Oscillating Residuals for flow around an airfoil -ANSYS FLUENT. What should I do?
 

I am using a structured mesh on a D-field having length of 12.5 chord lengths for an airspeed of 33m/s.
I tried both for ke-realisable and kw-sst, the residuals are wildly oscillating. I even tried giving URFs but it didnt work.
As the iterations proceed, back flow is occuring and turbulent viscosity ratio is being limited.
I am attaching an image of my mesh.

The mesh looks really good, and it seems odd to me that your residuals would be oscillating even after you enabled a turbulence model. Is this a steady-state or transient simulation?

Its a steady-state simulation.

Alright, so you're residuals are all oscillating, but how about other quantities? Did you set any report definitions?

I am interested in calculating the coefficients of lift, drag and moment. These I defined in the monitors tab. As for the Report Definitions tab, I haven't given anything.

Report Definitions are what the Monitor tab monitors. If you are monitoring your solution, you probably already set Report Definitions without even knowing it haha.



When you simulate, I assuming you are looking at the Solution Monitors every iteration to see how they perform. Are they oscillating as well, or do they seem to be okay?

I got the problem of back flow and viscosity limitation corrected.
I had missed an edge while defining inlet and outlet.

I ran using ke realizable for first order and second order.
The first order converged but the second order residuals has become more or less constant.
What shall I do?

For that, I would check out the topic of "Convergence" on the forum. It's not the easiest thing to judge, and you might actually be looking at a perfectly good simulation and not even know it. I know that's not the answer you were looking for, but it's the best one I can provide.

Getting oscillations in residuals is usually normal. This is because the shock, if tansonic, changes its location (a bit) in every iteration. This is due to inherent unsteadiness in the phenomenon and due to so called the pseudo steady simulation, the solution tries to converge with every time new results. This causes an unsteady pattern (oscillations) in residuals, A higher order residual might show steady pattern that the mesh was enough to converge for frist orders, but for the 2nd order the error is constant and need more refined mesh to converge more in case of 2nd order