1. Introduction

The United States, along with much of the world, lacks standardized practices for utilizing glass as a structural material in buildings. Over the past decade, several challenges have become evident:

• Glass’s inherent brittleness.
• Historical acceptance of breakage and fallout in non-structural applications as a trade-off for transparency.
• That glass is now much larger than was traditionally available.
• The historical use of glass in buildings predating the development of suitable interlayers.
• Glass’s vulnerability to damage from unforeseen loading events.
• The context-dependent acceptability of glass applications, complicating the formulation of universal guidelines.

In the absence of standardization, Authorities Having Jurisdiction (AHJs) have adopted varied approaches, from deferring to design experts to implementing ad-hoc regulations or outright bans, and/or mandating costly testing protocols.

Voluntary specifications and design guides play a crucial role in establishing standardized practices that will promote the use of glass as a structural material in buildings. The Structural Glass Design Manual serves as a valuable resource for architects who may not specialize in glass design, providing engineers and AHJs with a reference for considerations in their projects. While creating such a document is no small feat—otherwise, it would have already been accomplished—even an initial, imperfect version lays the groundwork for future refinement. This is an essential step toward achieving standardization and broad acceptance of glass as a viable structural material in construction.

Glass possesses high compressive strength but is brittle and significantly weaker in tension. Recent advancements in lamination technology and increased capabilities in glass fabrication suggest that glass has considerable potential as a structural material. The conventional use of glass for non-structural windows does not fully capture its current range of applications. While some projects have successfully used glass structurally, the lack of standardized design guidelines has limited its broader application and acceptance by Authorities Having Jurisdiction (AHJs).

The Structural Glass Design Manual addresses the specific considerations for using glass as a structural material, contributing to its wider acceptance for such applications. For glass to be utilized to its fullest potential, it must be easily specifiable by architects and include common engineering practices and safety considerations for AHJs. The Manual provides information to support these objectives.

This paper presents the main elements of the Structural Glass Design Manual and refers to other papers from this conference. The scope of a conference paper limits the inclusion of all details from the design manual. At the time of the conference it is anticipated that the Structural Glass Design Manual will be available in public draft format for comment.

2. Structural vs Non-structural

For the context of this paper, ‘structural glass’ refers to applications where the glass element or system supports other elements or systems, or where failure would impact safety, function, or have consequences beyond repair or replacement costs. Conversely, ‘non-structural’ glass is defined as an element or system whose failure results primarily in repair or replacement costs with minimal other consequences.

ASTM E1300 is used for the statistically acceptable use of glass under uniform load with continuous support on one to four edges. The underlying glass failure prediction model does not consider the peak tensile stress, rather it integrates stress over an area with a probability function that accounts for flaw distribution in weathered glass. In this method small areas of high stress are acceptable. This standard is effective for defining acceptable usage in windows and enables efficient design by recognizing that the most critical flaw is unlikely to coincide with the maximum stress point, as empirical testing often shows. However, this assumption may not be suitable for structural elements where failure could lead to significant consequential damage. Therefore, design for these systems should prioritize reliability over statistical acceptability. To address this, a structural glass design manual has been developed to guide design approaches that differentiate between window glass and structurally critical glass applications.

The Structural Glass Design Manual (SGDM) prioritizes reliability in design, assuming that the critical flaw could coincide with the maximum stress point, leading to a limit state stress-based design approach. It also acknowledges that glass components may be vulnerable to reduced capacity due to various factors such as inclusions, surface damage, or impact which are not part of the standard load and resistance calculations. The SGDM emphasizes good design practices that enhance robustness and safety, ensuring that if one component fails, the overall structure remains secure.

While many design guides concentrate on the glass’s capacity before fracture, the SGDM also considers the robustness required for an element to withstand loads in a damaged state, an area with limited existing guidelines. CEN/TS 19100 (parts 1~3) outlines tests for the unfractured state, fracture event, and post-fracture state but does not provide comprehensive guidance on their application.

