The fragmentation test is the primary destructive method used to verify the safety performance and tempering quality of tempered glass installed in curtain wall systems. By fracturing representative specimens under controlled laboratory conditions and analyzing the count, size, and distribution of the resulting fragments, accredited third-party laboratories can determine whether the glass complies with GB 15763.2, EN 12150-1, and related international standards. This article explains the testing principle, specimen requirements, procedures, acceptance criteria, and practical applications of tempered glass fragmentation testing for building envelopes.
Technical Overview
Tempered (thermally toughened) glass used in curtain walls is a safety glazing product manufactured by heating annealed glass to approximately 620–650 °C and then rapidly cooling its surfaces with quench air. This process locks a compressive stress layer into both surfaces, balanced by tensile stress in the core. The magnitude of this residual stress field determines how the glass breaks: fully tempered glass shatters into small, relatively blunt particles rather than the large, sharp shards produced by annealed or heat-strengthened glass.
The fragmentation test, sometimes called the breakage pattern or fragment state test, exploits this behavior as a direct measure of tempering quality. A specimen is deliberately fractured with a pointed tool, and the resulting particles are counted and measured within a defined reference area. A higher fragment count indicates a higher surface compressive stress level and a greater degree of temper. Because inadequate tempering can cause curtain wall glass to break dangerously in service, fragmentation testing is a mandatory requirement in most building safety glazing specifications.
The test is governed by several recognized standards frameworks, including:
- GB 15763.2 — Chinese national standard for tempered glass used in architecture, widely applied to curtain wall projects in China and Asia-Pacific markets.
- EN 12150-1 — European standard for thermally toughened soda-lime silicate safety glass, harmonized under the EU Construction Products Regulation.
- ASTM C1048 — North American specification for heat-strengthened and fully tempered flat glass.
- ANSI Z97.1 and 16 CFR 1201 — North American safety glazing standards that evaluate fragmentation behavior after impact loading using a weight-based criterion.
- JIS R 3206 — Japanese standard covering safety glass performance requirements.
Fragmentation testing is a destructive, laboratory-based type test performed by accredited third-party laboratories during product certification, factory production control audits, and project-specific compliance verification for facade contractors and glazing fabricators.
Testing Samples
Specimen Dimensions and Quantity
Each standard defines its own specimen geometry, and the laboratory must prepare samples to match the governing specification:
- EN 12150-1: test pieces of 1,100 mm × 360 mm, with a minimum of four specimens per glass configuration.
- GB 15763.2: square test pieces of 610 mm × 610 mm, typically with three specimens per configuration.
- ANSI Z97.1 / 16 CFR 1201: large panels of approximately 1,930 mm × 864 mm subjected to impact followed by fragmentation evaluation.
Samples must represent the actual production article: identical nominal thickness, coating type, edgework, and thermal treatment as the glass supplied to the curtain wall project.
Specimen Condition and Preparation
Specimens are cut to size before tempering, since fully tempered glass cannot be cut or worked after processing. Samples must be free of edge damage, chips, or pre-existing cracks that could initiate uncontrolled fracture. Prior to testing, specimens are conditioned at room ambient temperature, cleaned, and supported on a flat, cushioned surface such as a soft backing layer that retains particles in place for counting.
Sampling Basis
Samples are drawn from representative production lots of the same glass composition, thickness, and furnace recipe. For project-based verification, samples are typically taken from the same batch as curtain wall panels destined for installation, ensuring the test result is genuinely indicative of the delivered product.
Testing Items
The fragmentation test evaluates the following performance characteristics:
- Minimum particle count within a 50 mm × 50 mm reference square, compared against thickness-dependent acceptance limits.
- Maximum fragment size, including the length of the longest particle in the counting area.
- Presence of prohibited fragment types, such as long, dagger-shaped slivers or needle-like shards exceeding permitted lengths.
- Uniformity of fragment distribution across the specimen, which reveals stress imbalance or uneven quenching.
- For ANSI Z97.1 / 16 CFR 1201 evaluations, the aggregate weight of the ten largest fragments within a designated region, which must not exceed the weight of an equivalent area (approximately 6,450 mm² or 10 in²) of the original glass at the same thickness.
