Optical Surface Scratch-Dig Inspection Standards

Topic: MIL-PRF-13830B scratch-dig grading, visual inspection conditions, defect counting rules, active aperture exclusion, laser optics defect limits

Overview

Scratch-dig is the universal cosmetic surface quality specification for optical components, defined under MIL-PRF-13830B. It controls scattered light caused by surface defects, directly impacting laser stability, imaging contrast and sensor signal noise. Felix Glass strictly implements visual comparison against master submasters and separates acceptance criteria for active optical areas vs non-clear mounting zones to balance yield and optical performance.

Basic Definition: Scratch & Dig

• Scratch (S value): Linear surface damage; the number corresponds to relative visibility compared with standard submasters, not direct physical width. Lower numbers mean fewer light scatter.
• Dig (D value): Local pits, dents or inclusions; the value equals the maximum allowable diameter in hundredths of a millimeter (dig 10 = 0.1 mm max pit diameter).
• Full specification format: S-D, e.g. 20-10, 40-20, 60-40.

Standard Inspection Environment (Non-Negotiable)

1. Light source: 100W incandescent lamp, no ambient stray light.
2. Viewing distance: 25–30 cm, normal incidence viewing in transmission or reflection.
3. Magnification: No optical magnification unless otherwise specified; only visual comparison with certified submasters.
4. Boundary rule: Defects outside the clear active aperture (mounting edge zone) are graded one level looser.

Grade Selection & Defect Counting Rules

• 10-5 / 20-10 (High-Power Laser Optics): Very low scatter; no long continuous scratches; total dig area limited to 0.01% of clear aperture; no defects on laser beam footprint.
• 40-20 (Imaging & Medical Optics): Standard precision for lenses, windows and touch panels; short isolated scratches are permitted; small pits must be scattered randomly without clustering.
• 60-40 / 80-50 (Industrial Windows & Protective Covers): Loose cosmetic requirement; only prohibit deep gouges and clustered defects that cause mechanical failure.
• Counting rule: Total length of all scratches cannot exceed the diameter of the clear aperture; digs cannot overlap or form clusters within a 5 mm circle.

Common Pitfalls During QC Inspection

1. Confusing scratch visibility with physical width: S value is based on light scattering, not micrometer-measured width. 2. Applying the same strict scratch-dig standard to both active area and mounting edge, unnecessarily lowering production yield. 3. Mixing surface roughness Ra with scratch-dig: Ra measures micro-profile smoothness, while scratch-dig controls macroscopic discrete defects; critical optics require both specs together. 4. Ignoring post-coating defect inspection: AF or AR coating can turn tiny substrate scratches into visible scatter sources.

Dimensional Tolerance & Parallelism Measurement

Topic: linear dimensional tolerance, thickness TTV, flatness (wavefront error), parallelism angular deviation, interferometer testing, batch consistency control for plane-parallel optics

Overview

Dimensional and geometric tolerances control bonding gaps, optical path difference and wedge error of plane-parallel glass parts. Thickness variation, flatness and parallelism directly cause Newton rings, beam deflection and inconsistent OCA lamination results. This section covers tolerance classification and industry-standard measurement equipment for optical glass components.

1. Linear Dimensional Tolerance

• Standard Precision (General Windows): Length/width ±0.2 mm; cutouts & holes ±0.15 mm.
• High Precision (Touch & Imaging Panels): Length/width ±0.1 mm; CNC cutouts ±0.08 mm.
• Ultra-Precision (Laser Plane Parallel Plates): ±0.05 mm with edge polishing to eliminate chipping.
• Measurement tool: Digital caliper for routine checks; optical comparator for tight-tolerance contour inspection.

2. Thickness Uniformity: TTV (Total Thickness Variation)

1. Definition: The maximum thickness minus minimum thickness across the full clear aperture.
2. Grade benchmark: - Consumer electronics touch glass: TTV ≤ 5 μm; - Industrial optical windows: TTV ≤ 10 μm; - Laser etalon plates: TTV ≤ 2 μm.
3. Measurement: Multi-point thickness gauge sampling across 5 grid points (center + four corners) on every sheet.

3. Flatness Test (Wavefront Error)

• Inspection method: Fizeau laser interferometer, testing against a certified optical flat reference plate.
• Unit: Fringes (λ/10, λ/20, λ/50 at 632.8 nm He-Ne wavelength).
• Application split: - Cover glass: λ/2 ~ λ/4 flatness; - Precision optical windows: λ/10 ~ λ/20; - High-grade laser flats: λ/50 or better.

4. Parallelism (Wedge Error) Measurement

Parallelism describes the angular deviation between two polished faces, expressed in arcseconds or wedge angle. Excessive parallelism error tilts the transmitted light beam and creates inconsistent optical path difference.

• Measurement method 1 (Interferometry): Analyze fringe tilt on the interferogram; convert fringe offset into angular deviation.
• Measurement method 2 (Thickness mapping): Measure thickness at opposite edges and calculate the wedge angle from thickness difference and part diameter.
• Tolerance reference: - Ordinary protective windows: ≤ 30 arcseconds; - Imaging plane-parallel plates: ≤ 10 arcseconds; - Laser beam splitter substrates: ≤ 2~5 arcseconds.

Batch Consistency Control Checklist

1. First article inspection (FAI): Full dimensional, flatness, parallelism and scratch-dig testing before mass production. 2. Sampling plan: 100% TTV and visual scratch-dig; 10% random sampling for interferometric flatness & parallelism. 3. Environmental control: Keep temperature stable (±2°C) during measurement to avoid thermal expansion distortion. 4. Edge quality audit: Inspect edge chips and microcracks; edge defects will propagate during chemical strengthening and cause dimensional warpage.