It is not practical to expect all structural glass to function fully in a cracked state as this would increase costs significantly. Glass failures that occur are often due to impact, inclusions, or installation errors rather than load capacity exceedance. Since extraordinary events rarely occur simultaneously with peak variable design loads, it is often reasonable to design for reduced loads. The causes of failure are generally unrelated to design load distribution; therefore, applying load and resistance factors alone cannot prevent initial failure from these causes. The goal is to ensure that the glass can bear a reasonable load for a reasonable duration with an acceptable level of damage without disproportionate collapse. While some cases are reasonably evident, the matrix of possibilities is complex and requires sound judgement as to when requirements or restrictions should apply. The Design Manual is intended to provide guidance.

3. Load and Resistance Factor Design

Structural use of glass includes many load types, each with their own load distribution. A design standard needs to consider the distribution of both load and resistance to maintain a consistent reliability and 'safety factor'. ASCE 7 notes that Allowable Stress Design (ASD) does not achieve consistent target reliabilities. To be consistent with the load practices in ASCE 7 and the overall target reliabilities for various failure modes, the procedures for 'other materials' in ASCE 7 sections C1.3 and C2.3 are used as a baseline. These procedures are also consistent with the tension-stress-based strength models in EN 16612 and CEN/TS 19100, which have been adapted in the SGDM to be consistent with ASCE 7 practices and nomenclature. For the potential complexity of structures and analysis required for glass structures, engineers are more familiar with stress-based analysis than surface wide probability integrals, such as adopted in ASTM E1300.

The LRFD results are based on achieving a pre-established acceptable reliability of outcome. Reliability for loads is achieved by using load factors and load combinations. ASCE7 provides acceptable loads and load combinations (which are purposefully established independent of material load resistance) and a generalized methodology to utilize failure test data to determine appropriate values for ultimate load resistance.

Capturing the state of practice for loads, resistance, and reliability are all essential steps toward achieving similar performance to other structural materials using the same loading standards and broad acceptance of glass as a viable structural material in construction. We have recognized that building design professionals and authorities are more likely to accept glass as a structural material if the material follows consistent, reliable, and well understood engineering and design properties on par with other building materials used within structural design. To achieve this, the SGDM uses familiar target reliabilities, excluding extraordinary events, that are already defined within the ASCE 7 and special practices for handling extraordinary events. (See also separate paper at this conference.)

The SGDM methodology presented herein is similar to other building materials, as it uses many of the same fundamental principles of structural mechanics. Below is a summary of principals upon which the manual is based:

1. Establish a series of assumptions that may be used in developing a limit state LRFD design methodology for structural glass. The determination of a glass element’s usability is based on reliability theory as defined within ASCE7 for various limit states.
2. Use previously established glass failure test data as a baseline, then use the process described above within ASCE7 to calculate the ultimate load resistance values for recognizable structural engineering limit states when using annealed, heat-strengthened, and fully tempered structural glass.
3. For in service (undamaged) the SGDM follows LRFD load combinations in ASCE7 with few exceptions.
4. There are special load cases for extraordinary loads and post-damage limit states.
5. Loads/actions are converted into their respective three principal axes and then compared to the glass material section properties along those three principal axes.
6. Capacities are determined within separate limit states and compared to the loads for determination of adequacy in design.
7. Element properties within the limit states are presumed flat with a uniform thickness.
8. The glass material is isotropic, completely elastic, and follows Hooke’s law.
   
9. Glass does not strain-harden or yield. Glass will fail in a brittle fashion according to the Griffith mechanism of fracture mechanics (1920).
10. Surface tension controls capacity.
11. The critical flaw initiating fracture is assumed to occur at the point of largest stress (differing from ASTM E1300) and once initiated, is presumed to propagate through the entire ply of a glass element.
12. Monolithic glass element sections perpendicular to the axis of bending, which are in-plane before bending, remain in the plane after bending (aka “Plane sections remain plane”.)
13. Laminations (aka interlayers) are often located between glass plies to prevent the propagation of fracture between glass plies within the same glass element.
    • Laminated glass sections perpendicular to the axis of bending, which are in-plane before bending, may not remain in plane after bending, particularly for lateral bending and torsion.
    • Interlayers are viscoelastic materials of uniform thickness in between glass plies. Interlayer shear deformations are non-linear according to temperature and load duration.
14. Glass strength levels account for “weathering effects”, e.g. load duration and stress corrosion. This means that non-design load mechanics contribute to glass fracture.
    
