I am simulating an airfoil under the effect of sinusoidal gust having the following profile (Vg=kg*sin(omega*t)) and at the same time i am pitching the airfoil using a circular domain around the airfoil and using a sliding mesh (alpha-pitch=ka*sin(omega*t)), omega=2*pi*f.
The problem is that I am not having a smooth sinusoidal result of the forces coefficients especially at the peaks , when I decrease the time step from 0.01 to 0.001 , I have more un-smoothness in the result and a lot of small fluctuations as shown in the attached figures , what exactly i should do in this case ?
Dear Hesham Gaballa , did you have a look at the flow field? The amplitudes of the gust as well as the pitch motion have to be sufficiently small to ensure a linear behaviour. There might be also a problem with the quality of the grid.
https://www.sciencedirect.com/search?qs=Unsteady%20CFD%20simulation%20for%20gust%20over%20an%20airfoil&show=25&sortBy=relevance
Vladyslav Rozov The amplitudes of gust ad pitching motion are so as to have a maximum total angle of attack around 12 degrees and the airfoil is NACA0015 and it stalls at around 15 degrees. and the minimum element quality i have is 0.89.
Maximilian Winter yes i first conducted the grid study before the time step study and even with the fine grid i still have these fluctuations when i decrease the time step to 0.001 (T/1000), T=1/frequency.
Dear Hesham Gaballa , if the amplitude of the angle of attack oscillation is around 12 degrees, I doubt that the case can be ragarded as linear. I suggest that you depict the trends of the coefficients as Lissajous curves to see if you get an ellipse that typically arises for linear cases.
IN application, this the phenomena that occurs in VAWT , the wind are unsteady and if youI can alter the attack angle of the aerofoils then you can extract more energy from the wind...very interesting though. . I was working on this problem late 1976, very interesting phenomena, but the unsteady rotation was soved by just adding a flywheel.
You may interested to look into one of my paper working with gust on NACA 4 series aerofoil. Article Aerodynamics of Harmonically Oscillating Aerofoil at Low Rey...
Vladyslav Rozov , I plotted the lift coefficient variation with the angle of attack and I am having a bad ellipse shape at the peaks , it is not smooth and it is not having a perfect ellipse shape, does that mean that i am not in the linear range of the lift anymore ?? can you explain more ?
I mean every time step , the AOA is increasing so the lift is supposed to increase but having these fluctuations when i use dt=0.001 means that the lift increase and then decrease and so on ?
Dear Hesham Gaballa , I assume, with such big deflections you are not within a linear range of the problem any more. Hence, the dependence of the lift on the AoA can't be described by some proportional factor and a phase shift. Do you necessarily need large values of AoA?
Dear Vladyslav Rozov yes i need to increase my AOA in order to achieve a high (Lift/drag) ratio , because this increases my efficiency of reducing the drag.
What do you suggest i should do ?
Dear Hesham Gaballa , not the reference AoA is of concern, but the amplitudes of the gust and AoA oscillations around the reference AoA. The amplitudes have to be sufficiently small such that the lift changes dynamically linear with the amplitudes.
Dear Vladyslav Rozov , my reference AOA is 0, because the the AOA varies following a SINE pattern. even when the max total AOA is around 5 degrees, i still have these oscillations, IS it something numerical then ?
- Badges
- Science topic