NACA0012 moving airfoil by Michael Alletto - Revision history

Jozsef Nagy at 07:00, 14 May 2020

2020-05-14T07:00:23Z

Jozsef Nagy at 06:59, 14 May 2020

2020-05-14T06:59:19Z

Jozsef Nagy at 06:58, 14 May 2020

2020-05-14T06:58:34Z

Jozsef Nagy at 06:56, 14 May 2020

2020-05-14T06:56:59Z

Jozsef Nagy at 06:52, 14 May 2020

2020-05-14T06:52:19Z

Jozsef Nagy: Created page with "* '''contributor''': Michael Alletto * '''affiliation''': WRD MS * '''contact''': click here for email address
2020-05-14T06:38:54Z

Created page with "* '''contributor''': Michael Alletto * '''affiliation''': WRD MS * '''contact''': <mail address='michael.alletto@gmx.de' description='author'>click here for email address</mai..."

New page
* '''contributor''': Michael Alletto
* '''affiliation''': WRD MS
* '''contact''': <mail address='michael.alletto@gmx.de' description='author'>click here for email address</mail>
* '''OpenFOAM versions''': v1912
* '''Published under''': CC BY-NC-ND license ([https://creativecommons.org/licenses creative commons licenses])

Go back to [https://wiki.openfoam.com/NACA0012_by_Michael_Alletto NACA0012 tutorial page].

=NACA0012 Moving airfoil=

Files can be downloaded [https://github.com/jnmlujnmlu/OpenFOAMTeaching/blob/master/MichaelAlletto/moving.tar.gz here].

@@ Line 30: / Line 30: @@
 The next figure shows the velocity magnitude for different angles of attack. It is evident that the shocks moves from the lower surface to the upper surface. For angles of attack of &alpha; = 1.09° and &alpha; = -1.25° the shock is weakest. In general it seams that the qualitative picture of the shock moving from the upper to the lower surface during the periodically pitching motion is well represented.
-[[File:us_alettonaca.png|250px|center|first row left: velocity magnitude [m] at &alpha; = -2.0°; first row right: velocity magnitude [m] at &alpha; = -0.54°; second row left: velocity magnitude [m] at &alpha; = 1.09°; second row right: velocity magnitude [m] at &alpha; = 2.34°; third row left: velocity magnitude [m] at &alpha; = 2.01°; third row right: velocity magnitude [m] at &alpha; = 0.52°; fourth row left: velocity magnitude [m] at &alpha; = -1.25°; fourth row left: velocity magnitude [m] at &alpha; = -2.41°; ]]
+[[File:us_alettonaca.png|550px|center|first row left: velocity magnitude [m] at &alpha; = -2.0°; first row right: velocity magnitude [m] at &alpha; = -0.54°; second row left: velocity magnitude [m] at &alpha; = 1.09°; second row right: velocity magnitude [m] at &alpha; = 2.34°; third row left: velocity magnitude [m] at &alpha; = 2.01°; third row right: velocity magnitude [m] at &alpha; = 0.52°; fourth row left: velocity magnitude [m] at &alpha; = -1.25°; fourth row left: velocity magnitude [m] at &alpha; = -2.41°; ]]
 ==References==

@@ Line 30: / Line 30: @@
 The next figure shows the velocity magnitude for different angles of attack. It is evident that the shocks moves from the lower surface to the upper surface. For angles of attack of &alpha; = 1.09° and &alpha; = -1.25° the shock is weakest. In general it seams that the qualitative picture of the shock moving from the upper to the lower surface during the periodically pitching motion is well represented.
-[[File:us_alettonaca.png|750px|center|first row left: velocity magnitude [m] at &alpha; = -2.0°; first row right: velocity magnitude [m] at &alpha; = -0.54°; second row left: velocity magnitude [m] at &alpha; = 1.09°; second row right: velocity magnitude [m] at &alpha; = 2.34°; third row left: velocity magnitude [m] at &alpha; = 2.01°; third row right: velocity magnitude [m] at &alpha; = 0.52°; fourth row left: velocity magnitude [m] at &alpha; = -1.25°; fourth row left: velocity magnitude [m] at &alpha; = -2.41°; ]]
+[[File:us_alettonaca.png|250px|center|first row left: velocity magnitude [m] at &alpha; = -2.0°; first row right: velocity magnitude [m] at &alpha; = -0.54°; second row left: velocity magnitude [m] at &alpha; = 1.09°; second row right: velocity magnitude [m] at &alpha; = 2.34°; third row left: velocity magnitude [m] at &alpha; = 2.01°; third row right: velocity magnitude [m] at &alpha; = 0.52°; fourth row left: velocity magnitude [m] at &alpha; = -1.25°; fourth row left: velocity magnitude [m] at &alpha; = -2.41°; ]]
 ==References==