15. The thickness of any glass element is presumed to always be small in comparison with the other dimensions of the element. Therefore, weak axis deflections are potentially large, therefore nonlinear geometric deflections must be considered within every limit state.
16. Imperfections may interact with loading and have a significant effect on surface tension.
17. Because nonlinear geometric deflections must always be considered, manufacturing, installation, and design tolerances must be explicitly stated within the project documents as design criteria that the contractor must abide by for a successful project.

4. Design Manual Structure

Since glass frequently interfaces with materials like aluminum, steel, stainless steel, and cold-formed steel — all of which adhere to standards with a uniform chapter format — it is logical for the glass design guide to adopt a similar structure for consistency.

Figure 1: Table of Contents

GLOSSARY

A. GENERAL PROVISIONS

A.1 Scope

A.2 Exclusions

A.3 Reference Specifications Codes and Standards

A.4 Materials

A.5 Professional Use and Limitations

B. DESIGN REQUIREMENTS

B.1 General Provisions

B.2 Design Basis

B.3 System Categorization

B.4 Loads and Load Combinations

B.5 Material Properties

B.6 Design Strength of Glass

C. DESIGN FOR STABILITY AND DIRECT ANALYSIS

C.1 General Requirements

C.2 Calculation of Required Strengths

C.3 Calculation of Available Strengths

C.4 Braces

D. DESIGN OF MEMBERS FOR TENSION

D.1 Principal Tensile Stress Analysis

D.2 Principal Tensile Stress Capacity

E. DESIGN OF MEMBERS FOR COMPRESSION

E.1 Monolithic Flat Plates

E.2 Elastic Critical Buckling – Laminated Flat Plates

E.3 Imperfect columns and eccentric loads

E.4 Cruciform and Angle Columns

F. DESIGN OF MEMBERS FOR FLEXURE

F.1 General Provisions

F.2 Moment Capacity of Beams in Minor Axis Bending

F.3 Moment Capacity of Imperfect Beams in Strong Axis Bending

F.4 Elastic Critical Buckling Moment

G. DESIGN OF MEMBERS FOR SHEAR

G.1 Shear Strength

G.2 Shear Stability

G.3 Torsional Shear

H. DESIGN OF MEMBERS FOR COMBINED AXIAL AND FLEXURAL FORCES

H.1 for P/Pcr ≤ 0.15 and Mx /Mcr ≤ 0.4, the following condition is adequate:

H.2 for P/Pcr ≤ 0.15 and Mx /Mcr > 0.4, the following condition is adequate:

H.3 If P / Pcr > 0.15 or Mx/Mcr > 0.4

I. DESIGN OF COMPOSITE MEMBERS

I.1 Introduction

I.2 Extended Enhanced Effective Thickness (EEET)

J. CONNECTIONS

J.1 General

J.2 Methods of Analysis

J.3 Empirical/Proprietary Data

J.4 Justification by Testing

J.5 Justification by Rational Analysis

J.6 Countersunk Holes for Transfer of Out-of-Plane Loads

J.7 Isolation of In-Plane Loads

J.8 In-Plane Bearing Holes Through Glass

J.9 Friction Type Connections

J.10 Adhesives

K. Compatibility

K.1 Deformation compatibility

L. USE OF TEST DATA FOR DESIGN

L.1 GENERAL

L.2 PROOF TESTING – Validation of Strength

L.3 PROTOTYPE TESTING - Design by Testing

M. DESIGN FOR SERVICEABILITY (Non Mandatory)

M.1 Typical Deflection Limits

N. Other Considerations (NON MANDATORY)

N.1 Standards

N.2 Tolerances and Imperfections

N.3 Durability and Maintenance

AA. APPENDICES

AA.1 Recommended Minimum Glass Risk Categories

AA.2 Discussion of LRFD Principles as Applied to Structural Glass Design

AA.3 Advanced Method for Determining the Critical Buckling Moment of Laminated Beams with Continuous Restraint

AA.4 Discussion of Balustrade Loadings

AA.5 Other Useful References

------------------------------------------

5. Section Descriptions

Each chapter is briefly discussed, emphasizing the key technical principles employed.

