Meeting Update 04/10/2012 - Eddie

DM cooling added to Radiative Shock module

Due to the initial conditions of the problem, the immediate post-shock temperature is near the peak of the DM cooling curve. As a result, the simulations that use DM cooling cool much faster than the analytic simulations where the rate ~ T2. I found that, on average, the DM rate is approximately 250 times greater than the analytic rate within the cooling region. So for the DM simulations, I scaled the length down by a factor of 250. This resulted in a cooling region that was about 10% of the problem domain as it was in the analytic case.

However, even with the same resolution analytic and adiabatic simulations(~40 cells across the cooling region), the DM simulation was not very steady. I ran the DM simulations 10x longer than the analytic and adiabatic simulations that we looked at last week. I also looked at the analytic simulation at this longer runtime, and it was still steady. The snapshots below are all at time 0.12 (the final runtime of the analytic and adiabatic runs). Here is the DM 400 cell simulation:

movie​

I therefore doubled the resolution to ~80 cells per cooling region. Here is the 800 cell simulation:

movie​

I again doubled the resolution to ~160 cells per cooling region. Here is the 1600 cell simulation:

movie​

It looks like I am again running into resolution issues. However, I think that the interpolation of the DM cooling curve might be to blame?

Update

Decreasing the ambient velocity (aka shock velocity) solved the problem. The ambient conditions used for analytic cooling only work for analytic cooling. They lead to an immediate post-shock temperature that is beyond the peak of the DM cooling curve. Therefore, with DM cooling, these radiative shocks would be unstable. Reducing the shock velocity reduces the post-shock temperature. I found that changing the length scale as previously mentioned, and changing the velocity from 100 km/s to 80 km/s keeps the post-shock temperature below the critical value for stability (which is approximately 2e5 K). Here is the simulation for DM cooling (400 cells):

movie​

Attachments (8)

Comments

No comments.