[cturutoglu]

Hi All,

I am trying to solve an axial fan problem and i realized that if i use mrf model, i couldnt calculate the real torque value because of the fluid motion in the mrf zone. I would like to calculate the real torque value. I mean if i use mrf model, the fluid in the mrf zone is rotating with the value i described. Than i thought that if i use sliding model or dynamic mesh model, it would be more accurate solution. But i tried to use sliding mesh model and i activate mesh motion instead frame motion(same rpm). When i tried to calculate again, fluent is giving me a mesh error. The error is negative volüme detected. But i check mesh before analysis, there was not any negative volume. Also when i do it with frame motion, there is no any problem about mesh. I can solve the problem within real velocity and pressure values.

What should i do to calculate the real torque value?
Can anyone give me a udf for dynamic mesh? or any advise about sliding mesh?

Thank you!
Have a good day

[Ahmed Saeed Mansour]

Have you solved the problem? Is it a 2D simulation?

[cturutoglu]

No, i didnt solve it yet. It is a 3D problem.

[Aurelien Thinat]

Hi,

For a sliding mesh, you need a perfect interface between the rotor and the stator. Then from your mrf setup, you can type in the fluent consol :

/define/boundary-conditions/modify-zone/mrf-to-sliding-mesh
rotor-cellzone-name
()

It should create the interface and you will be able to launch the simulation. Be aware that you need a full 360° model or the same angular subdomains between rotor and stator.

Aurelien

[cturutoglu]

I think i have good interface between rotational and stationary bodies. I dont have stator its only a blade in the mrf zone and inlet, outlet bodies which has pressure inlet&outlet with 0 pa. They are from new part zones and interface is created automatically. Should i make this connection by using fluent interface or is it enough to make in form new part?

Thank you Aurelien!

[Aurelien Thinat]

Well, if you don't have a stator cellzone, you didn't need any interface.

From what you are saying, you just need to :
1) activate the sliding mesh option
2) define the rotating velocity,
3) check the axis and the origin.
4) in the dynamic mesh menu, you shouldn't have anything activated.

EDIT : By stator, I mean "not rotating volume".

[cturutoglu]

Ah than yes i have stator volume.

I did that options but an error has been occupied when i ran the calculation. The error is negative mesh volume detected. But if i use mrf model and run calculation there isnt a problem.

[Ahmed Saeed Mansour]

Check out this.. It may help you
https://www.youtube.com/watch?v=KFrZoDWUC5s

[cturutoglu]

I solved it. In order to help other people i will explain how i solved it.

I refined the mesh and i used fluent interface instead form new part model.

However i already couldnt calculate the real torque value. In my opinion, it is only calculateable when use dynamic mesh model. And i cant write an udf for that case.

[cturutoglu]

I still could not calculate the torque value, searched on the forum but there is no any valid solve about this problem.

I tried to set the referance value to area: blade and hub surface area and length: blade Radius but still couldnt reach the experimental torque value.

I guess this topic is the biggest problem of fluent software.

If some super hero WHO has perfect fluent skills can solve this problem, please notifiy me on that topic.

[Ahmed Saeed Mansour]

contact with me on 12317@eng.asu.edu.eg please

[Ahmed Saeed Mansour]

Do you have a problem in contacting me?

[cturutoglu]

No, I've sent you an e-mail.