AD-A271 217 GE FOM PPRCVE[IEl~~ ~ ~~~~~~g 10j 1i Ill II11,1 t 'o er iooncia. rc.ong rig lime -or ov9wrng .rstruc:orS Seac', si

gatrieri. IQi cal liiH lE H I olcjon oI ntormation Sena comments regardinrg !ms owren gsi~mai (or axsin g rerw ! v'as~nrglon He ~iro- Services. Directorato for intormaiion OuoratIorrs arno Rboorrls. '215 rrearsorr

Bans ~ Managemnte'r anda BUCger PaoenliOrK Reduction P'Oleci ,O?04-3188f. Waosr.,ng. DC 205034AGNY USE ONLY iLaave blank) REOAT DAE 3 REPORT TYPE AND DATES COVERED

4TITLE AND SUBTITLE 5 FUDIG UMERVelocity ad Vorticity Distributions Over an Oscillating Airfoil UnderCompressible Dynamic Stall

7 PERFORMING ORGANIZATION NAMELSI AND ADDRESS, ESI 8 PERFORMING ORGANIZATION-av NA-' ic z ns7:Z.. REPORT NUMBERl

9 SPONSORING,MONITORING AGENCY NAME( S) AND ACDRESS( ES) 10 SPONSO~RIN., MCNITORiNGAGENCY REPORT NUMBER

U. S. Army Research OfficeP O. Box 12211Researcn Triangle Park. NC 277109-221 1

!I SUPPLEMENTARY NOTESThe view, opinions and/or findings contained in this report are those of theauthor(s) and should not be construed as an official Department of the Armposition, policy, or decision, unless so designated by other documentatD'J}I

12a. DIST RIBUTION/AVAILABILITY STATEMENT 121b. At 'f IoApproved for public release: distribution unlimited. Ib C F ' 0,199

13 ABSTRACT (Maximum 200 words)#

The velocity and vorticity fields around an oscillating airfoil in compressible dynamicstall are reported. Phase averaged, two component laser velocimetry data were obtainedat a freestream Mach number of 0.3 and a reduced frequency of 0.05. This is the first setof velocity data available at a high Reynolds number (540,000) undercompressible flow conditions and it serves as a good database for development andvalidation of co~mputer codes. Of particular interest is the formation of a separationbubble, which bursts coincidentally with the formation of the dynamic stall vortex,adding an extra degree of physical complexity to the problem. 93-2462 1

14 SUBJECT TERMS 15. NUMBEH 01,- HAUbb

Dynamic Stall, LDV Measurements, Compressibility Effects, 2Vorticity Field 16. PRICE CODE

17 SECURITY CLASSIFICATION 18. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACTOF REPORT OF THIS PAGE OF ABSTRACT

UNCLASSIFIED UNCLASSIFIED UNCLASSIFIED ULStandard Form 298 (Rev. 2-89)

N5N Prescribed by ANSI Sid. 239-187540-01 -280-5500 298-102

S* , - ---

,I

II0

Velocity and Vorticity Distributions Over anOscillating Airfoil Under CompressibleDynamic StallM. S. Chandrasekhara and S. Ahmed

0

ij !

SI -

I I

Reprinted fromAIMA Journal'Vn'ume 31. Number 6. June 1993. Pages 995-996

A publication of theAmerican Institute of Aeronautics and Astronautics. IncThe Aerospace Center. 370 L'Enfant Promenade. SWWashmngton. DC 20024-2518

.. .

Vol. 31, No. 6, June 1941

Velocity and Vorticity Distributions over an Oscillating AirfoilUnder Compressible Dynamic Stall

M. S. Chandrasekhara*Naval Postgraduate School, Monterey, California 93943

andS. Ahmed+

.MCA T Institute, San Jose, California 95127 7

Abstract A 7.62-cm-chord NACA 0012 airfoil was oscillated sinuT HE velocity and vorticity fields around an oscillating soidally with its angle of attack varying as, a = 10 deg-10 deg

