which can be seen in Figure 3(a). The piezometer openings were made
la rge in the interest of increasing the response t ime of the p res su re measuring
system. Smaller openings would have been used had the p res su re not been
measured on the la rge-bore mult i -manometers . The same p res su re tap
openings were a l so used for flow visualization purposes by connecting the
thin polyethylene tubing to a low p ressu re source of water which had
previously been colored by dissolving potassium permanganate in it.
Figure 3 (a) a lso shows the b r a s s tubing f r o m which a jet of colored water
was discharged into the bubble wall near the tip of the model. Cotton tufts a l so
were employed for flow visualization. The model was prepared by drilling
small holes in the bottom, then counter-boring these holes with a l a rge r
diameter dr i l l f rom the upper or "suction" side. In this way a short length
of string knotted a t the end, could be inser ted through the model and held in
place by a plug of wax p res sed against the knot a t the base of the l a rge r
diameter hole in the upper surface. This method was deemed preferable to
gluing the tufts to the model surface itself since the thread protrudes normal
to the model surface and i s , therefore, f r e e to a s sume any flow direction
without initial bias.
At the extreme forward tip of the model close examination during
prel iminary runs showed that the water flowed around the sha rp edge of the w
model, clinging to the upper surface of the model for a distance of a s much
a s 1 /2 inch before springing f r ee at the sha rp edge! Since the models had
been made of aluminum, and since no part icular ca re had been taken to
achieve an extremely sha rp edge on the model, i t was felt that fabrication
defects, combined with the low Reynolds number which existed near the
extreme tip, produced this disparity f r o m the ideal flow. It was believed
that this wetting of the model tip might a l so account for the clinging of the
septum which could be observed along the ent i re upper surface of the model
a t angles of attack l e s s than 21 degrees for a l l of the deeply submerged
models. Similar to the "teapot" effect, the septum would cling o r spring
f r e e under the influence of small flow fluctuations near this minimum angle.
The wetting of the upper surface of the model by the septum did not
noticeably affect the forces acting on the model because the septum was
extremely thin, but i t had been hoped to achieve the lowest possible angles
of attack and sti l l maintain the bubble geometry a s presented in Figure 4.
F o r this reason a model was made f r o m stainless s teel with edges accurately
lapped to a razor edge to the extreme tip, a refinement which could not be
achieved on the more fragile aluminum models; Both for convenience of
fabrication and for the purpose of exploring another model geometry, the
upper surface of this model (Model 3A, Figure 3 ) was made V-shaped with
a dihedral angle of 90 degrees.
These attempts resul ted in no noticeable reduction in minimum angles
of attack and the upper surface of the t ip of this model was wetted to the
same extent a s were those made f r o m aluminum. The water viscosity and
surface tension appear to control the flow in this region, where the Reynolds
number (based on f r ee s t r e a m velocity and distance f r o m the tip) i s
approximately 15, 000.
Another attempt to reduce the angle of attack a t which superventilated
flow could be achieved was made by blunting the model tip. As expected,
this change in model shape had the effect of causing a local cavity to spring
f r o m the tip much the same a s i t would behind a flat disc, in the immediate
vicinity of the model tip. The effects of this smal l cavity disappeared within
an inch o r so and the remainder of the cavity bubble appeared the same a s
did those with sha rp t ips. A smal l reduction in the minimum angle of attack - was achieved before giving r i s e to the before-mentioned "teapot" effect, but
the gain was l imited to one o r two degrees and this scheme was abandoned.
A possible defect of the model support sys tem was that the deflection
of the model and support s t ru t could resu l t in a change in angle of attack of
the model. To check this possibility the models and support sys tems were
fastened to a heavy surface plate and loads were applied equal to those
measured during the run. The resulting deflections were measured and
recorded. All data presented in this repor t have been cor rec ted for model
deflection, even though that model deflection was, in nearly al l ca ses , l e s s
than 1 /10 of a degree.
Selection of Model Size - - The choice of model s ize was dictated by the following considerations:
1. The model must be smal l enough s o that tunnel blockage will not
noticeably affect the fo rces o r the shape of the cavity.
2. The model s ize must be compatible with the force range of the
available balance system. The upper l imit of these forces i s
10 pounds for the electr ical- mechanical balance, whe r e a s the
lower l imit i s se t , not by the accuracy of the balance, but by the
dimensions of the model support system. If the model i s made too
smal l , the cavity bubble produced by i t i s not l a rge enough to
envelop the existing s t ru t support fairing, with the resul t that
spray s t r ikes the model.
3 . The model must be la rge enough to show no velocity scaling
effects. If the model i s made too smal l i t can be expected that
viscosity and surface tension effects will play a l a rge role in
determining the flow configuration, whereas models which a r e too
la rge will be operating a t a sufficiently low Froude Number, for
gravity effects to be important.
The final verification of co r rec t model s ize was based on the prel iminary
experiments using delta wings with an apex angle of 15 degrees. Models 11. 2
inches long and 8.28 inches long were tested and found to exhibit identical
flow geometr ies and lift and drag coefficients. The sma l l e r one was chosen
for the reasons enumerated above. The a r e a of this model was 0. 063ft2:
The models with apex angle other than 15 degrees were made approximately
equal in plan f o r m a rea .
