How to plan an optical glass sample cutting trial

1. Define the material before selecting parameters

Provide the exact glass grade, supplier, annealing condition, block dimensions, hardness or machinability information, and any coating or bonded layers. “Optical glass” is not one uniform material: crown glass, fused silica, filter glass, and specialty formulations can respond differently to abrasive size, feed rate, coolant, and clamping.

2. Write the acceptance criteria first

Before cutting, agree on the nominal thickness and the measurements that determine success. Typical acceptance fields include:

• Finished thickness and positive/negative tolerance.
• Total thickness variation (TTV) across a defined measurement grid.
• Flatness and parallelism if required by downstream processing.
• Surface roughness Ra with the instrument and sampling positions stated.
• Maximum edge chipping and the microscope method used.
• Kerf width, material yield, cut time, and wire wear.

3. Record the complete machine setup

A result is difficult to reproduce if the report lists only machine model and cut time. Record wire diameter and coating, loop length, linear speed, tension, feed rate, cutting width, coolant, fixture orientation, guide-wheel condition, compensation setting, and whether a new wire was stabilized before contact with the workpiece.

4. Use a measurement plan that separates different errors

Average thickness shows whether the process is centered on the target. TTV shows variation across one sheet. Ra describes the measured surface profile. Edge chipping indicates local fracture near the perimeter. These values answer different questions and should not be replaced by one general statement such as “good cutting quality.”

5. Compare multiple samples, not one best piece

A single sample can hide drift, wire wear, fixture variation, or measurement uncertainty. A short series allows the engineer to compare sample-to-sample consistency and determine whether a parameter correction improves the full run rather than one isolated result.

Example: H-K9L optical glass

An Ensoll trial cut ten H-K9L sheets at 6 x 70 x 70 mm using an ESO-GM endless diamond wire saw. The test reported TTV of 0.003-0.015 mm, sample-average thicknesses of 5.992-6.008 mm, average Ra of 0.72-0.94 micrometers, and maximum chipping below 10 micrometers for all samples. The complete report includes the nine-point measurements and process settings.

Decision after the trial

Approve the process only when the documented result matches the intended production conditions closely enough. If the sample geometry, fixture, coolant, or wire differs materially from production, schedule a second validation rather than assuming the first parameter set will transfer unchanged.

For related background, see the diamond wire saw for glass guide, the machine selection guide, and the sample cutting service checklist.