Flow Control Device Evaluation for an Internal Flow with an Adverse Pressure Gradient

10 Pages·2002·0.62 MB·English

by  Jenkins, Luther N.

AIAA 2002-0266
Flow Control Device Evaluation for an
Internal Flow with an Adverse Pressure
Gradient
L. Jenkins, S. Althoff Gorton, and S. Anders
NASA Langley Research Center
Hampton, VA
AIAA Aerospace Sciences
40 th
Meeting & Exhibit
14-17 January 2002/Reno, NV
II
For permission to copy or to republish, contact the copyright owner named on the first page.
For AIAA-held copyright, write to AIAA Permissions Department,
1801 Alexander Bell Drive, Suite 500, Reston, VA, 20191-4344.
AIAA-2002-0266
FLOW CONTROL DEVICE EVALUATION FOR AN INTERNAL FLOW
WITH AN ADVERSE PRESSURE GRADIENT
Luther N. Jenkins*, Susan Althoff Gorton +, and Scott G. Anders:
NASA Langley Research Center
Hampton, Virginia
Abstract aft section of the vehicle. This configuration of the BWB is
shown in Ref. 6and pictured in Figure 1.
The effectiveness of several active and passive devices to
control flo_ in an adverse pressure gradient with secondary When the engines are positioned near the surlhce, the BWB
flows present was evaluated in the 15 Inch l,ow Speed Tunnel engine inlet must be an S-duct inlet _ith the capabilit} to
at NASA Langley Research Center. In this stud)', passive ingest the large boundary layer that will build up over the
micro vortex generators, micro bumps, and piezoelectric aircraft body. The inlet must pertbrrn this task without
synthetic .jets x_ere evaluated for their flow control producing a significant engine performance penalb' in terms
characteristics using surface static pressures, flo_ of distortion or pressure recovery. Since the boundau' layer
visualization, and 3D Stereo Digital Particle Image on the BWB is expected to be on the order of 30% of the inlet
Velocimetu. Data also were acquired lbr synthetic .jet height, this presents achallenging task lbr inlet design.
actuators in a zero flox_ environment. It was tbund that the
micro vortex generator is very effective in controlling the The requirements tbr inlet pertbrmance under the severe
flow environment tbr an adverse pressure gradient, even in conditions of an adverse pressure gradient from the S-duct
the presence of secondary vortical flow. The mechanism by and a very large onset boundao, layer flow have led to the
which the control is effected is a re-energization of the consideration of active llow control devices in the inlet to
boundary la.ver through flow mixing. The piezoelectric control the flow. As reported in References 7-25, much
synthetic jet actuators must have sufficient velocity output to research is already underway to identi[_ and develop active
produce strong longitudinal vortices ifthey are to be effective llow control devices and technologies and this represents onl?
tbr flow control. The output of these devices in a laboratory a sampling of the available material on the subject. There
or zero flow environment will be different than the output in a have also been investigations showing the successful use of
flow environment. In this investigation, the output was higher passive and active flow control technologies applied to inlets.
in the flow environment, but the stroke cycle in the flo_ did Ret_erence 6 discusses work using passive devices lbr an S-
not indicate a positive inflow into the synthetic jet. duct with boundary layer ingestion (BLI), and References 18-
20 discuss both passive (microvanes) and active (micro.jet)
Introduction concepts applied to aggressive serpentine inlets.
The effect of aviation on the environment and in particular The purpose of the present investigation was to lay the
global warming has recently become a focus of study _. In groundwork for a future study of active llo_v control applied
response to environmental concerns and to tbster to a duct representative of a BWB with BLI. In the present
revolutiona_' propulsion technologies. NASA launched the stud3, the ett_ctiveness of several active and passive devices
Ultra Efficient Engine Technology (UEET) program in late to control flow in an adverse pressure gradient with secondau,
19992 . This program has several elements, one of which is to flows present was evaluated by examining pressure recovery.
