TIMESTEP SMALLER WITH LEVEL SET METHOD

Numerical methods and mathematical models of Elmer
Post Reply
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by Andrea_P »

Good morning,
I am running a simulation with Level set Solver and Navier-Stokes; I have used the model of viewtopic.php?t=4366, but unfortunately, it gives me after 60 iterations the following error:
LevelSetTimestep: Levelset timestep 3.373E-17
WARNING:: ExecSimulation: Obsolete keyword > Timestep Function < , use > Timestep Size < instead
ERROR:: ExecSimulation: Timestep smaller than epsilon: 3.373E-17

STOP 1
How can I resolve this problem? I tried to increase the Courant number but without any improvement. Thanks in advance for your answes.
Regards,
Andrea
Attachments
output.txt
(50.44 KiB) Downloaded 134 times
case.sif
(5.83 KiB) Downloaded 146 times
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by raback »

Hi

This is your problem:

Code: Select all

ComputeChange: NS (ITER=10) (NRM,RELC): ( 0.12759146E+13 0.69524210     ) :: navier-stokes
When maximum velocity goes very large then to fill the criteria v*dt<h the timestep becomes very small. So probably the routine is ok you just have your flow solution diverging.

-Peter
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

Re: TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by Andrea_P »

Hi Peter
I thank you, I apologize for the following banal question: how should I fix it?
Regards,
Andrea
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by kevinarden »

With Elmer your best approach would be to lower the velocity. You might try to add the turbulent flow solver but not sure if it integrates well with level set.
viewtopic.php?f=3&t=6995&hilit=turbulent

High velocity turbulent flow has not been a focus of Elmer. Most high velocity turbulent flow analyses use openfoam.
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

Re: TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by Andrea_P »

The velocity on the left wall goes from 0 m/s to 1 m/s. Hence, I should reduce the velocity, such as 0.1 m/s?
Regards,
Andrea
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by kevinarden »

You could start there and determine the change. Turbulence depends on Reynolds's number which is a function of the velocity, pressure, and volume of the flow.

finer meshes, and smaller time steps, improve results when the flow starts to get turbulent.
Andrea_P
Posts: 176
Joined: 22 Mar 2021, 18:39
Antispam: Yes
Location: POLYTHECNIC UNIVERSITY OF TURIN

Re: TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by Andrea_P »

Hi,
I also have tried to not consider the Courant number, but that does not produce any physical results. Eventually, I have reduced the velocity and Elmer always gives the same error. Are there any other numerical solutions that I can adopt?
Regards,
Andrea
POLYTECHNIC UNIVERSITY OF TURIN-DIMEAS
kevinarden
Posts: 2237
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: TIMESTEP SMALLER WITH LEVEL SET METHOD

Post by kevinarden »

You could try adding the k-epsilon solver which handles some higher velocities.
Post Reply