MRF: Answering New Demands in the Precision Optics Marketplace

10mm mrfToday’s optics customers are more demanding than ever before. Customers are requesting shorter lead-times, and they expect the quality to be both higher and repeatable. But… they don’t want to pay for better quality than they need. Customers are placing orders for smaller batches, larger variety of shapes and more complicated shapes. In addition, optics buyers are less tolerant of quality variation, long lead times, and missed deadlines.

New manufacturing capabilities are required to be able satisfy these new requirements. Small batches, short lead times, larger variety of parts with more complexity, require more speed and flexibility. Increased reliability of the manufacturing process would enable less quality variation, shorter lead times, and on-time deadlines. Greater determinism during manufacturing would help optics manufacturers meet deadlines, achieve more consistent quality on more complex parts. And customers would only have to pay for the quality they need.

Sounds great! The problem has been that traditional manufacturing processes fall far short of meeting these goals. In general, when fabricating precision optics, there are 3 basic steps that everyone follows: 1.) Grinding, to shape; 2.) Polishing, to remove damage and smooth; and 3.) Finishing, to reach figure and roughness specifications.   Historically, the finishing step has been the most challenging, the most time consuming, the most variable and unreliable, and ultimately, the most costly.

Conventional polishing has proved to have many challenges. It is a highly artisan-based skill. It is non-repeatable, leading to quality and schedule variation both day-to-day, and from optician-to-optician. There is short supply of necessary highly-experienced labor and a long learning curve to become master optician. Conventional polishing equipment has limited flexibility. Planos tend to be polished on continuous or planetary polishers and spheres are processed on full aperture spindles. A good solution for aspheric shapes does not exist. Additionally, a large inventory of dedicated and varied tools is required. A different lap is required for each different radius! Pitch and slurry are difficult to control precisely, which can result in variation from batch to batch.

The introduction of CNC (Computer numerically controlled) polishing addressed some of the challenges and has many improvements over conventional polishing. The process is more repeatable. Pads and slurries are more controlled. And operators can achieve faster cycle times. But… CNC polishing equipment has limited “flexibility”. In general, it is optimized only for spherical parts. Plano, prism, aspheres are difficult, if not impossible. Additionally, there is still the need for the large tooling inventory, and the requirement for dedicated and varied tools. A different lap is still needed for each different radius. CNC processes still lack determinism. As pads wear, results can vary and yield and throughput can drop. Long setup times are an issue and highly skilled operators are required for “difficult” parts, including thin edges, high aspect ratios, and thin meniscus parts. Higher precision (e.g. lambda/8-lambda/10) levels can be routinely achieved using CNC equipment but this is not precise enough to meet next generation demands.

 The Value of MRF

  • Speed and Flexibility: With MRF technology, you only need one machine/process to run all your parts, from planos and spheres, to aspheres, prisms, and cylinders, etc. MRF works well with many materials, including optical glasses, glass ceramics, and crystals. The MRF process has short set-up times and it is easy to switch between jobs.
  • Determinism: MRF is most well known for the determinism it brings to a formerly non-deterministic field. The MRF process is very well characterized. We can very accurately predict the outcome of the polishing run, and get the same results every time! MRF means predictable quality and predictable throughput.
  • Science-based technology: Faster learning curve to become a proficient MRF user. It is also more scalable and more transferable.