CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   NACA 0012 Cl values (Lift) not dropping post-stall? (https://www.cfd-online.com/Forums/fluent/232270-naca-0012-cl-values-lift-not-dropping-post-stall.html)

rste97 December 7, 2020 12:55

NACA 0012 Cl values (Lift) not dropping post-stall?
 
2 Attachment(s)
Hi, I'm new to fluent, I am conducting a study comparing wind tunnel data of Cl, Cp to Fluent data. my Cl values are all good up until the stall angle. The Cl does not converge (even after 5,000 iterations), and keeps going up and AoA increases when the wind tunnel Cl drops after the stall. This is really frustrating as I want to do the rest of my analysis but can't because of these obviously incorrect values.



My classmates have managed to get the correct Cl values.

I'm using K-Omega SST, Mesh Quality check was good, sensitivity check was good, approx 148k nodes used, Re approx 332k.
I have attached some screenshots, I am at a loss as to what is wrong here.

duri December 7, 2020 14:23

CL and residuals are not expected to converge post stall with steady state solver. But your statement "CL keeps going up and AoA increases when the wind tunnel Cl drops after the stall" is not clear. AoA is set in boundary condition and it is not expected to change.

rste97 December 7, 2020 14:39

Quote:

Originally Posted by duri (Post 789994)
CL and residuals are not expected to converge post stall with steady state solver. But your statement "CL keeps going up and AoA increases when the wind tunnel Cl drops after the stall" is not clear. AoA is set in boundary condition and it is not expected to change.

Hi, I apologise, that was a typo on my behalf. It was meant to read “Cl keeps going up with AoA when in comparison the wind tunnel’s cl drops after the stall”.
So if it’s not meant to converge, I’m rather confused as we were told to let the iterations run until all values were steady.
This may seem like a stupid question but is the steady state solver where I set “SST” in the modelling part? We were told to use Viscous SST k-omega and not to change this, that’s what my classmates have done and all their results are perfect. I’m not very intelligent compared to them so I’ve obviously done something wrong.

Thank you for trying to help. I really do want to learn and get this right.

LoGaL December 9, 2020 10:21

I don't get this, so you successfully obtained results for 0 to about 16 degrees angle of attack, then your simulation does not stabilize when you increase the angle of attack further?

rste97 December 9, 2020 16:56

Quote:

Originally Posted by LoGaL (Post 790194)
I don't get this, so you successfully obtained results for 0 to about 16 degrees angle of attack, then your simulation does not stabilize when you increase the angle of attack further?

Yes that’s exactly what happened. Even my lecturer doesn’t know why it isn’t working. What’s stranger is that the pressure contours, velocity contours and cp graphs are all seemingly what they should be.

LoGaL December 11, 2020 07:23

I mean, if your airfoil is stalling, you shouldn't expect a steady solution, so it's not so surprising that the residuals don't go down. Sometimes by tweaking the numerics of the solver you can get a steady solution, but that is not what nature should be.

147k mesh elements seems really a lot, are you running 2D simulation? I would expect an order of magnitude less. I remember running aviation combustors simulations with 90k elements, and that was complicated 3D geometry. Very high element count tend to make RANS solutions oscillate

I don't have your setup but,
You can try using coupled solver instead of SIMPLE, it should improve stability.

Set pressure scheme to standard, not second order

Turn off curvature correction in the turbulence model, in my experience it reduces steady solver oscillations.


All times are GMT -4. The time now is 08:23.