[wali]

hello dear
i am a post graduate student, doing the simulation over a double wedge airfoil in supersonic regime but i have found a very irregular shape of pressure contours.
but when i simulate for biconvex shape i found a accurate pattern of pressure contours
these are parameters which i have set in fluent
mac no.1.25
angle of attack 0 degree
boundary condition ---- pressure far field for both inflow & outflow
adiabatic wall for airfoil
free stream temperature 260k
minimum orthogonality 1.78355e-01
maximum aspect ratio 1.70674e+02
domain size 80m
o grid structure mesh
smallest grid size near airfoil .0001
density based solver
laminar flow
reynolds number 500000
implicit formulation
time step size .00001

i am enclosing some images that i have captured after doing simulation as well as some images are for the mesh near airfoil.

please help me , thanks in advance

[gfoam]

Hi, are you tryin to solve it with a transient formulation? Why don't you try a steady one? Can you post a plot of the residuals you get? If the problem doesn't converge then try to use Solution Steering (with a Steady formulation) and prove if it works. Also you can try usin the Convergent Acceleration for Stretched Meshes (steady solver) to improve your convergence. But, until now I was talking with the assuption that your problem is not converged (I need to see the residuals), if it is converged, then you must check the BC's, for example be careful with the Pessure Far Fiel BC's because they must be placed too far from the object of interest, because they can affect your solution, so you can try to place the farfield farther and run the case again, or may be something else, we must see what can we do. Regards.
Gonzalo

[wali]

Originally Posted by gfoam

Hi, are you tryin to solve it with a transient formulation? Why don't you try a steady one? Can you post a plot of the residuals you get? If the problem doesn't converge then try to use Solution Steering (with a Steady formulation) and prove if it works. Also you can try usin the Convergent Acceleration for Stretched Meshes (steady solver) to improve your convergence. But, until now I was talking with the assuption that your problem is not converged (I need to see the residuals), if it is converged, then you must check the BC's, for example be careful with the Pessure Far Fiel BC's because they must be placed too far from the object of interest, because they can affect your solution, so you can try to place the farfield farther and run the case again, or may be something else, we must see what can we do. Regards.
Gonzalo

hey brother ,
i read ur advice but actually my solution is converged and i am applying the same condition for biconvex circular arc airfoil and it gives the good results of pressure contours & i am using the transient formulation. why it is not applicable for double wedge shape airfoil and i am using the domain size of 80m while the chord of airfoil is 1m.

[flotus1]

Just out of curiosity and because I am trying to learn about supersonic flow simulations: Is it the usual procedure to simulate supersonic flows at high Reynolds numbers without a turbulence model?

[gfoam]

Hi, @wali, if you solution converges, the BC's are correct and the domain you are using is correct too then you can improve your results using higher orders for the discretization schemes like QUICK or Third Order Muscle, another thing you can try is to refine the mesh and see if your solutions are better. Regards and I hope this helps you.
@flotus1 I think it depends on the Re number based on the chord and flow cnditions like in the case of incompresible or low mach Nº flows (I don't know, is just a supossition, may be somebody with more experience in this kind of simulations can answer with more precision)
Gonzalo