Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Investigating flow stability, turbulence and flow control using PIV Julio Soria
[email protected]
Laboratory for Turbulence Research in Aerospace and Combustion Department of Mechanical and Aerospace Engineering Monash University (Clayton Campus) Melbourne, VIC Australia
1
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Acknowledgments (co-authors ...) J. Arnaud C.H. Atkinson F. Bertellino J. Cater S. Coudert K. von Ellenrieder J.M. Foucaut M. Gordon S. Herpin D.R. Honnery J. Kostas T.T. Lim T.H. New N.R. Panchapakesan
T.K. Sengupta M. Sheridan I. Shephard M. Stanislas S. Tomar A. Tuck M.C. Welsh J. Wu and some more ...
2
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Acknowledgments (cont.) CSIRO Australian Research Council (Discovery and LIEF grants) AFOSR/AOARD
3
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Outline Study of shear layer instabilities lead to multigrid/ iterative cross-correlation PIV methodology Studying ZNMF jets with application to separation control Turbulent wall-bounded flow studies using SPIV and TPIV Comments ...
4
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows circular cylinders and blunt flat plates
Re = 2700, Wu et al. ETFS (1996)
3D Spanwise Structures
Re = 900, Soria et al. JWindEngIndAero (1993)
5
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows circular cylinders and blunt flat plates
shear flow measurements require high spatial resolution high magnification at the time fixed frame speed video recording with high magnification resulted in large particle displacements in the free-stream requirement for: different size IW in different areas of the flow
Iterative Adaptive Multigrid CCDPIV to measure these flows - Soria, Int. Colloquium on Jets, Wakes and Shear Layers, Melbourne, Australia (1994), Soria ETFS (1996), Soria ACLDFMC (1996) 6
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows Velocity Gradient and Vorticity Measurement aim of increased spatial resolution Multigrid
CCDPIV
spatial resolution is also of paramount importance in deduction of Velocity Gradients and Vorticity from PIV Fouras & Soria EiF (1998) investigated transmission of PIV experimental uncertainty into uncertainty in vorticity for I. finite difference vorticity calculation method (FD), II.
adpative Gaussian window smoothing of the velocity field with finite difference vorticity calculation method (AGW-FD) and III. local 2 velocity fit with analytic differentiation vorticity calculation method (2 x ) analytical expressions for uncertainty in PIV uncertainty in vorticity bias error due to spatial resolution, e.g. vortex ring Re = 2000 (Cater et al. JFM (2004)) y
8
egamI VIP noitisiuqcA noigeR
p
1
3 mm 005
= 32 px = 48 px = 64 px = 96 px SWS = 64px, = 64 px Oseen Vortex
x
WEIV EDIS 7
6
9
4
z
01
/max
mm 0011
1.0 0.8 0.6
mm 006
5
2
WEIV NALP
mm 005
x
0.4
1
0.2 0.0 -3.0
-2.0
-1.0
0.0
(x-x0)/L
1.0
2.0
= 32 px = 48 px = 64 px = 96 px SWS = 64px, = 64 px Oseen Vortex
/max
bias error can be more important than experimental uncertainty!
1.0 0.8 0.6 0.4 0.2 0.0 -3.0
-2.0
-1.0
0.0
1.0
2.0
(y-y0)/L
7
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows Velocity Gradient and Vorticity Measurement mathematical analysis of the velocity gradient calculation methods (Soria in “Lecture Notes on Turbulence and Coherent Structures in Fluids, Plasmas and Nonlinear Media”, World Scientific Publishing (2006)) showed that: for all numerical differentiation methods except 2 9 and FD significant attenuation of information and/or noise and phase error occurs if
0
10
k’x eff
10
92 and FD 132 212 , k’y = 0 212 , k’y = 0.25 k’x 212 , k’y = 0.50 k’x 212 , k’y = 0.75 k’x 212 , k’y = k’x AGW9FD, k’y = 0 AGW9FD, k’y = 0.25 k’x AGW9FD, k’y = 0.50 k’x AGW9FD, k’y = 0.75 k’x AGW9FD, k’y = k’x
1
⇒
kx
> 0.1
Δ
≤
λ 20
k
≡
kΔ π
Δ
=
IW sampling distance
λ
=
wavelength
where
10
2
10
2
10
1
k’x
10
0
8
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows MCCDPIV Applications ...
