Eurocode 10: Design Strength, Laminated Glass, and Post-Breakage

Material Properties and Strengths

Annealed glass: characteristic bending strength f_g,k

Glass Type	Product Standard	fg,k (N/mm²)
Float glass	EN 572-2	45
Polished wired glass	EN 572-3	33
Drawn sheet glass	EN 572-4	45
Patterned glass	EN 572-5	33
Wired patterned glass	EN 572-6	27

Pre-stressed glass: characteristic bending strength f_b,k

Glass Type	TTG (EN 12150)	HSG (EN 1863)	Chem. Strengthened (EN 12337)
Float / drawn sheet	120	70	150
Patterned	90	55	100
Enamelled float / drawn	75	45	—
Enamelled patterned	75	45	—

Partial safety factors

Factor	CC1	CC2	CC3
γM (basic material)	1.6	1.8	2.0
γP (surface pre-stress)	1.1	1.2	1.3

Reliability basis

The partial safety factors are calibrated against the reliability classes defined in EN 1990. The target failure probabilities and corresponding reliability indices are:

Consequence	Reliability Class	pf (1 year)	pf (50 years)	β (1 year)	β (50 years)
Small (CC1)	RC1	10−5	5 × 10−3	5.2	4.3
Normal (CC2)	RC2	10−6	10−4	4.7	3.8
Extraordinary (CC3)	RC3	10−7	10−5	4.2	3.3

The value γ_M = 1.8 for CC2 targets a 50-year reliability index β = 3.8, consistent with EN 1990. This has been confirmed by both FORM analysis assuming lognormal distributions and fully probabilistic approaches using Weibull statistics with random scratch orientation.

Design Bending Strength Formula

The total design bending strength consists of two components: intrinsic glass strength and pre-stress contribution:

Factor	Meaning	Typical Values
kmod	Load duration factor	Permanent: 0.29 • Snow: 0.43 • Wind: 0.74
ksp	Surface profile	As-produced float: 1.0 • Sandblasted: 0.6
ke	Edge / hole finishing	As-cut: 0.8 • Seamed: 0.9 • Polished: 1.0
kp	Pre-stressing process	Horizontal: 1.0 • Vertical: 0.60
ki	Interference factor	Accounts for statistical interference between surface flaws and pre-stress (NDP)
λA	Area size effect	Relevant for panes > 18 m²
λl	Edge length effect	Relevant for edges > 6 m

The first term captures the intrinsic strength of annealed glass, reduced by load duration, surface quality, and edge quality. The second term adds the beneficial contribution of pre-stress from thermal or chemical strengthening.

Laminated Glass: Three Modelling Levels

The interlayer shear coupling is the single most important factor for reducing glass mass and cost in laminated glass design. EN 19100 provides three levels of sophistication for modelling this coupling:

Level 1: Bounding Approach (No Shear / Full Shear)

The simplest approach: assume either no shear transfer (layered, conservative when shear is favourable) or full shear transfer (monolithic, conservative when shear is unfavourable, e.g. for load attraction in IGUs).

Level 2: Enhanced Effective Thickness (EET)

Based on the work of Galuppi & Royer-Carfagni (2012, 2014), the EET method calculates an equivalent monolithic thickness that produces the same deflection or stress as the actual laminate. It is an improvement over the older Wölfel/Bennison approach because it accounts for boundary conditions, load type, and multi-layered laminates.

The key parameter is the shear transfer coefficient η, which ranges from 0 (layered, no coupling) to 1 (monolithic, full coupling). It depends on the interlayer shear modulus G_int, which in turn depends on temperature and load duration.

This is where Prony series and time-temperature superposition become essential: the designer needs the interlayer shear modulus at the specific temperature and load duration of each design scenario.

Level 3: Numerical Models (FEM)

Level 3a: Linear elastic FE model with a constant shear modulus value for each load case (corresponding to the relevant temperature and load duration).

Level 3b: Transient viscoelastic FE model using Prony series and time-temperature superposition. This is the most accurate approach, but also the most computationally demanding.

