Divergence problem when airfoil exceed pitch angle
I've got a 2D FVM fractional step structured solver. It's used to simulate moving (oscillating, pitching) airfoil with grid deformation. Arc based TFI is used for the deforming of grids. It works well for pitch angle up to ~20deg. However, at higher angles or highly cambered airfoil, the results starts to diverge and the simulation has to be stopped. This divergence usually occur near the leading edge of the airfoil.
I've inspected the grid. It's deformed but it still seems ok (i guess). I wonder what can be done to prevent this from happening. Here's what I'm considering:
1. decrease time step 2. increase time step (avoid the high angle?) 3. increase grid no. 4. change deformation algorithm - suggestions?
I've done some initial testing but it's not too good. I wonder if there's other mtds available?
Re: Divergence problem when airfoil exceed pitch a
Before you jump into costly resolution studies I would try to diagnose the problem a little more. You say the grid looks ok even at the highest angle. Have you tried using this deformed grid to get a (quasi-) steady state solution at the high angle? This isn't a sufficient or even necessary condition for success, but it will help you sort out the effect of unsteadiness. In particular, you should take a look at the effect of the reduced frequency (choose a very small frequency of motion to get a quasi-steady solution, i.e. a solution that depends only on temporal angle of attack but not on rotational velocity of the airfoil).
And then, there's another revolutionary idea (wink): Look at the unsteady flow field (snap shots, movie) until the point when it starts to diverge. Also look at the unsteady pressure distribution. Do you see anything that stands out as a potential problem?
Furthermore, I would take out number 2 in your list of remedies as a very, very questionable (and probably useless) approach.
Hello, Your site is great. Regards, Valintino Guxxi
|All times are GMT -4. The time now is 02:19.|