Ansys Speos allows suppliers the ability to provide encrypted black box systems to users to protect their intellectual property. These Ansys Speos black boxes are a feature of the Ansys Speos Light Box that was introduced into the tool last year. If a vendor provides you with a black box of a camera system that you would like to integrate into a system-level simulation? What can you do with that black box?
The answer is, quite a lot! Now, remember, when working with a black box, the goal is not to redesign or enhance what is within the black box. The goal is to integrate a black box into a larger optical system and analyze it at a system level. However, it begs the question, how can we be sure that the objects in the black box are accurately modeled and light is being traced correctly? To answer this question, I went down the endeavor of running a stray light analysis on a camera system, replacing the lenses with a black box component, and comparing the results.
An Example of Stray Light Simulation with an Ansys Speos Black Box
I ran five different stray light simulations within Speos, all using the same Smartphone Camera Stray Light Analysis available on the Optics Ansys website:
Stray Light Analysis – Smartphone Camera – Ansys Optics
If you would like to follow a detailed overview of how Ansys Zemax OpticStudio and Ansys Speos can exchange files, please visit the linked article that explores the topic in more detail:
After running the initial stray light simulation on the smartphone camera, I went back and replaced all the optical surfaces (lenses, windows) with an Ansys Speos Black Box of the exact same optical system. In other words, I created a Speos black box with the optical surfaces of this system and used it in the simulation. To learn how to create your own Speos black box, check out my latest post on our website: How to Create an Ansys Speos Black Box – PADT.
I then went ahead and ran the same simulation, again, except this time including the full mechanical housing of the system, leaving just the aluminum enclosure and sensor mount. I then compared the results of all 3 simulations and found that there was less than a 1% difference between the average illuminance, flux, and maximum illuminance between all three simulations. The results are summarized in the following table.
I took the study a bit further by using the black box of the optical system, optics and mechanical housing included, in a Physical Camera Sensor. I did this after learning that starting in the 2025 R1 release of Speos, Physical Camera Sensors can use black boxes as their camera system input.
The Physical Camera Sensor in Speos is used to efficiently run a Stray Light simulation by propagating rays that converge onto the sensor. It does so by using Speos’s own ray aiming algorithm to select rays that interact with the ray aiming area. The ray aiming area in a Physical Camera Sensor is set as the first optical surface in the optical path.
There are two different methods for using a Physical Camera Sensor to run a simulation. The first one uses a standard Monte Carlo ray propagation engine to simulate the 3D scene, and the ray aiming area is made to be the first face of the optical system. This method does consider specular and scattering interactions and can therefore be just as time-consuming as using a regular sensor for the simulation if there are multiple optical surfaces interacting with the propagated rays.
The second method consists of two separate simulations. The first generates a sequence file that contains the full list of ray path sequences that are all sorted by energy. This list of sequences predefines the number of sequences in the simulation. The second simulation uses the sequence file from the previous simulation to increase the speed of the simulation. The simulation takes advantage of the ray aiming area and performs an optimized ray propagation on the most energetic ray paths. A deterministic algorithm then is used with only specular interactions being considered. This approach traces the same number of rays per sequence to achieve a higher ray.
Below is a quick example demonstrating the efficiency of the Physical Camera Sensor when simulating a simple display source:
I then went ahead and compared the simulation results from each method to the original stray light analysis results, just as I did before. The results yielded a percent difference less than 3% for the average illuminance, flux, and maximum illuminance. The results are summarized in the following table.
In summary, using a Speos Black Box in a simulation like a stray light analysis does not limit the type of results that users can get from a Speos simulation. It enhances an engineer’s understanding of their optical system while still protecting proprietary information. If you’d like to discuss how your team can benefit from using a Speos Black Box in your simulation, please do not hesitate to reach out! Until next time!
Are you exploring the use of Ansys Speos or other Ansys Optics tools? If you are lucky enough to be a PADT customer, reach out to our support team. And if you are not a PADT customer, no worries, we are here to help with our custom training, mentoring, and consulting capabilities. Reach out, let’s talk. You could be receiving this level of support and knowledge very day.
You must be logged in to post a comment.