Controlling Optical Surface Ripples via Gaussian Removal Profiles

16 concepts3 papers1 proposal

When manufacturing high-precision optics—such as aspheric lenses for cameras or massive mirrors for telescopes—the smoothness of the surface dictates performance. While engineers have long known how to correct large-scale form errors (the overall shape of the lens) and microscopic roughness, an intermediate category of defects presents a persistent challenge: mid-spatial-frequency (MSF) errors.

MSF errors manifest as periodic surface ripples or "tool marks." They are smaller than the lens itself but larger than microscopic scratches. These ripples act like a diffraction grating, scattering light, degrading image sharpness, and severely undermining optical performance. Recent research focuses on eliminating these ripples by optimizing the geometry of manufacturing tools, the physics of material removal, and the paths the tools take across the optical surface.

The Root of the Ripple: Tool Influence Functions

To understand why surface ripples form, we must look at how computer-controlled polishing tools interact with glass. As a polishing tool moves across a surface, it removes material based on its influence function—the characteristic geometry of its pressure and material removal profile.

If a tool exerts uneven pressure, it creates sharp, non-uniform removal profiles, such as "W-shapes" or trapezoids. When these sharp profiles are dragged across a lens in overlapping rows, the mathematical convolution of the tool's shape and its path spacing creates highly visible, periodic ridges.

Sharp/W-Shape Profile:   /\_/\   +   /\_/\   --->   /\_/\_/\_/\  (High MSF Ripples)
Gaussian Profile:       /   \  +  /   \  --->  ___________ (Smooth Overlap)

To solve this, researchers have targeted the mechanical design of the polishing tools themselves. In a study published in Photonics, Chi Song et al. investigated how to transform unstable, edge-heavy removal profiles into predictable, smooth distributions. In Improving smoothing efficiency of rigid conformal polishing tool using time-dependent smoothing evaluation model, the authors demonstrated that by physically lowering the drive pin hole or backplate within a rigid conformal polishing tool's structure, they could mitigate severe pressure spikes at the tool's edges. This geometric modification reshapes the tool's influence function into a smooth, predictable Gaussian-style distribution.

Why Gaussian Profiles Prevent Ripples

The intuition behind why a Gaussian profile suppresses ripples lies in how overlapping shapes blend. A smooth, bell-shaped curve lacks the sharp discontinuities of a trapezoid or a W-shape. When two Gaussian profiles overlap, their slopes complement each other mathematically, resulting in a highly uniform combined material removal depth.

In Micromachines, Zhaohao He et al. explicitly evaluated this phenomenon. In Research on the Influence of the Material Removal Profile of a Spherical Polishing Tool on the Mid-Spatial Frequency Errors of Optical Surfaces, the authors confirmed that a smooth, Gaussian-like profile leads to substantially lower MSF errors compared to sharper, non-uniform profiles.

However, changing the tool profile is only half the battle; the spacing between the polishing paths must also be meticulously managed. If the paths are too far apart, ripples reappear; if they are too close, manufacturing efficiency plummets, making the process unviable for industrial production.

To balance this trade-off, Zhaohao He et al. developed an optimized path spacing strategy. Their algorithm iteratively calculates the largest possible distance between tool paths that still keeps the cumulative root-mean-square (RMS) surface error below a strict quality threshold. By pairing a Gaussian removal profile with this optimized step-over distance, they successfully suppressed surface ripples without incurring excessive polishing times.

Mitigating Edge Deviations Before Polishing

While polishing acts as the final corrective step, surface errors can also be mitigated earlier in the manufacturing pipeline during the initial shaping process. For molded glass optics, surface deviations naturally tend to increase toward the outer edges of the lens due to uneven material displacement and thermal contraction.

Writing in the Journal of Engineering Science and Technology Review, Shuxian Shi et al. addressed this edge-heavy error distribution in Two-stage Glass Molding of Aspheric Lens. Instead of forcing a flat glass preform directly into a complex aspheric shape, the authors introduced a two-stage molding process. The glass is first shaped into a "best-fitting sphere" before undergoing final aspheric molding.

By breaking the deformation into two steps, the total material displacement required during the final, high-precision stage is drastically minimized. This geometric staging prevents the compounding of errors at the periphery, yielding a highly uniform error distribution across the entire optical surface and reducing overall surface errors by nearly 40%.

A Unified Approach to Surface Integrity

Whether at the molding stage or the final polishing stage, the underlying principle of modern optical fabrication remains the same: preventing localized, abrupt variations in material processing. By utilizing multi-stage molding to ensure a uniform starting surface, modifying tool geometry to achieve a smooth Gaussian pressure profile, and optimizing path spacing to ensure seamless convolutional blending, manufacturers can systematically eliminate mid-spatial-frequency ripples and produce ultra-precise optics efficiently.

Key Foundational Work

Improving smoothing efficiency of rigid conformal polishing tool using time-dependent smoothing evaluation model (2017) — contributed 4 concept elements to this direction.
Two-stage Glass Molding of Aspheric Lens (2020) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_convolutional_overlap_of_polishing_paths_dd77fce5,
  title = {convolutional overlap of polishing paths},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/convolutional-overlap-of-polishing-paths/rdr_dd77fce5d6fd4d1ea3590ba6b17f3443},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-20},
}

This exact version (version 1, frozen text):

@misc{lacuna_convolutional_overlap_of_polishing_paths_dd77fce5_v1,
  title = {convolutional overlap of polishing paths},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/convolutional-overlap-of-polishing-paths/rdr_dd77fce5d6fd4d1ea3590ba6b17f3443/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-20},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake