[nskelly]

Hi all,

Not sure if I should post here, or a thread more specific, but here goes.

I have done LES simulations (on OpenFOAM) and I am postprocessing the data and wanted some opinions on how I go about looking at transition to turbulence.

So one of the main characteristics of a turbulent flow is that it is three dimensional. So I want to see if the flow is transitioning to turbulent flow, I could look at some 3D aspect of the flow.

Would anyone have idea of what characteristic I should look at? Is there a way of quantifying 3Dimensionality of a flow?

Any thoughts would be great.

[FMDenaro]

Originally Posted by nskelly

Hi all,

Not sure if I should post here, or a thread more specific, but here goes.

I have done LES simulations (on OpenFOAM) and I am postprocessing the data and wanted some opinions on how I go about looking at transition to turbulence.

So one of the main characteristics of a turbulent flow is that it is three dimensional. So I want to see if the flow is transitioning to turbulent flow, I could look at some 3D aspect of the flow.

Would anyone have idea of what characteristic I should look at? Is there a way of quantifying 3Dimensionality of a flow?

Any thoughts would be great.

What flow problem did you solved? A clear parameter of the three-dimensonality of the flow is the stretching term.

[nskelly]

Hi FMDenaro,

I am looking at a backward facing step, but with two different rheologies.

I have not heard of the stretching term. Is this vortex stretching?

[FMDenaro]

Originally Posted by nskelly

Hi FMDenaro,

I am looking at a backward facing step, but with two different rheologies.

I have not heard of the stretching term. Is this vortex stretching?

yes, is the stretching term in the vorticity equation, it is zero in 2D flow.

[nskelly]

That makes senses.

So I could treat it as a velocity signal through time. Where I look at the vorticity components to see if it fluctuates in all three components at different Re.

[FMDenaro]

Originally Posted by nskelly

That makes senses.

So I could treat it as a velocity signal through time. Where I look at the vorticity components to see if it fluctuates in all three components at different Re.

Yes, a 2D flow would have only one vorticity component normal to the plane.

However, be careful to the fact that you cannot evaluate the vorticity in time only at some points (the flow could be locally 2D) but you need to consider all the domain.

[nskelly]

Would recommend a certain domain position (for BFS) which might be best?

I have been looking at instantaneous velocity close to the wall.

Is there other ways of examining 3D flow?

[FMDenaro]

Have a look here
https://www.researchgate.net/publica...ical_procedure