Circular Cylinder Flow Red = 769 (Soria ETFS (1996)) Flow
Flow 9
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows MCCDPIV Applications ...
Blunt Flat Plate (flow control) Reh = 1000 (Panchapakesan & Soria (2000))
ZNMF Jet Forcing
10
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows MCCDPIV Applications ...
accelerated flow over an airfoil at high angles of attack, Rec = 8,000 (Soria et al. ETFS (2003), Sengupta et al. JFM (2007)) a = 50 mm/s2
a = 100 mm/s2
α = 300
11
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Shear Layer Instabilities in Bluff Body Flows MCCDPIV Applications ...
accelerated flow over an airfoil at high angles of attack, Rec = 8,000 (Soria et al. ETFS (2003), Sengupta et al. JFM (2007)) a = 50 mm/s2
a = 100 mm/s2
α = 300
11
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Studying Zero-Net-Mass-Flow (ZNMF) jets with application to separation control ...
round free ZNMF jet (Cater & Soria JFM 2002) Laminar Rings Re = 104 @ x = 60Do o
Turbulent Jets Transitional Jets
Re = 3,336 St = 0.03
St = 0 St = 0.0015
- - - Gaussian Hussein et al. 1994
Re = 7,784 St = 0.012
Re = 104 o
Re = 104 St = 0.0015 Re = 11,121 St = 0.009
Re = 66,991 St = 0.009
St = 0 St = 0.0015
theoretical gradient for a top-hat initial velocity profile - Bremhorst & Hollis (1990), fully pulsed jets x = 60Do o
Re = 104, St = 0.0015 Re = 103, St = 0.0072
Re = 104 St = 0 12
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Studying Zero-Net-Mass-Flow (ZNMF) jets with application to separation control ...
phased-averaged vorticity Re = 103, St = 0.0072 t=0
t = T0/2
t = T0/8
t = 5T0/8
Stationary Vortex Ring t = T0/4
t = 3T0/4
Spreading Rate Increases
t = 3T0/8
t = 7T0/8
Mean vorticity
13
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Studying Zero-Net-Mass-Flow (ZNMF) jets with application to separation control ...
ZNMF Jets in Cross-flow (ZNMF-JICF) (Gordon & Soria EiF (2002), Tomar & Soria PIV03 (2003), Gordon et al. PoF (2004)) Do = 10 mm, AR = 1, VR = 3, St = 0.56
Do = 10 mm, AR = 1, VR = 3, St = 0.08
Do = 10 mm, AR = 3, VR = 3, St = 0.56
Do = 10 mm, AR = 3, VR = 3, St = 0.08
Elliptic ZNMF-JICF 14
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Studying Zero-Net-Mass-Flow (ZNMF) jets with application to separation control ...
ZNMF Jets in Cross-flow (ZNMF-JICF) (Gordon & Soria EiF (2002), Tomar & Soria PIV03 (2003), Gordon et al. PoF (2004)) Do = 10 mm, AR = 1, VR = 3, St = 0.56
Do = 10 mm, AR = 1, VR = 3, St = 0.08
Do = 10 mm, AR = 3, VR = 3, St = 0.56
Do = 10 mm, AR = 3, VR = 3, St = 0.08
Elliptic ZNMF-JICF 14
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Studying Zero-Net-Mass-Flow (ZNMF) jets with application to separation control ...
mean structure of AR = 1 ZNMF-JICF
two types of structures of AR=1 ZNMF-JICF
0.56
VR = 2
VR = 3
VR = 4
VR = 5
Freestream
St
0.08
15
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Studying Zero-Net-Mass-Flow (ZNMF) jets with application to separation control ...
Freestream
MCCDPIV at Re = 1066, R = 3, St = 0.56
MCCDPIV at Re = 1066, R = 3, St = 0.19
√
√
U2 + V 2 Uj
U2 + V 2 Uj
16
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Studying Zero-Net-Mass-Flow (ZNMF) jets with application to separation control ...