Observations and Flow Visualization - - Before placing the superventilated delta wing in the water tunnel i t was
not known what the resulting flow and bubble configuration would be. Resul ts
of tes t s performed in air"), and therefore corresponding to the fully-wetted
flow, indicated that a pair of vort ices would f o r m along the edges of the
delta wing and t r a i l downstream. Exactly what would become of these
vort ices a t cavitation numbers low enough to resul t in supercavitation o r
superventilation was a mat te r for conjecture. A plausible guess was that
a l a rge simple bubble such a s the one a s sumed by Cumberbatch and Wu (7 )
would envelop the upper surface of the hydrofoil and extend severa l chords
downstream. The actual bubble geometry observed was strikingly different
f rom the expected one. F i r s t , i t became apparent that moderate angles of
attack were necessary to achieve completely superventilated flow, in which
the upper surface of the model was not touched by the water. At angles of
attack grea ter than 21 degrees, with moderate quantities of a i r supplied in
the wake of the model, a c lear single bubble was obtained. The appearance
of this bubble i s best shown by the sketch in F igure 4, although cr i t ical
examination of Figure 2 and other photographs appearing in this report
reveals the same distinguishing features .
The drawing in Figure 4 i s a composite made f rom many photographs
and sketches and i s intended only to demonstrate the general features of the
flow ra ther than specific details associated with par t icular operating con-
ditions. Accurate survey techniques were not used to determine the shape
of the bubble c ross-sec t ion and so i t must be regarded a s an a r t i s t ' s sketch
only. The lines in the drawing represent the flow direction a s determined
by dye s t reak photographs and by photographs of the tufts on the model. The
angles depicted in the side view show how the flow begins to deviate f rom
being purely conical a s the base of the model i s approached.
Starting with the fully-wetted hydrofoil, and gradually increasing the
quantity of a i r which i s bled into the wake region aft of the hydrofoil, thin
frothy tip vort ices became visually apparent a s a i r was drawn into vortex - cores . Force measurements could not be accurately made with the existing
s t ru t and shield assembly for this flow condition since the model support
s t ruc ture was directly exposed to the s t r e a m of water between the vort ices .
As the a i r flow ra te was increased the p r e s s u r e within the pair of cavities
increased and the aera ted vortex co res grew l a rge r in s ize.
With fur ther reduction in the value of cavitation parameter , the vortex
co res were seen to coalesce and the wall of water between the vort ices
degenerated into a thin septum growing in thickness a s i t was convected
downstream, finally striking the fairing which surrounded the model support
s t ru t . This septum i s difficult to observe visually in the Free-Surface
Tunnel because i t must be seen through the cavity walls. Because of back-
ground tunnel turbulence, the cavity wall i s not perfectly smooth but shows,
instead, smal l waves and ripples. The resulting optical distortion makes
i t relatively difficult to see details within the cavity. To help increase the
contrast between the septum and the ambient s t r e a m and, incidently, to
obtain qualitative information about the flow in the vicinity of the cavity
walls, a jet of dye was directed against the bubble wall about 314 of an inch
behind the model t ip and about 1 18 of an inch above the point where the
bubble springs f r ee f rom the wetted lower surface. (Model 1C in Figure 3).
This dye s t reak can be seen in the sketch of Figure 4 and in the photograph
of Figure 5. After striking the cavity wall, the dye i s c a r r i e d along a
s t r e a m line over the "top" of one of the cavity lobes into the septum.
Observation of the dye s t reak in the septum i s difficult because it must be
viewed through the curved and "wrinkled" cavity wall and because the dye
has begun to diffuse.
That the septum i s not peculiar to this par t icular model o r to any
part icular model orientation can best be seen by inspection of F igures 6
and 7. Figure 6 shows the 15 degree delta wing mounted in an inverted
position within the tunnel working section, while Figure 7 presents models
of var ious apex angles within the range of 10 degrees t o 90 degrees a s well
a s V-bottom and the 15 degree diamond plan f o r m models. The appearance
of the septum at moderate angles of attack i s substantially the same for all
of these models.
The side views of the cavity, a s well a s the sketch of Figure 4 show an .- interesting feature of the flow. Tufts which were attached to the bottom of
the model and which were too long, slipped over the sha rp edge and s t reamed
along the cavity wall, thereby indicating the direction of flow in this region.
Although careful angle measurements were not ~ ~ d e r t a k e n , inspection of
the photographs shows that the flow angle i s not sensitive to the cavitation
number nor to the presence of the f r e e surface (F igure 8) . For the
15 degree apex angle delta wing, these tufts make an angle with the
undisturbed s t r e a m direction approximately a s shown in Figure 4, and a r e
independent of the model angle of attack for angles f r o m 21 degrees to
30 degrees.
When seen f r o m below, the tuft studies (F igure 9) show the relative
insensitivity of the flow direction, along the model surface, to cavitation
number and to model submergence. Since i t was not possible to obtain
high cavitation numbers a t low submergences o r low cavitation numbers a t
deep submergences, the spaces where the corresponding photographs belong
have been left blank. It should be noted that the photographs in F igure 9
were taken vertically, f r o m below the water tunnel. To eliminate the
apparent foreshortening due to the 30 degree angle of attack of the model,
the pictures were rectified during enlargement to produce a t rue view of the
plane of the model.
