Dynamic Mesh - Fluent - Savonius Wind turbine

[pathak5]

Hey guys, new user here.

I had a question about a project I am working on with a professor. Unfortunately he does not have much experience with fluent but is a very intelligent and experienced person when it comes to the theory.

So I am basically trying to develop a chart of the power factor of a VAWT for different angular velocities. My inlet velocity is at 25m/s(i know this is huge...but it's just a number i picked randomly to build the model).

Now I created in my mesh a separate interior zone of fluid in a circle around the blades that has a finer mesh than the rest of the 2d model.

I have read a few papers and some people have modeled for various application in such a manner that they created a rotating reference frame around the blades and modeled the blades (wall) as a rotating wall with 0 rad/s, but because it is adjacent to the reference frame it will rotate too theoretically. But obviously since the mesh is stationary, nothing is moving except the fluid. Now according to papers this is an acceptable assumption. I have gotten my power factor values for this for various speeds. Obviously they were high due to the inlet velocity but nonetheless my numerical model was created.

Now, I have investigated more, and I want to make the blades move, and animate it and so on. So I modeled the fluid section as a moving mesh and put in the dynamic mesh condition.

I keep crashing and having errors of "negative volumes" . I have found few items online to fix this erroneus function, but I am having absolutely no success.

Can someone help me here? What am I doing wrong? What else should I do? Do I have to create a user defined function program?

If any serious users want to help I am willing to show/send my case and data file if needed, but please no spammers. If sufficient help is provided I will attach person as co-author in the paper. I have msn, please let me know if it would be better to communicate there.

Any form of help is appreciated.

Thank you

regards

Joy

 

[richpaulcol]

Hi,

I've played around a bit trying to model a vawt in fluent and managed to get some pretty animations if not any usable data.

If i understand your question right i think you were nearly right.

I built two separate 2D models in gambit, a finely meshed circle around the blades of the turbine, then a larger less finely meshed outer section with a hole in it that the inner section fits into. I think you need to set the outer boundary of the inner circle and the inner boundary of the outer region to be 'interface' zones. You then need to export both meshes separately and combine them using the tmerge program (which can be found in the /Fluent INC/utilities/tmerge2.1/ directory).

This will output a single mesh file which can be loaded up as usual in fluent. You have to then set up the interface between the two regions in the ->Define->Grid Interfaces menu, i don't have fluent in front of me at the moment and can't quite remember how this bit works, but i think you just select the two interfaces from the list and click create. You can then set the internal region to be a moving mesh, and set your desired rotation speed. If it all works out ok when running you should see the blades spinning nicely in your model.

I got most of the information from the fluent tutorial guide on sliding meshes (i think it is tutorial 10, the one with the 3d stator blades) and modified it for 2d.

Hope this helps

Richard

 

[pathak5]

Hi,

I've played around a bit trying to model a vawt in fluent and managed to get some pretty animations if not any usable data.

If i understand your question right i think you were nearly right.

I built two separate 2D models in gambit, a finely meshed circle around the blades of the turbine, then a larger less finely meshed outer section with a hole in it that the inner section fits into. I think you need to set the outer boundary of the inner circle and the inner boundary of the outer region to be 'interface' zones. You then need to export both meshes separately and combine them using the tmerge program (which can be found in the /Fluent INC/utilities/tmerge2.1/ directory).

This will output a single mesh file which can be loaded up as usual in fluent. You have to then set up the interface between the two regions in the ->Define->Grid Interfaces menu, i don't have fluent in front of me at the moment and can't quite remember how this bit works, but i think you just select the two interfaces from the list and click create. You can then set the internal region to be a moving mesh, and set your desired rotation speed. If it all works out ok when running you should see the blades spinning nicely in your model.

I got most of the information from the fluent tutorial guide on sliding meshes (i think it is tutorial 10, the one with the 3d stator blades) and modified it for 2d.

Hope this helps

Richard

Hey thanks for the reply

I have been doing that. Unfortunately it gives me an error when I try to initialize.

Error: no periodic zones adjacent to grid interface
Error Object: ()


Any help?

regards

Joy

 

[filip-1986]

hello,im also making Savonius.im new with this,so can u tell me how to make dynamic mesh?do i use Udf function? or?if yes,how to make her?
paul i made model in gambit little differently, but thanks for advice.

anyway if u want we can compare results

 

[sushil pareek]

To phatak5,
Hey we have started working on a similar project. I need some assistance from you. please help us with some guidance about cfd analysis. We are stucked up in the begining. I need the steps how you performed analysis in static mess for 2d savonius rotor blades. please help!

 

[pri979]

Hello Pathak,

You have to define the boundary conditions in Gambit as periodic so that you won't get that error in Fluent.