Title:
|
Numerical and experimental studies of cold-formed thin-walled steel studs in fire
|
Worldwide, cold-formed thin-walled steel members have been widely used as
primary load bearing members, such as wall studs, floor joints, columns and beams,
in low to medium-rise buildings such as offices, hotels, flat blocks and houses. A
potential problem of using cold-formed thin-walled steel in building structures is fire
exposure. Steel has high thermal conductivity and thin-walled sections have high
section factors, both leading to rapid steel temperature rises in fire. Despite their
increasing use, our understanding of the performance of cold-formed thin-walled
steel sections in fire has been limited, mainly to the standard fire resistance testing by
manufacturers of cold-formed thin-walled steel sections. Such a practice is
expensive and time-consuming and it is difficult to develop fundamental
understanding of thin-walled steel structural behaviour in fire.
Against this background, this thesis presents the results of a major experimental and
numerical study of cold-formed thin-walled steel structures in fire. It includes the
thermal performance of thin-walled steel channel sections in planar systems under
fire attack from one side, the axial strength of cold-formed thin-walled channel
sections and rectangular hollow sections at ambient and under high temperature
conditions, and the axial strength of cold-formed thin-walled steel channel section
panels under fire expose from one side.
This project includes both experimental and numerical studies. The experiments
include fire testing of eight small steel stud panels (300x300mm) with different types
of steel section (lipped channel or unlipped channel), different numbers of gypsum
boards (one layer or two layers) and either with or without interior insulation; 52
tests on short cold-formed lipped and unlipped channels at ambient and various
elevated temperatures; and eight tests on full-scale steel stud panels (2200x200mm),
two at ambient temperature and six under fire.
The numerical studies of this project include an assessment of different design
methods, ABAQUS simulation of various tests and numerical parametric studies of
different systems. Modifications have been made to enable BS5950 Part 5 (BSI
1998) and ENV 1993-1-3 (CEN 2001) to predict the ultimate strength of thin-walled
columns with service holes, and to extend the ability of BS5950 Part 5 (BSI 1998),
ENV 1993-1-3 (CEN 2001) and AISI (1996) to deal with distortional buckling
failure and the change in the strength and stiffness of steel at uniform elevated
temperatures. For columns with non-uniform temperature profiles, the ambient
temperature design method for cold-formed thin-walled columns in ENV 1993-1-3
(CEN 2001) is also modified to take into account the thermal bowing effects.
The finite element program ABAQUS (HKS 2001) has been validated first against
the test results. The validated software is used to carry out a number of parametric
studies to investigate the performance of different systems, including thermal
performance of steel stud and cassette systems with different numbers of gypsum
boards on the exposed and unexposed sides; structural behaviour of short cold-formed
thin-walled channel columns; structural performance of rectangular hollow
columns of different lengths under uniform high temperatures, but with different
initial imperfections; and structural performance of cold-formed thin-walled columns
under non-uniform high temperatures.
The results presented in this thesis should be valuable to both users and researchers
of fire safety engineering and the cold-formed steel industry.
|