Verification formulae for structural glass under combined variable loads. (15th January 2015)

• Record Type: Journal Article
• Authors: Franco, Annalisa; Royer-Carfagni, Gianni
• Is Part Of: Engineering structures. Volume 83(2015:Jan. 15)
• Page Start: 233
• Page End: 242
• Publication Date: 2015-01-15
• Subjects: Structural glass -- Glass strength -- Subcritical crack growth -- Static fatigue -- Load combinations -- Verification formulae -- Load history; Structural engineering -- Periodicals; Structural analysis (Engineering) -- Periodicals; Construction, Technique de la -- Périodiques; Génie parasismique -- Périodiques; Pression du vent -- Périodiques; Earthquake engineering; Structural engineering; Wind-pressure; Periodicals; 624.105
• DOI: 10.1016/j.engstruct.2014.10.049
• Languages: English
• ISSNs: 0141-0296

Abstract

Highlights: Glass strength depends upon characteristic duration of applied actions. A strength criterion for variable load histories is theoretically derived. A verification formula for multiple loads with different durations is proposed. Comparisons are made with the prescriptions of other technical documents. Prestressing due to tempering should be treated as an external applied action. Abstract: It is well known that, depending upon the thermo-hygrometric environment, surface flaws in glass can grow over time even when they are well below the critical size, eventually leading to failure of the stressed material. This phenomenon, usually referred to as subcritical crack growth, or static fatigue, implies that the macroscopic strength of glass depends upon the characteristic duration of the applied loads. Various criteria have been proposed to evaluate the effects of the simultaneous combinations of actions applied at different times of the load history. Here, starting from a consolidated model of subcritical crack growth, an analytical approach to this problem is presented. Safety domains are calculated and compared with the approaches prescribed by recent proposals for standards. The analysis of a few case studies confirms that some approaches are not on the safe side, whereas other approaches can be too conservative. A proposal for new verification formulae is presented.