Negative moment coefficient
 

Hi,
I'm running transient analysis on a three bladed vertical axis wind turbine in 2d using fluent, the issue i'm having is that the moment coefficient is fluctuating between positive and negative values, See attached.

I also have attached plots of the mesh and a pressure contour which shows no large discontinuities.

Im using the transitional turbulence model with a "simple" scheme and second order upwind.

I have my reference values set up as follows
Depth = 1
Length = radius of turbine
Area = diameter of Turbine X length of blades.

If anyone could provide me with help it would be greatly appreciated, thanks in advance

Can anybody assist? it's been wrecking my head for a long time now,

thanks

??????????

Hi Thompson,

i am not sure/conclude with data you have provided. but for your better understanding please go through this tutorial.

http://www.cfd-online.com/Forums/ans...-tutorial.html

hope this will help.

cheers,
bharat

Hi Bharatt,
Thank you for your reply. unfortunately the link to the web site provided is dead, however when i was initially setting up my model I followed the instructions provided by daysley in the .doc file.

This is why I'm thinking the issue may be down to how I have defined my reference values? Could you have a look at how I have them defined and let me know.

Very much appreciate your help

If you're unsure of your reference values, you can always look at the raw moment data and calculate the moment coefficient yourself. If it doesn't match what Fluent reports, then something's off.

In this case, though, reference values definitely aren't the problem. Reference values are just scaling dimensions; they won't change the sign. It looks to me like your oscillations are decreasing each cycle, do they get better if you solve farther? I don't know much about VAWTs, but it seems plausible that that could be the correct behavior. How does the time-averaged power match the theory?

I agree with Kokemoor. The reference values are not the problem. I had a similar problem and I found that reference frame from the solution initialization tab required to be absolute instead of relative to cell zone. Hope this helps.

You said it is a transient problem. How long are you running this for?

Try increasing the time steps for your solution to stabilize as it looks like the differences are decreasing.

Hi Folks thanks for your replies to my problem

[QUOTE
In this case, though, reference values definitely aren't the problem. Reference values are just scaling dimensions; they won't change the sign. It looks to me like your oscillations are decreasing each cycle, do they get better if you solve farther? I don't know much about VAWTs, but it seems plausible that that could be the correct behavior. How does the time-averaged power match the theory?[/QUOTE]

Thanks for explaining why the reference values aren't the problem, I have run the solution for sufficient time for 4 full turbine rotations to occur, and my solution doesn't really get much better, I am not looking for the solution to converge just that I get a positive cm average for the last rotation as I'm neglecting the first three.

I personally taught my mesh was sufficient what is your opinion on this?

where else could I be having a problem.

Thanks for this but it didn't fix my problem unfortunately

While analysing a 3D wing what is reference radius in moment calculation of a wing?