The strength of annealed, heat‐strengthened and fully tempered float glass

Abstract

A B S T R A C T In the last two decades architectural glass has made an enormous leap from a secondary material to a material that combines structural and cladding roles. The structural role is a new and problematic one. In contrast to most other engineering materials the strength of glass is not a material parameter but a parameter dependent on processing quality and damage to the glass surface. There is also no real agreement on how strong glass is. There is a concept Euronorm for structural glass that has values for the characteristic strength for annealed, heat-strengthened, fully tempered and chemically toughened glass. There is however no real agreement on the validity of these values for design of glass beams or columns. To provide an independent set of values a statistically significant series of four-point bending tests on glass were conducted using both lying and standing positions resulting in a set of values for the characteristic strength.

The structural strength of glass: hidden damage

“…For the more common float glass this is more complicated. Results of bending experiments by various authors suggest that the processing and specimen size influence the results and suggest systematic data deviation from the Weibull statistic distribution [3][4][5].…”

“…For the more common float glass this is more complicated. Results of bending experiments by various authors suggest that the processing and specimen size influence the results and suggest systematic data deviation from the Weibull statistic distribution [3][4][5].A likely explanation for this is that the usual processing of float glass results in multiple types of defects which provides a multilinear Weibull plot [6,7].To investigate this systematically it was decided to look in more depth at the effect of processing on float glass strength. The initial step, which is described in this paper, deals with the effect of cutting and breaking quality on the strength of processed float glass.…”

A numerical method for analysis of fracture statistics of glass and simulations of a double ring bending test

“…(8) in one form or another (prEN 16612:2013). A number of studies, however, have indicated that the Weibull distribution does not provide a superior fit compared with a lognormal or normal distribution (Lü 1997;Calderone et al 2001;Veer et al 2009;Huerta et al 2011;Kinsella and Persson 2016). It has been noted that the estimated value of the Weibull shape parameter varies quite significantly from one sample to another in experiments (Ritter et al 1985;Carre 1996;Huerta et al 2011).…”

“…First, glass fracture statistics tend to produce bimodalities in the probability plots according to e.g. Veer et al (2009). In fact, the experiment considered in Sect.…”

“…There is disagreement among researchers as to which prediction model is the correct one to use (Fischer-Cripps and Collins 1995). A range of experiments have shown a consistent bilinearity in the probability plots when the Weibull distribution is used for modelling the strength of annealed glass (Veer 2007;Veer et al 2009). As regards the GFPM, it has been said that it "is best suited to representing glass strength for specific test conditions."…”