explore the feasibility of the Blended-Wing-Body (BWB) flow topology, and flo_-field velocity and vorticity
concept as an efficient alternative to conventional transport characteristics. These data were obtained for passive micro
configurations. The BWB concept has been considered in vortex generators, micro bumps, and synthetic jets using
various tbrms for several years _-_. Studies have shown that in surface static pressures, |1o_ visualization, and 3D Stereo
order to make the largest impact on the vehicle perlbrmance, Digital Particle Image Velocimetry.
the engines and inlets should be placed near the surface on the
Experimental Apparatus and Methods
*ResearchEngineer, Flow Physics andControl Branch
_Research Engineer, Flow Physics andControl Branch, Member AtAA
Facility and Model
;Research Engineer. FlowPhysics andControl Branch
Copyrigh_ © 2(102 b_ the American Institute of Aeronautics and ]'he experiment was conducted in the NASA Langley 15-Inch
Astronautics, Inc No copyright isasserted m lhe United States under Low Speed Tunnel. This tunnel is a closed return,
Title 17, U.S Code The US Government has aroyalw-free license to atmosphcric facility' used primarily tbr fundamental flow
exercise all rights under the copyright claimed herein [brgovernment
purposes Allother rights reserved bythecopyright o_71er
I
American Institute of Aeronautics and Astronautics
incidence of the cameras to the light sheet, all threc Baseline flow field
components of velocity were measured in each PlY Thc baseline configuration flow visualization topology is
measurement plane through stereoscopic vector shown in Figure 7 tor the freestream velocity of 140 it/see.
reconstruction, The light sheet was produced by a pulsed, Although flow along the splitter plate in the tunnel is two-
frequency-doubled, 300mJ Nd:YAG laser operating at 10 dimensional for the most part. two large spiral nodes reveal
Hz. The laser could also be triggered phase-locked to the the lormation of vortical structures. This occurs when the
synthetic jet input signal. In this mode, the laser would fire sidewall boundau' layer reacts to the adverse pressure
on multiples of the synthetic jet cycle, as the laser gradient near station 61.75 on the ramp. The vortical
physically could not fire at afaster rate than 10 Hz. structures are similar to what might be expected from
secondau, flow and vortex liftoff in a duct. so no attempt
At each measurement location, the PlY field of view was was made to control the vortices tbr this investigation.
approximately 4 inches wide by 3 inches tall, centered Rather, it was thought that the challenges of the strong
along the centerline of the tunnel. The measurement vortical flow field would provide a better indication of how
location was carefully aligned with the model system, and the flow control devices would work in a realistic inlet
the cameras were calibrated with an in-situ target for each configuration. It should be noted that the vortices are
location. The tunnel was seeded with atomized mineral oil highly unsteady and appear to have atrajectory that departs
injected into the flow in the tunnel settling chamber, and from the surthce of the ramp and extends downstream in
the particle size was approximately 5-10 microns. For all the tunnel. In addition to the vortical structures, Figure 7
conditions, at least thirty samples of PlY data were also highlights other significant flow features such as a
obtained over a 3 to 6 second period and averaged. The separation node, an attachment node, and evidence of
low rms of the mean data indicated that this was enough reverse flow in the center of the ramp.
data to capture the relevant flow features lbr this
investigation. The algorithm used to process the images Figure 8 presents the centerline and spanwise surface static
acquired in this investigation is described in Reference 30. pressure distributions tor the ramp. In Figure 8a, the
Estimating the accuracy of the stereo PlY measurements is repeatability of the baseline pressure profile over a time
itself a matter of instrumentation research at this time: the period of two months and after two major model removals
best estimate the authors can provide tbr the accuracy of is also shown.
the PlY velocity measurements is included in Table I.