airfoil in compressible dynamic stall are reported. Phase sin wt at a reduced frequency [k = (sfc C'_] of 0.05 in a flowaveraged, two component laser velocimetry data were ob- with a freestream Mach number of 0.3. The details of the windtamed at a freestream Mach number of 0.3 and a reduced tunnel used are given in Ref. I. A standard two component (I Lfrequency of 0.05. This is the first set of velocity data available and V) frequency shifted laser velocimeter was used for theat a high Reynolds number (540,000) under compressible flow measurements. Optical encoders mounted on the drive pro-conditions and it serves as a good database for development vided the pertinent time-dependent information. A novel ap-and validation of computer codes. Of particular interest is the proach of freezing the encoder data when coincident U and Iformation of a separation bubble, which bursts coincidentally velocity samples were present served to provide reliable phasewith the formation of the dynamic stall vortex, adding an angle data, which were later ensemble averaged by sorting inoextra degree of physical complexity to the problem. bins covering 3 deg in phase angle. The velocities were com-

puted for bins containing at least 50 samples. When this SContents condition was not satisfied, suitable interpolation methods

The performance of helicopters and the maneuverability of were used. if data were found in the neighboring bins. One tmfixed wing aircraft is limited by dynamic stall and the gross polystyrene latex particles suspended in alcohol were used forseparation of the unsteady flow that accompanies it. The seeding the flow. Further documentation of the experimentaldynamic stall flowfield is a complicated combination of a details can be found in Ref. 2.multitude of fluid dynamic effects that include tremendous Velocity Measurements in the Separation Bubblefluid acceleraiioua aruuid zh, lcaia, ed.- formation ofstrong suction peaks; development of the local boundary layer Some interesting flow features can be seen in the variationunder strongly adverse pressure gradients; transition of the of the horizontal component of velocity U with phase angle o.leading-edge laminar boundary layer, its separation, and reat- in Fig. 1. for different heights (y/c) at the streamwise locationtachment resulting in a separation bubble that grows and x/c =0.083. At y/c = 0.067, phase angle of 160 deg, a =6.6eventually bursts just when the dynamic stall vortex forms; deg, the velocity drops rapidly as a separation bubble formsunder compressibility conditions (M>:0.3), formation of over the airfoil. (It should be noted that the flat portions ofshock(s) and the induced separation due to it; addition of large the distributions for y/c =0.067 and vlc = 0.083 are causedamounts of coherent vorticity into the flow and its coalescenceinto the dynamic stall vortex that later convects over the airfoil A Y/C = 0.200 Y'C = 0.150 X Y'C = 0.100upper surface and interacts with the trailing edge separated + Y/C = 0.183 V Y/C = 0.133 0 Y'C = 0.083flow that propagates toward the leading edge; and so on. The x YC = 0.167 H Y'C = DA17 9 YC = 0.067complex interactions governing the physical problem have 4.0hitherto made understanding the origin of dynamic stall andattempts at modeling the flow very challenging. The studybeing reported provides a comprehensive set of velocity data 3.0 0........... . ....... ." didAA " ........... . .... .along with vorticity distributions derived from it, which can be ++ ++Qused to model the flow. In addition, it also serves as a data-.base to verify computational results and enables development 2.5 .......of new codes.

p 2.0 ..

Presented as Paper 91-1799 at the AIAA 22nd Fluid Dynamics, Plasmadynarnics. and Lasers Conference, Honolulu, HI. June 24-26,a1991; received Sept. 20, 1991; synoptic received Sept. 25, 1992; ac- 1.5 .. ............................ ......cepted for publication Nov. 2. 1992. Full paper available from AIAALibrary. 555 West 57th St.. New York. NY 10019. Copyright cc 1991by the American Institute of Aeronautics and Astronautics, Inc. No ,0 ...........copyright is asserted in the United States under Title 17, U.S. Code.The U.S. Government has a royalty-free license to exercise all rightsunder the copyright claimed herein for Governmental purposes. All 0.5 ............................... 0other rights are reserved by the copyright owner.

*Associate Director and Research Associate Professor, Navy-NASA Joint Institute of Aeronautics, Department of Aeronautics and 0140 180 220 260Astronautics, Mailing Address: M.S. 260-1. NASA Ames ResearchCenter, Moffett Field, CA 94035-1000. Associate Fellow AIAA, Phase angle

tResearch Scientist; on leave from The National Aeronautical Lab- Fig. I U component velocity measurements at x/c = 0.083 (velocityoratory, Bangalore, India. offset by 0.2 (U/U..) for each successive y/c location).

