Abstract
This paper presents an experimental study on shear-critical fibre-reinforced lightweight aggregate concrete (FRLWAC) beams under fire conditions. This study was designed and conducted due to scarcity of similar tests in the literature, highlighting the urgent need for FRLWAC shear tests at elevated temperatures. A series of fire tests was therefore performed on six shear-critical FRLWAC beams. Five of these beams were continuous, while the remaining specimen was simply-supported. These beams were designed with different shear-span-to-depth ratios (a/d) and steel fibre contents. Cross-section temperatures were measured and reaction forces were monitored throughout the fire exposure to track redistribution of forces. The test results showed that significant force redistribution took place from the sagging to the hogging regions throughout the fire test. It was also observed that continuous FRLWAC beams had better structural behaviour under fire over the simply-supported specimen, mainly due to the force redistribution during the fire exposure. Beams with a lower a/d ratio tended to last longer under fire compared to those with a larger a/d, despite all the beams being loaded with similar relative load ratios. The steel fibre contribution to shear resistance was found to be sensitive to elevated temperature exposure. An analytical model was developed based on the strut-and-tie analogy prescribed in the recently published EN 1992–1–1:2023 to predict the failure time of FRLWAC beams exposed to elevated temperatures. This model is both simple and accurate enough to be used as a quick prediction tool for structural engineers.
Original language | English |
---|---|
Article number | 121059 |
Journal | Engineering Structures |
Volume | 343 |
DOIs | |
Publication status | Published - Nov 15 2025 |
Externally published | Yes |
Bibliographical note
Publisher Copyright:© 2025
ASJC Scopus Subject Areas
- Civil and Structural Engineering
Keywords
- Fibre-reinforced concrete
- Fire behaviour
- Lightweight concrete
- Shear behaviour
- Truss model