The centerline pressure distribution indicates separation
Test Conditions occurring near station 64 but does not show the dramatic
The main test condition was established by setting the flow features that the flow visualization revealed. In fact,
tunnel velocity to 100 ft/sec. This corresponded to a local the spanwise pressure distribution in Fig 8b indicates a
velocity of 140 it/see at station 57 due to the acceleration of fairly unilbrm and symmetric pressure pattern. In the
the flow above the splitter plate. Station 57 was the thnhest absence of the other information from the flow
aft static surface pressure port location on the flat pan of visualization and PIV. this type of pressure distribution
the splitter plate. For this reason, Station 57 conditions are could easily be interpreted to be representative of unilbrm.
used to define the onset flow to the adverse pressure two-dimensional flow.
gradient ramp, The boundary laver was measured at station
57 and lbund to have a thickness. _,of approximatelk 0.87 Figure 9 shows velocities measured using PlY along the
inches. The boundary profile was converted to wall centerline of the ramp geometD, at four longitudinal
coordinates and compared with Spalding's Law. Based on stations. Each li'ame consists of at least thirty samples of
the agreement between the two profiles, the boundary layer data acquired at the laser internal trigger frequency of 10
was determined to be turbulent. Hz. The contours clearly indicate a thin region of reverse
flow in the center of the ramp at station 68.00. The
Discussion of Results velocity measured vet) close to the surlhce on the
centerlinc is plotted in Figure 10 and shows very slow
moving and even reversed flow atthese locations.
Data _ere obtained tbr man), different configurations and
test conditions during this investigation. In this paper, the
Flow Control Devices
basic flow over thc ramp will be presented to define the
As described earlier, several flO_r control devices were
baseline flow environment with pressures, flow
applied to the ramp in order to assess their relative ability
visualization, and flow field velocity measurements.
to control the flow. Figure I1shows the flow visualization
Comparisons among the differcnt flow control devices will
obtained along the ramp tbr the MVG's and the synthetic
then be presented with respect to the baseline to emphasize
the effect of the devices on the flo_ environment. In the ,jets. There was no flo_ visualization obtained lor the
micro bumps. Note how the MVG's create a series of
tinal section, details of several attempts to optimizc the
strong vortices, as indicated by the dark separation lines.
synthetic jet output will be given, and the ,jet perlbrmance
which reduce the influence of the sidewall vortices and
in a no-flow environment will be presented.
allow the flow in the center of the ramp to remain attached.
3
American Institute of Aeronautics and Astronautics
variationwsereattemptteodincreastheeperlbrmanocfe
actuators operating at 700 Hz and an input amplitude of 92
thejets.Althougthhesynthejteictshadbeeonptimizeodn
VAC. At the lowest velocity, 45 l_sec, the actuators
thebencthop,itwasthoughthtattheoptimufmorflow
appeared to improve in perlbrmance. Figure 19 shows
contromlighntotcorrespotnod
the optimum for synthetic vectors lbr the phase-locked output of the jets tbr atunnel
.jet operation. Additionally, it was possible that the jets velocity of 140 Wsec and 45 fl/sec. In both cases, the
were not operating in the same manner in a flow freestream vertical velocity bias has been removed to show
environment as they did on the bench. In order to sort out the operation of the jet. The vectors indicate that the jet
these issues, some limited parametric variations were output has essentiaffy doubted fbr the higher speed
evaluated and are discussed belo_ _. condition, and the ratio of the maximum jet output to the
freestream has increased from 14% in the 140 ft/sec case to
Hole Size - Originally the synthetic .jet output holes were 18% in the 45 ft/sec case.
0.040 in. in diameter. Because the vortex generation was
not strong enough, the hole diameter was increased to No Flow Operation of the Jets - ]he question of whether
0.094 in. in order to increase the mass flow through the the jets were operating as efficiently in the flow
holes. This was the largest size hole possible lbr the environment as they did in a laboratou_ environment could
current geometu,. Figure 17 shows that there was little only be answered by measuring the output of the jets in situ
effect on the pressure recover) due to increasing the hole with noonset flo_v.
size, although it was noted during the testing that the mass
flox__had increased substantially. Figure 20 presents the zero flow operation lbr the synthetic
.jets at 700 Hz and 300 ttz with 92 VAC and zero
Backpressure - It was hypothesized that perhaps the reason backpressure. Note that the output magnitude of the jets is
the .jet output was lower than expected was that the far less than Ref. 21 reports and also less than the output
synthetic jet could not adequately pull in air mass during shown under the onset flow conditions in Figure 19.
