Experimental study on fire response of double glazed panels in curtain walls

Yu Wang, Qingsong Wang*, Yanfei Su, Jinhua Sun, Linghui He, K.M. Liew

*Corresponding author for this work

Research output: Journal Publications and ReviewsRGC 21 - Publication in refereed journalpeer-review

14 Citations (Scopus)

Abstract

Double glazing is increasingly employed in glass curtain walls, but very little is known about its thermal response under fire conditions. In this work, a total of twenty seven 600 × 600 mm2 double glazing units, with 6 mm, 9 mm and 12 mm air spacings, were tested. Three different installation types were adopted: 1) exposed frames, 2) horizontal-hidden frames and 3) vertical-hidden glass in order to study their potential effects on the glazing breakage behavior. A 500 × 500 mm2 n-heptane pool fire was used to heat the glass panes. The breaking time, surface temperature, total heat flux, heat release rate, and crack and fallout morphology were measured and analyzed. The test result suggests that different installation types have a significant effect on the fracture behavior of double glazing, especially for glass panes located at the fire side. Glass in exposed framing is more prone to cracks than those in semi-exposed frame, but fallout may more easily occur in semi framing façades. When no, or very limited fallout occurs at the fire side pane, thicker air space leads to a longer breakage time of ambient glass panes, thus improving the unit's integrity during fire. The experimental results obtained in this study is valuable for the development of practical guidelines for fire safety design.
Original languageEnglish
Pages (from-to)53-63
JournalFire Safety Journal
Volume92
Online published3 Jun 2017
DOIs
Publication statusPublished - Sept 2017

Research Keywords

  • Air space
  • Double glazing
  • Fire condition
  • Glass crack in fire
  • Installation type

Fingerprint

Dive into the research topics of 'Experimental study on fire response of double glazed panels in curtain walls'. Together they form a unique fingerprint.

Cite this