A - SCOPE

The Structural Glass Design Manual (SGDM) is developed for the use of glass as a structural component in buildings and does not aim to supersede standards designed for non-structural uses. ‘Structural’ in this context implies that failure of the element carries significant consequences, whereas ‘non-structural’ pertains to elements where failure presents an acceptable level of risk. When the SGDM does not cover a specific aspect within its scope, it directs users to other reference documents. Unlike ASTM E1300, which applies to window glass with statistically acceptable usage methods, the SGDM adopts an approach aligned with Load and Resistance Factor Design (LRFD), similar to that used for other structural building materials, to manage risks associated with design loads and extraordinary events.

B - DESIGN REQUIREMENTS

Robustness and Glass Risk Category

Designing with glass presents a significant challenge in determining the appropriate level of robustness for different situations. There is a wide range of design approaches, from ASTM E1300, which does not consider post-failure behavior, to ASTM E2751 for glass walkways, which demands retention, redundancy, and robustness, ensuring no additional breakage after one critical layer fails. A separate presentation at this conference delves into the philosophy of robustness classification based on occupancy, situation, and potential impact of structural glass failure.

The SGDM’s design requirements are guided by the principles of ASCE 76, which assesses the risk of disproportionate collapse in buildings as Risk = Hazard Likelihood × Vulnerability × Consequences. For in-service conditions, it follows the standard load demand and capacity model. Post-damage vulnerability is managed through robustness requirements determined by the Glass Risk Category, which considers occupancy and potential consequences. Robustness levels vary from no specific requirements to allowances for one or two layers broken under reduced loads. Another conference paper further explores this aspect of glass design.

The Glass Risk Category (GRC) is based on principles akin to those in ASCE 7 and IBC. However, while the highest occupancy dictates the standards for an entire building in those codes, the GRC specifically applies to the occupancies in the proximity of the glass and their egress routes.

Figure 2: Table B‑1 Glass Risk Category Descriptions

Robustness is determined from a combination of element type, application and Glass Risk Category.

Some definitions:

System – a group of structural and/or non-structural elements, assemblies, or both, interacting to serve a common purpose, which may include glass, interlayers, assemblies, structural adhesives, reinforcing elements, and connections providing a structural load path.

• infill, n – elements or assemblies that are only required to support themselves and applied loads.

Other than those that are part of the primary structural system.

• secondary, n – a structural element, assembly, or system that is not considered a primary or infill system:

• primary, n – an element, assembly, or system that constitutes one of the following:

• Columns, including glass fins and glass walls which support trafficable roof and/or floor elements;
• Structural members having direct connections to the columns, including glass fins and walls, and are required for stability of the columns;
• Elements, assemblies or systems that support secondary systems and the failure of which would cause the collapse of the secondary system.
• Any Element, assembly, or system which is required to maintain the lateral stability of a structure.
• Bracing member(s) that are essential to the stability of the above elements.

The minimum amount of robustness for each type of element can be determined from flow charts for each element type.

Figure 3: Robustness Requirements Flowchart

Figure 4: Table B‑2 Robustness Requirements

Each type of glass element has specific robustness requirements based on the Glass Risk Category, as well as factors like height, orientation, and use. Unlike the Building Risk Category, where high risk in one part impacts the whole building, the Glass Risk Category only influences glass near the relevant occupancy or its escape route.

There’s substantial built precedent with vertical glass walls ranging from 12 to 20 feet (4 to 6 meters) high, without redundancy or retention features, which have been largely successful with minimal injuries. Like all glass standards, there’s a balance between reducing costs and ensuring safety. For Risk Category II designs, there’s been debate over the appropriate height at which to mandate glass retention. It’s widely accepted that glass under 10 feet (3 meters) is safe without retention, while glass over 26 feet (8 meters) must have it. For heights between 10 and 26 feet (3 to 8 meters), potential compromises include allowing designer discretion if certain conditions are met, such as using an AN/HS fin element with structural silicone on one edge or a heat-soak tested Fully Tempered glass fin protected from edge impact.

For simplicity, a nominal height of 18 feet (6 meters) was chosen as the cutoff between Robustness Categories R-2 and R-3, which requires additional checks when one layer is broken. A voluntary specification provides flexibility for those familiar with project specifics to set suitable limits.

