wind.swf​

• Found an error in the code that has been affecting cooling. Here is a plot that shows the temperature profile for a shock model before and after the error was fixed.

• This fix will affect any simulations that I have done recently or in the past that use cooling. This includes the pulsed jets and mach stems projects that I've been working on.

• I reran the 2.5-D hydro pulsed jet model that I used in my paper and compared the new results to the old. The old model is on the left, and the new model is on the right…

movie​

The overall morphology doesn't look too different. As expected, the big differences are in the small scale structures and the head of the jet.

• The high resolution 2-D mach stem runs are done, but again the Z cooling models are with the old version of the code which has the aforementioned error. I suspect that rerunning these would make the Z cooling models look even closer to the gamma = 1.01 models. These high resolution Z cooling sims took a very long time on bluestreak, but I'm not really in any hurry so I should be ok to rerun.

Would like to discuss this paper:

​http://www.ita.uni-heidelberg.de/~chfeder/pubs/outflow_model/Federrath_outflow_model.pdf

*Ablative RT

1. checked the growth rate of the front along the middle line of x which Rui thought it's simplified too much. Gave Rui the 2D code working with AMR and let him try on LLE's machines.

1. Mach Stems- all but one run done. Made a very silly mistake with restarts, should be finished within 4 days.

2. Finished Chapter 13 & 14 of Toro. I have some ideas for the scope of my final paper- a possible New User's Guide to Numerics?

3. PN Poster. Received template for poster from Martin, I will begin to work on constructing this poster to present at the CIRC symposium. What are the timelines?

Starting from equation:

(v^2-a^2)\frac{1}{v}\frac{dv}{dr}=\frac{2a^2}{r}-\frac{A}{r^2}

It is:

\frac{dv^2}{2dr}-a^2\frac{dlnv}{dr}=\frac{2a^2}{r}-\frac{A}{r^2}

Integrate with respect to r:

v^2-2a^2lnv=2a^2lnr+\frac{A}{r}+C (1)

Where the sonic point implies:

A=2a^2 r_c

Substitute above relation in to equation (1) we can get:

\frac{v^2}{a^2}-ln{\frac{v^2}{a^2}}=4ln{\frac{r}{r_c}}+4\frac{r_c}{r}+C

For wind start subsonic and accelerate to supersonic, apply boundary condition that:

v=a,r=r_c

Then C can be determined to be -3

(\frac{v^2}{a^2}-1)-ln{\frac{v^2}{a^2}}=4ln{\frac{r}{r_c}}+4(\frac{r_c}{r}-1) (2)

Given a certain radius and time dependent A(t), we can find the velocity at the radius by solving (2)

In the limit that r\rightarrow\infty,v^2\gg a^2

v\frac{dv}{dr}=\frac{2a^2}{r}

v^2=4a^2ln\frac{r}{r_c}+C

v will grow indefinitely but very slowly. This implies that if wind is to fall back, right hand side should change sign, or it will just keep growing. So, gravity (and radiation driving force) together should give a non-constantly accelerating wind solution—it should decelerate at some radius beyond the sonic radius.

Circumstances might be: Optically thick AGB? Or binary star located at radius larger than sonic point s.t. deepen the potential well significantly.

2D Simulation A:

Argon ambient at 300K, wire density contrast 10⁵, 40 km/s Shock. Resolution is 100 cells across the wire diameter, fixed grid.

Left to right: at 50 ns, 100 ns, 160 ns, 240 ns.

Top to bottom: wire field 0, 2T, 20T, 50T.

Movies:

​Hydro

​Wire Field = 2T

​Wire Field = 20T

​Wire Field = 50T

See FluxLimitedDiffusion#no10 for a description…

This image shows the absorption time scale, the equilibrium time scale, the diffusion time scale, and the resulting 'safe' time step for the 1D planetary atmosphere setup.

If we use the single temperature solution, then we can switch to just the diffusion time scale which looks like it is 1000x longer? This might work for the two temperature model as well?

TSF
Finished the low res MTSF runs on Kraken. The contained poloidal case is slow (only about 10 frames per day after particle formation) on 1200 cores, 256 zones per radius resolution. The toroidal case is much faster (30 frames per day).

The images: MTSF Contained Toroidal, Beta = 12, Column Density

​movie
Cut through with inverse beta countour:

​movie
Here we see consistent behavior compared to hydro cases. The cut movie takes a few hours to make, should be done by tonight.

MTSF Contained Poloidal, Beta = 12, Column Density

​movie
Cut through with inverse beta countour:

​movie
It is very clear that the material ejection region overlaps the magnetic cocoon.
I think the comparison is promising, the next step will be choose proper parameters to start production runs.

• 2D Ablative RT

Density
Temperature

Choose the star mass = 1 solar mass. Mass loss rate 1E-5 solar mass per year. Radiation force to be 92.83% of gravitational force and applied to all space. Temperature is 2000 Kelvin. Isothermal wind.

Result from Mathematica is shown as below:

Below is the result from AstroBEAR:

The relative difference is about 16%. The reason we have this difference is that we have to specify the wind speed and density on the edge of the star and AstroBEAR module will linearize them within the star and apply them as ghost cell in Riemann Solver. Thus create self-inconsistent boundary conditions.

Another factor is the resolution, which is obvious.

If we want to get a more realistic result:

Solve the problem to give the boundary condition. However, the problem may become tricky because the solution is singular at sonic point.

The code behavior starts to drop when running more than 1024 cores on Stampede. This is consistent with Jet module (which was done for the proposal of 10/15/2013 and only shown up to 1024 cores for better scaling) and Colliding Flows.

Colliding Flows
Triggered Star Formation (512 hydro)
Jets

1 solar mass, mass loss rate 1e-5 solar mass per year, mean molecular weight 1.3 proton mass, cs=3.1km/s, T=1500K isothermal, dust form around 2.4AU, sonic point at 2.5AU, radiation force off around 25AU(should have turned off gradually and smoothly, here is for simplicity). radiation force strength around 95% of gravity, constant.

Red: Wind velocity

Blue: Escape velocity

Actually wind velocity may be just above the escape velocity. If we make it a binary system, the deepened potential well will trap material and have some phenomenon like fall back disk, accretion disk around secondary.

Simulation requirement: For binary star, 10AU separation use 40 cells to resolve, therefore, finest cell correspond to 0.25AU. 200AU*200AU*100AU simulation box. This can define the roughest cell, like 2AU.

Base cell: 100*100*50 Level: 3

Star usually need extra resolution to see the disk. Then use 4 level AMR.