I am going the run the unsteady simulations on airflow through the Human Respiratory tract and have done the steady one. I am confusing about the selection of time step size and number of time steps in FLUENT.
Dear Bholu
Cornell University has a perfect data base for ANSYS fluent tutorial. this link is about Unsteady Flow Past a Cylinder and at the 6th step it helps you to choose time step.
https://confluence.cornell.edu/display/SIMULATION/FLUENT+-+Unsteady+Flow+Past+a+Cylinder
Good Luck
Generally in the simulation of dynamic systems following parameters need to be considered while selecting time step (dt) and length of simulation time (T)
1) Lowest time constant of the system (tau). The relaxation time to free response is approximately 4*tau. So dt < tau/4.
2) Highest frequency of fluctuation or vibration fu (Hz). According to sampling theorem dt < 0.5*fu.
3) Lowest frequency of fluctuation or vibration fl(Hz) which is equal to the excitation frequency. T > 1/fl or T > 1/fn. fn=lowest natural frequency of the system.
There are two ways to find these parameters
a) Analytical- Model the system, find eigen values, then tau = abs(1/real part) and fn = sqrt(imaginary part).
b) Experimental modal analysis.
The above suggestions are based on my experience in simulation of vibration systems and there might be some additional aspects.
Hi
I want to investigate the time step independency for a laminar flow around a cylinder with Re=100 in viscoelastic fluid. I've choosed 0.01 , 0.005 , 0.001 and 0.0005 . but I couldn't get independency!
What is my mistake?!
I think that as I have laminar flow, I need a time step as long as 0.01 or 0.005. but 0.0005 is not enough!
- Badges
- Science topic