Closed bruhwiler closed 1 year ago
Based on the email received today from Jeroen van Tilborg, I believe it's important to first implement the 'enhanced' propagation algorithm, where n2 is constant out to a specified radius, but then drops to zero.
JVT wants to compare the simulated wavefront with his WFS data. These comparisons are very problematic, if the simulation assumes linear focussing, based on an n2 value that is independent of radius.
The longitudinal variation is implemented in the compare_experiment
notebook.
The radial variation is only implemented for the n0n2_srw
propagator at this point. I haven't looked at the others yet and will wait so we can merge this today.
This could be a new notebook.
Take the n2 values from the RHS plot in Boaz's presentation, where there is a max value of approximately 30, which then drops exponentially. Use enough crystal slices to be reasonably sure it's converged.