Limit States

In addition to the usual strength ultimate limit state (ULS) and serviceability limit state (SLS) the SGDM includes guidance for post-damage limit state and, where appropriate, damage-event limit state. Where possible, the SGDM follows ASCE 7 extraordinary event load combinations whist providing additional information suitable for structural glass design.

Strength Model

Structural glass supports various load types, not just wind load or purely live load, and to utilize load distribution data from LRFD codes like ASCE 7, a limit-state strength model is employed. This model prioritizes reliability over statistical usage, assuming a flaw exists at the most critical point. ASCE 7’s Table 1.3.1 outlines target reliability indices for different failure modes, with the required reliability increasing with Building Risk Category and as the failure mode becomes sudden or progressive. Notably, the target failure probability for non-sudden failure that does not progress in building risk category I is still greater than in E1300 for 50-year wind event. Moreover, these targets don’t account for extraordinary events, to which glass is particularly vulnerable. Structural glass needs both a lower probability of failure then windows and control of extraordinary events.

The SGDM aims for a reliability index of β=3.5 and employs robustness indices to prevent extensive damage spread. This is similar to other structural materials for building risk category II. The chosen strength model aligns with peak principal tensile stress, similar to EN16612 and CEN/TS 19100 formulas. Adjustment factors for this model cater to different reliability targets as per ASCE7 equation C2.3-2, considering the different coefficients of variation for annealed (AN), heat-strengthened (HS), and fully tempered (FT) glass types. These adjustments standardize to Glass Risk Category II (G-II), acknowledging the correlation between glass risk and building risk categories, with reliability adjustments made through load modifications across risk categories.

Loads and Load Combinations

Load combinations for an undamaged state mirror those in ASCE 7 and the International Building Code (IBC), with an additional combination for removing snow from roofs with skylights, a scenario identified by NiOSH/OSHA as having caused multiple fatalities. Post-damage load conditions are reduced, based on the low probability that significant damage to a glass ply will occur and be followed by a design-level event, like a 50-year wind, before repair. However, exceptions include constant loads like dead load and scenarios where damage and loading might coincide, such as with live load.

The likelihood of such synchronous events depends on occupancy and is addressed by making residual capacity dependent on the Glass Risk Category. Since glass’s response to loads can vary over time and ASCE7 doesn’t define load durations, a range of durations is proposed, allowing design professionals to make informed choices based on project specifics. These suggestions align with ASTM E1300 standards, plus an additional 0.3 seconds for sudden load redistribution in damage events.

Post-Damage Loads are reduced from typical load combinations due to the principle that damage is a rare event and the damaged state has conditional probability. There is a presumption that damage to the glass will be repaired in reasonable time and prior to a peak event.

Post-Damage Residual Capacity Combination

α_D D + α_W W + α_S S + α_T.T

α_D D + α_L L + α_S S + α_T.T

Figure 5: Table B‑5 Post-Damage Load Factors

*Note that in the United States, wind load, W, is a factored limit state load, for regions where W is at an ASD (~50 year mean recurrence interval) divide the coefficients by 0.6. e.g. α_W G-4: 0.6/0.6=1,0.

C - DESIGN FOR STABILITY AND DIRECT ANALYSIS

Glass is elastic to the point of brittle fracture, but interlayers are visco-elastic with deflection increasing as a function of time and temperature. Glass is also often used at slenderness ratios that are much greater than typical construction materials. Glass is much stronger in compression than in tension. The usual compression yield failure-based design principles are inadequate for structural glass design. The slenderness makes amplification of imperfections under load, with resulting tensile stress due to minor axis deflection important to design. The lack of ductile yield plateau makes those tensile stresses critical.

The Direct Analysis Method is applicable to glass design with care. As with other structural materials, imperfections can be accounted for by modelling the imperfection explicitly or applying notional loads. Models need to account for P-Δ non-linearity, while P-δ effects may be accounted for within the model or the relevant chapter for the load type.

D - DESIGN OF MEMBERS FOR TENSION

The absence of ductility in glass means that analyzing local stress concentrations in tension members is vital. The strength model accounts for this by reducing the stress capacity for edges under uniform tension. There’s also guidance on how to handle areas around holes, re-entrant corners, and connections.

