More colliding Al flows

Changed the set up a little bit so that there is no gap now (no ambient). The perturbation is now pseudo-random. The bottom half of the domain is initialized with either a density or velocity perturbation which contains multiple sinusoidal modes with random phases. Below is an image of an initial 5% amplitude density perturbation.

I ran 4 models: density and velocity perturbations, 5% and 10% amplitudes. All simulations were run to 300 ns.

Density Pert. Velocity Pert.
5%
movie​ movie​
10%
movie​ movie​

I also made lineouts of density and temperature across the middle of the cooling region.

Density Pert. Velocity Pert.
5%
movie​ movie​
10%
movie​ movie​


No cooling

For comparison, here is a 10% velocity perturbation run without cooling.

movie​ movie​


Inflow perturbed

One idea that we had was to keep the inflow from the bottom boundary perturbed with the same perturbation that is initialized in the bottom half of the domain. This drives the perturbation for a longer time and should lead to more dynamical effects. The perturbation is again 10% on velocity.

movie​ movie​

Attachments (25)

Comments

No comments.