Why shear coupling matters: a practical example

For a 1.0 × 1.0 m two-side simply supported laminated glass panel (2 × 6 mm glass, 0.76 mm interlayer) under wind design load w_d = 2.3 kN/m²:

Interlayer	G (MPa)	Γ	σEd (MPa)	Stress Reduction
No shear bond	0	0	23.4	—
Standard PVB	0.4	0.21	16.2	−31%
Stiff PVB (Trosifol ES)	7	0.82	10.9	−53%
SentryGlas (ionoplast)	100	0.98	10.4	−56%

Even standard PVB with G = 0.4 MPa provides a 31% stress reduction compared to the no-shear assumption. Increasing from stiff PVB (7 MPa) to ionoplast (100 MPa) has almost no additional effect. This demonstrates that even modest interlayer stiffness delivers significant structural benefit, and that the conservative “no shear” approach used in DIN 18008 leaves substantial design efficiency on the table.

Enabling thinner glass

The impact of shear coupling on glass thickness selection is dramatic. For the same panel under the same loads, the “no shear bond” assumption requires 6 mm glass to pass the design check (σ = 23.4 MPa < R_d = 27.7 MPa per DIN 18008). With 4 mm glass and no shear coupling, the stress would be 52.7 MPa, nearly double the resistance.

However, with a stiff PVB interlayer (G = 7 MPa), the same 4 mm glass achieves σ = 22.8 MPa, passing the design check. This means the use of 4 mm glass becomes possible only when shear coupling is properly accounted for, which EN 19100 enables through its three-level approach. For large facades, this translates directly into significant material and cost savings.

Design resistance across different codes

The design resistance for the same glass (annealed float, seamed edges, short-term wind load) varies significantly across codes:

Code	Rd (MPa)	Notes
DIN 18008 (Germany)	27.7	kmod=0.7, kc=1.8, γM=1.8
prCEN/TS 19100 (Eurocode)	20.8	kmod=0.7, ke=0.7, RM=0.9, γM=1.8
ASTM E1300 (USA)	18.3	From tabulated values for seamed annealed glass

The German standard allows a higher design resistance due to the construction factor k_c, which accounts for the positive effect of the construction type on post-breakage behaviour. In the Eurocode, the consequence class factor R_M serves a similar but more conservative role.

German building approvals (abZ): interlayer shear modulus values

In Germany, since DIN 18008 does not allow favourable shear transfer in the base code, product manufacturers must obtain a general building authority approval (allgemeine bauaufsichtliche Zulassung, abZ) that specifies the shear modulus values permitted for design. These approvals provide either individual G values for specific load cases or full Prony series:

Product	Approval No.	Prony Series	G Wind (MPa)	G Snow heated (MPa)	G Snow unheated (MPa)	G Dead (MPa)
Trosifol ES (stiff PVB)	Z-70.3-236	Yes	7	0.58	100	0
Saflex DG 41 (stiff PVB)	Z-70.3-230	Yes	2	0.44	20.4	0
EVASAFE (EVA)	Z-70.3-197	No	3.6	2.5	2.5	0
SentryGlas SGP 5000	Z-70.3-170	No	100	60	60	0
Glaschobond SGP 5000	Z-70.3-175	No	100	60	60	0
GEWE Composite (cast resin)	Z-70.3-240	Yes	1.23	1.23	4.81	0
Lamex X-Strong (std PVB)	Z-70.4-137	No	0.4	0	0	0
SGT extra safe (std PVB)	Z-70.4-165	No	0.4	0	0	0

Shear Coupling Across National Codes

One of the most significant differences between national glass design standards is how they handle interlayer shear coupling. The following table, based on Kuntsche et al. (2019), summarises the approaches:

