Glass Anisotropy — Complete Visual Guide

The physics of birefringence, how polarised light inspection is performed, retardation grading, and reference to ASTM C1048 / EN 12150.

ASTM C1048 §7.4 EN 12150 §4.4 ASTM F2328 (SCALP) Inherent — not a defect per code
▸ 01 — What is anisotropy?
The physics of birefringence

Ordinary glass is optically isotropic — its refractive index is identical in all directions. Light passes through it uniformly with no colour splitting.

During tempering or heat-strengthening, glass is heated above the transition temperature (~620°C) then rapidly quenched by air-jets. This locks in unequal stress distributions — high surface compression, central tension — across the panel.

These stress gradients make the glass birefringent: polarised light entering the glass is split into two rays (ordinary and extraordinary) travelling at different speeds parallel to the two principal stress axes. When these two rays recombine, they are out of phase — producing constructive or destructive interference for specific wavelengths, which the eye sees as colour.

ASTM C1048 §7.4 — "A strain pattern, also known as iridescence, is inherent in all heat-strengthened and fully tempered glass… It is a characteristic of heat-treated glass and should not be mistaken as a defect."

Visual appearance — "leopard spots"
Simulated facade — polarised daylight view

Pulsing coloured patches (amber, violet, green) — the "leopard spots" — visible when polarised sunlight hits tempered glass at a glancing angle. More intense on clear blue-sky days.

▸ 02 — Physics: how birefringence produces colour
1. Unpolarised light 2. Partial polarisation 3. Glass — birefringent zone 4. Ray split + phase shift 5. Colour fringe SUN Sky polarises ~40% of light Tempered glass (birefringent zones) Ordinary ray (n₁) Extraordinary ray (n₂) Phase difference = retardation Rays recombine Eye sees colour Retardation (nm) = (n₁−n₂) × thickness
▸ 03 — Optical retardation grading (nm scale)

Anisotropy severity is quantified as optical retardation in nanometres (nm) — the phase difference between the two split rays. Drag the slider to understand each level.

60 nm ≤ 60 nm — Premium quality
0 nm5090120160200+ nm

RetardationIsotropyVisible appearanceProject suitability
≤ 60 nm > 95% Invisible — even under polarised sunglasses Premium facades, museums, flagship retail
60 – 90 nm 85–95% Faintly visible only at acute glancing angles High-end commercial glazing
90 – 120 nm 75–85% Colour fringes visible on clear blue-sky days Standard commercial buildings
> 120 nm < 75% Rainbow "leopard spots" clearly visible Non-critical / industrial use only

Note: Retardation limits (e.g. ≤ 90 nm or ≤ 60 nm) are project-specific contract requirements — they are not codified in ASTM C1048 or EN 12150, which only say anisotropy "is not a defect". Always include retardation limits in project specifications for façade glass.

▸ 04 — Polarised light inspection methods
Visual / site
Circular polariscope
Online scanner
SCALP (stress)
Site visual inspection — polarised sunglasses
  • Equipment: Standard polarised sunglasses or a 100 mm circular polarising filter.
  • Optimal conditions: Clear blue sky, low sun angle (morning / late afternoon), water or snow reflections nearby (increases polarised light proportion).
  • Viewing angle: 30–60° from perpendicular (glancing angle). Perpendicular view often hides anisotropy entirely.
  • What to look for: Oval rainbow patches, grey/white stripes, iridescent shimmer that shifts as you tilt your head. Spots typically 100–400 mm diameter.
  • Edge zones: Higher intensity near glass edges (within ~50 mm) is normal — evaluate the body of the panel.
  • Standard: ASTM C1048 §7.4 — observe; record qualitative severity.
  • Limitation: Qualitative only — cannot produce an nm retardation value. Highly dependent on lighting conditions.
Inspection protocol (site)
  • Step 1 — Choose a clear day with blue sky in the direction of observation. Avoid overcast conditions (low polarised light).
  • Step 2 — Stand 5–10 m from the installed panel. Put on polarised sunglasses.
  • Step 3 — Tilt your head 45° and rotate slowly (rotates the polarising axis). Anisotropy will appear and disappear as you rotate — this confirms it is anisotropy, not a contamination.
  • Step 4 — View from multiple angles (30°, 45°, 60° from glass normal). Note worst-case view.
  • Step 5 — Photograph from worst-case angle through polarising filter held in front of camera lens. Rotate filter to maximise visibility.
  • Step 6 — Record: location, time, weather conditions, angle, photo reference, severity rating (mild/moderate/severe).

For a reliable mock-up assessment, always review full-size samples outdoors — indoor artificial light lacks sufficient polarisation to reveal anisotropy accurately.