the instroke cycle in the presence of the onset flow and its Itowevcr, in Figurc 20 there is a clear inflow and outflow
pressure field. With no air ingested during the instroke, stroke of the actuator that is not apparent with the flox_ on
there would be little air available to pump out on the (Figure 19). Also the flow generated by thc.jets penetrates
outstroke. In order to ensure that the actuator had mass further awa_ from the jet inthe no flow condition.
available to pump out. the actuators were modified by
installing small air pressure teed lines directly to the .jets. The difference between the jet output in zero flo_ lbr this
A high-resolution regulator controlled the air in the lines, configuration and that of Ref. 21 may bc due to the small
and various backpressures were applied to the plenum in this configuration that increases the distance
configuration. between the jet output slot and the surface of the ramp.
However, such plenums ma_ be necessa_ for realistic
Figure 18 shows the pressure distribution for the zero applications and the pertbrmance of the actuator must be
baekpressure case and two cases with backpressurc applied improved to account lbr this. The results of the no onset
at 60 psi and 80 psi. Analysis of PlY velocity data for the flow measurements also show that the actuator output in a
zero backpressure case and the 60 psi backpressure case zero t'1o_'environment is lower than what is achieved in an
also showed that backpressure has a minimal effect _br the onset flow setting, but the onset flow condition affects the
140 fl/sec case with the actuators operating at 700 Hz, penetration of the velocity into the llov_ as well as the
There was some slight effect of backpressure when the phasing and stroke cycle of the actuator. These factors
tunnel speed was lowered to 45 h/see and the actuators must be considered when the requirements Ibr flow control
were run at 300 Hz. Stead), blowing through the actuators are determined.
backpressure tubes without the synthetic jets operating also
had no effect. These data lead to thc conclusion that lack Conclusions
of air mass was not the primary reason lbr the low output of
the synthetic jets in the onset flow. The effectiveness of several active and passive devices to
control flow in an adverse pressure gradient with secondau,
Frequeno,- With the freestream velocit) at 140 fl/sec, the flows present was evaluated. In this stud), passive micro
operating frequency of the .jets was swept through a range vortex generators, micro bumps, and piezoelectric synthetic
from 200-1001) Hz with no noticeable effect on the pressure jets were evaluated tbr their flo_v control characteristics
recoveu, along the ramp. using surface static pressures, fio_ visualization, and 3D
Stereo Digital Particle Image Vclocimetr)'. Data also were
Amplitude The amplitude of the synthetic jet input signal acquired for synthetic jet actuators in a zero flow
was swept through a range of 40-92 VAC at a freestream environment. The conclusions are summarized as Ibllo_s:
velocity of 140 ft/sec with no significant effect on the
pressure recoveu' data. l, The micro vortex generator is ver3 effective in
controlling thc flow environment for an adverse
Freestream IelociO' - The freestream velocity _as pressure gradient, even in the presence of
changed in a range from 45 fl/sec to 140 ft/sec with the
American Institute of Aeronautics and Astronautics
24 GreenblaDtt,, andWygnanskI,i, "'Parameters 28 Bryant, R G, Fox, R l.,Lacho_icz, J 1-,andChen.
AffectingDvnamiSctallControbly Oscillatory. F J, "'Piezoelectric Synthetic Jets tbr Nircrafi Control
ExcitationA,"IAA09-31I,2June1t,)QQ Surfaces," SPIE Proceedings. Vol 3074, 1000. PP
25 Wygnans[k,i,'SomeNewObservatioAnffsecting 220-227
theControolr"SeparatiboynPeriodiEcxcitation,'" 29 /,struments and Apparatus "+PactI Measurement
A[AA2000-23J1u4n.e2,000
Uncertainty," ANSI/ASME PTC Io I-1_)85, American
2(3Lin,J C, "'ControolfTurbulenBtoundary-Layer National Standards [nst, 1085
Separati[o;sningMicroVortexGeneratorAs,I"AA 30 Lourenco, L M., and Krothapalli, A. "lme
90-340J4u,ne1,9o9 Resolution PIV: AMesh-Free Second Order Accurate
27 Joslin,R D, ttorta,[, G, andChen,F J, Algorithm," Proceedings of the I0a' International
"q'ransitioning Active Flow Control to Applications," Symposium on Application Techniques in Fluid
AIAA 00-3575. June, 100,3 Mechanics, Lisbon, July 2000
Table 1 Measurement Uncertainty
_0001
Temperature, de.g F ±0. I Ct,
Density, slu_,fi' _0.00001 PlY velocity components, fl/sec
±001 Streamwise =5 2
Total pressure, psi
Dynamic pressure, psi ±0,01 Vertical =26
Tunnel velocity, ft/sec 1.3 Lateral z26
Figure 4 Micro vortex generator (M\:G's) configuration.
