Documentation: Synrad facet parameters

Image

Sticking factor: The probability (between 0 and 1) that an incident photon will be absorbed on the selected facet(s).

Reflection:

Diffuse: the reflected photon will be leaving the surface with phi angle with cos(phi) probability (Lambertian radiator)

Mirror: the reflected photon will leave the surface with the incident angle

Material (copper, aluminium, Al2O3): The reflection probability (1-sticking_factor) depends on the photon's energy and incident angle according to the following eperimental data. Forward- and backscattering is optionally supported, and you can define material reflection tables yourself (CSV files in the parameters directory)

Image

Roughness:

You can optionally enable rough surface scattering.

In Tools / Global Settings, you can choose between the old and new reflection models.

In the old (default) reflection model, reflection is specular (mirror-like), but as surface roughness is taken into account, the theta and phi angles (between the reflected direction and the surface normal) are perturbated with a gaussian distributied offset:

thetaOffset=atan(sigmaRatio*tan(PI*(rnd()-0.5)));
phiOffset=atan(sigmaRatio*tan(PI*(rnd()-0.5)));
where thetaOffset and phiOffset are the random angles added to the calculated ones (those of mirror reflection), rnd() is a uniform random number between 0 and 1, and sigmaRation is the value the user inputs in the Material Roughness field (the ratio of the sigmaH and sigmaX values, where sigmaX is the standard deviation of the size of cavities on the surface, and sigmaH is the st.dev. of the depth of those). Essentially, the larger the "roughness" value (theoretically the ratio of RMS roughness and autocorrelation length), the more the reflection direction is perturbated around the perfect mirror direction.
 
Please note that Synrad has a new reflection model, an approximation of [1], where there is a probability (called Debye-Waller factor) of a perfect mirror-like reflection, and in case of rough scattering, RMS roughness and autocorrelation length are individually taken into account (not just their ratio). The exact model is detailed in chapter 5.2 (Photon tracing / rough surface scattering) of the algorithm behind Synrad document.
 
In a polygonized geometry, the old algorithm results in a "smooth" flux texture, whereas the new reflection algorithm has artifacts due to the perfect reflection from discretized curved surfaces. At CERN, we tend to use the old model.

Sides: One sided facets are transparent from their back (back is the opposite side of where the normal vector is pointing), while 2-sided facets are opaque from both sides. In SR applications, 2-sided facets might always work.

Opacity: The probability that a photon will interact with the facet. (If opacity and sticking factor are both 0.5, the facet will let through 50% of incident photons, reflect 25% and absorb 25%).

Teleport: You can "teleport" photons, a function you might want to use for systems with periodic boundary conditions. See the relevant help section for Molflow to learn to set the up.

Structures: Identically to Molflow, you can speed up the simulation by dividing the structure into smaller structures. Not really used.

Profile: if enabled, the facet will be divided into 100 slices along their local U or V vector, and flux and power will be calculated for each slice. Use Tools/Profile Plotter to visualize the curves.

Record spectrum: for speed and memory reasons, the spectrum of absorbed photons are calulcated only for facets where you enable this feature. If you enable it, use Tools/Spectrum plotter to visualize it.

Mesh button: will pop up the Facet Mesh dialog where you can create textures:

Image

To add a texture to the selected facet(s), tick Enable, tick the absorbed (or reflected) mode, and set resolution. Resolution is in samples/cm, so if the resolution is 10, each texture cell will be 0.1cm*0.1cm.

The view settings panel allows you to make the volume and texture display of selected facets transparent (without stopping the simulation).

 

[1] GF Dugan, KG Sonnad, R Cimino, T Ishibashi, and F Sch¨afers. Measurements of x-ray scattering from accelerator vacuum chamber surfaces, and comparison with an analytical model. Physical Review Special Topics-Accelerators and Beams, 18(4):040704, 2015.