Circular polariscope — lab / QC inspection
  • Setup: Light source → Linear polariser → Quarter-wave plate → Glass sample → Quarter-wave plate → Linear analyser (90° to polariser) → Camera / eye.
  • Why circular (not linear)? A linear polariscope produces "isoclinic" fringes (dark lines where principal stress axis aligns with filter axis) that mask the actual stress map. Circular polarisation (quarter-wave plate) eliminates isoclinics, leaving only "isochromatic" colour fringes that directly show stress magnitude.
  • Output: Full colour fringe map — each colour band corresponds to a specific retardation value. Counting fringes × fringe-order constant → surface stress (MPa) or retardation (nm).
  • Standard: Referenced in ASTM F2228; standard glass QC tool. SCALP is the quantitative successor.
  • Fringe colours: Black (0 nm) → grey → white → yellow → orange → red → violet → blue → green → yellow (next order) in isochromatic order.
  • Limitation: Offline / lab-based; sample inspection only (not 100%); requires skill to interpret fringe orders correctly.
Polariscope optical arrangement
White light source Linear polariser Quarter-wave plate (λ/4) Tempered glass sample Quarter-wave plate (λ/4) Analyser + camera
Online anisotropy scanner — 100% production inspection
  • Position: Installed on the tempering line immediately after the quench / cooling zone.
  • Light source: Three-colour (RGB) circular polarised LED array below the glass conveyor.
  • Camera: Line-scan or area camera array above, capturing isochromatic fringe images as glass passes through.
  • Resolution: 0.5 – 1 mm per pixel across full panel width.
  • Output per pane: Full retardation map (nm/pixel), average retardation, maximum spot value, "predictive iridescence index", edge stress profile.
  • Speed: Real-time — 100% of production inspected.
  • Feedback loop: Data feeds directly to furnace operator for zone-by-zone heating correction and quench nozzle adjustment.
  • Leading systems: LiteSentry Osprey 10, Viprotron StrainScanner 3, Softsolution/Softscan.
  • No recalibration: Most systems work across glass types, thicknesses, and tints without recalibration.
What the retardation map looks like

The map shows each pixel coloured by retardation value — typically a false-colour scale from deep blue (low, near-zero nm) through green, yellow, orange, to red and purple (high nm values).

0 nm 60 120+ nm Retardation map

Red/purple areas = high retardation (problem zones); blue = near-zero (ideal). Operator uses this to tune furnace heating zones in real time.

SCALP — Scattered Light Polariscope (surface stress)
  • Purpose: Measures absolute surface compressive stress in MPa — the safety-critical criterion per ASTM C1048 §4.
  • Principle: A polarised laser beam enters the glass edge at the surface. Internal stress causes the beam to scatter polarised light through the glass depth. The polarisation state of scattered light is captured by a camera viewing perpendicular to the beam. The stress profile (surface compression, mid-depth, central tension) is calculated from the resulting intensity pattern.
  • ASTM Limits verified by SCALP:
    — Fully Tempered (FT): surface compressive stress ≥ 69 MPa (10,000 psi)
    — Heat-Strengthened (HS): 24 – 52 MPa
  • Portable: Handheld device used on the factory floor or on-site. No sample cutting required — measures through the edge.
  • Standard: ASTM F2328; widely used in glass industry QC per ASTM C1048 §4 requirements.
  • Key distinction: SCALP measures stress (MPa) = safety criterion. Anisotropy scanners measure retardation (nm) = aesthetics criterion. Both required for full QC.
Summary — all four methods compared
MethodMeasuresUnit100%?
Polarised sunglasses Appearance Qualitative Site
Circular polariscope Fringe / retardation nm (rel.) Sample
Online scanner Full retardation map nm/pixel 100%
SCALP Surface stress MPa Portable QC
▸ 05 — What makes anisotropy worse?
Glass thickness & laminate

Thicker glass (10–15 mm) has higher thermal mass — uneven quench creates larger stress gradients. Laminated units (2+ panes) compound the effect; each pane's spots can be visible through the other.

Panel size, shape & cut-outs

Large panels, extreme aspect ratios (e.g. 600 mm × 3000 mm), holes, notched corners, and triangular shapes are harder to heat and quench uniformly → higher stress differentials.

Furnace heating non-uniformity

Uneven heating zones, roller-temperature gradients, and glass pausing on rollers create "cold spots". These locked-in thermal non-uniformities become visible stress patterns.

Quench nozzle irregularities

Blocked, worn, or misaligned air-blast nozzles in the quench box create differential cooling across the panel width — a direct cause of localised high-retardation spots.

Low-E coatings

High-reflectivity soft-coat low-E coatings act as a partial mirror for internal stress patterns. They do not cause anisotropy but can amplify its visual appearance significantly for observers outside the building.

Building location & sky conditions

Blue sky + low sun angle maximises polarised daylight. High-rise buildings and sites near water (sea, river, rooftop pools) reflect polarised light upward. North-facing facades in the southern hemisphere experience intense sky-polarisation at certain times.

▸ 06 — Code references
StandardClauseRequirement / StatementImplication
ASTM C1048 §7.4 "Strain pattern (iridescence) is inherent in all heat-strengthened and fully tempered glass… shall not be mistaken as a defect." Cannot reject glass on anisotropy grounds under ASTM alone — project spec must add a retardation limit.
EN 12150-1 §4.4 Anisotropy "is not considered a defect or deficiency". Same limitation as ASTM — project specification must define acceptable retardation in nm.
EN 1863-1 §4 Same as EN 12150 — inherent property of heat-strengthened glass. Applies to HS glass (Class 2 equivalent).
ASTM C1048 §4 FT: surface compression ≥ 69 MPa. HS: 24–52 MPa. Verified by SCALP per ASTM F2328 — safety criterion, distinct from anisotropy aesthetics.
ASTM F2328 Full doc SCALP method for surface stress measurement via scattered light polariscope. Use SCALP to verify tempering level meets ASTM C1048 §4 MPa criteria on every lot.
Project spec Retardation ≤ 60 nm (premium) or ≤ 90 nm (standard) — typical contract limits. Must be stated explicitly in the specification and purchase order to be enforceable.