@@ Line 30: / Line 30: @@
 The next figure shows the velocity magnitude for different angles of attack. It is evident that the shocks moves from the lower surface to the upper surface. For angles of attack of &alpha; = 1.09° and &alpha; = -1.25° the shock is weakest. In general it seams that the qualitative picture of the shock moving from the upper to the lower surface during the periodically pitching motion is well represented.
-[[File:us_alettonaca.png|750px|center|Cp (left), Cf (right)]]
+[[File:us_alettonaca.png|750px|center|first row left: velocity magnitude [m] at &alpha; = -2.0°; first row right: velocity magnitude [m] at &alpha; = -0.54°; second row left: velocity magnitude [m] at &alpha; = 1.09°; second row right: velocity magnitude [m] at &alpha; = 2.34°; third row left: velocity magnitude [m] at &alpha; = 2.01°; third row right: velocity magnitude [m] at &alpha; = 0.52°; fourth row left: velocity magnitude [m] at &alpha; = -1.25°; fourth row left: velocity magnitude [m] at &alpha; = -2.41°; ]]
 ==References==

@@ Line 30: / Line 30: @@
 The next figure shows the velocity magnitude for different angles of attack. It is evident that the shocks moves from the lower surface to the upper surface. For angles of attack of &alpha; = 1.09° and &alpha; = -1.25° the shock is weakest. In general it seams that the qualitative picture of the shock moving from the upper to the lower surface during the periodically pitching motion is well represented.
-[[File:OtherScheme_alettonaca.png|750px|center|Cp (left), Cf (right)]]
+[[File:us_alettonaca.png|750px|center|Cp (left), Cf (right)]]
 ==References==
-[1] Kazem Hejranfar, Hossein Hashemi Nasab, and Mohammad Hadi Azampour. Arbitrary lagrangian-eulerian unstructured �finite-volume lattice-boltzmann method for computing two-dimensional compressible inviscid flows over moving bodies. Physical Review E, 101(2):023308, 2020.
+[1] Kazem Hejranfar, Hossein Hashemi Nasab, and Mohammad Hadi Azampour. Arbitrary lagrangian-eulerian unstructured finite-volume lattice-boltzmann method for computing two-dimensional compressible inviscid flows over moving bodies. Physical Review E, 101(2):023308, 2020.
 [2] RH Landon. Naca 0012. oscillating and transient pitching. Data Set, 3, 1982.
 [3] Daniella E Raveh. Numerical study of an oscillating airfoil in transonic buffeting flows. AIAA journal, 47(3):505-515, 2009.

← Older revision		Revision as of 06:52, 14 May 2020
Line 10:		Line 10:

