Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

Rotating Machine with turbulent flow

Patrick Namy Certified Consultant

Please login with a confirmed email address before reporting spam

Hello everybody,

I would like to know if anybody has already tried the rotating machine application with a turbulent flow (RANS ke).
I am facing a lot of difficulty in starting the computation, and I am wondering about the best strategy to initialize it.

I used the segregated solver (2 groups and a lower limit on ep and k), and I tried to used a "ramp" function (flc2hs in fact) to make the rotating machine accelerate from 0 to the targeted value of rpm.

In the model library, the baffled_mixer model is very useful, but unfortunately, it is implemented in a laminar flow. In my case, the Re number is around 320000.

thanks a lot,
kind regards,

Patrick

7 Replies Last Post Jun 4, 2012, 9:41 a.m. EDT

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 23, 2011, 5:34 a.m. EDT
Hi,

If you manage to get it working let me know. In the extremely unlikely case that I ever manage to get it working I will let you know, I tried almost anything just short of defining new wall functions.

Cheers
Hi, If you manage to get it working let me know. In the extremely unlikely case that I ever manage to get it working I will let you know, I tried almost anything just short of defining new wall functions. Cheers

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 23, 2011, 7:17 a.m. EDT
I had such a simulation and even worked with (almost) default settings of the solver. But I have to mention that it was just a 2D geometrie (elipse rotating in a circle with little stoppers at the sides). And I used a really fine mesh (comperable with physical used mesh, extremely fine).

To start the mixer I didn't use a ramp function, but a step function (from 0 to 5 m/s in 10 seconds). Step functions are a little bit smoother then ramps.

If you don't have an inlet and outlet in your tank, then DON't forget to set a pressure point as a boundary condition somewhere in your model. The position doesn't matter.
I had such a simulation and even worked with (almost) default settings of the solver. But I have to mention that it was just a 2D geometrie (elipse rotating in a circle with little stoppers at the sides). And I used a really fine mesh (comperable with physical used mesh, extremely fine). To start the mixer I didn't use a ramp function, but a step function (from 0 to 5 m/s in 10 seconds). Step functions are a little bit smoother then ramps. If you don't have an inlet and outlet in your tank, then DON't forget to set a pressure point as a boundary condition somewhere in your model. The position doesn't matter.

Patrick Namy Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 23, 2011, 7:55 a.m. EDT
In my case, it is a step function too... and the 3D geometry is quite complex... But I think that, for the ke model, the mesh need not to be very fine (not as fine as for Navier Stokes for instance)

Best,

Patrick
In my case, it is a step function too... and the 3D geometry is quite complex... But I think that, for the ke model, the mesh need not to be very fine (not as fine as for Navier Stokes for instance) Best, Patrick

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Sep 23, 2011, 8:32 a.m. EDT
Hi,

My case is a 3D too and no matter whether it is a ramp a step or user defined. Regarding the mesh, it must be fine because it is still navier stokes equations plus some equations to take care of turbulence. Well rather than fine I would say the mesh needs to be balanced depending on your geometry.

Cheers
Hi, My case is a 3D too and no matter whether it is a ramp a step or user defined. Regarding the mesh, it must be fine because it is still navier stokes equations plus some equations to take care of turbulence. Well rather than fine I would say the mesh needs to be balanced depending on your geometry. Cheers

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Nov 23, 2011, 12:04 p.m. EST
Hi

I'm trying to simulate a stirring tank, Re is 23000, I dont think this fully turbulent flow, so I use laminar flow, but the model doesnt converge. I tried with turbulent flow and is very slow.

I really dont know what happen, some have any clue.


Why models does not converge for big time?

I did the simulation with the tutorial, t= 5s, and runs, but when i change to t=100, doesnt converge

Hi I'm trying to simulate a stirring tank, Re is 23000, I dont think this fully turbulent flow, so I use laminar flow, but the model doesnt converge. I tried with turbulent flow and is very slow. I really dont know what happen, some have any clue. Why models does not converge for big time? I did the simulation with the tutorial, t= 5s, and runs, but when i change to t=100, doesnt converge

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Dec 11, 2011, 6:25 a.m. EST
Hi all,

Did anyone solved problem stated in subject?

Would you please help?

What model did you use?

Would you please attach file solved in Comsol 3.5a.

Thanking you.
Hi all, Did anyone solved problem stated in subject? Would you please help? What model did you use? Would you please attach file solved in Comsol 3.5a. Thanking you.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jun 4, 2012, 9:41 a.m. EDT

Hi,

My case is a 3D too and no matter whether it is a ramp a step or user defined. Regarding the mesh, it must be fine because it is still navier stokes equations plus some equations to take care of turbulence. Well rather than fine I would say the mesh needs to be balanced depending on your geometry.

Cheers


Hi Fadel,

I am stucking with a mesh quality on my 3D turbulent flow. So I was wondering what does 'mesh needs to be balanced depending on you geometry' mean?

Thank you.


Atit
[QUOTE] Hi, My case is a 3D too and no matter whether it is a ramp a step or user defined. Regarding the mesh, it must be fine because it is still navier stokes equations plus some equations to take care of turbulence. Well rather than fine I would say the mesh needs to be balanced depending on your geometry. Cheers [/QUOTE] Hi Fadel, I am stucking with a mesh quality on my 3D turbulent flow. So I was wondering what does 'mesh needs to be balanced depending on you geometry' mean? Thank you. Atit

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.