[gnut1989]

Dear everyone,

I am confusing with the boundary condition in 3D geometry. I used symmetry for the plane which I showed in the picture below but I think that it may make a wrong results because the 3D geometry is only has 1/6 the real geometry. Is it right?

Could you please help to give me some suggestions about this issue? Can I change from symmetry to wedge for boundary condition?

Thank you and best regards,
Duong Tung

[gnut1989]

Nobody has some suggestions for my issue?? I really need all of your helps because I have to solve this problem soon.

If anybody has any suggestion, please let me know to help me solve my problem.

Thank you and best regards,

[alexB]

Hi Duong Tung,

I am soon facing similar problems, because I have to build a case for cavitation in an injector too.

At the stage where I am now, I think the symmetryplane could be replaced by simple walls, because there should not be high velocities or transfer of matter.

Some sort of symmetryplane would sure be more elegant.

regards
Alex

P.S.: Until which date do you have to finish your work?
You'll get faster help if you share more information and files, because some FOAMer just try to run your case to find errors and the solutions.
It is also useful to know which OF Version you are working with... see http://www.cfd-online.com/Forums/ope...-get-help.html (I haven't read the whole thread myself )

[gnut1989]

Originally Posted by alexB

Hi Duong Tung,

I am soon facing similar problems, because I have to build a case for cavitation in an injector too.

At the stage where I am now, I think the symmetryplane could be replaced by simple walls, because there should not be high velocities or transfer of matter.

Some sort of symmetryplane would sure be more elegant.

regards
Alex

P.S.: Until which date do you have to finish your work?
You'll get faster help if you share more information and files, because some FOAMer just try to run your case to find errors and the solutions.
It is also useful to know which OF Version you are working with... see http://www.cfd-online.com/Forums/ope...-get-help.html (I haven't read the whole thread myself )

Hello Alex,

Thank you for your suggestion!

As you said, you faced with the similar problem and did you overcome this issue? And how did you set up the boundary for your geometry if it has the symmetry plane?

As your suggestion, I need to ask you a question that I can change the type of boundary condition in the constant/polymesh directory from symmetry to wall for both of planes which I showed in the previous post??? Is it right??

I will send to you the checkMesh log file and also some set up files and hope that all of you can help me to solve this problem. I also attached some pictures which show the results when I ran by cavitatingFoam with my old boundary condition.

P/S: My main purpose is that I have to predict the cavitation region inside the injector by using cavitatingFoam and my due date is the end of this month.

Thank you for your help and best regards,

Duong Tung

[alexB]

Hi Duong Tung,

I wanted to start these days with my injector model in 3d with shm .

What I did before where 2d Models with moving meshes and InterPhaseChangeDyMFoam (the DyM Version of cavitatingFoam seems to be bugy and IPCF runs faster).

You can have a look at my cases:
https://www.dropbox.com/sh/rpwe8cno8...6wHZihMga?dl=0

To change from komegaSST to kepsilon in the RAS Model you have just to change the entry in RASProperties file.
The same in the LES model (LESproperties) I tested laminar, OneEqEddy and SpalartAllmarasDDES.

regards
Alex

[gnut1989]

Originally Posted by alexB

Hi Duong Tung,

I wanted to start these days with my injector model in 3d with shm .

What I did before where 2d Models with moving meshes and InterPhaseChangeDyMFoam (the DyM Version of cavitatingFoam seems to be bugy and IPCF runs faster).

You can have a look at my cases:
https://www.dropbox.com/sh/rpwe8cno8...6wHZihMga?dl=0

To change from komegaSST to kepsilon in the RAS Model you have just to change the entry in RASProperties file.
The same in the LES model (LESproperties) I tested laminar, OneEqEddy and SpalartAllmarasDDES.

regards
Alex

Hi Alex,

Thank you for your quick response!

In my topic, I have to use cavitatingFoam to predict the cavitation regions inside the injector so the important thing I want to ask you that can I use the wall boundary condition for both the symmetry planes? Does it affect to the final result?

Moreover, Do you think that using interPhaseChangeFoam can give the more exact results than cavitatingFoam and I should change my solver from cavitaingFoam to IPCF?? Please give me some suggestion about this issue.

Could you please check my set up file and evaluate my results which I sent to you?

Thank you and best regards,

Tung

[alexB]

Hey Tung,

Just as I said,

Originally Posted by alexB

At the stage where I am now, I think the symmetryplane could be replaced by simple walls, because there should not be high velocities or transfer of matter.

Some sort of symmetryplane would sure be more elegant.

I will test it in a few days. If I can implement a symmetry plane, it would be my first choice.

For the comparison of CavFoam vs InPhaChaFoam I found this pdf.
http://web.student.chalmers.se/group...PosterOFW5.pdf

I will take a look on your case, but you haven't sent anything yet.
I can't say if results are good or not... I just can compare with studies available in the internet. As http://www.researchgate.net/publicat...JECTOR_NOZZLES

regards
Alex

[alexB]

I give a short progress report.

My suggestion is, symmetryPlane should work, but you have to split your symmetry plane in two parts. SymmetryPlane seems just to work with mostly planar boundaries.

I have to change my geometry as well.
And try it again soon.

regards
Alex

[alexB]

Now I've got it.

I splitted the symmetry boundary in two... but that didn't change much.
I still got the error that "symmetryPlane not planar blabla" .
I changed the boundary type from symmetryPlane to symmetry and everything works fine. (I hope it won't crash )

Perhaps it is not necessary to split the symmetry boundary in two if you change from symmetryPlane to symmetry.

regards
Alex

[gnut1989]

Originally Posted by alexB

Now I've got it.

I splitted the symmetry boundary in two... but that didn't change much.
I still got the error that "symmetryPlane not planar blabla" .
I changed the boundary type from symmetryPlane to symmetry and everything works fine. (I hope it won't crash )

Perhaps it is not necessary to split the symmetry boundary in two if you change from symmetryPlane to symmetry.

regards
Alex

Hi Alex,

Thank you very much for your suggestion and testing my problem.

In my set up file which I sent to you, I also has already set the boundary type is symmetry and it worked. I hope this do not affect so much to the final results.

One more thing, can you share to me your results and some pictures about your geometry to compare with mine???

Thank you again and best regards,

Duong Tung