High-Res MHD Shear

• Runs are coming along - Beta10, Shear15 (95/200) Shear30 (92/200) Shear60 (100/200) and Beta1, Shear60 (75/200)

• Keeping track of the restarts on this CF page (updated),-​http://astrobear.pas.rochester.edu/trac/wiki/u/erica/CFRunStatsHighRes

• Going to need a solution for storing this data. I am estimating needing an extra TB locally, this will come right to the edge of available storage on our local machines as of now. So maybe folks need to delete old stuff, or we should decide on how many of the frames to keep (only 1/10 frames or so..?)

• Only saw nans in one of the simulations (B10, S60), but they were resolved, only saw restarts in 2/4 of the sims, and only 1-2 times each; the sims seem to be progressing fine

• Restarting with a less number of cores resolved memory issue — so guess on 4096 cores I hit a limit with memory per core. The sim can run fine on 2048 cores

• Am post-processing data now. Getting the different pdfs, histograms, and column density maps Jonathan got for his CF paper, and will have Marissa work on fly-throughs

• Going to do a resolution-convergence test for one of the runs, run out to frame 50 with 1 level more and 1 level less for the paper

High-res Hydro runs

• Asked Christina for location of hydro sims. She said she would work on getting me an account on her local machines. Will ping her again this week.

• She is teaching for the summer. I am wondering if I should offer to do the analysis for the hydro runs as well? Try to get 2 papers out on the CF, MHD and Hydro? Just thinking it would be nice to have a hydro version to compare to for the MHD effects..

• When should plan a visit to NC?

Allocations

I called NICS the other day as I was having trouble logging into the Kraken replacement machine. They said that our allocation time had expired just on that day, leaving >½ a million cpu hours. He said starting July 1st we have allocations starting on San Diego & Tack. Said I might need to be added to these grants.

Outreach

The horizons program for innercity youths is starting next Tuesday. This is an 8 session summer workshop series that we are participating in to teach kids about astronomy. Marissa, Jonathan, and I are going to lead workshops for the kids. We each can come up with our own lesson plans, but please let me know what you plan to do with them by Monday.

Bonnor Ebert Paper

The forms have been received and the paper is awaiting publishing

In the previous group meeting, we discussed the comparison plots I made with the different box sizes. Turns out, that by changing the GmX ratio you're affecting the resolution. So given the original ratio of dx = 1.5625, I took Erica's box size of L_x = 200 and applied a GmX of 128, yielding the same ratio. Here is the result.

​GIF

This is a test run. Constant mass loss rate at 10E-4 solar mass per year, the primary star's mass is 1 solar mass and the secondary is 0.5 solar mass. Separation is 3.999 AU, computation box is 32 AU * 32 AU * 16 AU(z dimension). Gas temperature is 2000K. According to Parker's solar wind solution, the sonic radius of the primary star is 2.5 AU. I specify the wind velocity at 0.5 AU which is 0.23v_{sound} thus can get the density at that radius from constant mass loss rate.

circular.gif​

Mass loss rate is 10E-4 solar mass per year, separation is 4 AU but it seems that the orbit is elliptical?

Binary.gif​

Ran a simple photo ablated clump model in 2D - AMR - parallel

Click here for a movie​

This latest run is at the full resolution (160 cells/rclump), with gamma = 5/3, and a separation distance d = 200 AU.

movie​

​http://quarknova.ucalgary.ca/MHD.html

• 3D heat diffusion solver: #243
• PN study: ​3D; will try on darter for scaling test

High-Res MHD Shear Flows

Here is my page on CF runs stats - ​http://astrobear.pas.rochester.edu/trac/wiki/u/erica/CFRunStatsHighRes

Shear15 - 63/200 done

movie​

Shear30 - 28/200

movie​

Shear60 - 39/200

movie​

Binary simulation encounter segmentation fault. (in the attachment) If I don't let the stars move, CFL number will be OK and will not trigger restart and the program is OK. But when I add velocity to stars, it encounter segmentation fault.

Stellar Winds shows that if we fix mass loss rate while changing wind density and wind velocity, the wind will not excite shock wave. Only when mass loss rate is increased will the successive winds generate a shock. (I had some problem on bluehive2 and can not make the movie, will ask for help, but this is the conclusion.)

Here we go, as I promised my last post:

​GIF

I have these lines in my .bashrc:

module purge

module load PrgEnv-intel

module load fftw mercurial cray-hdf5 cray-mpich

ulimit -s unlimited

The Hydro Shear 60, x = 200, y = z = 64 data set has the smallest height and width, so we're seeing whether or not the flow leaving the box was subsonic or supersonic as this case would be thought to be more problematic (due to it being more smooshed). To do this we consider the ratio between the magnitude of the velocity vector \vec{v}, with components v_{i}, and the sounds speed, c_{s}.

