[thezack]

Hi there! My name is Zack. I am relatively new to fluent and CFD in general. I have a general idea of what is going on but I need some help with a project I am working on. I am trying to run a simulation of a water turbine moving by the motion of the water. I have a few questions regarding how I can best set up this simulation. I have run the simulation for a while but I am not sure if I am doing it the best (quickest and most accurate) way. I would LOVE some advice and help. I would be forever grateful!



In this picture you can see the outline of the blades in blue. This is a 3D simulation. Water flows from left to right. There is a pressure outlet at ambient at the bottom right. The channel is 8 meters long and has gravity running from left to right. Since for some reason internal flows induced by gravity are a problem in fluent, I had to change the operating density to zero (still not sure what exactly this does).

I used the Ansys Workbench mesher to mesh the model. When I first meshed the model, the cell count was around 3million. Tetrahedral mesh. I have a pretty great computer so this is not a big issue for me. I am using the 6dof solver constrained to rotation around the negative x axis. I provide a resisting torque around the x axis to simulate a generator. I assume this is the best way to predict power output. (correct me if you can think of a better way) I am using remeshing and smoothing in the dynamic mesh options. My first question is related to my 6DOF dynamic mesh settings, pictured below.



The waterwheel part is the turbine blades. Why are there 4 separate dynamic mesh zones? I just selected all of them and had the same settings for each. One of the zones is a "shadow wall". Is the presence of these zones a problem?

With my current settings, all of these zones activated using the same 6dof UDF, with a 2500 NM resisting generator torque I am getting a rotational speed of around 1 rad/sec. I am not sure whether to trust this number. First off, below is a diagram of the torque plot of the wall surface around the blades.


What are these spikes all about? Should I ignore them? I think it may be happening each time it remeshes. It looks like it averages out to the same value.



Above are the remeshing settings for the dynamic mesher. Is there anything wrong here? Should I be using the local face or region face meshers?

I am assuming that it would be advantageous (if not crucial) to have the region around the blades be a finer mesh. I am imagining that it would be advantageous as well to somehow set up a fluid mesh region in the ANSYS mesher which has the same look as the blades (but thicker so that the fluid region around the blades has a finer mesh) which would rotate with the 6DOF solver water turbine. I can make a model of that region and set it up as a fluid region in the ANSYS mesher, obviously, but I am unsure as how I can make this rotate with the blades. Any ideas on that? Or, is my approach to this relatively legitimate? The mesher can make the mesh more fine near the blades, but that region is static relative to the rotating blades!


Above is the meshing tab in the dynamic meshing options window. Is it essential to set up this "adjacent zone cell height"? If so, what should it be? the same as my cell base size, or some arbitrary finer value?

Lastly any advice regarding turbulence models would be appreciated but the above issues are more crucial.

Thank you all so much for your help! It is very important for me to get some accurate results from this simulation.

[thezack]

Any thoughts?

[saiarjun]

I am to ansys cfd 6dof Can you teach me how to create rotating 6dof thank you

Sent from my GT-I9300I using CFD Online Forum mobile app