Figure I Blended Wing Body configuration.
Figure 2 Adverse pressure gradient ramp installed in the 15- Figure 5, Micro bump configuration
Inch Love Speed Tunnel
13/32"
Figure _ Piezoelectric synthetic jet configuration.
Figure 3 Micro vortex generator profile
American Institute of Aeronautics and Astronautics
FLOW DIRECTION 3 Streamwise
veloclty,fu_¢
._c2.5 III III '
-5 8 21 :1447 60 73 8G 99 112125
1 _ L5
0
125 t2_,
0.5
_ 0
-2 -1 0 1 2
Lateral distance from centedine, inches
Streamwise
Vek_ity,lt/xec
1
0
-2 -1 0 1 2
Lateral distance from ¢entedine, inches
1
234
1
-2 -1 0 1 2
Lateral distance from centedine, inches
Figure 9. Velocity contours for baselinc ramp at measurement stations X =61.75, X= 66.90. X= 68.00, and X= 69.50.
9
American Institute of Aeronautics and Astronautics
Synthetic Jets
le 3
=o
Streamwise
.o¢: 2.5 Velocity, ftJe.c
-5 8 21 34 47 80 73 86 99 112 i25
-5 8 21 34 47 60 73 86 1)9 112125
1.5 125
_ 0.5
-2 -1 0 1 2 -2 -1 0 1 2
Lateral distance from centadine, inches Lateral distance from centerline, inches
X=61 .75
_ 3 Streamwise
Velocity, fU_c
o l Streamwise
._ 2.5 :_2s_- ,..o,,._.o
-5 8 21 34 47 60 73 86 99 112 125
m 2
2[-- -S258 21 34 47 60 73 86 9g 112125
_ 1,5
0
8
_ 0.5 7 `¸
.__
_ 0 -2 -1 0 1 2 a o -2 -1 0 1 2
Lateral distance from centedine, inches Lateral distance from centerline, inches
X = 66.90
Stmamwise m_ Streamwise
velocity, ft/sec
c 2. .c 2
t
2
J
1.5 /
.... 2
°
n 1
8 8
_0. "0.5
_1 0 _ 0
-2 -1 0 1 2 -2 -1 0 1 2
Lateral distance from centedine, inches Lateral distance from centedine, inches
X = 68.00
.c 2.5
1.5
c: 05
-2 -1 0 1 2 -2 -1 0 1 2
Lateral distance from centedine, inches Lateral distance from centerline, inches
X = 69.50
Figure 14. Comparison of stream_ise velocity contours fi)rMVG's and synthetic .ietsatstations X =61.75. X =66.90.
X = 68.00. and X =69.50.
I1
American Institute of Aeronautics and Astronautics
1.5 1.5
¢0 Phase =0deg. m Phase =0 deg.
JGo:) Vorticity, sac-1 llqr'rff J¢ Vertical velocity, ft/sec
.=_ .G
-340-260.180 -1oo-2Q 60 140 :)20 300 380 460
'I¢
¢o
¢D
>
j_ O,S j_ 0.5
8 8
c c
..q
•= 0
¢3 Synthetic Synthetic atlocations
-2 -1 0 -2 -1 0
Labaraldistance from centedine, inches
1.5 15
Phase = 90 deg. ¢0 Phase =90 (:leg.