Effect of Model Submergence and Specification of Cavitation Number
Since the ear ly experiments of ~ e i c h a r d t ' ~ ) i t has been recognized that
the hydrodynamic forces and moments acting on a body which produces la rge
open cavities in i t s wake a r e the same whether that cavity i s filled with water
vapor o r with a i r , so long a s the cavitation parameter i s given in t e r m s of
the measured cavity pressure . The cavitation parameter i s usually expressed
in the f o r m
When the cavity p r e s s u r e and tunnel velocity a r e measured by means of
water-fi l led manometers it i s convenient to express the cavitation number a s
where h i s the reading of a water-fi l led manometer (in feet) measuring k
the cavity p res su re , and h i s the p r e s s u r e head (in feet) f a r ahead of the 0
model in the undisturbed flow. Since only the difference ho - h i s important, k both p r e s s u r e s may conveniently be measured with respect to the atmosphere,
and ho = Y i s mere ly the model submergence i n f e e t . The denominator 0
equals the reading of a water-fi l led manometer , r e f e r r e d to the moving
water surface in the working section, and connected to a tap located a t any
convenient point ups t ream of the tunnel nozzle.
F o r air-supported cavities a t velocities l e s s than 25 fps, the value of
hk i s usually within a few tenths of a foot of a tmospheric p res su re . Under
these conditions the specification of the cavitation number becomes
ambiguous since the value of ho will necessar i ly vary over a wide range
fo r relatively long models inclined to the flow. If the notion of a "local"
cavitation number
i s adopted to examine the importance of the difference in s ta t ic p res su re
at the ex t reme ends of the model, and i f Reichardt ' s well known formula
C D = C D (1 + K) for drag applies to elementary s t r ip s a t varying depths, 0
then it follows that the normal forces acting on the cavity-producing body
can be regarded a s being made up of three t e r m s
where only the l a s t t e r m depends on submergence and i s equal to C A Y;:. Do
F o r the conditions of velocity and p res su re examined during these tes t s
"local" cavitation numbers a r e actually negative a t the tip of the model. w
Using s imi lar procedures , i t can be shown that the moment coefficient
o r the center of p res su re location should a l so be cor rec ted for the effect
of difference in depth along the model, but these can be neglected since in
the worst case (in which the model tip just breaks' through the water sur face)
the e r r o r introduced by assuming that the "hydrostatic" fo rces act at the
centroid i s l e s s than the experimental sca t te r .
Lift and drag measurements taken a t an angle of attack of 21 degrees,
show no significant variation even when the model broaches the f r ee surface,
and a s much a s one quar te r a r e a of the model l i e s above the undisturbed
water surface. Under these conditions the septum no longer ex is t s ;
instead the cavity bubble exhibits a longitudinal gap which s t a r t s a t the
model t ip and inc reases in width downstream, creating a V-shapld opening
which connects the bubble with the outside a i r . If the force coefficients a r e
based on the plan f o r m a r e a of the model below the undisturbed water
surface, these constant force values mean that the lift and drag coefficients
increase slightly when the model p ierces the f r e e surface.
Force and P r e s s u r e Measurements
The resu l t s of the force measurements made on the superventilated
delta wing models a r e presented in F igures 10 through 15 for models with
increasing apex angle. In these f igures lift and drag coefficients a r e plotted
a s a function of model cavitation number, K, a s i s the position of the center - of p res su re x . The force coefficients and cavitation (or ventilation) number
have been defined in the usual way ( see l i s t of symbols on pages i and ii).
F igure 11, which presents the r e su l t s of the measurements performed
on the 15 degree delta wing, presents data taken over a grea ter range of
operation pa ramete r s than has been done for the other models since this
model had been singled out for exploration of the effect of variations in sub-
mergence, model scale , etc. The resu l t s for angles of attack l e s s than
21 degrees and those for cavitation numbers l e s s than 0.05 were obtained by
running the model nea re r the f r e e surface than the 0.65 ft. value used f o r
the main tes t sequence. As pointed out previously, in some cases the model
t ip was permit ted to broach the water surface.
The effect of apex angle on the force and moment coefficients a t a
constant value of cavitation number and angle of attack i s presented in the - c r o s s plot shown in F igure 16.
Some indication of the effect of model dead r i s e and plan fo rm, a s well
a s of deviations f r o m standard orientation and configuration, a r e shown in
F igure 17 for models with a nominal apex angle of 15 degrees and a t a
30 degree angle of attack.
A comparison of F igure 1 7 with F igure 1 1 shows that inverting the model
in the tunnel had no significant effect on the measured force coefficients o r
on the position of the center of p res su re . Likewise, the effect of blunting
the model, which had been done in a n attempt to secure better cavity bubble
formation a t the lower angles of attack had no significant effect on the
measured coefficients.
Running the model backwards so that the 15 degree apex t rai led
downstream resul ted in lift and drag coefficients ve ry close to those
obtained for the orthodox configuration. The center of p res su re , measured
in a l l ca ses f r o m the trail ing edge of the model shows a more pronounced
shift upstream f r o m the centroid of the plan f o r m and the re-entrant jet o r
septum i s not formed when the model i s in this attitude. Instead, the upper
surface of the cavity bubble soon i s flat in t r ansve r se section where i t
leaves the leading edge of the model and soon becomes more concave
(downward) a s the section i s taken f a r the r downstream.
