[George44]

Hello,

I am currently trying to run a simulation for the standard ONERA M6 wing benchmark test, which is transonic flow over a wing. However, after I set the simulation running (I'm using rhoSimpleFoam) I get the error "maximum number of iterations exceeded". I have looked around these forums in an attempt to alleviate the error but unfortunately I have had no luck so far. From what I have picked up, the most likely cause of the issue is a poor mesh, incorrect thermophysicalProperties file or incorrect boundary conditions. Hence I have checked my boundary conditions and thermophysical properties but I can see no issue with them. The only potential issue I can see is that when I run checkMesh I have 10 non-orthogonal faces but I am still given the Mesh OK statement at the end. Just to make sure this was not the problem, I used a different mesh, which had a maximum non-orthogonality below 70 degrees, but I still had the same issue when running the simulation. I would greatly appreciate if someone could point me in the right direction please?

Please find attached my case files in the zip folder, I have also attached an image of the error message and my checkMesh output. Thanks in advance.

[geth03]

in fvSolution for p try the solver PCG and preconditioner DIC,

btw your tolerance for U can be tighter i.e. 1e-8 and relTol 0.

[George44]

Hi geth03 - thank you very much for your reply. I have implemented the changes that you have suggested but unfortunately I still get the same error at the same time step.

[geth03]

1. try a run with laminar flow instead of turbulence to ensure that turbulence model is not the problem here.

2. your mesh:your aspect ratio is way too high, recommended is normally 5, in cases where gradients are small it could be as much as 10, but 86 is way too much. it means at least one cell has a side that is 86 times larger than the smallest side.

3. can you post a screenshot of your mesh? also post a screenshot of the pressure and velocity field in the timestep before the crash occurs.

[George44]

Thank you again for your time and patience with me.

1. I have run the simulation as a laminar flow but I get the same error at the same time step again. I guess this confirms your suspicion that my mesh is the problem as opposed to the turbulence model.

2. Thank you for the guidelines, I will start working on improving my aspect ratio. I think the inflation layer I have used is what is driving my aspect ratio so high, hence I will focus my attention there.

3. Please find attached photos of my mesh in this reply and I will attach my p and U plots in the next reply.

Thanks again!

[George44]

And here are my U and p plots.

[geth03]

lets clarify a few things for my understanding:

you are simulating a density varying fluid in dependence of pressure,
also as EOS you are using the perfect gas equation.
you pressure field has negative values,
so what happens with your density? does it have negative values too?

[George44]

Yes, I am simulating a density varying fluid and I am using the perfect gas equation. Neither my density nor temperature plots have negative values (please see attached), which I guess is good because this would be non-physical. However, I am not entirely sure how my pressure then has negative values, because considering the perfect gas equation, if rho, temperature and the gas constant are positive, then pressure should also be positive.

[geth03]

yeah thats what i am concerned about, looks like the pressure equation can not converge and maybe thats the issue.

[George44]

Ok, thank you for identifying that. I think my best bet then is to improve my mesh quality and also review my initial conditions for pressure. I may also try changing my divergence scheme for pressure to "Gauss upwind" instead of "Gauss limitedLinear" to help with convergence and then change back to the linear scheme after pressure has started to converge.

[geth03]

good luck !

if you find a solution pls write what the matter was.