Error using h_misalign for Laue diffraction

[luweb]

Hello, I am trying to implement h_misalign for Laue diffraction and I am getting the following error:   [INFO] bmad_parser: Parsing lattice file(s). This might take a minute or so... [FATAL | 2024-JUL-02 12:07:26] lat_sanity_check: ELEMENT: CRYST1 (3) HAS GRID TYPE H_MISALIGN BUT THIS IS NOT IMPLEMENTED FOR LAUE DIFFRACTION!

I just wanted to double check with you if this is correct and h_misalign is not implemented for Laue diffraction. Is there a workaround?

The reason for using h_misalign is that we think that “b_param” is set to a fixed value for the center photon energy but for a spectrometer it varies with energy. So, perhaps instead it would make sense to set the asymmetry angle, which is fixed for the crystal, if possible.

Thank you for your attention.

[DavidSagan]

@luweb It looks like the problem is that the code was never extended from the Bragg case where the H vector is primarily pointing in the z-direction to the Laue case where the H vector is primarily pointing in the x-direction. I will work on this.

[luweb]

Thank you for looking into this @DavidSagan.

[DavidSagan]

@luweb I modified the code to allow h_misalign with Laue. I also removed the restriction that h_misalign and displacement cannot be used at the same time. Finally, I found and fixed a bug where a surface grid with negative index for the lower bound was not getting handled correctly. This was a substantial modification of the code so I would suggest you make some runs and compare the output of the old code with the new code to make sure no bug has accidentally crept in.

[sstoupin]

Thank you, David. A recipe to calculate rot_y and rot_z can be helpful. We have a reciprocal vector (hx,hy,hz) defined on a crystal surface in local crystal coordinates (origin at center ray). If I use rot_y = arcsin(hz), rot_z = arcsin(hy) to calculate H_Misalign grid and use b_param = 1 I get some sagittal focusing but something is not right: the rays don't cross in the meridional plane where they should (our crystal also has curvature in the meridional plane).

[DavidSagan]

@sstoupin H_misalign misaligns from the H calculated without any misalignment. Also the nominal H is in the negative x-direction so the formula rot_y = -arcsin(hz) is only approximate.