[hitzhwan]

Originally Posted by vinerm

It appears that you are using either VOF or Mixture model. That implies there is single set of Navier-Stokes equation. In this case, there would be single velocity field, hence, single total pressure and total temperature. So, ignore my previous post. About the unavailability of total values, it is strange. However, it is possible that you wrote selected variables. If you loaded Fluent data file, then total values should be there. However, you can just create total values yourself using expressions.

I use the VOF model and workbench to calculate, I think it may be because that the solution did not finish calculation. If it finished， it will show in it.
Thank you ，do you know how to delete a surface that have been formed in the Design Model, I try everytime, but it does not work.Snipaste_2020-04-22_10-15-18.png

[vinerm]

You cannot delete a surface or any other feature if it is being used by some other entity. You have to first delete the entity that is based on this surface and then you can delete this surface. E.g., if any volume is cut using this surface, you have to remove that cut before you can delete the surface. But you can suppress it easily.

[hitzhwan]

Originally Posted by vinerm

You cannot delete a surface or any other feature if it is being used by some other entity. You have to first delete the entity that is based on this surface and then you can delete this surface. E.g., if any volume is cut using this surface, you have to remove that cut before you can delete the surface. But you can suppress it easily.

Thanks, I got it. I have another question, do you have met the phenomena that the static pressure in the calculation domain is not continuous， do you think it is reasonable？ But the relative difference is not very big？ Do you we can neglect the thing？

[vinerm]

What do you mean by static pressure not being continuous?

[hitzhwan]

Originally Posted by vinerm

What do you mean by static pressure not being continuous?

You see the static pressure in the picture, it has obviously discontinuity，do you think it is reasonable？

[vinerm]

That's not discontinuity; its just the scale that you are using. Change the scale and you will observe something else.

[hitzhwan]

Originally Posted by vinerm

That's not discontinuity; its just the scale that you are using. Change the scale and you will observe something else.

I change it， but it also has the same phenomenon

[vinerm]

Static pressure, in case of VOF and mixture, even total pressure should be continuous. However, it appears that you have discontinuity. How do the volume fraction contours look?

[hitzhwan]

Originally Posted by vinerm

Static pressure, in case of VOF and mixture, even total pressure should be continuous. However, it appears that you have discontinuity. How do the volume fraction contours look?

Yes ，you are right， it is continuous by changing the scale. Do you know how to show the worst quality mesh position in the Ansys mesh? And how to improve it ?

[vinerm]

Meshing tool shows a histogram below the graphics window when a quality criterion is selected under the Mesh. You can click on Control in the histogram, then define the limits of the criterion, such as, 0.96 - 1.0 for skewness, and then click on Update for number of cells. Then close the control and let the histogram display again. Now click on any bar in the histogram and you can see the cells that have bad cells. You can click on the bar(s) even without changing the limits but if you have many bad cells then it will show a huge number of cells in graphics.

[hitzhwan]

Originally Posted by vinerm

Meshing tool shows a histogram below the graphics window when a quality criterion is selected under the Mesh. You can click on Control in the histogram, then define the limits of the criterion, such as, 0.96 - 1.0 for skewness, and then click on Update for number of cells. Then close the control and let the histogram display again. Now click on any bar in the histogram and you can see the cells that have bad cells. You can click on the bar(s) even without changing the limits but if you have many bad cells then it will show a huge number of cells in graphics.

Yes， I can find it now， but how can I improve it，does there have any function to improve it? Secondly, do you think there need any layers for the laminar flow?

[vinerm]

There are hard and fast rules for mesh improvement. You learn it with practice. For laminar flow, you may not require a very fine mesh near the boundary because the gradients are lower than those in turbulent flows.

[hitzhwan]

Originally Posted by vinerm

There are hard and fast rules for mesh improvement. You learn it with practice. For laminar flow, you may not require a very fine mesh near the boundary because the gradients are lower than those in turbulent flows.

But when I refinement a mesh, I find it do not reach a steady state, what is the reason? I do not agree with you, if the mesh is not fine, the film will broke down, and if I do not give the boundary layer, it enhanced.

[vinerm]

A pre-requisite of using VOF is that the mesh should be much smaller than the interfacial length scale. So, yes, the mesh has to be fine enough to resolve the interface but that is not defined using . Though can be defined for both, laminar and turbulent, flows, it is not available as a field variable in Fluent when laminar flow is solved. The reason for that is the velocity profile in laminar flow. So, to resolve velocity boundary layer, you do not need a fine mesh or inflation or boundary layer. However, if you have a thin film of a fluid, then you have to resolve it but the reason and scales are different.

[hitzhwan]

Originally Posted by vinerm

A pre-requisite of using VOF is that the mesh should be much smaller than the interfacial length scale. So, yes, the mesh has to be fine enough to resolve the interface but that is not defined using . Though can be defined for both, laminar and turbulent, flows, it is not available as a field variable in Fluent when laminar flow is solved. The reason for that is the velocity profile in laminar flow. So, to resolve velocity boundary layer, you do not need a fine mesh or inflation or boundary layer. However, if you have a thin film of a fluid, then you have to resolve it but the reason and scales are different.

Yes， you are right，we can not get the y+ in the laminar flow of fluent, so how can we decide the first mesh size of our model. Additionally, as see in the figure, I can see the min orthogonal quality is 0.456, but I cannot find it by clicking the bar in the bottom. I only can find the orthogonal quality=0.55 elements. You see , I want to improve these, but I do not know how?

[vinerm]

The orthogonal quality is already good enough. No need to improve it nor will it help. But do check for skewness; it appears that skewness is high. But if the maximum is below 0.96, then that is alright as well.

[hitzhwan]

Originally Posted by vinerm

The orthogonal quality is already good enough. No need to improve it nor will it help. But do check for skewness; it appears that skewness is high. But if the maximum is below 0.96, then that is alright as well.

Yes， the maximum skewness is below 0.96，but I cannot accept it, can we improve manually?

[vinerm]

Whether it can be improved or not depends on the geometric model. But you won't get any difference in results even if you bring it down from 0.96 to 0.94 or 0.92 until and unless it is brought down to 0.5. Even going to 0.5 may not show much improvement in results.

[hitzhwan]

Originally Posted by vinerm

Whether it can be improved or not depends on the geometric model. But you won't get any difference in results even if you bring it down from 0.96 to 0.94 or 0.92 until and unless it is brought down to 0.5. Even going to 0.5 may not show much improvement in results.

What is reason? Do you mean the mesh quality has no such high influence on the results? There are only few elements are bad， why can not improve it？

[vinerm]

The quality of the mesh is certainly important, however, once the quality is above a certain threshold, further improvements do not change the results significantly.