Commit graph

21 commits

Author SHA1 Message Date
bc2c29092a Injection of particles from surfaces direction
Now, if no normal is provided to an injection in the input file, the
velocity direction of the particles is chosen to be the surface normal.

This allows to inject particles from curves, corners... without having
to provide a direction or declaring multiple injections.
2023-02-03 14:03:22 +01:00
7ce1b7a4dd Reducing overhead when no collisions are present
Particles are added to lists only if there are MCC collisions. Hopefully
this will reduce overhead when OpenMP is used and no collisions are
active.
2023-01-07 12:12:37 +01:00
2486ef6316 Reduction in pushing
Reduction in 10-20% of time spend in pushing in 2DCyl thanks to
rewriting fPsi and dPsi.
2023-01-05 16:47:13 +01:00
c7d205e4dd fix possible issue in particle injection 2022-12-12 17:57:18 +01:00
924ba4e20e Alphie grid case and issues
Output for the example ALPHIE_Grid.

Found an issue when multiple injections were used with species with
different time steps.

Modification to the way to compute the ionization boundary:
  The maximum number of ionizations is computed by eRel/eThreshold
  (relative energy / threshold of ionization)
  For each possible ionization, the probability of ionization is
  computed based on the density of neutrals, cross section and effective
  time divided by the number of maximum ionizations.
  If an ionization takes place, the ionization energy is substracted
  from the relative energy.
2022-12-11 22:39:28 +01:00
23e2fe9bae I'm dying with hay fever but I have to commit
I'm feeling awful but I have work in my desktop that I need to commit so
I can work with my laptop while I'm at the IEPC 2022 in Boston.
2022-06-10 16:07:14 +02:00
35bd61fda9 Common scatter subroutine
All subroutines of scattering particle properties to the nodes of a
volume have been converged into one in moduleMesh.
2021-04-21 23:40:58 +02:00
7dccba5906 The normal vector in an injection is now normalized to 1 and it is no
longer responsability of the user to ensure that the norm is 1.

This is a very simple fix and I do not why I did not implemented early.
2021-04-14 20:56:53 +02:00
a2631f6b78 Impliementation of a collision mesh which is independent for the mesh
used to scatter particles and compute the EM field.
2021-04-03 09:20:46 +02:00
ec128902ad The integer part%sp that referenced the species index has been
substituted for a pointer to the species.
2021-03-28 15:55:26 +02:00
2a843547b8 Ionization boundary condition fully tested.
Documentation updated properly.

3D Cartesian geometry also tested.
Documentation updated properly.

Added weighting probability in the injection of particles.
2021-03-27 11:38:18 +01:00
db6b0a2c03 Fixed an issue with reflection of particles in all geometries and also
assigning the normal vector in 2D and 3D.

3D Cartesian geometry is working properly, although it needs testing.

Still issue with ionization boundary.
2021-03-22 12:39:34 +01:00
3b01099aba Merge branch 'development' into feature/3DCart 2021-03-20 13:08:55 +01:00
8cf50cda68 First implementation of ionization boundary. Still some work to do. 2021-03-20 13:08:01 +01:00
c236c5e0e2 Fixing some issues in 3D Cartesian coordinates. Included 3D pusher.
Still there are issues linking a volume to an edge.
2021-03-15 10:00:34 +01:00
20bc1abc29 Fixed an issue in which the logical coordinates of the particle (xi)
were not being initialized and was causing particles to get stuck in
  the iteration loop.
2021-01-27 09:59:37 +01:00
5516e46ec3 Particles are injected in each iteration and pushed with tauMin when
injected. Then, particles are pushed as normal in their own time step.
2021-01-02 16:31:31 +01:00
9e0d1a7cc7 Final implementation of ionization process by electron impact.
Possibility to input initial species distributions (density, velocity
    and temperature) via an input file for each species.

New moduleRandom includes function to generate random numbers in
different ways (still uses) the implicit RANDOM_NUMBER().
2020-12-26 22:45:55 +01:00
e50cc3325b Implementation of ionization process.
Now collisions can have a different time step.

Added species name to output names as it was starting to get confusing
in Gmsh for multiple species.

Output filenames adapted to match any number of iterations.
2020-12-25 23:08:59 +01:00
37b0139b1f Implementation of different distribution functions for velocities.
Maxwellian and Diract Delta distributions have been implemented.

The input for injection of particles should be rewritten to allow more
clear input file.
2020-12-13 13:56:48 +01:00
af74205932 Files renamed and makefile make compatible with ifort. 2020-12-10 19:25:17 +01:00