Hello everyone! I'm working on VOF modeling of vegetated open channel flows. The geometry and meshing has been done on GAMBIT. The conditions in fluent are listed below:
- Steady flow simulation
- VOF model with open channel flow
- Implicit body formulation with sharp inter-facial modeling
- K-epsilon turbulence model with standard fall function
- Pressure-velocity coupling scheme is SIMPLE
- Discretization scheme for pressure as Body force Simulation and Compressive for Volume Fraction
Kindly guide me if there is some mistake in setting these conditions. Moreover, there is great variation in turbulence parameters i-e k and epsilon. Variation of residuals with iterations is shown in attached picture.How can I manage my model to get better results with reasonable accuracy .and convergence.Thanks
Dear/Kamran
I would suggest to do the followings:
1-Specifiy the goemetry
2-Begin with regular medium mesh
3-Check the boundary and initial conditions
4-Begin with 2-D
5-What about Reynolds Number(laminar or Turbulent to what extend)
6-Try one equation turbulence model
7-Try under relaxation for stream-wise velocity components
Regards
Try to use the coupled solver instead of SIMPLE and to switch to the pseudo-transient solution scheme for your steady-state flow. Potentially this could improve your convergence. The Higher-order term relaxation might be switched on as well (HOTR option).
Regards,
Dr. Th. Frank.
Dear sir
Thomas Frank, what should be usual value of High-order term relaxation for supersonic type of flow problems
Thank you.
Gangadhar Venkata Ramana Pinapatruni : ???
High-order term relaxation is just a check box in ANSYS Fluent pseudo-transient solution scheme. Value can be yes/no.
Thank you for helpful reply Thomas Frank sir.
I have attached a screenshot from Ansys Fluent theory guide 15. As mentioned in the image, the factor I am changing is 'f' the under-relaxation factor. If 'f' is too low, that is less than 1e-4 then the new value is almost equal to old value. I don't think the obtained solution is accurate. Kindly help me to know whether my statement is correct or not. And how it is relating "High order term relaxation"
Regards
GKL
Gangadhar Venkata Ramana Pinapatruni : Please see attached screenshot from a setup in ANSYS Fluent 2019.R3, selection of HOTR option. The default value of the underrelaxation factor is set to 0.25 under the "Options" button, which I usually remain unchanged. It is not at all recommended to decrease URF's below approx. 0.1. The better approach is to decrease in the pseudio-transient solution method the pseudo-timescale or in a real transient simulation to decrease the time-step size.
Regards,
Dr. Th. Frank.
- Badges
- Science topic
- Similar topics
- Correction