# Unsteady Results-Flow over a sphere

 Register Blogs Members List Search Today's Posts Mark Forums Read

 March 28, 2019, 11:11 Unsteady Results-Flow over a sphere #1 New Member   mike Join Date: Aug 2013 Posts: 11 Rep Power: 11 Hi everyone, i have a question about my unsteady cfd results. I am simulating a time-dependent (unsteady at inlet velocity 20 m/s) flow over a sphere with Star CCM+ which is in the middle of the wind tunnel. Also some pictures are saved in every time-step. After some time-steps I mean the time= about 0.5 s, the velocity and also vorticity distributions change really minimally, as opposed to the time between 0-0.5s. So that i cant see the turbulent structures on these distributions. The solution is like an end result of a steady state! (Time-Step is chosen= 0.002s) I have expected that the periodic fluctating behaviour could be clearly seen. Is it normal that the flow can be stable in this case? Or is it unrealistic how I define the problem? Thanks in advance.

March 28, 2019, 12:06
#2
Senior Member

Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,259
Rep Power: 67
Quote:
 Originally Posted by raskolnikov Hi everyone, i have a question about my unsteady cfd results. I am simulating a time-dependent (unsteady at inlet velocity 20 m/s) flow over a sphere with Star CCM+ which is in the middle of the wind tunnel. Also some pictures are saved in every time-step. After some time-steps I mean the time= about 0.5 s, the velocity and also vorticity distributions change really minimally, as opposed to the time between 0-0.5s. So that i cant see the turbulent structures on these distributions. The solution is like an end result of a steady state! (Time-Step is chosen= 0.002s) I have expected that the periodic fluctating behaviour could be clearly seen. Is it normal that the flow can be stable in this case? Or is it unrealistic how I define the problem? Thanks in advance.

 March 28, 2019, 12:32 #3 New Member   mike Join Date: Aug 2013 Posts: 11 Rep Power: 11 Sorry for the missing informations about my model. It is an air flow over a sphere with 220 mm diameter like a soccer ball. The mesh size in the wake is 6 mm. I think it is a general question for me, because the same thing occured when i tried to simulate a vehicle aerodynamic problem in transient case. I observed, it doesnt come to almost no changes when i looked at velocity distributions after certain time-steps. So, could it be a problem at the my problem definition or is the time step i have chosen a wrong value? Or Do I have to make the mesh size finer in the wake? Thanks

March 28, 2019, 12:41
#4
Senior Member

Filippo Maria Denaro
Join Date: Jul 2010
Posts: 6,259
Rep Power: 67
Quote:
 Originally Posted by raskolnikov Sorry for the missing informations about my model. It is an air flow over a sphere with 220 mm diameter like a soccer ball. The mesh size in the wake is 6 mm. I think it is a general question for me, because the same thing occured when i tried to simulate a vehicle aerodynamic problem in transient case. I observed, it doesnt come to almost no changes when i looked at velocity distributions after certain time-steps. So, could it be a problem at the my problem definition or is the time step i have chosen a wrong value? Or Do I have to make the mesh size finer in the wake? Thanks

No, the first and more relevant think is about the mathematical set of equations you are solving. What kind of formulation are you using?

Be aware that your Re is O(10^5). This is a very critical regime, see https://www.grc.nasa.gov/www/k-12/ai...ragsphere.html

This flow is very challenging, requires a proper formulation and a very refined grid