	Files can be downloaded [https://github.com/jnmlujnmlu/OpenFOAMTeaching/blob/master/MichaelAlletto/moving.tar.gz here].		Files can be downloaded [https://github.com/jnmlujnmlu/OpenFOAMTeaching/blob/master/MichaelAlletto/moving.tar.gz here].
		+
		+	==Moving Airfoil==
		+
		+	Having found a setup which gives reasonable results for different angles of attack and Mach numbers, now it is time to move to the periodically pitching airfoil. Also for the moving airfoil four different simulations are performed: Three simulation using the Spalart-Allmaras model. One of this used the grid2 from the grid independence study (labeled as Ma0.75omega42.5grid2), one with grid1 from the grid independence study (labeled as Ma0.75omega42.5grid1) and the last one using the first order upwind scheme instead of the second order linear scheme for the convective terms in the pressure equation (labeled as Ma0.75omega42.5grid11Orderp). The fourth simulation uses the k-omega SST model. The purpose was to show that the conclusions drawn regarding the influence of the grid, the turbulence model and of the discretization of the convective term in the pressure equation for the stationary airfoil can be applied also for the moving case.
		+
		+	The next figure shows the evolution of the lift (left) and of the moment coefficient about a quarter chord (right) over the airfoil angle α. The data are compared with the experiments of [2]. Note that they are extracted from the paper of [1] for convenience. It is evident that the lift coefficient agrees reasonable with the experiments during the nose down movement (going from a positive angle towards a negative angle in counter clockwise direction). During the nose up movement however, the phase lag between the angle and the lift force is more pronounced compared with the experiment leading to a much higher hysteresis. The elliptical form of the diagram is a result of the phase lag between the airfoil movement and the lift force. If both would be in phase the diagram would show a straight line. The maximum Cl is captured reasonably well. Similar to the fixed airfoil, only the case where the convective term in the pressure equation is discretized with a first order upwind scheme deviates from the other tree cases. It is evident that the hysteresis curve is narrower respect to the other cases.
		+
		+	[[File:moving_alettonaca.png\|750px\|center\|Cp (left), Cf (right)]]
		+
		+	Looking at the evolution of the moment coefficient about c/4 we see that at the maximum and minimum angle α is around zero in agreement with the experiment. This means that at this angles the center of pressure is at around c/4. What we can also observe is that the change of the moment coefficient with the angle α is too fast after the minimum or maximum of <math>\alpha</math>. That means that the movement of the shock position towards the trailing edge after the minimum or maximum angle of attack is too fast. Also for the moment coefficient major deviations among the cases simulated can be observed only for the first order upwind simulation. For this case we see a much more elliptical form of the hysteresis curve.
		+
		+	The next figure shows the distribution of the pressure coefficient over the airfoil for different angles of attack. The experimental values are from the measurements of [2]. Note that they are extracted from the paper of [1] for convenience. The agreement is reasonable for almost all angles except for α = 1.25° where the experiments predict the shock location at the upper surface of the airfoil while for the simulation the shock is still on the lower surface. This is in agreement with the bigger hysteresis of the lift coefficient compared with the experiments. As for the functional dependence of the lift coefficient of the angle of attack, also for the distribution of the pressure coefficient along the airfoil there are only small differences between the coarser and the finer grid and the simulation with the k-OmegaSST turbulence model. Remarkably is that for the case where a first order upwind scheme is used for the convection term in the pressure equation, the shock is completely smoothed out.
		+
		+	[[File:cps_alettonaca.png\|750px\|center\|first row left: Cp at α = -2.0°; first row right: Cp at α = -0.54°; second row left: Cp at α = 1.09°; second row right: Cp at α = 2.34°; third row left: Cp at α = 2.01°; third row right: Cp at α = 0.52°; fourth row left: Cp at α = -1.25°; fourth row left: Cp at α = -2.41°; ]]
		+
		+	Note that other simulations (see [1,3]) which resolved the wall (the first grid point lied at y+ =< 1) obtained a narrower hysteresis curve for the lift coefficient and a smoother transition of the moment coefficient from zero to its maximum or minimum value with the angle of attach. This may be a hint in which direction an interested reader could go to improve the results.
		+
		+
		+	The next figure shows the velocity magnitude for different angles of attack. It is evident that the shocks moves from the lower surface to the upper surface. For angles of attack of α = 1.09° and α = -1.25° the shock is weakest. In general it seams that the qualitative picture of the shock moving from the upper to the lower surface during the periodically pitching motion is well represented.
		+
		+	[[File:OtherScheme_alettonaca.png\|750px\|center\|Cp (left), Cf (right)]]
		+
		+	==References==
		+
		+	[1] Kazem Hejranfar, Hossein Hashemi Nasab, and Mohammad Hadi Azampour. Arbitrary lagrangian-eulerian unstructured �finite-volume lattice-boltzmann method for computing two-dimensional compressible inviscid flows over moving bodies. Physical Review E, 101(2):023308, 2020.
		+
		+	[2] RH Landon. Naca 0012. oscillating and transient pitching. Data Set, 3, 1982.
		+
		+	[3] Daniella E Raveh. Numerical study of an oscillating airfoil in transonic buffeting flows. AIAA journal, 47(3):505-515, 2009.