The diamond-shaped model Number 8 and the 90 degree V-bottom
Model 3A, were tested only to explore, in a prel iminary way, the effect of
the change in model plan f o r m and dead r i s e on the measured force
coefficients and on the observed flow geometry.
The p res su re distributions of F igures 18 and 19 supplement the infor-
mation provided by the force measurements and provides bet ter insight
into the effect of change of angle of attack and cavitation number on the
15 degree model. The p res su re coefficient a s used he re i s r e fe r red to
the cavity p res su re p instead of the f r e e s t r e a m stat ic p res su re p a s i s k 0 '
customarily employed for fully-wetted flows. Using this notation the
stagnation p res su re i s equal to 1 t K and the p res su re coefficient anywhere
within the cavity bubble i s taken a s zero. The plan f o r m of the model i s
included in these f igures to show the approximate location of the p res su re *
taps and to define the distance coordinates used in the p res su re distribution
plots. F r o m this drawing i t can be seen that the p res su re taps a r e located
along selected r ays drawn through the t ip of the model. In the left hand
curves both the longitudinal center line p res su re distribution and the
individual t r ansve r se section p res su re distributions (shown rotated into the
plane of the paper) give a graphical representation of the p res su re forces
which ac t normal to the model. F o r clar i ty of presentation, p res su re
readings f r o m only the left hand side of the model have been included in
these t r ansve r se views, and the t r ansve r se tap position, although drawn to
the same scale a s the longitudinal position, has not been identified for each
of the sections. The representation on the right hand side of the f igures
shows the p res su re coefficient a t each of the t r ansve r se stations drawn
superimposed and normalized to the value of p res su re coefficient a t the
center line . Within the accuracy of the p ressure measure ments the
t ransverse pressure distributions a r e not affected by the longitudinal
position of the station along the model, nor i s the nature of the pressure
distribution significantly affected by changes in the cavitation number or
angle of attack of the model over the range of values of these variables
shown in the figures.
When the angle of attack of the model was reduced to 1 5 degrees, and
the model was permitted to approach the water surface in order to secure
low cavitation numbers, an interesting change in the pressure distribution
was observed:
(1) Because of an apparent tunnel flow asymmetry near the surface,
the pressure distribution was not symmetrical with respect to the model
centerline. (Compare tufts in Figure 9. )
(2) A pressure distribution, more nearly in agreement with those
in Reference 2 were obtained, that i s , the pressure coefficient increased,
rather than decreased, for rays outboard of the model centerline.
These latter distributions have not been included since the effect of the
flow asymmetry may be producing non-representative values.
ACKNOWLEDGMENTS
This report represents the combined efforts of many individuals:
Mr. Murphy Lum, who carr ied out most of the force and pressure
measurements; Mrs. Laura Gaard, who performed the calculations
and prepared the curves; Mr. Carl Eastvedt, who took the photographs;
and Mr. George Lundgren, who fabricated the models and model
attachments. The helpful suggestions of Professors Acosta and Wu were
instrumental in determining the scope and direction of the experiments.
REFERENCES
1. Jones, R. T. , "Propert ies of Low-Aspect Ratio Pointed Wings at Speeds Below and Above the Speed of Sound", Technical Note No. 1032, Langley Memorial Aeronautical Laboratory, March 1946.
2. Lee, G. H. , "Note on the Flow Around Delta Wings With Sharp Leading Edges", Aeronautical Research Council R and M No. 3070, 1958.
3. Tulin, M. P. , "Supercavitating Flow P a s t Slender Wingstt, Journal of Ship Research, Vol. 3, No. 3, p. 17, 1959.
4. Reichardt, H. and Sat t ler , W., "Three-Component Measurements on Delta Wings With Cavitationu, Max- Planck Institut fuer Stroemungsforschung , Goettingen, July 1962.
5. Knapp, R. T . , Levy, J . , O'Neill, J. P., and Brown, F. B. , "The Hydrodynamics Laboratory of the California Institute of Technology", Trans . ASME, July 1948.
6. Kic e niuk, T a r a s , "An Experimental Study of the Hydrodynamic F o r c e s Acting on a Family of Cavity- Producing Conical Bodies of Revolution Inclined to the Flow", California Institute of Technology, Report No. E 12.17, June 1954.
7. Wu, T. Y. and Cumberbatch, E. , "Cavity Flow P a s t a Slender Pointed Hydrofoilu, Vol. I1 Journal of Fluid Mechanics, pp. 187- 208, 1961.
m
8. Reichardt, H. , "The Laws of Cavitation Bubbles at Axially Symmetr ical Bodies in a Flow1f, R and T No. 766, Ministry of Aircraf t Production, August 15, 1946.
(a) 0 a = 30 K = ,120 V = 14 fps
(b) 0
a = 30 K = ,121 V = 14 fps
0 Fig. 2 - 15 Delta Wing Supported f r o m the Three -
Component F o r c e Balance in the Working Section of the F ree -Sur face Water Tunnel. Electronic F l a s h was Used in (a) Above; T ime Exposure in (b) .