Standard	Design Concept	Shear Modulus Approach
Germany DIN 18008	No favourable shear transfer; only “no shear” and “full shear” limit states	None (only via abZ approvals)
Austria ÖNORM B 3716	Partial shear bond possible; no specific method	One G value for PVB (0.4 MPa for wind)
Belgium NBN S 23-002	Effective thickness method, similar to prEN 16612	One ω value, one load case, PVB only
France NF DTU 39 P4	Reduction of equivalent required thickness	None
Italy CNR-DT 210	Three levels: (a) EET, (b) FEM constant G, (c) FEM viscoelastic	G values from manufacturers or testing
Netherlands NEN 2608	Partial shear bond, effective thickness method	Prony series + WLF for PVB and ionoplast
Norway NS 3510	Partial shear bond possible	Tabulated G for different load cases
USA ASTM E1300	Non-factored load charts + effective thickness	One G value, formula for ω (A = 9.6 always)
Europe prEN 16612	Effective thickness via stiffness families	Stiffness families 0, 1, 2 (ω tabulated)
Eurocode prCEN/TS 19100	Levels 1–3b: bounds, EET, FEM constant, FEM viscoelastic	G from EN/ETA or experimental testing

Germany is the most conservative: the base code prohibits favourable shear transfer entirely. At the other end, the Netherlands (NEN 2608) already includes Prony series and WLF constants for two interlayer types directly in the standard. The Italian CNR-DT 210 and the Eurocode share the most sophisticated approach with three modelling levels. EN 19100 will harmonise these disparate approaches into a single European framework.

Post-Breakage Behaviour

The FLS and PFLS are what make EN 19100 fundamentally different from other Eurocodes. Three resistance mechanisms are distinguished for laminated glass after breakage:

Mechanism I: All glass plies are sound. The interlayer provides shear coupling and composite plate theory applies. This phase ends when the first ply breaks.

Mechanism II: One ply is broken, the remaining ply or plies carry the full load. The interlayer retains the broken shards, and if the distance between cracks is large enough, the polymer still transfers shear between intact zones. This mechanism works better with annealed or heat-strengthened glass (large shards) than with toughened glass (small fragments).

Mechanism III: All plies are broken. Glass fragments carry compression by contact, while the polymer interlayer provides the tensile force (acting as a membrane). The load-bearing capacity depends heavily on fragment size:

• Annealed glass: Large shards → good post-breakage capacity
• Heat strengthened glass: Medium shards → good post-breakage capacity
• Toughened glass: Tiny fragments → “wet towel” effect → poor post-breakage unless a very stiff interlayer (ionoplast) is used

References

1. Feldmann, M. et al. (2023). “The new CEN/TS 19100: Design of glass structures.” Glass Structures & Engineering, 8:317–337. doi:10.1007/s40940-023-00219-y (Open Access)
2. Feldmann, M., Kasper, R. & Laurs, M. (2025). “EN 19100 Design of Glass Structures: Key changes and benefits through design examples.” JRC Workshop “The Second Generation Eurocodes”, 5 June 2025.
3. Feldmann, M., Kasper, R. et al. (2014). “Guidance for European Structural Design of Glass Components.” JRC EUR 26439 EN. Free PDF from EU Publications Office
4. Kuntsche, J., Schuster, M. & Schneider, J. (2019). “Engineering design of laminated safety glass considering the shear coupling: a review.” Glass Structures & Engineering, 4:209–228. doi:10.1007/s40940-019-00097-3
5. Siebert, G. (2025). “Benefits of revised German code for glass design.” Glass Structures & Engineering, 10:5. doi:10.1007/s40940-024-00286-9 (Open Access)
6. Galuppi, L. & Royer-Carfagni, G. (2014). “Enhanced effective thickness of multi-layered laminated glass.” Composites Part B, 64:202–213. doi:10.1016/j.compositesb.2014.04.018
7. Galuppi, L. et al. (2020). “BAM approach for IGU cavity pressure calculation.” Based on Betti’s Analytical Method using Green’s functions.
8. CEN/TS 19100-1:2021 — Design of glass structures — Part 1: Basis of design and materials.
9. CEN/TS 19100-2:2021 — Design of glass structures — Part 2: Out-of-plane loaded glass components.
10. CEN/TS 19100-3:2021 — Design of glass structures — Part 3: In-plane loaded glass components.