Wind performance of in-service roofs

Volume 17, Number 1, March 2012

A flexible membrane covers a large commercial roof

For the past 16 years, SIGDERS (Special Interest Group for Dynamic Evaluation of Roofing Systems), an industry/NRC partnership, has developed test standards and guidelines to improve the wind performance of mechanically attached flexible membrane roofs roofs (see Construction Innovation, December 2005).

The research carried out by SIGDERS revealed a missing link in the understanding of the life cycle performance prediction - the loads induced by wind on these modern roofs compared to conventional, built-up roofs. Roof cladding loads specified in the National Building Code of Canada (NBC), American Society of Civil Engineers (ASCE) standards and other wind standards are mainly derived from wind tunnel tests on models. These codifications, developed 30 years ago, do not take into account the recent advancements in membrane technology.

To understand the behaviour of modern roofs, SIGDERS recently completed a field monitoring study on a large commercial roof (167 000 m2 or 1.8 million ft2) that uses poly[vinyl chloride] (PVC) as the waterproofing component. Wind loads on the membrane were measured using pressure sensors embedded at the corner, edge, and field zones of the roof in accordance with building code specifications. In parallel, the response of the roof to these wind loads was quantified by measuring the forces on the fasteners and observing fluttering behaviour of the flexible membrane.

An in-service roof with sensors installed on a roof corner to measure wind pressures

The monitoring results show that modern flexible membrane roofs experience higher wind-induced dynamics compared to conventional rigid roofs. More specifically, membrane roofs behave differently compared to current code practice as follows:

  • The wind load intensity at the corner and edge zones is similar: the entire roof perimeter behaves as a single, high-intensity zone. This means that current design underestimates the loads for edge zones.
  • Induced forces on the fasteners are much lower than the normal engineering predictions using tributary area, indicating that design should consider the elastic nature of the membrane in the load calculation.

The information from this monitoring will be used to update code requirements and manufacturing standards, which will now be based on the influence of various roof coverings on induced loads rather than solely on wind-tunnel testing. This change will help designers provide mechanically attached membrane roofs that are more durable and better able to handle high wind loads.

The next phase of the SIGDERS program will expand the scope of field monitoring by testing different roof configurations and identifying changes in wind load patterns that might result from the growing trend of affixing photovoltaic panels to roofs.

For more information

Inquiries from prospective industry partners are welcome. For more information, contact Bas Baskaran at 613-990-3616 or