( a ) Model I C
Apex Angle - 15'
Length - . 6 9 3 ft
Area - . 064 ft2
(b) Model 2 C
Apex Angle - 15O
Length - . 6 5 3 ft
Area - . 058 ft 2
( c ) Model 3A
Apex Angle - 15O
Length - . 6 0 0 ft
Area - . 048 ft2
Fig. 3 - Delta W i n g Models
(d) Model 4
Apex Angle - 45'
Length - . 375 f t
Area - . 058 ft 2
( e ) Model 5
Apex Angle - 90' Length - . 250 f t
Area - . 063 ft2
( f ) Model 6
Apex Angle - 60°
Length - . 333 ft ,
A r e a -.064 ft 2
Fig. 3 -(Continued) Delta W i n g Models
(g) Model 7
Apex Angle - 30°
Length - .458 f t
Area - . 056 ft 2
(h) Model 8
ex Angle - 15' ngth - . 9 06 ft ea - . 054 ft2
(i) Model 9
Apex Angle - l o 0
Length - . 828 ft
Area - . 060 ft 2
r i g . 3 -(continued) Delta Wing Models
Fig . 4 - Diagrammatic Sketch of Superventi lated Flow P a s t a 15O Delta Wing a t a Low Cavitation Number and a n Angle of Attack of 30°.
0 Fig. 5 - 15 Delta Wing Showing Injscted Dye
S t r eak and Septum. a = 30 , V = 14.7 fps , K = .13.
0 Fig. 6 - 15O Delta Wing Inver ted in Tunnel. a = - 30 ,
V = 16 fps.
(b) y = 15O a = 30 o
Fig. 7 - Superventilating Delta Wings of Various Apex Angles a t Moderate Angle of Attack.
(e) y = 60' 0
a = 30
(g) Y = 15O (90' V-Bottom) 0
a = 25
(h) Diamond plan f o r m y = 15' 0 a = 30
Fig. 7 - (Continued) Superventilating Delta Wings of Various Apex Angles a t Moderate Angle of Attack.
Fig . 8 - Compar i son of Flow Direct ion i n Cavity Wall Near Edge of Del ta Wing with Apex Angle = 15O and Angle of At tack of 25O. Upper Model is Near Sur face ( submergence t o cen t ro id - .087 ft). Lower Model is at Modera te Depth ( submergence t o centroid- . 57 f t ) . Model Length is . 69 ft.
.v,
Submergence, Y"' = . 250 f t . -7,
Submergence , Y". = = 650 ft .
Cavitat ion Number not Attainable
Cavitat ion Number Not Attainable
Ful ly- Wetted Ful ly- W-etted
F i g . 9 - Bottom View of Superventi lat ing Delta Wing a t Varying Cavitat ion Numbers and a t Two Submergences ( . 250 fee t , lef t hand c o l u y n and . 650 fee t , r igh t hand column) . Angle of At tack = 30 , Velocity = 15 fps .
.05 . I 0 . I 5 C A V I T A T I O N NUMBER, K
Fig. 10 -Lift, Drag and Center of P r e s s u r e Location for a Super- ventilated lo0 Delta W i n g a s a Function of Cavitation Number.
.05 .I 0 . I 5 C A V I T A T I O N N U M B E R , K
F i g . 11 -Lif t , Drag and Cente r of P r e s s u r e Location f o r a Super- venti lated 15 O Delta Wing a s a Function of Cavitation Number.
0 0 . 0 5 .I 0 . I 5 .2 0 .2 5
C A V I T A T I O N N U M B E R , K
Fig . 12 -Lif t , Drag a%d Cente r of P r e s s u r e Location f o r a Super - venti lated 30 Delta Wing a s a Function of Cavitation Number .
.O 5 .I 0 1 5 .20 . 2 5
C A V I T A T I O N NUMBER, K
Fig. 13 -Lif t , Drag and Cente r of P r e s s u r e Location for a Super- venti lated 4 5 O Delta Wing a s a Function of Cavitation Number .
Fig. 14 -Lift, Drag a@ Center of P r e s s u r e Location for a Super- ventilated 60 Delta Wing as a Function of Cavitation Number.
.05 .I 0 .I 5
C A V I T A T I O N NUMBER, K
Fig. 15 -Lift, Drag an$ Center of P r e s s u r e Location for a Super- ventilated 90 Delta Wing a s a Function of Cavitation Number.
MODEL APEX ANGLE, y , I N D E G R E E S
Fig . 16 -Effect of Apex Angle f o r a Supercavi ta t ing Del ta Wing a t an Angle of At tack of 30' and Cavitat ion Number of . l .
M O D E L 2 C
M O D E L 3 A
- - - -
I I I I I I I I I
0 5 .I 0 1 5
C A V I T A T I O N N U M B E R , K
.@a-
n .,
V
- -
- I I I I I I I I I
Fig. 17 -Lift , Drag and Cente r of P r e s s u r e Location fo r Various Superventi lated Delta Wings a t an Angle of Attack of 30°. Model descr ipt ions a r e given in F ig . 3. Flow i s f r o m le f t to r igh t fo r models a s shown.
* v - n V
MODEL NO.
a
2 C -
a ' 2 8 -
1°C -
e
3 A -
-7 I
- -
- - l I I 0 I l l l
h "
+
0 Fig. 18 - P r e s s u r e Distribution of a 15 Superventilated Delta Wing a t
21° Angle of Attack.