So we defined

v_i = \frac{p_{i}}{\rho}, \quad i = x, y, z

and

c_s = \sqrt{\frac{\gamma P}{\rho}}.

Note that P is the Hydro/3DPressure expression in VisIt, i.e.

P = (\gamma - 1)\left(E-\frac{1}{2}\frac{(p_{x}^2 + p_{y}^2 + p_{z}^2)}{\rho}\right),

and that \gamma = \frac{5}{3}.

​GIF

Given that fluid would be sonic at a ratio of 1, we can see that anything orange and red is super/sonic, otherwise it is subsonic. So a lot of the fluid leaving the box is subsonic, however the clumps we see being flung off from the interface of the colliding flows is supersonic. This might be because in this simulation we are taking the magnitude of the velocity, and not just in the z-direction.

​GIF

Above is a simulation for \frac{|v_{z}|}{c_{s}}, we still see a lot of subsonic fluid.

Objective: Going to make a 200x200x200 file, so by the 200th Chombo, the affects won't reach the boundary. Then I'll check the ratio of times for the different box sizes and determine how economical it'll be to have a box size that huge and to what extent we should shrink it.

Now I've included chombos 150 and 200.

With this one, I turned on hviscocity. Still gamma = 1.01, and all other parameters the same as before.

movie​

I also decided to run a different gamma to see what how different it looks. This one is gamma = 1.4.

movie​

Continuing with the previous most I made titled, "Why no sinks?!" Today I played around with some contour plots in VisIt to see if that could be any bit illuminating. The first picture is the 3D contour, looking down the barrel, comparable to the .bov file I attached in the previous post. Then the second image is a 2D slice from the same perspective.

GIF​

GIF

So I think that this might actually be a physical density. Going to make some longer plots as soon as I can.

Here I've made another visual illustrating box comparisons. The min: 0.5, max: 55. For chombo files 0, 50, 100. Now you can compare side-by-side more clearly.

Erica and I are hoping to decide what sizing we think works for the high res runs we're going to do. Similarly for fly-bys.

Consider ​this image that Erica posted on her ​Low Rest MHD Shear Flows page. Note how the Beta 1, Shear 60 simulation becomes the most density at its semi-major axes before any of the other sets.

GIF​

The other two Beta 1 cases (shear 15 and 30) see sinks occur beginning at around 16.5 and 17.17 Myr respectively, however the shear 60 case does not see sinks occur at all. My question is why don't we see sinks forming at those two regions?

As you can see by the picture above, by the end of the simulation the maximum is above 6000, and the minimum is ~24. The maximum is in the ball park for sinks to potentially form, however the minimum is not. Around 17Myr, the minimum is at ~28, and the max is at ~3000.

• Perhaps the density we observe here is physical. Maybe elongating the time on this simulation might result in sinks popping up. Perhaps it just takes them longer to form with a higher shear angle?
• Perhaps the density we observe is not actually physical, but a result of the nature of the .bov. Given the higher shear angle, the more elliptical the projection, and more area for the projection to pick up on at the semi-major axes. This might just be a area density the .bov is picking up on.

I was testing the program on bluehive2.

This is the wind vprofile when it reaches steady state. However, it is very weird that when I use color plot, it shows that the lowest velocity is 0.005691 which is exactly the boundary condition I specified. Besides that, a considerable region is below unity, but in the lineout plot, the lowest velocity is above unity.

The boundary condition would agree with the color plot, but why there is a difference?

The asymptotic behavior of the wind is very close to Parker's solution (

comparing the final velocity, the relative diff is only 5%

), the global quantitative difference has not been determined yet since there maybe a mistake in the lineout plot.

They are not very symmetric…

Erica asked me to do some tests for the box sizes in global.data with the CollidingFlows problem. Did some side-by-sides for the ~/Hydro_control/Shear60/ set (x = 62.5, y = z = 75), as well as Erica's ~/Testing_Grid_Domain/Shear60 (beta = 0.0, and x = 200, y = z = 64). I also figured maybe we're missing some affects in z, so I increased Erica's test box to x = 200, y = z = 80. Here are the comparisons done in VisIt:

So the first two images seem to be illustrating the same sort of physics — just that perhaps we can see more of what is happening with y = z = 80. The last image (x = 62.5) has the characteristic "hydro" wisps and fingers that we would expect, where as we don't really see these features in the others. The beta = 0 for all of them, and the physical boundary conditions are the same… not sure where these differences are coming from. Perhaps I'll play around more with some smaller sizing. Also a fly through will be coming to all of you soon~

Ambient

• density = 10³ 1/cc

• temperature = 10⁴ K

Clump

• density = 10⁷ 1/cc

• temperature = 10³ K

Wind

• density = 5 x 10³ 1/cc

• temperature = 10⁴ K

• velocity = 50 km/s

Also, gamma = 1.01, and the effective resolution is 160 cells/rclump.

movie​

The next image/movie is the same run, but with a different color bar.

movie​

Fixed Grid

No AMR, still 160 cells/rclump. I made 2 different movies of the same run, both are of density. The only difference is the color bar.

movie​

movie​

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

is solved by Newton-Raphson method.

