Stray Light

Stray Light

Stray light in an optical system is an undesired effect due to reflections from glass-air interfaces and internal mechanical components and surfaces. In imaging and non-imaging systems, it can be difficult to trace back to the location that is the cause of the unwanted light at the image or detecting plane. Artifacts can include ghost images, flaring, and reduction in image quality. By analyzing an optical system before production, known issues can be isolated to determine whether stray light is at an acceptably low level, or measures must be taken to reduce or eliminate it.
In the 3DOptix cloud-based simulation tool, analysis of stray light using detectors and the analysis tool are easily applied to determine effects at the image/detector plane and internally. We will overview methods to analyze stray light and some common component locations that produce the effect.
Our optical system will consist of a Cooke Triplet with the following components:
  1. Thorlabs LA 4725, plano-convex
  2. Thorlabs LD 1357, bi-concave
  3. Thorlabs LA 4052, plano-convex
  4. Thorlabs SM1L10 Lens Tube
  5. Light Source
    • Plane Wave
    • Circular, 5mm radius
    • 633 nm wavelength
    • Power, 1 W
    • Unpolarized
  6. Detector: Input/Output
    • Spot: Coherent Irradiance
    • Analysis Rays: 1 million
    • 500×500 pixels
You can see the image of our optical system below. The 3DOptix simulation file can be downloaded to see additional information about the optical system such as component spacing and analysis detectors.

For the analysis we have built up two identical Cooke Triplets with the lenses in the first without anti-reflective coatings and the second having the Thorlabs A- coating (RAVG<0.5% from 350-700 nm). The different systems will be denoted with BBAR or NC.

When we look at the light throughput before and after the optical components, we can get an idea of how much light is lost to Fresnel reflection since we are on-axis and not scattering on any mechanical components. This is our first step to determine stray light purely from the lenses.

The uncoated (NC) Cooke Triplet optical system loss due to Fresnel reflections amounts to ~19% which matches well with theory (~4% loss per surface).

The coated Cooke Triplet (BBAR) optical system loss amounts to ~0.4233% which matches well with manufacturer specifications (~0.1% loss per surface).

Adding an anti-reflective coating is an obvious way to reduce unwanted stray light in the optical system by reducing Fresnel reflections. This is a simple but effective method to analyze how much light is being lost.

Next, we may want to determine how much light is reflected from a single surface, either a lens or a mechanical component. Instead of placing a detector after all optical components and measuring the light loss, we can place it at the surface from which we want to measure reflections. In this case the Fresnel reflections from the last lens.

As a reminder 3DOptix detectors have an “F” for the front face and a “B” for the back. No light will be analyzed if incident on the back surface. We want to measure the back reflected light so the detectors need to be flipped.

We now have our analysis detector in place and will measure the Fresnel reflections.

First, we need to turn on reflections to get the stray light to show on the analysis detectors. We will go into the optical settings for the uncoated and coated lens 3 and unselect the NO REFLECTIONS check box.

The MAX BOUNCES will stay at 1 since we only care about the first reflection at the surface.

Now we can run the analysis and see what the reflected power is for the plano surface of lens 3.
We can do this sequentially by turning on reflections for one surface at a time or every surface to analyze the cumulative effect of stray light. Let’s turn on reflections for the curved surface of lens 3 to see how much the stray light increases.

There are now two surface contributions to the Fresnel reflections that the detectors will measure. Notice that the left detector above (BBAR) is now in log scale instead of linear. This is because the curved surface of lens 3 produces a large relative central reflection, but the actual power is still very small.

In addition to the Fresnel reflections, we also have scattering that can occur from the mechanical enclosure and optical surfaces. Although the lens tube is aluminum coated with a black powder, scattered light can still be a larger contributor.

The surface scattering will be turned off for lens 3 and turned on for the lens tube. We can also specify a scattering-type for the lens tube depending on the type we want to simulate.

For this application, we will turn on Lambertian scattering for surface 24, which is the top inner section of the lens tube.

The transmission was set to 0% since this is a metal component and 10% reflectance to simulate the scattering. The rest of the power will be absorbed. We are going to simulate some light source coming from an external location, so we will change the light source parameters:

Light Source:
  • Plane Wave
  • Circular, 5mm radius
  • 633 nm wavelength
  • Power, 1 W
  • Unpolarized
The first change we want to make is to change the ray colors to be directional based. This way we can visually determine the different scattering directions. In a more complex optical system, we could trace the light through the system, and based on the ray color it would be clear where scattering or light loss is happening.
We now have a small external light source entering our optical system and Lambertian scattering on the lens tube. We will analyze this at the image/detector plane.
Notice that the uncoated Cooke Triplet has more scattered light reaching the image/detector plane than the BBAR. This is due to the coating allowing more light through the lenses. This is an important factor to realize when building optical systems, as unanticipated effects can occur. Stray light is an important parameter to consider for optical systems and can be difficult to understand unless light sources and optical systems are properly analyzed.

3DOptix works
only on desktop!

Please go to 3doptix.com on a
desktop device, using the
Chrome or Edge browser

Available on January 30th, 2023