Source term in NS eqns
 

hi,

may I know if there is a need to add a source term which contains rotational/coriolis effect?

I'm trying to solve the NS eqns in ALE formulation with moving/deforming meshes due to plunging & pitching airfoil motion.

thanks!

Re: Source term in NS eqns
 

If the coordinate system you are solving the equations in is rotating then yes you need to add a term to account for the motion of the coordinate system.

Re: Source term in NS eqns
 

thanks ag...

in other words, if i 've a pitching aifoil and the grid deforms at the airfoil surface but it is fixed at the outside boundary, there is no need for the additional source term, right?

similarly, if the airfoil is accelerating up together with the grid, there is also no source term required, cause there's no angular velocity/acceleration, right?

thanks again...

Re: Source term in NS eqns
 

Dear Zonexo,

There is no need to add any source term corressponding to the coriolis effect, as long as your frame of reference is inertial. Coriolis componenet comes in when there is a non-inertial frame( such as rotating) of reference.

To reduce the complexity, you could first try it out for a few Euler cases and then can easily incorporate into N/S

Regards

Ganesh

Re: Source term in NS eqns
 

See ganesh's response. Grid motion has nothing to do with the existence of a Coriolis source term. The Coriolis acceleration arises if the basis vectors you are using to describe your vector and tensor quantities are rotating. An additional source term would appear if they are linearly accelerating. The easiest way to see if this is the case is to consider a uniform freestream in the coordinate system you are using. Are the velocity components of the uniform freestream changing in time? If so then your coordinate system is accelerating and you will need to capture that in the formulation of the governing equations. But to emphasize - grid motion has nothing to do with the need for a Coriolis source term.

Re: Source term in NS eqns
 

I don't know very well your problem and i didn't read your message with necessary attection. When you suppose that Coriolis force acts in the problem is always good rule estimate Rossby Number effect!!!

Ro=u/(2omega*L)