CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Mesh Quality is not improving after Dynamic Remeshing! (https://www.cfd-online.com/Forums/cfx/163789-mesh-quality-not-improving-after-dynamic-remeshing.html)

shivasluzz December 6, 2015 02:29

Mesh Quality is not improving after Dynamic Remeshing!
 
Hello everyone...
I am trying to simulate a dynmaic remeshing problem..
I set the remeshing condition using minimum orthogonality angle..
( minval(Ortho angle)@ Defaultdomain < 20 [deg])
Set the expert parameter to solve only for the mesh deformation and when the solver starts solving, the mesh orthogonality angle written by the solver is near 55 degree..
As the mesh deforms, the mesh orthogonality angle decreases steadily and when it comes below 20 degree, the remeshing (through icem replay file) started..
But after the remeshing, the orthogonality angle is still very nearby 20 degree.. so after 1 or 2 timestep, again remeshing starts and the process continues till the solver fails with negative volume error..

I tried the following:
1. Increased the quality of initial mesh by refinement and smoothing (and recorded the script for replay)
2. Reduced the timestep as minimum as possible (0.001 s)
3. decreased the model exponent of mesh stiffness from the default value of 10 to 1

But still the same problem occurs :(:confused:...
Anbody have any idea about this? please help..
Thanks..

ghorrocks December 6, 2015 05:14

It sounds like you need to check that your dynamic remeshing setup is giving reasonable mesh quality. It sounds like your original mesh is OK by the remeshed ones are poor.

shivasluzz December 6, 2015 05:29

thanks for the reply ghorrocks...
the settings of the replay file i have given for dynamic remeshing is same as that of initial mesh..

ghorrocks December 6, 2015 05:51

Something is making the new mesh not be very high quality. You are going to have to look around and find what it is. Check what the new mesh looks like - it might give you a clue to the problem.

shivasluzz December 6, 2015 06:50

thanks ghorrocks...
i created a deformed geometry and used the replay script to create the mesh.. the mesh created was fine.. However with the dynamic remeshing, it doesnot mesh...
some of my observation are:
1. whatever method i use to create unstructured tetra mesh, 100% elements are not satisfactory... In the solver, mesh statistics shows that 99% OK, <1% ok and 0% !
as per the help file, 'capital' OK means element is fully satisfactory in terms of quality.. 'smallcase' ok means mesh are satisfactory and 'exclamatory mark' is unacceptable mesh elements...
only in structured hexa mesh i am able to create 100% OK elements.. when using using unstructured tet mesh there is atleast <1% ok elements..
I think this <1% ok elements is creating the problem, since i am using 'minimum orthogonality angle' for interuption control

shivasluzz December 6, 2015 07:07

files are attached
 
am just working on a sample problem to understand the dynamic remeshing..
i am working on this problem for more than 2 weeks and still could not resolve this problem :(.. I might have missed something important to specify in the post and so i attached files here (cfx file, geometry tin file, mesh replay file, CCL files of ansys 15 format).. Please help..

here is the dropbox link

https://www.dropbox.com/s/i5sfg75ghu...ement.zip?dl=0

Thanks

chiragsvnit December 11, 2015 07:15

Did you check this tutorial?
https://www.youtube.com/watch?v=UiXXFud4QGs

shivasluzz December 11, 2015 08:33

thanks for the reply chiragsvint..

that was just an animation!

shivasluzz December 11, 2015 08:40

my remeshing process during solving was very quick...
that raised doubts because in icem it took some time to create the mesh using the same replay file..
then after going through all the related files in help file, i found out the reason of the problem...
when we set 'Mesh control' as automatic, during remeshing icem applies its own default part mesh sizes... If we set 'none' then the sizings are as per the scripts in replay file...

so i set the 'Mesh control' as none, then the quality after remeshing was very good!

but it ended up with another problem :-( the solver doesnot respond.. 'solver waited for 60 sec and terminated' error came..

windfire February 3, 2016 07:02

can you give me some help examples about mesh deformation using ICEM replay in CFX,i am a newcomer!thank you


All times are GMT -4. The time now is 23:16.