Updating Documented Examples#459
Conversation
|
To get this working, you are going to need to add "sphinxcontrib-video" to the list of python-dependencies are installed as part of our "docs" dependency-group (this list is tracked as part of the Here is an exact link to the line in pyproject.toml where this list is defined |
| **Important:** This test must be run with diode boundaries [disabled](https://github.com/alwinm/cholla/tree/main-diode) in order to perform as expected (thank you @alwinm!). This branch also uses the Van Leer integrator. | ||
|
|
||
| ## Parameter file: (**modified** from `cholla/examples/1D/Shu_Osher.txt`) | ||
| Modified to add yl_bcnd, yu_bcnd, zl_bcnd, and zu_bcnd=0. Xmin changed to -1.0 and xlen to 2.0. |
There was a problem hiding this comment.
I think this has been fixed, so this line is no longer needed.
There was a problem hiding this comment.
This line should be removed - the parameter file in the repository matches the one shown here. Also remove the header formatting of the parameter file line so it is not so large, and matches the formatting in the 1d blast test, for example.
| This test initializes a square wave density pertubation. The setup consists of an initial density and pressure of 1.0 and 0.01, respectively. A square wave is initialized with amplitude 1.5. Gamma is set to 1.666666666666667. This test was performed with the hydro build (`cholla/builds/make.type.hydro`) and Van Leer integrator. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Square_Wave()`. | ||
|
|
||
| ## Parameter file: (modified from`cholla/examples/1D/square_wave.txt`) | ||
| Modified to add yl_bcnd, yu_bcnd, zl_bcnd, and zu_bcnd=0 |
There was a problem hiding this comment.
Same two comments here
| This test is similar to the Sod shock tube but has higher initial pressure and density differences. This shows the ability of a code to limit oscillatory behavior in areas of high density and pressure contrasts. The setup consists of a density and pressure of 10.0 and 100.0, respectively, for 0 \< x \< 0.5 and density= pressure = 1.0 for 0.5 \< x \< 1.0. Gamma is set to 1.4. This test was performed with the hydro build (`cholla/builds/make.type.hydro`) and Van Leer integrator. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Riemann()`. | ||
|
|
||
| ## Parameter file: (modified from`cholla/examples/1D/strong_shock.txt`) | ||
| Modified to add yl_bcnd, yu_bcnd, zl_bcnd, and zu_bcnd=0 |
There was a problem hiding this comment.
Same two comments here.
|
Looking good so far (and the movies show up now, hooray!). A couple notes: |
| @@ -0,0 +1,23 @@ | |||
| # 1D Shu and Osher test | |||
| This test (Shu & Osher, 1989) highlights the ability of a code to resolve small scale smooth flow and shocks simultaneously. Further, it shows how lower resolution solutions can cut off some of the amplitude of maxima due to the slope limiters. Parameters are from Stone et al., 2008, Section 8.1. The test consists of left and right states separated at x = -0.8. On the left, density is set to 3.857143, pressure to 10.33333, and velocity to 2.629369. On the right, density is sinusoidally varying: $\rho(x)$ = 1.0 + 0.2 $\sin(5.0\pi x)$. Pressure is set to 1.0 and the velocity is 0.0. Gamma is set to 1.4. This test is performed with the hydro build (`cholla/builds/make.type.hydro`). Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Shu_Osher()`. | |||
|
|
|||
There was a problem hiding this comment.
The warning about diode boundaries that is in the wiki page for this test should be added here. See https://github.com/cholla-hydro/cholla/wiki/1D-Shu-Osher
| # 1D Sod Shock Tube | ||
| This test highlights the ability of a code to resolve shocks and contact discontinuities over a narrow region. Parameters from Sod (1978). The setup consists of a density and pressure of 1.0 for x \< 0 and 0.1 for x \> 0.5. Gamma is set to 1.4. This test was performed with the hydro build (`cholla/builds/make.type.hydro`) and Van Leer integrator. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Riemann()`. | ||
|
|
||
| ## Parameter file: ({repository-file}`examples/1D/sod.txt`) |
There was a problem hiding this comment.
Please update the formatting here to match that on the 1D blast page (link to the parameter file on a line with normal font size).
| # value of gamma | ||
| gamma=1.4 | ||
| ``` | ||
| Upon completion, you should obtain two output files.The initial and final density, pressure, velocity, and internal energy (in code units) of the solution is shown below. Examples of how to extract and plot data can be found in `cholla/python_scripts/plot_sod.ipynb`. |
There was a problem hiding this comment.
Only the density, velocity, and pressure are shown now.
There was a problem hiding this comment.
Also, is that the correct path for the plotting example? Or do you want to reference the new markdown pages you added in the Python Analysis Examples tab?
| This test initializes a compression/rarefaction wave across the grid. The setup consists of an initial density and pressure of 1.0 and 0.6, respectively. A sound wave is initialized as a sinusoidal wave with amplitude 1e-4 and wavelength of 1.0. Gamma is set to 1.666666666666667. This test was performed with the hydro build (`cholla/builds/make.type.hydro`) and Van Leer integrator. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Sound_Wave()`. | ||
|
|
||
| ## Parameter file: ({repository-file}`examples/1D/sound_wave.txt`) | ||
| Modified to add yl_bcnd, yu_bcnd, zl_bcnd, and zu_bcnd=0 |
There was a problem hiding this comment.