Testing Methods
Particle Counting Method (GB 15763.2 / EN 12150-1)
The specimen is placed horizontally on a suitable backing surface. A sharp-pointed fracturing tool, such as a spring-loaded punch or pointed steel instrument, is applied at the initiation point prescribed by the standard, typically near the midpoint of the long edge. Fracture must occur immediately on contact, and the fragment pattern is documented promptly, usually by photography, before secondary cracking can develop.
Particles are then counted within one or more 50 mm × 50 mm transparent counting templates positioned at the locations specified by the standard, excluding zones close to the initiation point and near the edges where fragment patterns are naturally atypical. In principle, the least fragmented area is assessed against the minimum particle requirement, ensuring the entire pane meets the safety threshold rather than only the most densely fragmented region.
Weight-Based Method (ANSI Z97.1 / 16 CFR 1201)
Under the North American approach, the specimen is first subjected to impact by a standardized shot bag from specified drop heights. Following impact and fracture, the ten largest particles within a designated measurement region are collected and weighed collectively. The combined weight is compared with the calculated weight of an equivalent area of the original glass. This criterion limits the size of individual fragments and thereby the injury potential of broken glazing.
Equipment and Test Environment
- Pointed fracturing instrument with sufficient hardness to initiate immediate fracture.
- Transparent 50 mm × 50 mm counting template with defined edge boundaries.
- Soft, contrasting backing material (typically black or white) that traps fragments in position for accurate counting.
- Controlled lighting, including oblique illumination, to make particle boundaries clearly visible.
- Calibrated laboratory balance for weight-based evaluations under ANSI Z97.1 / 16 CFR 1201.
- Camera equipment for permanent photographic records of every fragment pattern.
Acceptance Criteria
Acceptance limits vary with nominal thickness. As a general reference, fully tempered glass of common curtain wall thicknesses (approximately 6–12 mm) must produce a minimum of 40 particles within the 50 mm × 50 mm counting square under both GB 15763.2 and EN 12150-1, while reduced minimum counts are permitted for very thin or very thick products. Standards also restrict the number and length of elongated particles, since long slivers indicate locally insufficient temper or stress non-uniformity. Any specimen producing large, angular shards with a low particle count constitutes a failure of that configuration.
Applications
Fragmentation testing is applied throughout the curtain wall and facade industry:
- Product certification and CE marking of thermally toughened safety glass under EN 12150-1.
- Compliance verification against GB 15763.2 for building projects in Chinese and Asian markets.
- Factory production control and tempering furnace validation, helping processors optimize heating and quench parameters.
- Project-specific material submittals for unitized curtain walls, stick-built facades, and structural glazing assemblies.
- Verification of safety glazing used in balustrades, glass railings, canopies, skylights, overhead glazing, entrance doors, and interior partitions within curtain wall buildings.
- Failure investigation and root-cause analysis when installed glazing breaks in service.
Because curtain walls expose glazing to wind loads, thermal stress, seismic movement, and potential human impact, verified fragmentation performance is a fundamental element of facade quality assurance and public safety.
FAQ
Is the fragmentation test destructive? Yes. The test requires the complete fracture of the specimen, so it can only be performed on spare samples and never on glass that will be installed. For installed curtain wall panels, non-destructive alternatives such as surface stress measurement may be used as an indirect assessment.
How many samples are required? Typically three to four specimens per glass configuration, depending on the governing standard. All specimens must pass for the configuration to be accepted.
What causes a fragmentation test failure? Failures usually stem from insufficient quenching, low heating temperature, uneven air balance across the furnace nozzles, or glass exiting the tempering process below the required surface compressive stress. Corrective action typically involves furnace parameter adjustment followed by retesting of new samples.
Is fragmentation testing the same as a heat soak test? No. Heat soak testing (per EN 14179) is a separate screening process that accelerates the fracture of glass containing nickel sulfide inclusions to reduce the risk of spontaneous breakage. Fragmentation testing instead verifies the degree of temper and the safety of the break pattern.
Can coated or printed glass be tested? Yes. Samples must reflect the actual product, including low-emissivity coatings, ceramic frit, or tinted substrates, since coatings and glass composition can influence quench behavior and the resulting fragment pattern.
How often should fragmentation testing be performed? It is conducted during initial type testing and certification, at defined intervals for factory production control, and whenever the glass thickness, coating, furnace recipe, or processing line changes significantly.