E - DESIGN OF MEMBERS FOR COMPRESSION

Glass is incredibly strong under compression, but its tensile strength on a flawed surface can be less than 1% of its compressive strength. When considering compression, it’s important to note how the load is applied; uneven support can deform the glass and create tension near the load application point. Additionally, slenderness and imperfections matter; compression on slender, imperfect members can cause significant secondary bending and tensile stress before reaching buckling loads. The guidelines include formulas to calculate incremental displacements and the tensile stresses they cause due to compression. These formulas are based on Luible’s work, also featured in ‘Structural Use of Glass²¹’.

F - DESIGN OF MEMBERS FOR FLEXURE

Extending on a separate paper at this conference by Green, Bedon and Galuppi (2022) is dedicated to the design of beams and cantilevers including: calculation of St Venant torsion stiffness and warping stiffness for both monolithic and laminated sections, effective thickness modelling, continuous elastic restraint from structural silicone, and the effect of imperfections. The use of Kala’s equation relating (Euler) critical elastic buckling moment, section properties, elastic stress capacity and imperfection magnitude allows an analytical process for various levels of imperfection and production/installation specification.

Notes:

• as glass members are frequently slender, the stress capacity of fully tempered glass may exceed the Euler buckling stress.
• The design curve Mn asymptotes to Mcr, whereas the finite element method captures post buckling capacity. The design curve including safety factors asymptotes to Φ.Mcr.

G - DESIGN OF MEMBERS FOR SHEAR

While it is extremely rare that shear controls design at a material strength level, significant amounts of commentary is provided regarding tension arising from applied shear loads and the potential for buckling causing secondary bending and tension stress.

H - DESIGN OF MEMBERS FOR COMBINED AXIAL AND FLEXURAL FORCES

(Section still under development) The combination of axial forces and bending moments in glass structures is complex, especially because glass is often used in slender forms and the limits are set by resulting tension rather than the usual compression yield or crushing. For cases with low axial load and moment, simply adding up the stresses is considered sufficient. When axial stress is low but the moment is moderate, a more detailed approach is used: calculate the amplified tensile stress due to axial load (from Chapter E) and divide it by the tensile stress capacity, then calculate the applied moment and divide it by the moment capacity (from Chapter F). The sum of these ratios is then compared to unity.

For high load combinations (at the time of writing this is still being validated) the following process is used:

1. The design stress capacity for the type of glass is calculated
2. The stress demand and amplified imperfections due to axial force are calculated
3. The residual stress capacity, the amplified imperfections and a reduced buckling moment allowing for the axial compression are used as input to calculate the reduced moment capacity using a process similar to Chapter F.
4. Where the reduced moment capacity, considering the axial load, is greater than moment demand, then the section is acceptable.

I - DESIGN OF COMPOSITE MEMBERS

The Extended Enhanced Effective Thickness Method is thoroughly discussed in another paper at this conference by Green, Bedon, and Galuppi. It provides processes for generalizing the laminate sandwich which can be analysed; includes St Venant torsional stiffness for stability (as well as proposals which are currently being validated for warping stiffness).

J - CONNECTIONS

Connections are one of the most interesting and challenging aspects of glass structures. To attempt to provide prescriptive solutions would potentially limit future innovation in this critical area. One of the philosophies of the Design Manual is to provide multiple paths to problem solving without limiting designs to prescriptive approaches. Multiple paths are offered/suggested: 1 Empirical/proprietary capacity data; 2 Testing and; 3 Rational analysis, including modelling or relevant formulas. By offering multiple paths it promotes innovation and is a valuable resource for practitioners.

The empirical route allows design through past experience and testing, or data developed by proprietary suppliers; it also encourages a process of product certification.

Testing is outlined in Chapter L with options for validation of design by rational analysis (small sample) or design by testing, with requirements to achieve statistical significance (larger sample with variance and confidence intervals.)

Rational analysis is an available option with guiding commentary on areas to be wary of, particularly when used by engineers who may be less experienced with glass.

