Fiber- and Photonics Alignment Algorithms: Every Fiber-Optical Power Optimization Starts with a First-Light Search

Fiber- and Photonics Alignment Algorithms: Every Fiber-Optical Power Optimization Starts with a First-Light Search


First Light Search in photonics and fiber optics is typically a time-consuming and, from an industrial mass production standpoint, expensive procedure.

There are several first-light search algorithms used in photonics alignment.

Coarse and fine spiral algorithms, which execute a spiral motion pattern, have been utilized for a long time. These algorithms read the optical power of a power meter connected to the fiber that’s spiraled across a photonic component, either to characterize the power distribution or until a predefined threshold is reached. Here, the user typically sets the radius, spacing, and threshold above the noise floor.

Spiral Scan Fiber-Alignment Routine

Raster scans and sinusoidal scans are equally popular. If a complete scan is conducted, the center of optical power distribution can be calculated as the starting point for fine alignment.

Meander Scan Fiber-Alignment Routine

These traditional first-light search algorithms are straightforward but rather slow. In the case of dual-sided alignment, they are extremely slow – taking minutes to hours. Consider the scenario of sending light from one fiber through an optical circuit on a Silicon Photonics wafer into a second fiber coupled to the output. What are the chances of having both input and output fibers in the correct location at the same time, especially when dealing with sub-micron structures?

Breakthrough in First-Light Capture - PILightning

A new algorithm dubbed PILightning solves the first-light search problem. It utilizes AI and can reduce the time to find first light by orders of magnitude. This groundbreaking technology slashes alignment time scales, delivering precision and speed like never before. Find out more here:

Video: PILightning Algorithm explained

After first light has been found – the power optimization

Once first light has been found, the power optimization algorithms can start.

Power optimization alignment algorithms fully optimize fiber alignment to the position where maximum optical power is reached. The fastest way to achieve optimal power is to run the PILightning first-light search algorithm, automatically followed by a gradient-search hill-climb algorithm and a tracking algorithm to maintain optimal power, even when external influences such as mechanical drift caused by temperature changes or forces from curing epoxy change the physical location.

Optical power distribution and hill-climb algorithm (gradient search routine)

Other, slower methods to find the optimum location involve a complete spiral or meander scan followed by an automatic move of the alignment system to the location with the maximum optical power reading.

FMPA – Fast Multi-Channel Photonics Alignment Algorithm

FMPA technology consists of embedded commands running on highest-performance motion controllers for motorized stages, air bearings, hexapods, or piezo nanopositioning scanners. FMPA allows the fastest coupling optimization across multiple degrees-of-freedom, inputs & outputs, elements & channels, between photonic and other optical devices and assemblies. By running the optimizations in parallel, even if the individual optimizations interact, FMPA can yield vast reductions in test and assembly time, and cost reductions of 99% are routinely seen.

Top-hat coupling The FMPA controller can automatically determine the position of the centroid

FMPA controllers can also automatically calculate a centroid position, which can be useful for power distributions with multiple peaks or for top hat (flat) distributions.

More on the latest Photonics Alignment Systems for Silicon Photonics Test and Assembly.



要查看或添加评论,请登录

PI USA - PI (Physik Instrumente) L.P.的更多文章

社区洞察

其他会员也浏览了