This is the velocity profile

• Last week I mainly worked on fixing my 2.5D pulsed jet simulations, and regenerating the figures for the paper.

• I have also been working on the Al colliding flows problem with Francisco. The most recent simulation looks better than this one from last week: ​http://astrobear.pas.rochester.edu/trac/blog/ehansen06072014.
  • The small scale structures and explosion went away when I imposed a static mesh.
  • However, there is still an odd asymmetry.

movie​

• I am also working on the Mach stems project. I need to get some good runs completed so I can write the HEDLA proceedings. Right now, I am simply running a wind over a clump to see if I can get a nice bow shock. My previous simulations looked somewhat strange. This is queued up on bluestreak.

Ablative RT

1. 2D: fixed an error in periodic boundary condition. ​new result
2. 3D: ​http://astrobear.pas.rochester.edu/trac/wiki/u/bliu

I partly resolve the firewall issue by using VPN, but the connectivity is still poor.

I read Chapter 4,5 of Astrophysics in a nutshell and get some basic idea and questions about star formation, PN and PPN. It is interesting that the birth and death of stars in a molecular cloud is connected by turbulence, magnetic field and radiation.

I did not manage to see the Dean of Astronomy of Beijing Normal University because I felt uncomfortable in Beijing after staying for 5 days and went back to Wuhan.

Below is the research in stellar wind:

The next goal is to change the intensity of radiation force. Restate the governing equation first. To change the radiation force means to change the sonic radius and perhaps the temperature.

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

The first way is to solve this algebraic equation by Newton-Raphson method (This is one of the many methods). A given r will correspond to 2 v values as solutions. Then plug the r vs. v, r vs. \rho profile (from the wind base up to sonic point) back to outflow source and then change the wind.

Another way is to solve the hydrostatic equation:

\nabla P=-\frac{GM}{r^2}

M is constant mass because I assume the wind base is at moderate radius s.t. the inner mass dominate. Use isothermal assumption:

P=\frac{2\rho k T}{m_p}

The gas is assumed to be consisted of fully ionized hydrogen. m_p is the proton mass and \rho is the gas density.

The final solution of \rho is:

\rho=\rho_0 Exp(\frac{m_p G M}{2 k T}(\frac{1}{r}-\frac{1}{r_0}))

where \rho_0=\rho(r_0)

Hydrostatic solution is appropriate only when \nabla P \gg \rho u \cdot \nabla u i.e. in much inner region than the sonic radius.

The strategy is to specify the wind base v and \rho profile and let AstroBEAR build self consistent atmosphere and wind via Riemann solver.

Remark: People may think the first way is easier and more computationally economic, but the second way also has certain advantage: it is more self consistent, we can further specify the nonuniform magnetic field at the wind base. If we go deeper into the AGB star I think we will touch the regime of dynamo theory and polytrope.

The adiabatic case is not very interesting, so I only posted the density movie. The flows collide and form reflection shocks which propagate all the way to the boundaries.

adiabatic density movie​

I ran the cooling case for longer time. Now each frame is equal to 10 ns. There is a lot going on in this simulation. The small structures that form due to cooling look nice and are believable. However, there is a strange explosion near the end of the simulation. My intuition tells me that this is not physical, but we should discuss this further and figure out why the code produced this result.

There is some asymmetry between the top and bottom halves of the grid which is also curious. The inflow at both of the boundaries are the same. It looks like the bottom half of the grid cools more than the top half.

movie​

movie​

I also included an enlarged image of the first frame of the explosion. The image shows pressure, temperature, velocity, and density. This might have been caused by AMR effects, so I will try a run that has a static fine mesh in the collision region.

Resolution is twice as high as the old runs, and the cooling is stronger because of the error that I fixed in the temperature calculation.

It is interesting that the hydro and beta = 5 cases are so different. You might expect the runs to approach the hydro case as you increase beta, but this is not what we see here.

The runs are (from left to right): beta = inf (hydro), 5, 1, 0.4.

movie​

Here are some CDMs for beta 10 data that I got done today. Done for all shear angles in x-field. You'll notice that sink particles form toward the end of each gif (when we'd expect star formation).

GIF​

GIF​

GIF​

Objectives:

• CDM plots with sink parks for beta 1: 15, 30 in x, as well as hydro.
• CDM plots for the other two shear angles to help complete the page Erica is creating.
• Fly through movie for AstroBEAR YT channel with some of these x-field plots.

