Hello, I am trying to simulate air flow over a NACA 0012 airfoil with a 60 degree flap. The wing is set to 16 degree angle of attack.
I am trying to check the yplus values, by using a yplus filter and the “point value” feature in the post-processing pane. In this way, by clicking on the boundary layer I can get wall shear stress and yplus values.
However, the yplus values I am getting are way higher than what they should be. When I calculate yplus using yplus = y * rho^0.5 * shear stress^0.5/ dynamic viscosity, I am getting 0.63, when the “point value” is saying about 800! I am pretty sure my formula is correct.
Am I doing something wrong in using the “point value” feature of the post processor?
And another question: is it totally crazy to use wall functions given the massive separation occurring? I can’t seem to get better accuracy in Cl and Cd than about 30%.
Hello @perezs , and thanks for reaching out to us via Forum!
Are you using the correct first layer height to achieve y+ = 0.63? If possible, could you please share a URL to your project here so that we can take a look at it?
Regarding your second question, I would opt for the Full Resolution approach rather than Wall Functions, as we can assume the wing is fully stalled at 16 degrees of angle of attack.
At the high pressure side of the foil, the “inspect point” gives a yplus of about 1000. To get a plus of 30, I input the first cell location during pre-processing as 2.6e-3 m (the online yplus calculator had estimated 1.3 e-3 m, based on the 5 m chord length and velocity of 9.15 m/s in air and yplus = 30). Big difference between 30 and 1000!
The “inspect point” value of wall shear tau is 5.9 e-2 Pa. When I use the yplus formula with the 5.9e-2 Pa, I get a reasonable yplus of 9.4.
I’m actually seeing the expected y+ value on the high pressure side as follows, please correct me if I’m missing something:
Other than that, I have a few recommendations to improve the solution. Please make sure that your far field is actually far enough from the wing - from all directions. Academic papers usually recommend having a distance of 10 times of the chord at least on downstream.
Transition from the boundary layers cells to core cells looks a bit abrupt. Please make sure you have a decent growth size in boundary layer cells, so that the volume ratio between the last boundary layer cell and the first core cell remains a reasonable value (1 to 5 I would recommend to capture the velocity gradient better)
You may want to define a pressure outlet on the upper face since the angle of attack is quite high, and you wouldn’t want upper wall to affect your solution. Make sure wind tunnel walls are set to a slip condition.
You may want to go with a transient solution as well, steady-state RANS approach may not be the best when it comes to post-stall analysis.
The error was all mine in that I was using the “inspect point” feature improperly (I was using it on a cutting plane in the post-processor). I see now that the Yplus values are totally reasonable.
And thank you also for your very insightful comments about the tunnel size and boundary conditions. I will be trying them out very soon.
