Use pyamrex in spacecraft charging example #4817
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
RemiLehe
changed the title
ax3l
reviewed
Sep 25, 2024
Examples/Physics_applications/spacecraft_charging/inputs_test_rz_spacecraft_charging_picmi.py
Outdated
Show resolved
Hide resolved
ax3l
reviewed
Jan 24, 2025
Examples/Physics_applications/spacecraft_charging/inputs_test_rz_spacecraft_charging_picmi.py
Outdated
Show resolved
Hide resolved
Until we add `__len__` via AMReX-Codes/pyamrex#407
ax3l
approved these changes
Jan 25, 2025
Use the one that does not rely on naming conventions but has named arguments. (The other one worked as well, but is deprecated.)
RemiLehe
commented
Jan 28, 2025
Examples/Physics_applications/spacecraft_charging/inputs_test_rz_spacecraft_charging_picmi.py
Outdated
Show resolved
Hide resolved
…rz_spacecraft_charging_picmi.py
ax3l
added a commit
to AMReX-Codes/pyamrex
that referenced
this pull request
Jan 29, 2025
We already have `.size` and we can iterate the `FabArray`, so we should support `len(mf)`, too. Seen in BLAST-WarpX/warpx#4817
ax3l
reviewed
Jan 29, 2025
| 2 * len(phi) | ||
| nr = phi.shape[0] | ||
| r = np.linspace(rmin, rmax, nr, endpoint=False) + (rmax - rmin) / ( | ||
| 2 * nr | ||
| ) # shift of the r points because the derivaties are calculated in the middle | ||
| face_z0 = ( | ||
| 2 * np.pi * 1.0 / dz * ((phi[:, 0] - phi[:, 1]) * r).sum() * dr |
There was a problem hiding this comment.
Looks correct/as before: the : is without ghost cells, () is with ghost cells.
https://pyamrex.readthedocs.io/en/latest/usage/compute.html
ax3l
reviewed
Jan 29, 2025
Comment on lines
+81
to
+84
| self.normalized_Er = warpx.multifab("Efield_fp", dir=self.dir_r, level=0).copy() | ||
| self.normalized_Er.mult(1 / q_v, 0) | ||
| self.normalized_Ez = warpx.multifab("Efield_fp", dir=self.dir_z, level=0).copy() | ||
| self.normalized_Ez.mult(1 / q_v, 0) |
There was a problem hiding this comment.
This is now perfectly parallel in MPI contexts, no global comms.
ax3l
reviewed
Jan 29, 2025
| phi = PhiFPWrapper(include_ghosts=True) | ||
| phi[...] += (q - q_v) * self.normalized_phi | ||
| Er = warpx.multifab("Efield_fp", dir=self.dir_r, level=0) | ||
| Er.saxpy(Er, (q - q_v), self.normalized_Er, 0, 0, 1, 0) |
There was a problem hiding this comment.
This is now perfectly parallel in MPI contexts, no global comms.
ax3l
reviewed
Jan 29, 2025
Examples/Physics_applications/spacecraft_charging/inputs_test_rz_spacecraft_charging_picmi.py
Outdated
Show resolved
Hide resolved
…rz_spacecraft_charging_picmi.py Co-authored-by: Axel Huebl <[email protected]>
dpgrote
reviewed
Apr 18, 2025
| r = np.linspace(rmin, rmax, len(phi), endpoint=False) + (rmax - rmin) / ( | ||
| 2 * len(phi) | ||
| nr = phi.shape[0] | ||
| r = np.linspace(rmin, rmax, nr, endpoint=False) + (rmax - rmin) / ( |
There was a problem hiding this comment.
This works, but I had to stare at it for a moment to figure out what it is doing. I'll often do
nr = phi.shape[0] - 1
dr = (rmax - rmin)/nr
r = np.linspace(rmin + dr/2., rmax - dr/2., nr)
atmyers
pushed a commit
to atmyers/WarpX
that referenced
this pull request
Jul 3, 2025
# Overview This PR modernizes the spacecraft charging example to use the newer `pyamrex` interface for multifabs, instead of the fields wrapper. The `pyamrex` interface is in principle more flexible (since it gives access to the AMReX functions that operate on `MultiFabs` and should have all the functionalities that the old field wrappers had. (See also AMReX-Codes/pyamrex#370). This is also the interface that is documented in the [WarpX documentation](https://warpx.readthedocs.io/en/latest/usage/workflows/python_extend.html#fields) # Resetting the checksum I had to reset the checksum, as they differed by `~1e-2`, which is not negligible! Manually looking at the fields seem to indicate that the old interface and new interface differ in how they were handling the guard cells (at least in the way I use them in this test). While I don't fully understand the differences, this PR seems to produce more correct results. (In hindsight, it is striking that we missed the unphysical `Ez` field when we implemented this test. The reason might be that we were focusing `phi`, which does not exhibit this unphysical feature in such an obvious way.) <img width="1251" alt="Screenshot 2025-04-18 at 8 55 31 AM" src="https://github.com/user-attachments/assets/683235aa-c03b-425c-8b55-d98016d77bdb" /> --------- Co-authored-by: Axel Huebl <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Overview
This PR modernizes the spacecraft charging example to use the newer
pyamrexinterface for multifabs, instead of the fields wrapper.The
pyamrexinterface is in principle more flexible (since it gives access to the AMReX functions that operate onMultiFabsand should have all the functionalities that the old field wrappers had. (See also AMReX-Codes/pyamrex#370). This is also the interface that is documented in the WarpX documentationResetting the checksum
I had to reset the checksum, as they differed by
~1e-2, which is not negligible!Manually looking at the fields seem to indicate that the old interface and new interface differ in how they were handling the guard cells (at least in the way I use them in this test). While I don't fully understand the differences, this PR seems to produce more correct results.
(In hindsight, it is striking that we missed the unphysical
Ezfield when we implemented this test. The reason might be that we were focusingphi, which does not exhibit this unphysical feature in such an obvious way.)