helical modes in swirling jets

daniel

Hello everyone, I got this problem.

I have a swirling jet with vortex breakdown. The simulation has been done by axisymmetric DNS code. I use the steady solution obtained by axisymmetric calculation for a full 3D run.

In order to highlight the effect of three dimensionality, I subtract from the instantaneous flow field computed the steady axisymmetric solution, so that I would immagine to see the perturbation's evolution.

My simulation reveals azimuthal modes 1 and (sometimes, depending on the Re and swirl) 2 growing in linear regime. Fine. Suppose the dominant mode is 1. When I plot (for example) the isosurface of radial component (of perturbation) at a fixed time (be carefull, this is the key point...at a fixed time) I find a double helix which is rolling in the opposite direction of the main flow when advances in positive streamwise direction. The same is if I plot the vorticity isosurface.

However,when I consider the evolution in time (!) in one section, I see the perturbation rotating in the same direction of the main flow.

So, the complete structure of perturbation is of a double helix, spatially rolling in the opposite direction of the main flow, but rotating in the same direction of the flow.

A simple picture of what happens: put the middle finger of your right hand upon the forefinger and turn the hand in the clockwise direction (clockwise from your point of you)

This is what happens. Now,

1) do you think is reasonalble?

2) if, so...what's happing to the perturbation?

Thank you very much