This is the same setup as before: ​http://astrobear.pas.rochester.edu/trac/blog/ehansen05302014.

The only difference is this is with an updated cooling table, so I had to change the scaling on the legends. I also turned on h_viscosity to help get rid of any nonphysical carbuncles.

movie​

movie​

Vx, Linear-Linear
Vx, Log-Linear

The equation of \alpha\sqrt{kg}-\beta kv_a was found in Takabe's paper: ​http://scitation.aip.org/content/aip/journal/pof1/28/12/10.1063/1.865099

with \alpha=0.9, \beta=3-4, v_{a}=3000m/s

plots of density, velocity and temperature.

shell density: 10⁴ /cc, velocity: 50 km/s, temperature: 10 K, width: 0.005 pc

wind density: 1.0 /cc, velocity: 150 km/s, temperature: 10⁶ K, gradually slowing down (you can see the slight slope of the wind velocity at the final frame, however the linear decrease spans 20 pc, our box is only 0.5 pc so no noticeable slow down can be observed.

density: in log-linear plot. normalized to the shell density (10⁴ /cc)

temperature: in log-linear plot. normalized to the wind temperature (10⁶ K)

velocity: in linear-linear plot. normalized to wind velocity (150 km/s)

under the new shell and wind speed, the clump crushing time is around 10kyrs, so the following movie shows the shock structure over 1 clump crushing time.

This looks OK in general however one thing I'm not sure is that there is an expanding buffer region between the shell and the wind.

movie:

​http://www.pas.rochester.edu/~shuleli/tsf_new_shock/shock_structure.gif

same assumption as above, with 0.035 pc shell:

​http://www.pas.rochester.edu/~shuleli/tsf_new_shock/shock2.gif

​movie

Here's a new simulation setup for the next round of TSF sims (rho and temperature scales are log10 based):

For these, the wind velocity is subject to an exponential decrease since we have linear feature for the velocity-distance plot (Chevalier 1974), we can compute the time dependence by:

dv/ds = -c (where c is some slope counting from the shell front to the left)

thus dv/dt = dv/ds ds/dt = dv/ds v = -cv

thus v is an exponential decay on t: v = v0exp(-c t) where v0 is the initial wind velocity 150 km/s, c is the slope in the velocity-distance plot, which is about 7.884 if we count t in million years. This leads to the wind drops to subsonic at about 0.11 million years. In the code, the wind injection is shut off when Mach number of the wind drops below 1.

Also, I found in the previous paper the Mach 1.5 is using 2x ambient density while the Mach 3 run is using 4x ambient density. So I'm inclined to redo the Mach 3 run to make the comparison even for the revision. The rest of the revision are done.

• High resolution 2.5-D pulsed jet sims are finished. I am now just working on regenerating the figures for the paper. Below is an image and movie comparing the old hydro run to the new hydro run.

movie​

• I'm having trouble running the 2.5-D random velocity model on stampede. It always quits before it starts time-stepping, and I'm not sure why. It should only take a day or two to run once I figure out the problem.

• Colliding flows simulations are working: ​http://astrobear.pas.rochester.edu/trac/blog/ehansen05302014
  • Francisco sent me an updated cooling table this morning which I will implement, and then I will rerun the same setup.

• Ablative RT

To study the perturbation growth rate, the difference between the front positions along the center line and edge is calculated and plotted versus time. This seems different from the three stages of the normal RT instability: the exponential growth stage ends when the bubble starts. Not sure it's just because ablation or something wrong.

Middle line, Linear-Linear, 200 Extended zones
Middle line, Log-Linear, 200 Extended zones
Quarter line, Linear-Linear, 200 Extended zones
Middle line, Linear-Linear,5 Extended zones

• PN Jets: test Martin's 2.5D module on Stampede

​movie

• New users Report: Czarships/NewUsers/AskingForCode

China has firewall against USA. I can not login into bluehive or alfalfa as smooth as in USA. I have downloaded some astrobear f90 files to read, I am reading Clumps and PlanetaryAtmospheres to learn how to generate density, velocity profiles.

I am also reading a book: Astrophysics in a nutshell. I have read 3 chapters. It is interesting and helpful in understanding the process of radiation and binary evolution (as in chapter 4)

here is its amazon link.

​http://www.amazon.com/Astrophysics-Nutshell-In-Princeton/dp/0691125848/ref=sr_1_1?ie=UTF8&qid=1401716251&sr=8-1&keywords=astrophysics+in+a+nutshell

I am now in Tsinghua University (meeting friends and learning here) and will go to Beijing Normal University next week to meet Zonghong Zhu. He is the dean of Astronomy of Beijing Normal University.

​http://astrowww.bnu.edu.cn/CN/index.php/2010-04-28-10-49-23/122-zzh