thus laminar. The flow separation at the low and low. RC is simply referred to in the following as "laminar separation". The effect of excitation on this separation.
NASA Technical AIAA-89-0565
of "Laminar by Acoustic
K.B.M.Q. Zaman and D.J. Lewis Research Center Cleveland,
(hASA-_M-1C13IgJ ._tPAaATItN CVE_ hXCI2A_IC_
6_ IA_ISA_ BY _CC[_'IIC p
Prepared for the 27th Aerospace Sciences Meeting sponsored by the American Institute Reno, Nevada, January 9-12, 1989
CONTROL OF "LAMINAR
BY ACOUSTIC EXCITATION
K.B.M.Q. Zaman and D.J. McKinzle NASA Lewis Research Center C]eveland, Ohio 44135
The effect of acoustic excitation in reducing "laminar separation" over two-dimensional airfoJ]s at low angles of attack is Investigated experimentally. Alrfoils of two different cross sections, each with two dlfferent chord lengths, are studied in the chord Reynolds number range of 25 000 < Rc < 100 000. While keeping the amplitude of the excitation Induced velocity perturbation a constant, it is found that the most effective frequency scales as U_ 3/2 The parameter St/Rcl/2, correspondlng to the most effective f for all the cases studied, falls in the range gf 0.02 to 0.03, St being the Strouhal number based on the chord. Nomenclature angle C1
tunnel cross resonance frequency with m sound pressure nodes In y, and n sound pressure nodes in z
mean veloclties respectively
mean velocity measured hot wire approximating
ficiently low Rc, on the other hand, extensive separation on the suction side may take place even at low _. This is accompanied by a rapid deterioration of the airfoil performance with decreasing R C, approximately In the range RC < lO0 000. The low _ separation at R c = 40 000 is illustrated in Fig. 1 by visualization pictures taken from Ref. 5. Note that the flow on the upper surface is separated for all the lower _'s but has reattached at the highest _, presumably due to earlier transltion of the separated shear layer in that condition. Stability analysis, carried out in Ref. 5, indicated that the boundary layer prior to separation for the low _ cases must be stable and thus laminar. The flow separation at the low and low R C is simply referred to in the following as "laminar separation". The effect of excitation on this separation is the focus of the present study.
towards separation in the flow over an airfoil and thereby improve its performance. I-7 The separatlon process, and the effect of excitation thereupon, has been noted to be d_fferent depending on the ranges of the angle of attack and the Reynolds number. = While at all Rc the flow separates ultimately at large _ (poststall), an unsteady separation may occur around the static stall condition. 5,8 At suf-
with a sinqle (U2 + V2) I/2
rms velocity fluctuations directions; subscript r ues at reference locatlon spectrum
in x,y,z denotes val-
rms total and fundamental fluctuation in the dlrectlon of , as measured by a single hot wire
(u;2 _ v;2)1/2
there is small slnce the single hot wlre primarily senses the amplltude in the direction of the mean flow. As the leading edge of the airfoll is approached u' For 342 Hz becomes large, even larger than the amplitudes for the other two fp's. Downstream of the leading edge the amplitude variations show standing wave patterns, reminiscent of the acoustically excited boundary layer data of Ref. II. This occurs due to the interference of the excited instability wave and the exciting acoustic wave when the amplitudes due to the two are comparable. The wavelength of the standing wave should exactly equal the shorter hydrodynamic (Instability) wavelength (_). X for the three fp's were obtained from Fig. 14" X and fp provided the phase velocity of the instabllity wave. These quantities and the Strouhai number based on e at x'/c = 0.3 are listed below. fp _68 253 342
0.14 0.10 0.077
0.51 0.55 0.57
fpe/U= 0.008 0.012 0.016
Further downstream, one observes that the amplitude at 168 Hz grows to the largest value. This appears anomalous as, referring back to Fig. 8, 253 Hz is found to be the center Frequency in the band of effective fp's. Thus intuitively one would expect a larger amplitude g_owth at 253 Hz. The reason for this remalns unclear, but differences in the tunnel resonance condltlons could be a contributing factor. One also notes from Fig. 14 that the amplitudes rise sharply for a11 f#'s past the 50 percent chord location. Referr_ng back to the boundary layer profiles in Flg. 11, it is apparent that the amplificatlon of the imposed disturbance takes place in the separated shear layer. Conclusion Small amplitude acoustic excitation at an appropriate frequency can effectively reduce laminar separation occurring on the suction side of airfoils at low _ and low Rc. This results in a significant improvement in the lift coefficient. It is inferred From data wlth alrfoiIs of two crosssectional shapes, each with two different chords, that the optimum effect occurs when the parameter St/Rcl/2, corresponding to the excitation frequency, falls in the range of 0.02 to 0.03. Detailed flow field data recorded for a specific case, indicate that a separated region still exists under the excitation, and the amplification of the imposed perturbation takes place primarlly in the downstream shear layer rather than in the upstream boundary layer. References t.
The fundamental amplitude growth along the 70 percent velocity point was measured for three
x'/c = 0.2. Note that the reference amplitude (u_) was held constant at 0.5 percent of U,. However, at 342 Hz, only v' is induced upstream of the leading edge, u' being very small. Thus, the meas-
Collins, F.G. and Zelenevltz, J., "Influence of Sound Upon Separated Flow Over Wings," AIAA Journal, Vol. 13, NO. 3, Mar. 1975, pp. 408-410.
