Shear-lag effects in composite glazing panels

Journal Article (2025)
Author(s)

Carlos Pascual (University of Cambridge)

R.C. Hartwell (TU Delft - Structures & Materials)

P.J. Buskermolen (TU Delft - Teachers of Practice / AE+T)

Mauro Overend (TU Delft - Structures & Materials)

Research Group
Structures & Materials
DOI related publication
https://doi.org/10.1016/j.engstruct.2025.120863
More Info
expand_more
Publication Year
2025
Language
English
Research Group
Structures & Materials
Volume number
342
Reuse Rights

Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.

Abstract

Composite glass sandwich panels, consisting of glass face sheets bonded to linear stiffeners (spines) in the core region, can provide significant benefits in material efficiency, reduced thickness, and greater overall transparency. However, current analytical models of their mechanical performance fail to account for the non-uniform longitudinal stress distribution caused by shear-lag effects in wide structural panels. This study redresses this by means of experimental research on composite glazing panels with different loading and geometrical configurations. Six 4-point bending experiments were performed on 34 mm thick, 1000 mm long, and 700 mm wide composite glazing panels, made from soda-lime silica glass face sheets bonded to glass fibre-reinforced polymer core spines. Two types of adhesives were tested: a relatively low stiffness silicone-based adhesive, and a relatively high stiffness epoxy-based adhesive. The shear-lag effects are quantified in terms of effective width ratios (EWR). The study showed that the epoxy-bonded panels provided a significant degree of composite action (DCA = 0.85) whereas the composite action in the silicone-bonded panels was negligible. Furthermore, it was found that applying the EWR values from this study in a recently published analytical model yields predictions of maximum strains at mid-span that deviate by no more than 16 % from the experimental results.