Learning Forum

[Kevin Gnanaraj]

Hi all,

I am trying to run a simulation in CFX to determine the static thrust of a propeller. I have not been getting results that match experimental data, so I tried to compare FR to stage. However, with the exact same settings on both, I keep getting floating point overflow errors when using the stage interface when compared to FR. What would I need to change to make it work? I want to make this comparison because I will be using valid results to run a coaxial contra-rotating configuration, and frozen rotor will not be ideal for that simulation.

What boundary conditions do I need to add to have a good simulation to get good results? I have tried total pressure, static pressure, velocity for the inlet, and pressure and opening outlets that haven't provided good results. You can see the images here:

[CFD_Friend]

Hi,

Can you check the mesh quality? The mesh quality statistics can be found in the OUT file. Please see if the mesh is OK.

[Kevin Gnanaraj]

Which specific statistics do you want? It was meshed with a target skewness of 0.9

[CFD_Friend]

Also, you can turn on double precision for the run.

[Kevin Gnanaraj]

That was also done, I have asked some people, who have said that my boundary conditions are wrong, I will report back once once simulations with the new conditions are complete.

I was using a symmetry condition and opening pressure of 1 atm and a total pressure of 1 atm, which resulted in little fluid motion.

[Kevin Gnanaraj]

My simulations are even worse, I am getting a force of 1400 N when I expect 45 N. Also, I'm running on a compute cluster, and am trying to run parallel, nd have set a flag 

module load ANSYS/2021R1

 

nodes="sed -e :a -e N -e 's/\n/,/' -e ta $TMPDIR/mpichnodes"

/apps/Common/software/ANSYS/2021R1/v211/CFX/bin/cfx5solve -def default_mesh.def -double

 

-par -par-dist $nodes -start-method "HP MPI Distributed Parallel"

 

but I see serial when editing run in progress. What did I write wrong?