CSA A123.21

CSA A123.21, Standard test method for the dynamic wind uplift resistance of membrane-roofing systems, is published by CSA Group in Canada. Unlike static tests such as UL 1897 or ANSI/FM 4474, it applies a sequence of fluctuating pressures derived from records of real wind, so that the roof experiences the gusting and cycling that can fatigue fasteners and seams over time.

The method was developed from research by the National Research Council of Canada (NRC) and an industry consortium known as SIGDERS (Special Interest Group for Dynamic Evaluation of Roofing Systems). It has been revised several times: the 2004 edition (reaffirmed 2009) was replaced by a 2010 edition, then by a 2014 edition, and a 2020 edition is listed by standards databases.

What it covers

The standard has two methods:

• Method 1 – mechanically attached, reinforced membrane systems, within limits on fastener row spacing (up to 2,896 mm / 114 in) and fastener spacing within rows (up to 610 mm / 24 in).
• Method 2 – systems where components are adhered to resist uplift, without mechanical fasteners or ballast.

The test applies load cycles at increasing pressure levels until the system fails or completes the sequence. The highest level completed sets the system's rated resistance.

Code status

The National Building Code of Canada references CSA A123.21 for determining the wind uplift resistance of membrane roofing systems. Designers compare the rated resistance with the factored wind load calculated for the building under Part 4 of the NBC, considering roof zones (field, perimeter and corners).

The Canadian Roofing Contractors Association (CRCA) has published an application policy on how the standard should be used, reflecting debates over interpreting test results and applying them to different roof areas.

Why dynamic testing matters

Mechanically attached single-ply membranes billow and flutter in wind. Static pressure tests can overstate their capacity because they do not reproduce repeated loading. Dynamic tests aim to capture failure modes such as fastener pull-out after many cycles, seam peel, and membrane tearing at plates.