Freestream
MCCDPIV at Re = 1066, R = 3, St = 0.56
MCCDPIV at Re = 1066, R = 3, St = 0.19
√
√
U2 + V 2 Uj
U2 + V 2 Uj
16
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Application of ZNMF jets Separation control of NACA 0015 airfoil, Rec = 3.1 x 104 (Tuck & Soria AEAT (2008)) Uncontrolled Airfoil = 18o
Controlled Airfoil = 18o F+ = 1.3 cμ = 0.14% VR = 0.67
uncontrolled
F+ = 1.3 c = 0.14% VR = 0.67
anomaly of averaging over 4 phases only!
μ
17
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Application of ZNMF jets Separation control of NACA 0015 airfoil ... Extended to LES and Biglobal stability analysis: Kitsios et al. CTR Summer School Proc. Stanford (2006), Kitsios et al. JCP (2009)
18
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Application of ZNMF jets Separation control of NACA 0015 airfoil ... Extended to LES and Biglobal stability analysis: Kitsios et al. CTR Summer School Proc. Stanford (2006), Kitsios et al. JCP (2009)
18
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow interested in studying the topology and dynamics of coherent structures in wall-bounded flows Re = 1000 DA Iso-surface Flores et al. JFM (2005)
Focal Structures with DA = 1 - 10 Chong et al. JFM (1998)
Re = 395 DA Iso-surface, RA < 0 Blackburn et al. JFM (1996)
19
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow interested in studying the topology and dynamics of coherent structures in wall-bounded flows Re = 1000 DA Iso-surface Flores et al. JFM (2005)
Focal Structures with DA = 1 - 10 Chong et al. JFM (1998)
Re = 395 DA Iso-surface, RA < 0 Blackburn et al. JFM (1996)
19
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow contrast DNS and PIV experiments ...
PIV Reynolds number: Type of flow:
HRNBLWT (University of Melbourne):
DNS Limited by CPU power :
Re = 2 x 104
Re < 1.9 x 103 (Del Alamo (2006))
Boundary Layer
Boundary Layer
Channel flow
Channel flow
Pipe flow
Spatial dimensional order:
Well established: 2D (SPIV)
Spatial dynamic range:
Limited by : - size of CCD array
Getting there:
- size of IW
3D
3D (TPIV)
All relevant scales must be resolved
- noise
20
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow - SPIV high spatial resolution SPIV (Herpin et al. EiF (2008), Herpin PhD thesis (2009)) Flow characteristics:
Top-view of the setup: Flow
LTRAC water-tunnel
Data characteristics:
z
y x
XY plane of a turbulent BL Re = 2200; + = 1360 Wstr = 205
[Sx; Sy] = [2.6; 0.75] IW: 13+ Vector spacing: 5+ (64% overlap of the IW)
1
2
SDRth=125 3
4
SDR/Wstr = 0.6 - Four times higher than previous PIV experiments
large SDR achieved due to: large CCD array: 4008 x 2672 px2 (PCO.4000) experimental procedure designed to minimize sources of noise
21
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow - SPIV Snapshot of stereo-PIV data set ... Mesh points : Nx x Ny = 661 x 187 = 123 607 points per velocity field 2000 statistically independent velocity fields
Scaling in wall units
Longitudinal spectra of u’ at y+ = 100 using inner scaling Perry & Chong JFM (1986)
22
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow - TPIV Need for 3C-3D velocity field measurements:
Is 2C-2D or 3C-2D data sufficient to interpret 3D flow topologies unambiguously? Perry & Chong ASR (1994): “In order to know fully the topology of a flow pattern about a critical point it is of great importance to have full three-dimensional information.” e.g. misinterpretation of Burger’s vortex Limit Cycle
SF/S
(a) Sectional velocity field and integrated streamline pattern around a Burgers vortex cut by a sectioning plane through the origin x3 = 0, where x3 is in the direction of the vorticity. The planes intersect at 60°. (b) The sectioning plane now intersects the vortex away from the origin, i.e. x3 0. After Steiner and Perry (1987).
