Analysis Surfaces & Ray Projection

Often we use analysis surfaces very simply, attaching them to a surface and generating analysis results for the rays on the surface of interest. However, unexpected results can arise if you're not aware of how FRED produces its analysis when the analysis surface is displaced from a geometry surface and/or when using the "All Rays" ray filter.

Displaced Analysis

Consider the case where we have an ideal lens focusing light to an absorbing plane... but the analysis surface is positioned behind the absorbing plane:

We can see that the rays do not reach the analysis surface. What is more, the analysis has a filter on it to only include rays on the absorbing surface:

In this case, we might conclude that running an irradiance analysis on the analysis surface will yield no result because no rays reach the analysis surface, but if we run an irradiance analysis we find there is a result:

The reason for this is that in this case FRED looks for rays meeting the filter (rays on the absorbing surface) and free space projects them, from their last known position, backwards or forwards along their last known directions, until they intersect the analysis surface:

And this is what generates the irradiance distribution of an out of focus spot.

The effect of ray projection is useful in cases where free space propagation is valid and intentional, such as examining Beam Propagation "down range" from a source to see a far field distribution, or Investigating ray distributions in other regions, such as around a focal plane (as per the example shown above).

In these cases we can just position analysis surfaces in the desired location(s) without additional geometry and gather our results without needing to retrace the system - any rays in the ray buffer that meet the prescribed ray filter are simply projected until they intersect the analysis and contribute to the result.

Note that only analysis surfaces support ray projection, and that the projection is not limited to being in the exact same direction the ray had... the projection can be forwards or backwards along the trajectory of the ray in order to reach the analysis surface. In this way we could also place an analysis surface at the virtual focus of a diverging lens and analyze the virtual image of the lens.

"All rays"

Lets modify our example such that the analysis surface is at the absorbing plane, but with the "All Rays" filter:

A small circular obscuring surface is placed between the source and the lens to block the central part of the beam:

In this case, we might expect the obscuring disc to block the central rays and for the irradiance on the image plane to be a perfectly focused spot. The irradiance result we see may initially look ok due to the focal spot having a great deal of power:

However if we rescale the result on a log basis we can see additional rays around the focused spot:

Where do these rays come from? The same free space projection as encountered before. This time there are rays absorbed by the obscuring disc, these rays meet the "All Rays" filter (because all rays in the system meet this filter, it's really no filter at all!) and so are free space projected to the analysis surface:

As we can see, in the case where the ray filter is left at the default "All Rays" setting the same free space projection occurs, but in this case every single ray in the ray buffer now potentially counts towards the final analysis result.

Summary

When using analysis surfaces the effect of ray projection is useful in cases where free space propagation is valid and intentional, e.g.

• Beam Propagation "downrange" from a source
• Investigating ray distributions around a focal plane.
• Finding and analyzing virtual focii

However, when using analysis surfaces the user must understand and account for ray projection, and exercise caution to make sure the ray filters in their system represent the rays they wish to analyze otherwise unexpected results are to be expected.