[FrankS]

Hi All,
I am simulating high speed rotors in non-newtonian fluids. So far my simulation results match up bang-on with my experiments. However, I would like to be able to monitor the contributions of viscous and pressure torque on the rotor. Is there a way to do this in CFX? I know that both are included in the "torque" function so how do I see the contribution of each? My rotors are complex. One has 188 faces. Obviously I am hoping there is a simple way to do this.
Thanks
Frank

[ghorrocks]

You can get the bits by integrating the pressure over the blade and then integrating the wall shear stress. Note that you will have to include a factor for the element normal vector.

Also note that it is unlikely this approach will sum to the torque value. But hopefully it will be close.

[FrankS]

I ran my test case in Fluent which automatically breaks things up. With water, the viscous contribution to the overall torque is 5%. I have continued running test cases in CFX for thicker, and thicker fluids and the results are interesting; the more I increase the viscosity, the lower the rotor torque as the degree of flow separation is reduced with increasing viscosity. This is interesting as it sheds some light on a lot of my experimental results using medium consistency pulp suspensions.

Thanks for the reply ghorrocks. I will see how difficult it is to implement your suggestion in CFX.

[Opaque]

Have you tried adding another Monitor Plot in the Solver Manager ?

There is the possibility to plot the normal and tangential component of the moments. The normal component should be mostly pressure (normal viscous stress is usually small), and the tangential component should be wall shear based.

Similar values are also written to the output file at the end of the run.

[FrankS]

I guess that makes sense. I will give it a try.

[FrankS]

Originally Posted by Opaque

Have you tried adding another Monitor Plot in the Solver Manager ?

There is the possibility to plot the normal and tangential component of the moments. The normal component should be mostly pressure (normal viscous stress is usually small), and the tangential component should be wall shear based.

Similar values are also written to the output file at the end of the run.

You are exactly right! It appears that CFD Post adds up the normal and tangential torques when reporting torque. By playing with the viscosity, I have also found that you appear to be correct that the tangential torque is the viscous component and the normal torque is due to pressure and flow. Thanks for the help.