UF/C
NEED 3C-3D experimental data! TOMO-PIV (Elsinga et al. EiF (2006)) or TOMO-DHPIV (Soria & Atkinson MST (2008)) OF TBL 23
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow - TPIV MLOS-SMART TPIV in LML HRBLWT MLOS-SMART TPIV (Atkinson & Soria EiF (2009), Atkinsion et al. PIV09 (2009)) Reconstruction time (seeding density 0.03 ppp) MART: 120 min / volume 1 CPU MLOS-SMART: 11 min / volume 1 CPU TPIV experiments conducted in LML HRBLWT Two volume thicknesses were examined: 3 mm & 8 mm 1 x 2 m2 cross-section 18 m downstream of initial boundary layer development zone Vertical mirror Tunnel floor Slit
y
U
Re
3 m/s
7800
0.33 m 0.115 m/s
Volume
CCD arrays
/ u
+
125 μm
2650
z
x 45 deg mirror Cylindrical Diverging Lens
u
4 Hamam Hamamatsu 2048 x 2048 px CCD arrays 105 mm M Micro Nikkor lenses Scheimpflug condition about the camera’s vertical axis Scheimpfl
Four CCD array cameras
Lens
CCD
Tunnel floor Calibration grid
Light-sheet parallel to the tunnel floor 300 mJ Nd:YAG laser 12-bit 2048x2048 px 105 mm Micro Nikkor lenses Mirror Improve light intensity
45 deg mirror
LASER
105 mm Scheimpflug Micro Nikkor adapter Lens Hamamatsu camera
Reduce scattering bias Flow seeding Poly-ethylene-glycol smoke generator 1 μm
24
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow - TPIV Instantaneous Velocity Field ... Re = 7800, thick Mean velocity subtracted Unfiltered velocity IV = 643 px3, 75% overlap
x+ ~ 330+
y x
Plane y+ ~ 25 u’ - contours z
z+ ~ 330+ y+ ~ 50+
Low speed streak Vortices from Q-criteria
IW W+ ~ 26 x+ ~ 5
25
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow - TPIV Instantaneous Velocity Field ... Re = 7800, thick Mean velocity subtracted 3x3x3 Gaussian Unfiltered velocity IV = 643 px3, 75% overlap
x+ ~ 330+
y x
Plane y+ ~ 25 u’ - contours z
z+ ~ 330+ y+ ~ 50+
Low speed streak Vortices from Q-criteria
IW W+ ~ 26 x+ ~ 5
26
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Turbulent wall-bounded flow - TPIV Statistics Re = 7800 ...
Averaged over 500000 vectors
100 velocity fields streamwise and spanwise p homogeneous g directions
27
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
Comments ... 2C-2D PIV (time-series) has proved a useful tool in flow instability studies 3C-2D SPIV (time-series) has provided useful data of turbulence statistics 3C-3D TPIV (time-series) is getting there .... in the unambiguous study of the structure (topology) and evolution of coherent structures in turbulent flow Tomographic digital HPIV may in the long term be the 3C-3D technique of choice once appropriate sensors become available ... 28
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
References Wu, J., Sheridan, J. Hourigan, K. & Soria, J. 1996 Shear layer vortices and longitudinal vortices in the wake of a circular cylinder. Experimental Thermal and Fluid Science 12, 169 - 174. Soria, J. Sheridan, M. & Wu, J. 1993 Spatial evolution of the separated shear layer from a square leading-edge at plate. Journal of Wind Engineering and Industrial Aerodynamics 49, 237 - 246. Soria, J. 1994 Digital cross-correlation particle image velocimetry measurements in the near wake of a circular cylinder. International Colloquium on Jets, Wakes and Shear Layers, 25.1 - 25.8. Melbourne, Australia. Soria 1996 An investigation of the near wake of a circular cylinder using a video-based digital cross-correlation particle image velocimetry technique. Experimental Thermal and Fluid Science 12, 221 - 233. Soria, J. 1996 An adaptive cross-correlation digital PIV technique for unsteady flow investigations. In 1st Australian Conference on Laser Diagnostics in Fluid Mechanics and Combustion, 29 - 48. Dec., Sydney, Australia. Fouras, A. & Soria, J. 1998 Accuracy of out-of-plane vorticity measurements using in plane velocity vector field data. Experiments in Fluids 25, 409 - 430. Cater J., Soria, J. & Lim, T.T. 2004 The interaction of the piston vortex with a piston-generated vortex ring. Journal of Fluid Mechanics 499, 327-343. Soria, J. 2006 Particle Image Velocimetry - Application to Turbulence Studies. In LEC-TURE NOTES ON TURBULENCE AND COHERENT STRUCTURES IN FLUIDS, PLASMAS AND NONLINEAR MEDIA (ed. Michael Shats & Horst Punzmann).World Scientic Publishing, 309 - 348. Panchapakesan, N.R., & Soria, J. 2000 The structure of a locally excited shear layer in a blunt leading edge separated flow. In 7th Australasian Heat and Mass Transfer Conference, Townsville, July 3 - 6. Soria, J, New, T.H., Lim, T.T. & Parker, K. 2003 Multigrid CCDPIV Measurements of Accelerated Flow Past an Airfoil at an Angle of Attack of 30o. Experimental Thermal and Fluid Science 27, 667-676.
