Projects and Solutions

Fatigue performance of metal roof cladding when subjected to cyclonic winds

Roofs of low-rise buildings are subjected to large pressures during windstorms. The creation of a dominant opening by debris impact, door failure, etc, in the windward wall can generate large positive internal pressures which in combination with large suctions at the edges of the roof will generate large net pressures. This scenario is commonly responsible for much of the damage to buildings. Furthermore, roof cladding and components are susceptible to fatigue failure under these fluctuating wind loads. Building products used in cyclonic regions have and are required to be evaluated against cyclic load tests, such as the TR440 and DABM tests and the Low-High-Low (L-H-L) test now specified in the Building Code of Australia.

During the mid 70’s to mid 90’s, in the derivation of the various test methods and criteria, extensive research was carried out in Australia by people such as Beck, Stevens, Mahendran and Xu on the fatigue failure of roof components. The testing methods generally used reduced span setups with the cladding typically loaded by a line load at mid span.

The wind load acting on a part of the roof of a building during a cyclone event will depend on the parameters of the cyclone such as the peak and mean wind speeds, size and translational speed of the cyclone, and also building geometry and orientation with respect to the passage of the cyclone. The L-H-L test was based on research by Mahendran (1995) and Jancauskas et al (1994) for a “design” cyclone using external pressures measured on a single storey gable ended building model and extensive sinusoidal loading tests. These studies did not incorporate the effects of internal pressure resulting from dominant openings in the building envelope.

In utilising the capabilities of the 3LP PLA, we are now for the first time able to apply highly fluctuating representative wind loads to our pierced fixed roofing systems and to compare to the L-H-L test.

An airbox has been constructed to allow either the application of positive and negative pressures to be applied to the cladding samples (Figure 1). In using the PLA, constant amplitude sinusoidal cyclic load tests have been conducted to generate data for S-N curves (load vs cycles to failure). This data has been valuable to compare and validate previous studies which used line loads on reduced spans. The airbox incorporates a 3 axis load cell to monitor the change in screw reaction with changing cladding properties (deformation, cracking etc). Point load tests have been conducted to quantify the influence of loads on adjacent crests to determine appropriate factors for area averaging pressures.

airbox connected to PLA and cracking of cladding

Wind tunnel time series data contained in the NIST aerodynamic database is used to derive load cycles (varying means and amplitudes from a rainflow analysis) for changing wind direction and speed. These are compared for different building geometries and dominant openings. The NIST time series data is also being used to assemble 4 to 5 hour “cyclonic wind traces”. In applying these fabricated wind traces, it has been interesting to note the variation in crack patterns and the differences with the standard cyclic product tests (Figure 2). The wind traces will be deconstructed to evaluate the relationship of the various magnitude peaks with respect to time along the trace and to the S-N curves. As an aid to this, tests are being conducted to compare damage vs load rate using triangular traces.


read more ...




Return to Western's home page
Faculty of Engineering • Spencer Engineering Building
The University of Western Ontario
London, Ontario, Canada, N6A 5B9 • Tel: 519-661-2128

Updated June 22, 2010 by contactWE
Privacy | Web Standards | Terms of Use