995 0

S... .. .. ... . . .. . . B ... ... .. t . . . .. ... . 1 . . , I . .... ab -,a

996 CHANDRASEKHAILA AND AHMED AIRFOIL COMPRESSIBLE DYNAMIC STALL

.45 by the blockage of the beams close to the airfoil.) Since thebubble is encountered later in the cycle at a higher y/c point,

zi .: the phase angle at which the drop occurs increases with yic.3-0 However, at y/c values outside the bubble, the velocity distri--- 0butions are parallel to each other. This leads to the conclusion

that the bubble height is 0.03 c-0.04 c above the airfoil uppersurface.

.15 6 An analysis of the corresponding vertical component ofvelocity V showed rapid increases 'n the velocity at o = 200

3- deg, a = 13.4 deg. This is associated with bursting of the2 separation bubble. It is worthwhile mentioning that the bubble

0 bursting is somewhat gradual and not as abrupt as is normallyperceived.

Vurticily Distributions-. 15 The z component of vorticity was calculated from the mea-

sured U and V components of velocity by first fitting a cubicspline curve to the data and interpolating the velocities in a

-. 30' grid at a resolution of 1.25 min, using a second-order central-. 30 -. 15 0 .15 .30 .45 differencing scheme. Thus, the noise level in the distributions

X/C is expected to be high at about 20% of the local maximum 0vorticity values (in both the positive and negative quantities).The following discussion about the vorticity field should still

.45 be valid, especially before the dynamic stall vortex begins to[ t v,, convect because no discontinuities such as shocks were en-

-- -,Lcountered within the measurement grid. Thus, the picture of- &= the flowfield is also quantitatively valid up to the point where

6 =a the particles were able to follow the flow properly.0M Figure 2a shows that at 0 = 171 deg (ot = 8.44 deg), a region

.2m of clockwise vorticity has developed over the airfoil, justaround the location of the separation bubble, with a peak

W f vorticity of - 8 units in it. A region of counter-clockwise7 66 vorticity could also be found above it, but the peak vorticity

\5. within it is only about 5 units. As the airfoil reaches an angle2 3of attack of 10 deg (Fig. 2b) the clockwise vorticity has in-

creased to - I1 units, whereas the anticlockwise vorticity is --05 still at 5 units. The extent of the vortical region has grown to

about 2507o chord in both the x and y directions. As the airfoilpitches to higher angles of attack, the vorticity should increasesteadily until the vortex begins to convect. Figure 2c showsthat at 0 = 198 deg, this is the case as the clockwise vorticityhas doubled to - 22 units, but the anticlockwise vorticity has

. only increased to about 10 units. Earlier experiments 3 have --.30 -.05 .0.45 shown that the vortex begins to convect at around this phase

b)IC angle. The separation bubble also bursts around the sameb) angle of attack. Thus, a combined effect is felt by the airfoil,

which should be seen in its vorticity field. A calculation of the.45 circulation 2 over the measurement region showed an increase

C Vuntil the stall vortex convection was initiated and dropped-2 slightly beyond this angle of attack.

2 'SIM The study leads to the following conclusions.3 --15IA

- 0 . 1) One of the salient features of the flow is the formation ofs --sin

- O.M a separation bubble. This bubble bursts (opens up) just.20 around the angie of attack at which and the location where the

-- 0-d dynamic stall vortex forms, complicating the flow physics.

2) The clockwise vorticity was found to increase in the flowuntil the vortex begins to convect.

2 '3 Acknowledgment

.05 This research was supported by the Army Research OfficeGrant (MIPR-ARO-132-90) to the Naval Postgraduate Schooland was monitored by Thomas L. Doligalski.

References'Carr, L. W., and Chandrasekhara, M. S., "Design and Develop-

ment of a Compressible Dynamic Stall Facility," Journal of Aircraft,.30 Vol. 29, No. 3, 1992, pp. 314-318.

-. 30 -05 .20 .45 2Chandrasekhara, M. S., and Ahrued, S., "Laser Velocimetry Mea-) X/C surements of Oscillating Airfoil Dynamic Stall Flow Field," AIAAPaper 91-1799, June 1991.

Fig. 2 -"wtumr of .ormglized z component of vouliy: a) 0 = 171 3Chandrasekhara, M. S., and Carr, L. W., "Flow Visualizationdeg, a - 8.44 deg, I) 0 = 130 deg, a = 10.0 deg, and c) = 198 degt, Studies of the Mach Number Effects on the Dynamic Stall of Oscillat-a= 1-3.69 deg. ing Airfoils," Journal ofAircraft, Vol. 27, No. 6, 1990, pp. 516-522.