29
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
References Sengupta, T. Lim, T. T., Sajjan, S., Ganesh, S. and Soria, J. 2007 Accelerated flow past a symmetric aerofoil: experiments and computations. Journal of Fluid Mechanics 591, 255 - 288. Gordon, M. & Soria, J. 2002 PIV Measurements of a Zero-Net-Mass Flux Jet in Cross-flow. Experiments in Fluids 33, 863-872. Soria, J. & Tomar, S. 2003 MCCDPIV measurements of ZNMF jet in cross-flow. In 5th International Symposium on Particle Image Velocimetry, September 22-24, Busan, Korea. Gordon, M., Cater, J.E. & Soria, J. 2004 Investigation of the Mean Passive Scalar Field in Zero-Net-Mass-Flux in Cross-flow using PLIF. Physics of Fluids 16(3), 794-808. Tuck, A and Soria, J 2008 Separation control on a NACA 0015 airfoil using a 2D micro ZNMF jet. Aircraft Engineering and Aerospace Technology 80(2), 175 -180. V. Kitsios, R.B. Kotapati, R. Mittal, A. Ooi, J.Soria and D. You 2006 Numerical simulation of lift enhancement on a NACA 0015 airfoil using ZNMF jets. In Studying Turbulence Using Numerical Simulation Database XI, Proceedings of the 2006 Summer Program (ed. Parviz Moin and Nagi N. Mansour). Center for Turbulence Research, Stanford University, 457 - 468. Kitsios, V, Rodríguez, D., Theofilis, V., Ooi, A., Soria, J. 2009 2009 BiGlobal stability analysis in curvilinear coordinates of massively separated lifting bodies. Journal of Computational Physics 228, 7181–7196. Flores. O & Jimenez, J 2006 Effect of wall-boundary disturbances on turbulent channel flows. Journal of Fluid Mechanics 566, 357 - 376. Chong, M.S., Soria, J., Perry, A.E., Chacin, J., Cantwell B.J. and Na, Y. 1998 A study of the turbulence structures of wall-bounded shear flows using DNS data. Journal of Fluid Mechanics 357, 225 - 248. Blackburn, HM; Mansour, NN; Cantwell, BJ 1996 Topology of fine-scale motions in turbulent channel flow. Journal of Fluid Mechanics 310, 269 - 292.
30
Symposium 25 Years of PIV in Aerodynamics, Göttingen, September 23 - 25, 2009
References Herpin S., Wong, C.Y., Stanislas M. and Soria J. 2008 Stereoscopic PIV measurements of a turbulent boundary layer with a large spatial dynamic range. Experiments in Fluids 45, 745 - 763. Perry, AE; Chong, MS 1994 Topology of flow patterns in vortex motions and turbulence. Applied Scientific Research 53 (3-4), 357 - 374. Elsinga, G. E., Scarano, F., Wieneke, B., van Oudheusden, B. W. 2006 Tomographic particle image velocimetry. Experiments in Fluids 41, 933 - 947. Soria J. and Atkinson C. 2008 Towards 3C-3D digital holographic fluid velocity vector field measurement tomographic digital holographic PIV (Tomo-HPIV). Measurement Science and Technology 19(7), 1 - 12. Atkinson C., Soria, J. 2009 An efficient simultaneous reconstruction technique for tomographic particle image velocimetry. Experiments in Fluids, DOI 10.1007/s00348-009-0728-0. Atkinson, C., Coudert, S., Foucaut, J.M., Stanislas, M. and Soria, J. 2009 Thick and Thin Volume Measurements of a Turbulent Boundary Layer using Tomographic Particle Image Velocimetry. 8TH INTERNATIONAL SYMPOSIUM ON PARTICLE IMAGE VELOCIMETRY - PIV09 Melbourne, Victoria, Australia, August 25-28.
31