Specific guidance is provided for:

1. Modelling of laminated glass with out of plane loading at point fixings
   1. Concentrated loads: 3-dimensional effects due to out of plane compression of interlayer
   2. Contraflexure: local effects on effective thickness due to reverse curvature generating shear at the interlayer that is not fully developed or not favorable.
2. Countersunk Holes for Transfer of Out-of-Plane Loads
3. Isolation of In-Plane Loads
4. In-Plane Bearing Holes Through Glass
   1. Stress Concentrations; Fabrication Tolerances; Edge Strength; Bearing
5. Friction Type Connections
6. Adhesives

K - COMPATIBILITY

Apart from extraordinary events such as vehicular impacts or bomb blasts, one of the few design cases that can potentially fracture multiple layers of glass simultaneously is the incompatibility between the glass structure and the surrounding structure. Due to its high in-plane stiffness, glass can attract significant forces when subjected to incompatible displacements.

The guiding principle for design should be to prevent damage or fracture at the Allowable Stress Design (ASD) level, which corresponds to a ~50-year mean recurrence interval; and there should be no collapse or fallout at the Strength Limit State (LRFD), which corresponds to a 700~1500 year mean recurrence interval.

It is crucial to maintain the engagement of connections and isolate differential movements. When displacements cause changes in the load path, the capacity of the new load path under displacement must be justified.

L - USE OF TEST DATA FOR DESIGN

The testing section is intended to provide general guidance for achieving statistical significance without limiting the scope to specific test methods. The section follows similar principles to AS1170.0 Appendix B with substantial specialization for structural glass. It has three sections: general guidance and reporting requirements; validation testing for systems otherwise numerically analyzed; and prototype testing for achieving statistical significance of adequacy by testing alone.

Note: Testing factor k_t is applied to LRFD load combinations.

M - DESIGN FOR SERVICEABILITY (Non Mandatory)

Design for Serviceability serves as a non-mandatory companion to the Compatibility chapter. It not only provides guidance on common displacement limitations but also includes sections on applicable loads, durability, temporary conditions, and considerations for ponding.

A collection of common serviceability and deflection criteria and references.

N - OTHER CONSIDERATIONS (Non Mandatory)

Materials such as steel have companion standards that deal with fabrication tolerances, installation, and quality control. The Design Manual does not attempt to create specifiable standards, rather it highlights the importance of coordination between the design assumptions and the specified fabrication and installation tolerance parameters.

Guidance is also provided regarding durability and maintenance, staining, mechanical abrasion, chemical attack, and moisture effects.

Long term success of all structures is dependent on correct maintenance; thus sections are dedicated to maintenance manuals and repair and replacement strategy documentation.

Appendices

AA.1 Recommended Minimum Glass Risk Categories

Appendix AA.1 offers non-mandatory guidance on choosing an appropriate Glass Risk Category. This format, similar to ASCE 7’s building risk category based on occupancy, differs in that the Glass Risk Category only affects glass in the proximity of the relevant occupancy or its egress path.

AA.2 Discussion of LRFD Principles as Applied to Structural Glass Design

Appendix AA.2 discusses why the Design Manual prefers limit state design (LRFD) over allowable stress design (ASD) used in standards like ASTM E1300, ASTM E2358, and ASTM E2751. A separate conference paper is dedicated to this topic. In summary:

The aforementioned standards focus on one predominant load type, while the Design Manual considers multiple load types.

• Standards like E1300 address load resistance to a probability of breakage for non-structural applications, offering statistically acceptable usage at reliability levels lower than other structural materials.
• The Design Manual aims for structural reliability comparable with other materials and considering potential damage due to non-design loading in consequential structural applications.

AA.3 Advanced methods for stability and known imperfections

Appendix 3 is reserved for Advanced Methods of Determining Elastic Critical Buckling Loads of beams and cantilevers. (Green et al. CGC9 2024)

AA.4 Derivation of design loadings for balustrades

Appendix 4 explains why the Design Manual adopts balustrade loadings from international standards like EN1991, AS/NZS 1170.1, BS 6993, ABNT NBR 6120, etc., instead of ASCE 7. This topic is also covered in a separate conference paper.

AA.5 Other useful references

Appendix 5 lists other useful references and standards.

