[h1773]

Hello,

i am performing transient simulation of gas discharging through nozzle from high pressure tank. I want to know how to perform grid independence study in such transient simulation?

Thank you for guidance.

[FMDenaro]

Use a constant CFL number for each refined grid

[h1773]

Hello,

Thank you for your response. I am unknown for this kind of simulation..so asking few more questions. Hope you will answer them

i am solving the problem with the segregated pressure based solver. Then what to do?

More over there is convergence issues in this problem i am not able to start solution with coupled or density based solver at even 10-6s time step so compelled to solve the problem with segregated solver and judging the convergence by the pressure, mass and tank variation inside the tank (when the variation below 10-5)…Any advice would be highly appreciable

[aero_head]

Originally Posted by h1773

Hello,

Thank you for your response. I am unknown for this kind of simulation..so asking few more questions. Hope you will answer them

i am solving the problem with the segregated pressure based solver. Then what to do?

More over there is convergence issues in this problem i am not able to start solution with coupled or density based solver at even 10-6s time step so compelled to solve the problem with segregated solver and judging the convergence by the pressure, mass and tank variation inside the tank (when the variation below 10-5)…Any advice would be highly appreciable

Whether you use a pressure or density based solver does not matter for the CFL condition.

Again, do not just guess your time step, calculate it from the CFL condition for your case. If you must, use adaptive time stepping to hone in on a time step.

[h1773]

Hello,
I am running simulation in fluent on SIMPLE algorithm the problem is same as above. Courant number is directly available in the case of pressure based coupled solver and density based solver but not in the segregated solver (which are SIMPLE,SIMPLEC etc.). How to find courant number in such cases?

[aero_head]

Originally Posted by h1773

Hello,
I am running simulation in fluent on SIMPLE algorithm the problem is same as above. Courant number is directly available in the case of pressure based coupled solver and density based solver but not in the segregated solver (which are SIMPLE,SIMPLEC etc.). How to find courant number in such cases?

It is based on your mesh and velocity:

C =

where u is the characteristic wave speed of the system, ∆t is the timestep of the numerical model and ∆x is the spacing of the grid in the numerical model.

[PenPencil]

I would recommend this for (rigorous) grid convergence: https://www.grc.nasa.gov/www/wind/va.../spatconv.html