Can remove the line about modification; please also update the formatting of the parameter file line to match that in the 1D blast page.
| gamma=1.666666666666667 | ||
|
|
||
| ``` | ||
| Upon completion, you should obtain five output files. By changing the outstep to 0.005, we can obtain the evolution of the density, pressure, velocity, and internal energy (here at 4 fps). Examples of how to extract and plot data can be found in `cholla/python_scripts/plot_sod.ipynb`. |
There was a problem hiding this comment.
Only density, velocity, and pressure are plotted now.
There was a problem hiding this comment.
Also make sure the path for the plotting script reference is correct.
| :align: center | ||
| :autoplay: | ||
| :loop: | ||
| ::: |
There was a problem hiding this comment.
If we're going to show a movie on this page it would be nice to show the evolution of the wave for a full period. So perhaps instead of modifying the outstep to 0.005, you could modify the final output time to 1.0.
There was a problem hiding this comment.
It seems like there might be an extra movie file in this folder that can be removed.
There was a problem hiding this comment.
Is this parameter file being used? If not it can be removed.
| # value of gamma | ||
| gamma=1.666666666666667 | ||
| ``` | ||
| Upon completion, you should obtain 101 output files. We can obtain the evolution of the density. Pressure is constant to the $10^{-14}$ level. Examples of how to extract and plot data can be found in `cholla/python_scripts/plot_sod.ipynb`. |
There was a problem hiding this comment.
Maybe instead of "we can obtain the evolution of the density" you could say something like "the evolution of the density is shown below".
| # value of gamma | ||
| gamma=1.4 | ||
| ``` | ||
| Upon completion, you should obtain two output files. The initial and final density, pressure, and velocity (in code units) of the solution is shown below.. Examples of how to extract and plot data can be found in cholla/python_scripts/plot_sod.ipynb. |
There was a problem hiding this comment.
Maybe "density, velocity, and pressure" to reflect the order of the plot.
| This test highlights a collision between two shocks. This test is from Toro's *Riemann solvers and numerical methods for fluid dynamics* Sec. 6.4, test 4.The test consists of left and right states separated at x = 0.4 with velocities 19.5975 and -6.19633, respectively. Density of the left side is 5.99924 and pressure is 460.894. For the right side, density is 5.99242 while pressure is 46.095. Gamma is set to 1.4. This test is performed with the hydro build (`cholla/builds/make.type.hydro`). Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Riemann()`. | ||
|
|
||
|
|
||
| ## Parameter file: (**modified** to add y and z boundary conditions = 0 from {repository-file}`examples/1D/two_shocks.txt`) |
There was a problem hiding this comment.
Same comment about formatting here.
| # value of gamma | ||
| gamma=1.4 | ||
| ``` | ||
| Upon completion, you should obtain 2 output files. The initial and final density, pressure, velocity, and internal energy (in code units) of the solution is shown below. |
There was a problem hiding this comment.
Something looks badly wrong with this test. Looks like it's a problem with the diode boundaries again: https://github.com/cholla-hydro/cholla/wiki/1D-Two-Shocks. I think maybe we should not include the new plots on these, and instead include the warning at the top, and use the old plots from the wiki (until we add the diode disabling back in).
There was a problem hiding this comment.
This file can be removed if it's not being used.
There was a problem hiding this comment.
Same for this one.
There was a problem hiding this comment.
Same for this file.
| # 2D Implosion | ||
| This tests a code's ability to resolve contact discontinuities for a non-grid aligned shock tube and can determine the symmetry of a code. It consists of a diamond region of density = 0.125 and pressure = 0.14 inside a square region of both density and pressure of 1. Parameters from Liska and Wendroff 2003. Gamma is set to 1.4. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Implosion_2D()`.This test is performed with the default hydro build (`cholla/builds/make.type.hydro`). | ||
|
|
||
| ## Parameter file: ({repository-file}`examples/2D/Implosion_2D.txt`) |
There was a problem hiding this comment.
Same comment here on formatting.
| # path to output directory | ||
| outdir=./ | ||
| ``` | ||
| Upon completion, you should obtain 251 output files. An evolution of the pressure overlaid with density contours is seen here at 10 fps. Examples of how to plot projections and slices can be found in `cholla/python_scripts/Projection_Slice_Tutorial.ipynb`. |
There was a problem hiding this comment.
I think here you want to reference your 2D plotting example, not the projection and slice examples (since those don't exist for 2D datasets).
| This test highlights the ability of a code to resolve mixing caused by shear flows, emphasizing the importance of an efficient, high order reconstuction method and a fast code. The level of mixing increases with the resolution. See Chandrasekhar 1961. It consists of a high density region (density = 1 and x velocity = 0.5) in the middle third of the grid sandwiched between two low density regions (density = 1 and x velocity = -0.5). Y-velocities are perturbed by a sine wave of amplitude 0.1. Pressure is 2.5 everywhere and gamma is set to 1.666667. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `KH()`.This test is performed with the default hydro build (`cholla/builds/make.type.hydro`) and Van Leer integrator. | ||
|
|
||
|
|
||
| ## Parameter file: ({repository-file}`examples/2D/KH_discontinuous_2D.txt`) |
There was a problem hiding this comment.