0 F ig . 19 - P r e s s u r e Distr ibution of a 15 Supervent i la ted Delta Wing a t 30° Angle of Attack.
APPENDIX I
SUMMARY OF FORCE DATA
2 Model vo -
X
K 2 g - Iesignation y A f t 2 a. y ::
C~ CD L
APPENDIX I
(continued)
Model Designation Y A f t2
APPENDIX I
(continued)
APPENDIX I
(continued)
Model Designation
APPENDIX I
(continued)
Model Designation Y A f t 2 a y ::: K
APPENDIX 11 Summary of P ressure Data
--t- z I - :--
-0- Z x 2
z 3 - 2 - - Z x 4
Apex Angle, y = 15O Velocity =18. 5 fps P r e s s u r e Tap No.
Angle of Attack, a = 30°
Angle of Attack, a = 30°
0 Angle of Attack, a = 2 1
Angle of Attack, a = 21°
DISTRIBUTION LIST
Chief, Bureau of Naval Weapons Depar tment of the Navy Washington 25, D. C. Attn: Codes DLl-3 (1)
RAAD-3 (1) RRRE (1) RRRE-7 (1) RUAW-4 ( 2)
Chief, Bureau of Ships Department of the Navy Washington 25, D. C. Attn: Codes 335 (1)
42 1 (2) 44 2 (1) 6 44 (1)
Chief of Naval R e s e a r c h Depar tment of the Navy Washington 25, D. C. Attn: Codes 429 (1)
43 8 (1) 46 6 (1)
Commanding Officer and Di rec to r David Taylor Model Bas in Washington 7 , D. C. Attn: Codes 142 (1)
500 (1) 526 (1) 59 1 (1)
Chief, Bureau of Yards and Docks Depar tment of the Navy Washington 25, D. C. Attn: R e s e a r c h Division (1)
Commanding Officer and Di rec to r U. S. Naval Engrg . Exper imen t Sta. Annapolis, Maryland Attn: L i b r a r i a n (1)
Superintendent U. S. Naval Academy Annapolis , Maryland Attn: L i b r a r i a n (1)
Superintendent U. S. Naval Pos tgraduate School Monte rey , Cal ifornia Attn: L i b r a r i a n (1)
Commander U. S. Naval Weapons Labora tory Dahlgren, Virginia Attn: L i b r a r i a n (19
Commanding Officer U.S. Naval Underwater Ord. Sta. Newport, Rhode Is land Attn: L i b r a r i a n (1)
Commander U. S. Naval Ordnance Labora to ry White Oak, S i lver Spring, Md. Attn: Desk DL ( L i b r a r y ) (1)
XL (Aerobal l i s t ics ) (1)
Commanding Officer and Di rec to r U. S. Naval Civil Engrg . Labora to ry P o r t Hueneme, Cal ifornia Attn: L i b r a r i a n (1)
Commander U.S. Naval Ordnance T e s t Station Pasadena Annex 3202 E. Foothi l l Blvd. Pasadena , Cal ifornia Attn: Codes P508 (2)
P804 (2) P8074 (1) P8076 P80962
(1) (1)
D i rec to r Naval R e s e a r c h Labora to ry Washington 25, D. C. Attn: L i b r a r i a n (1)
Commanding Officer and Di rec to r U.S. Navy Underwater Sound Lab. F o r t T rumbul l New London, Connecticut Attn: L i b r a r i a n (1)
Commanding Officer and Di rec to r U. S. Navy E lec t ron ic s Labora to ry San Diego 52, Cal ifornia Attn: L i b r a r i a n (1)
Commanding Officer and Di rec to r U. S. Naval A i r Developmt, Center Johnsvi l le , Pennsylvania Attn: L i b r a r i a n (1)
Commanding Officer U. S. Navy Mine Defense Lab. Panama City, F lo r i da Attn: L ib r a r i an (1)
Commander U. S. Naval Ordnance T e s t Station China Lake, California Attn: Code 7533 (1)
B r i t i sh Jo in t Se rv i ce s Miss ion (Navy Staff) , v ia Chief, Bu reau of Naval Weapons Depar tment of the Navy Washington 25, D. C. Attn: Code DSC-3 ( 4)
Defense R e s e a r c h M e m b e r (W) Canadian Joint Staff, v ia Chief, Bu reau of Naval Weapons Depar tment of the Navy Washington 25, D. C. Attn: Code DSC-3 (1)
ASTIA Refe rence Cente r Technical Informat ion Division L i b r a r y of Cong re s s Washington 25, D. C. (1)
Defense Documentation Cente r Cameron Stat ion Alexandr ia , Virginia ( 6 )
Office of Technical Se rv i ce s Depar tment of C o m m e r c e Washington, D. C. (1)
Scientif ic and Technical Informat ion Fac l l i f y Attn: NASA Re presen ta t ive P. 0. Box 5700 Bethesda, Maryland 20014
Director National! Bureau. of S tandards Washington 25, D. C. Attn: Fluid Mechanics Div. (1)
Coordinator f o r R e s e a r c h Mar i t ime Adminis t ra t ion 441 G St r ee t , N. W. Washington, D. C. (1)
Di rec to r Engineer ing Sc iences Division National Science Foundation 1520 W St r ee t , N. W. Washington, ID. C. (1)
Commander A i r R e s e a r c h and Developmt. Command Andrews A i r F o r c e B a s e Washington 25, D. C. (1)
A i r F o r c e Office of Scient . R e s e a r c h Mechanics Division Washington 25, D. C. (1)
Commanding Officer Office of Ordnance R e s e a r c h Box CM, Duke Stat ion Durham, North Caro l ina (1)
Commit tee on Undersea War f a r e Nat. Academy of Sc iences National R e s e a r c h Council 21 01 Consti tut ion Avenue, N. W. Washington 25, D. C . (1)
D i r ec to r U. S. A r m y Engineer Waterways Expe r imen t Stat ion Corps of Eng inee r s Vicksburg, Miss i s s ipp i (1)
Super intendent U. S . Merchan t Mar ine Academy Kings Po in t , Long Is land, N. Y. Attn: L i b r a r i a n (1)
Massachuse t t s Inst i tute of Tech. Cambr idge 39, M a s s . Attn: Dept. of Naval Arch i t e c tu r e
and Mar ine Engineer ing P r o f . L . T r o o s t (1) Hydrodynamics Labo ra to ry P r o f . A. Ippen (1)
Applied P h y s i c s Labo ra to ry Univers i ty of Washington Seat t le , Washington Attn: L i b r a r i a n (1)
D i r ec to r St. Anthony Falls Hydraul ic Lab. Univers i ty of Minnesota Minneapolis 14, Minn. (1)
Stanford Univers i ty Stanford, Cal i fornia Attn: Dept. of Mechanical Engrg .