6. Conclusions

In areas where there is a lack of standards, it is still important to have a document that has a collection of useful criteria that is in a form that is specifiable on a project-by-project basis. The Structural Glass Design Manual is not intended to be a comprehensive standard, however it addresses a number of the challenges to designing structural glass on a voluntary basis by providing 4 consistent levels of risk suitable for different occupancies and risk objectives. It has identified and addressed a number of deficiencies in common practices and existing standards. For some of the more challenging aspects of glass design it has proposed novel technical solutions.

The Structural Glass Design Manual is a document aimed a promoting the use of glass as a structural material by providing simply specifiable Glass Risk Categories for Architects, design practices for Engineers and robustness categories to promote safe usage for consideration by AHJs as a supplement to the usual code requirements.

At the time of the conference it is anticipated that the Structural Glass Design Manual will be available in public draft format for comment.

ANSI/AISC 360 Specification for Structural Steel Buildings. American Institute of Steel Construction

ASTM E1300–- Standard Practice for Determining Load Resistance of Glass in Buildings, ASTM International. West Conshohocken, PA

ASTM E2751 - Standard Practice for Design and Performance of Supported Laminated Glass Walkways, ASTM International. West Conshohocken, PA

Australian Standard AS1288 Glass in Buildings – Selection and Installation, Standards Australia GPO Box 476 Sydney, NSW 2001 Australia.

Bedon, Chiara, Belis, Jan, and Amadio, Claudio (2015). Structural assessment and lateral-torsional buckling design of glass beams restrained by continuous sealant joints. Engineering Structures, 102 (11): 214-229.

Bedon, Chiara (2021). Lateral-torsional buckling (LTB) method for the design of glass fins with continuous lateral restraints at the tensioned edge. Composite Structures, 266: 113790.

CEN/TS 19100-1 Design of glass structures–- Part 1: Basis of design and materials

D’Ambrosio, Gianmaria, and Galuppi, Laura (2020). Enhanced effective thickness model for buckling of LG beams with different boundary conditions. Glass Structures & Engineering, 5(2): 205-210.

EN 16612: Glass in building. Determination of the lateral load resistance of glass panes by calculation

EN 1993-1-1: Eurocode 3: Design of steel structures–- Part 1-1: General rules and rules for buildings

Galuppi, Laura, and Royer-Carfagni, Gianni (2014). Enhanced Effective Thickness of multi-layered laminated glass, Composites: Part B, 64: 202–213.

Galuppi, Laura, and Royer-Carfagni, Gianni (2020). Enhanced Effective Thickness for laminated glass beams and plates under torsion. Engineering Structures, 206: 110077.

Galuppi, Laura, Manara, Giampiero, and Royer-Carfagni, Gianni (2013). Practical expressions for the design of laminated glass. Composites: Part B, 45: 1677–1688.

Green, R., Bedon, C., Galuppi, L.: Design and stability of laminated glass fins with continuous lateral silicone restraint. Glass Structures and Engineering 8(3):1-20 http://dx.doi.org/10.1007/s40940-023-00224-1

Green, R., Bedon, C., Galuppi, L.: Moment capacity of imperfect slender beams and cantilevers with continuous silicone restraint, Challenging Glass Conference 9 (CGC9), 2024.

Griffith, A.A.: The Phenomena of Rupture and Flow in Solids. Phil. Trans. R. Soc. A221, 163-163 (1920)

Haldimann, M; Luible, A; Overend, M - Structural Use of Glass IABSE (2008)

Kala, Zdeněk (2013). Elastic Lateral-Torsional Buckling of Simply Supported Hot-Rolled Steel I-Beams with Random Imperfections. Procedia Engineering 57: 504-514.

NCCI SN-003-EN-EU NCCI: Elastic critical moment for lateral torsional buckling, by Access Steel. https://eurocodes.jrc.ec.europa.eu/doc/WS2008/SN003a-EN-EU.pdf

NCCI SN-006-EN-EU NCCI: Elastic critical moment of Cantilevers, by Access Steel.

Nizich, Adam J, La GReca, Andrea M and Galuppi, Laura. Advances in Effective Thickness For Laminated Glass Structural Design, Facade Tectonics Institute's biennial World Congress, 2022.

Simulia (2022), ABAQUS computer software, v.6.14, Providence, RI, USA.

Strand7 finite element software, R3.1.1, Sydney, NSW, Australia. https://www.strand7.com/