Same comment about the parameter file line formatting.
| # path to output directory | ||
| outdir=./ | ||
| ``` | ||
| Upon completion, you should obtain 201 output files. The initial and final density (in code units) is shown below. Examples of how to plot projections and slices can be found in `cholla/python_scripts/Projection_Slice_Tutorial.ipynb`. |
There was a problem hiding this comment.
Same comment here about the plotting script reference.
| :loop: | ||
| ::: | ||
|
|
||
| We can compare to the results of the [Athena](https://www.astro.princeton.edu/~jstone/Athena/tests/kh/kh.html) code on a 512x512 grid at t = 1.00. Density is shown on a linear color map between 0.9 and 2.1 |
There was a problem hiding this comment.
I think we can get rid of the Athena comparison here (it's not exactly the same test, so it's a little confusing).
| @@ -0,0 +1,71 @@ | |||
| # 2D Resolution Independent Kelvin-Helmholtz Test | |||
| This test highlights the ability of a code to resolve mixing caused by shear flows, emphasizing the importance of an efficient, high order reconstuction method and a fast code. In general, the level of mixing would increase with the resolution; however, this is a resolution independent version. The test consists of a region of higher density (100) sandwiched between two regions of lower density (1.0). The high density layer has a velocity of 10.5 and the low density layers have a velocity of 9.5. Pressure is set to 2.5 and a 1% pertubation (amplitude of 0.1) is added to the high density layer to provoke the instability. Gamma is set to 1.6666666666666667. This test is performed with the default hydro build (`cholla/builds/make.type.hydro`) and Van Leer integrator. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `KH_res_ind()`. | |||
There was a problem hiding this comment.
Can you add a reference here to Robertson 2010, after "this is a resolution independent version inspired by the test presented in Robertson et al. 2010".
| :::{figure} snapshots_khri_256.png | ||
| ::: | ||
|
|
||
| The solution does appear to be independent of resolution. We can also compare the time evolution of the density at the above resolutions. |
There was a problem hiding this comment.
I had forgotten that there was not much density contrast in these initial conditions. It may be worth it for us to adjust the initial conditions in the code. In the meantime, I think picking a lower maximum density for the color bar (perhaps 100) would help make the instability more visible.
| # 2D Noh Strong Shock | ||
| This test highlights the ability of a code to track a high Mach number shock. Parameters from Liska and Wendroff 2003. The test consists of an infinitely strong circular shock radiating from the origin. There is initially a constant density of 1.0 across the grid and a velocity of 1.0 towards the center everywhere. Pressure is set to $10^{-6}$ everywhere. Gamma is set to 1.66666667. This test is performed with the default hydro build (`cholla/builds/make.type.hydro`) and Van Leer integrator. Full initial conditions can be found in `cholla/src/grid/initial_conditions.cpp`under `Noh_2D()`. | ||
|
|
||
| ## Parameter file: ({repository-file}`examples/2D/Noh_2D.txt`) |
There was a problem hiding this comment.
Same comment about the parameter file formatting and removing the comment about boundaries.
|
|
||
| ## Parameter file: | ||
| ## Parameter file:({repository-file}`examples/2D/Rayleigh_Taylor.txt`) | ||
|
|
There was a problem hiding this comment.
You can just refer tp the parameter file normally here, since this documentation is tied to the dev branch.
|
|
||
| Cholla utilizes finite volume methods (FVM) to solve the partial differential equations related to various hydrodynamics problems. In FVMs, the average values of each conserved quantity (i.e. density, momentum, energy) are stored in each cell. However, the cell-average quantities are updated in time using fluxes, which are calculated at cell interfaces. The reconstruction method determines what the input values for solving the Riemann problem at cell interfaces should be. | ||
|
|
||
| This section will evaluate two piecewise linear methods (PLM) and one piecewise parabolic method (PPM). PLM is a second order reconstruction that models the shape of the fluid linearly, and thus representing the fluid more accurately. PPM is a third order reconstruction method that models the fluids parabolically and can be useful in problems where PLM proves to be too diffusive. |
There was a problem hiding this comment.
Maybe "and thus represents the fluid more accurately than a first-order method using the cell average values".
docs/sphinx/ChollaExamples/index.md
Outdated
| 2D-Rayleigh-Taylor/2D-Rayleigh-Taylor | ||
| 2D-Kelvin-Helmholtz-Discontinuous/overview | ||
| 2D-Kelvin-Helmholtz-Resolution-Independent/overview | ||
| Reconstruction-Methods/overview |
There was a problem hiding this comment.
Maybe adjust this order so the Reconstruction methods comes after the new 3D adiabatic test.
3 participants
1D Square
1D Shock
1D Blast
1D Shu Osher