P r o f . B. P e r r y (1) Head, Dept, of Math. ( 1 )
Cornell University Ithaca, New York Attn: Director, Graduate School
of Aeronautical Engrg. (1)
Harvard University Cambridge 38, Massachusetts Attn: Dept. of Engineering Sciences
Prof . G. F. C a r r i e r (1) Dept. of Mathematics Prof. G. Birkhoff (1)
Iowa Institute of Hydraulic Research State University of Iowa Iowa City, Iowa Attn: Prof . H. Rouse, Dir. (1)
Prof. L. Landweber (1)
Director Alden Hydraulic Laboratory Worcester Polytechnic Institute Worcester , Massachusetts (1)
University of Arizona Department of Mathematics Tucson, Arizona Attn: Prof. L. M. Milne- Thomson
(1)
Director Garfield Thomas Water Tunnel Ordnance Research Laboratory Pennsylvania State University P. 0. Box 30 University Pa rk , Pa . (1)
Davidson Laboratory Stevens Institute of Technology 71 1 Hudson Street Hoboken, New J e r s e y Attn: Dr. J . Bres l in (1)
Johns Hopkins University Balt imore 18, Maryland Attn: Prof. S. Corrs in , Head
Dept, of Mech. Engrg. (1)
Colorado State University Fo r t Collins, Colorado Attn: Prof. M. Albertson
Dept. of Civil Engrg . (1)
University of Michigan Ann Arbor , Michigan Attn: Prof . R. B. Couch (1)
Prof . V. S t ree te r (1)
Polytechnic Institute of Brooklyn 99 Livingston Street Brooklyn 2, New York Attn: Head, Dept. of Aero. Eng.
and Applied Mech. (1)
Brown University Providence, Rhode Island Attn: Div. of Applied Math. (1)
Div. of Engineering (1)
University of California Berkeley 4, California Attn: College of Engineering
Prof . A, Schade (1 Prof. I.V. Wehausen (1)
Webb Institute of Naval Architecture Crescent Beach Road Glen Cove, Long Island, N. Y . Attn: Librar ian (1)
New York State Marit ime College F o r t Schuyler, New York Attn: Librar ian (1)
University of Kansas Lawrence, Kansas Attn: Dean J. S. McNown (1)
Lehigh University Bethlehem, Pennsylvania Attn: Prof . J . B. Herbich
Civil Engrg. Dept. (1)
University of Notre Dame Notre Dame, Indiana Attn: Prof . A.G. Strandhagen (1)
Dept. of Engineering Mech.