Gr.) Vorticity, sac-1 J¢ Vertical velocity, Wsec
cJ (J
._¢ ._c
-340-260-180-100 -20 60 140 220 300 380 460 -20-16.12 -8 -4 0 4 8 12 16 20
¢o
'i=
-i
._ 0.5 j_O.5
8 8
c
•_ 0 .-_0
0 Synthetic etlocations
-1 0 -2 -1 0
Lateral distancefromcentedine, inches Lateral distancefrom centedine, inches
1,5
Phase = 180 (leg. Phase = 180 deg.
=
Vorticity, sac-1 J= Vertical velocity, Wsec
o
._=
-340-260-180-100 -20 60 140 220 300 380 460 -20-16-12 -8 -4 0 4 8" 12 16 20
,-i
A}
0,5
8
P
e_ etlocations
-2 -1 0 -2 -I 0
Lateral distancefrom centedine, inches Lateral distancefromcentedine, inches
w 1.5 Phase = 270 deg. 1.5['- phase =270 deg.
® vortic_, sec-_ ._ Vertical velocity, ff/sec
"_ -340-260-tSO-100 -20 60 140 220 300 380 460 '' -20-16-12 .8 -4 0 4 8 12 16 20
1
-'J
>_ >_
30. 3o._
-2 -1 0 -2 -1 0
Lateral distancefrom cenledine, inches Lateral distancefrom centedine, inches
Figure 16. Phase-locked worticity and vertical velocity contours of the synthetic.jet actuators operating in a 140 ll/sec onset-llm_.
13
American Institute of Aeronautics and Astronautics
700 Hz 300 Hz
Plmee =0deg. Reference vector, Phase =0dee. Reference vector,
30 ft/sec -- -_o.5E ..... :: 30 ft/sec -
_O-5F:::I:
._0.4_-i _.:: ii:i:i: '_"i,!_i):_:_::i::_..
f_ .3 ,
.@DO. 1[t": ,! .... " "'' "i '' ":';
_-olL_
r "_ _
_- " _r ; i I i Siynthietic+ l_i't locations + i l L i + __- " _r i i I i Siynthetic, j#|t lociationIs i , [ _ I I I
-0.5 0 0.5 -0.5 0 05
Lateral distance from centerline, inches Lateral distance from centerline, inches
Reference vector, Reference vector,
Phase =90 (leg. 30 ft/sec -- Phase =90 (leg. 30 ft/sec --
_0.5 :::, ;: ::::: -8°5r
0.4 iii: :i: :iii :::_i!!_--i_!\::::::._: o+F ,: : :,:, : ::_'b'::: :
0.2
0 0 r + "_".i : .,,_: ;?;,
m OE-. L;,:,: ; .....
0 ..... .......
_-0 th_ _z3 " L__-E] _3
locations " I£ Synthetic jet locations .
-0.5 0 05 -0_5 0 0.5
Lateral distance from centedine, inches Lamral distance from centerline, inches
Reference vector,
8°5FPhase = 180 deft 30 fUsec -
_o4_ : ; ; :!i:i!: + ::
_=o_ !i_:i
gothl
: +i :++
-o.t
locations c: _ , _ I _ SiynthIetic_ jeIt loc_ation_s I ' ] ' I '
-0.5 0 0.5 -0.5 0 05
Lateral dimnce from centedine, inches Lateral distance from centerline, inches
p + Reference vector, Reference vector.
8 I_se =270 oe9. 30 ft/sec 8°_rPhase =270 deg. 30 ft/sec --
05.... _ %................., :: , , I_';,,£_ _%_X X,_,' ',......'%'.... , 'k I ¢:o4_-
_o._[-
_, o_-r
-otF--"
__-- ' [rE _ I , Siynthietici jg/t loicationis _ I , , , _c I, _ I Synthetic _t locations
-0.5 0 0.5 -0.5 0 0.5
Lateral distance from centerline, inches Lateral distance from centedine, inches
Figure 20. Phase-locked velocity vectors lor synthetic.jet actuators operating at 700 tlz and 300 Hz inzero onset flow.
15
American Institute of Aeronautics and Astronautics