2. Carmichael, B.H., "Low Reynolds Number Airfoil Survey: Vol. I," NASA CR-165803, 1981.
3. Mueller,T.J. andBati]l, S.M.,"Experimental Studiesof Separation ona Two-Dimenslonal Airfoil at LowReynolds Numbers," AIAA Journal, Vol.
20, No. 4, Apr.
4. Ahuja, K.K. and 8urrin, R.H., "Control of Flow Separation by Sound," AIAA Paper 84-2298, Oct. 1984. 5. Zaman, K.B.M.Q., Bar-Sever, A., and Mangalam, S.M., "Effect of Acoustic Excitation on the Flow Over a Low-Re Airfoil," Journal of Fluid Mechanics, Vol. 182, Sept. 1987, pp. 127-148. 6.
Neuburger, D. and Wygnanski, I., "The Use of Vibrating Ribbon to Delay Separation on TwoD1menslonaI Airfoils: Some Preliminary Observations," Workshop on Unsteady Separated Flow, Air Force Academy, July ]987 (private communication).
7. Huang, L.S., Maestrel]o, L., and Bryant, T.D., "Separation Control Over an Airfoil at High Angles of Attack by Sound Emanating From the Surface," AIAA Paper 87-1261, June 1987.
Zaman, K.B.M.Q., McKinzie, D.J., and Rumsey, C.L., "A Natural Low Frequency Oscillation of the Flow Over an Airfoil Near Stalling Conditions," Journal of Fluid Mechanics, 1988 (submitted). (NASA TM-IO0213.) Goldstein, 1976,
lO. Baumeister, K.J., "Reverberation Effects on Directionality and Response of Stationary Monopole and Dipole Sources In a Wind Tunnel," Journal of Vlbration, Acoustics, Stress, and Reliabllity in Deslgn, VoI. 108, No. I, Jan. 1986, pp. 82-90. If. Leehey, P. and Shapiro, P., "Leading Edge Effect in Lamlnar Boundary Layer Excitation by Sound," Laminar Turbulent Transition, R. Eppler and H. Fasel, eds., Springer-Verlag, New YorK, 1979, pp. 321-331.
FIG. I - SMOKE-WIRE FLOW VISUALIZATION PICTURES FOR VARIOUS a FOR LRN AIRFOIL (c = 10.2 CM) AT
Rc = 4xi04, REF. .
2_.2:G_NAL PAGE OF POOR
T ± 76.2
3.8 D HOLE
/-- REF. MIC
TOP VIEW (2) ACOUSTIC DR]VER
FIG. 2(a) - PHOTOGRAPH OF WIND TUNNEL TEST SECTION. (b) - SCHEMATIC OF TEST SECTION; DIMENSIONS ARE IN CENTIMETERS.
>' ___--+____-+,o_o -.+i
,' I +] +-'----k\50 /..... ,,+ ,7 i _- "........
';'2 2.7 I
VERSUS R c.
.5 _llJIl_l_J_lI,Jlllllfl_,l 20 qO 60
Rc x 10-3 FIG. 4 - C1 VERSUS Rc AT a = 6o AIRFOILS; c = 12.7 CM.
FOR THE fgo
7......... ! t_,f
/_'\""I i /
fp, Hz FIG. 5 - REFERENCEAMPLITUDE PARN'IETERSVERSUS fp. HOT-WIREAT x = -11.q cM, y = 1.27 cM, AND z = O; MICROPHONE LOCATEDAS SHOWNIN FIG. 2(b). c = 12.7 cM LRN AIRFOIL AT _ = 8° WITH Rc = 50 000.
,i" " I
688 .... ;z_ "'-. i
............................. 342 1.2 _, 570 ,,
, o -3.o (b)
FIG. 6(a) - u' AND v' AMPLITUDESVERSUS y MEASURED AT x = -II.4 CM AND z = O, FOR INDICATED fp's. SAME FLOW AS IN FIG. 5. (b) - u' AND v' AMPLITUDESVERSUS z MEASUREDAT x = -11.4 CM AND y = 1.27 CM.
.- / 5zo
.... --7 -
¢ .... -
I i 100
I i 110
...._- /-7.--'/---------'-----253 .....................
1.0 ( u_)/Uoo,
AMPLITUDE EFFECT ON
SAME FLOW AS Ill
DATA OF FI6.
FUNCTION OF (u;).
CROSS PLOTTED AS A
' ' -_I _
Rc x I0-3
R c x I0-3
,.. ......... 8.o_.
50 / L_
St/R c x 2_rx 104 I
i '(b) ''_ ' I ' ' ' i I I I ' ' I I ' _ ' I '
Rc x 10-3 "".....
Rc x 10-3
-_.'-. .......... 6o_ .............
/Uoo= 0.005. LRN AIRFOIL_AT a = G0 , Rc = 50 000.
20 ..11 IIIIllllllll III x'/c = 0.3
/ _ I
III Ir / 0.7 I -
(U>/Uoo FIG. 11 - BOUNDARY LAYER PROFILES OF AT DIFFERENT X' FOR THE SAME FLOW AS IN FIG. 10. SOLID LINES FOR UNEXCITED FLOW, DASHED LINES FOR EXCITATION AT fp = 253 HZ AND (U_,)/Uoo= 0.0025. ABSCISSA APPLIES TO PAIR ON LEFT, OTHERS ARE SHIFTED TO TIIERIGHT SUCCESSIVELY BY ONE MAJOR DIVISION.
20 X'/C =
.OG FOR THE UNEXCITED FLOW; SHORT _SHED LINE• TOTAL