Rensselaer Polytechnic Institute Troy, New York Attn: Prof . H. Cohen
Dept. of Mathematics (1)
California Institute of Technology Pasadena, California Attn: Prof. F. C. Lindvall (1)
Prof. M.S. P l e s se t (1)
University of Illinois Urbana, Illinois Attn: College of Engineering
Prof . J. Robertson (1
Scr ipps Institution of Oceanography Universi ty of Cal i fornia L a Jo l la , Cal i fornia Attn: L i b r a r i a n (1)
Woods Hole Oceanographic Inst i tute Woods Hole, Massachuse t t s Attn: L i b r a r i a n (1)
Case Inst i tute of Technology Cleveland, Ohio Attn: L i b r a r i a n (1)
Inst i tute of F lu id Mechanics and Applied Mechanics Universi ty of Maryland College P a r k , Md. Attn: L i b r a r i a n
Yale Univers i ty Mason Labora to ry 400 Temple S t r e e t New Haven 10, Connecticut Attn: L i b r a r i a n (1)
Phi lco Corpora t ion 4700 Wissahickon Avenue Phi ladelphia , Pennsylvania Attn: Engrg . L i b r a r i a n (1)
Vitro Corpora t ion of A m e r i c a 96 2 Wayne Avenue Si lver Spr ings , Maryland Attn: Engrg . L i b r a r i a n (1)
Gibbs and Cox 2 1 West S t r e e t New York 6 , N.Y. Attn: Dr . S. H o e r n e r
Hydronautic s , Inc. Pindel l School Road Howard County Laure l , Md. Attn: M r . P. E i senbe rg
Technical R e s e a r c h Group Route 1 10 Melville, N. Y . Attn: L i b r a r i a n
Aero je t G e n e r a l Corpora t ion 6352 Nor th Irwindale Avenue Azusa , Cal i fornia Attn: M r . J. Levy
The Mar t in Company Ba l t imore 3 , Maryland Attn: Science Tech . L i b r a r i a n
Mai l No. 5398 (1)
Nor th A m e r i c a n Aviation, Inc. Inte rna t iona l A i r p o r t Los Angeles 45, Cal i fornia Attn: Engineering L i b r a r i a n
Dept. 56 (1)
Lockheed A i r c r a f t Corpora t ion 1555 N. Hollywood Way Burbank, Cal i fornia Attn: Engineering L i b r a r i a n
Bldg. 6 3 , F a c t o r y A1 (1)
Douglas A i r c r a f t Company, Inc . E l Segundo, Cal i forn ia Attn: M r . A . M . 0. Smith (1)
Bel l A e r o s y s t e m Company P. 0. Box 1 Buffalo 5, New York Attn: Engineering L i b r a r i a n (1)
McDonnell A i r c r a f t Corpora t ion P. 0. Box 516 St . Louis 3, Mis sour i Attn: Engineering L i b r a r i a n (1)
Chance Vought A i r c r a f t , Inc. P. 0. Box 5907 Dal las 22, Texas Attn: Engineering L i b r a r y (1)
Republic Aviation Corpora t ion Fa rmingda le , Long Is land, N. Y . Attn: Engineering L i b r a r i a n ( 1)
E D 0 Corpora t ion College Poin t , New York Attn: Engineer ing L i b r a r i a n (1)
The RAND Corpora t ion 1700 Main S t r e e t Santa Monica, Cal i fornia Attn: L i b r a r i a n (1)
E l e c t r i c Boat Division G e n e r a l Dynamics Corpora t ion Groton, Connecticut Attn: Engineering L i b r a r i a n (1)
Hydrodynamics Laboratory Convair Division General Dynamics Corporation P. 0. Box 1950 San Diego 12, California Attn: Mr. H.E. Brooke (1)
Goodyear Aircraf t Company Akron 15, Ohio Attn: Engineering Librar ian (1)
Grumman Aircraf t Engrg. Corp. Bethpage, Long Island, N. Y. Attn: Engineering Librar ian
Plant 5 (1)
Aeronutronic Division Ford Motor Company Ford Road Newport Beach, California Attn: Engineering Librar ian (1)
Director Department of Applled Mechanics Southwest Re sea rch Institute 8500 Culebra Road San Antonio 6, Texas (1)
Boeing Airplane Company P. 0. Box 3707 Seattle, Washington Attn: Aero- Space Div. Librar ian
Org. No. 2-5190, Mail Stop 1384 (1)
Hughes Tool Company Florence and Trole Culver City, California Attn: Librar ian Bldg. 2
Mail Station 5 (1)
United Technology Corporation P. 0. Box 358 Sunnyvale, California Attn: Dr. D. A. Rains (1)
Cleveland Pneumatic Ind. , Inc. Adv. Systems Development Div. 1301 El Segundo Blvd. E l Segundo, California Attn: M r . S. Thurston (1)
Mr . W. Ellsworth (1)
Westinghouse Elect r ic Corp. Baltimore Division P. 0. Box 1897 Friendship International Airport , Md. Attn: Engineering Librar ian ( I )
General Elect r ic Corporation Ordnance Department 100 Plas t ics Ave. Pittsfield, Massachusetts Attn: Engineering Librar ian (1)
Society of Naval Architects and Marine Engineers 74 Trinity Place New York 6, N.Y. (1)
Applied Mechanics Reviews American Soc. of Mech. Engineers 29 West 39th Street New York, N. Y . (1)
Engineering Societies L ibra ry 29 West 39th Street New York 18, N.Y. (1)
Oceanics, Inc. Technical Industrial P a r k Plainview, Long Island, N. Y . Attn: Dr. P. Kaplan (1)
General Elect r ic Corporation LME Dept. , Bldg. 28 No. 1 River Road Schenectady 5, New York ~ t t n : Engineering Librar ian ( 1)
Clevite Brush Development Clevite Research Center 540 E . 105th Street Cleveland, Ohio Attn: Engineering Librar ian ( 1)
AVCO Manufacturing Company 2385 Revere Beach Parkway Evere t t 49, Massachusetts Attn: Engineering Librar ian (1)
Inst . of the Aerospace Sciences Library 2 Eas t 64th Street New York 21, N. Y . (1)
New York Naval Shipyard Mater ia l Laboratory Brooklyn, N. Y. Attn: Mr . D. Kallas ( 2 )
Laboratorio Hidrotecnico Salturnino de Bri to Rua F e r r e i r a Pontes 637 Rio de Janeiro , Brazi l Attn: Mr. V. F. Motta ( 1 )
C ommande r Office of Naval Research USN, Fluid Dynamics Branch Washington 25, D. C. Mr. A. J. Coyle (6)