Institute for Catastrophic Loss Reduction

Building resilient communities

Home

About Us

Strategic Plans

Governance

Board of Directors

Management Committee

Advisory Committee

Research Associates

Staff Profiles

Homeowners

Be Prepared!

Research for Safer Homes

New Homes

Existing Homes

Earthquake Risk Tool

Outil de Cartographie.

Communities

RSVP for Cities

Understanding Hazards

Research Network

Businesses

Open for Business (TM)

Protecting Our Kids

Membership

Become a Member

General Membership

Associate Membership

Member Companies

Partners

Resource Centre

Magazine Articles

Hazard Research Papers

Cat Tales eNewsletters

Homeowner safety

Focus on

Workshop slide archive

Icestorm 98 Main page

Books

ICLR Health Presentations

Workshops

Friday Forums

Conferences

Media Centre

Speeches

Press Releases

Loading
Back to Tornado Papers
Back to Hazard Research Papers


Tornado papers
Wind Loads on Houses: Destructive Model Testing of a Residential Gable Roofed House
B. Visscher
G.A. Kopp
P.J. Vickery
March 24, 2004
ICLR Research -
Paper 37


Abstract

In August 1992 Hurricane Andrew struck southern Florida. Thirty people died and 250,000 were left homeless. Damage from this Category 4 hurricane was estimated at 30 billion dollars [1]. After this storm, there was a strong interest to identify the type of damage associated with the large insurance costs. Residential subdivisions located in this region that were relatively unsheltered from the full force of the storm incurred the most damage. The damage to these houses presented a unique opportunity to investigate the type and extent of damage to this type of structure.

Major roof failure was the single most common and abundant type of damage. The investigation concluded that the main types of roof failure were:
1. local failure (loss of sheathing);
2. discrete roof truss failure; and
3. global roof failure (whole roof) due to roof-to-wall connection failure.
In the high damage areas, sheathing failures were the most common type of roof failure recorded.

In the current study, a 1:20 scale destructive model of a low-rise gable roofed house was tested in the Boundary Layer Wind Tunnel Laboratory at the University of Western Ontario. The goal was to study the failure of roofs and roof elements. The term ‘destructive’ is used to describe the model because it was designed to fail in the wind tunnel. Both local (roof sheathing panel) and global (whole roof) failures were investigated. A number of different variables were studied to determine their effects on roof and roof panel failure, such as: dominant openings and their location with respect to the wind direction, background leakage, roof-to-wall connection strength, roofing panel hold down strength, the ratio of ridge and soffit vent area to the roof area, and the presence or absence of sheathing panels. The trajectory of airborne sheathing panels (after local roof failure) was studied as in a severe wind event airborne debris from damaged buildings could cause damage to neighbouring structures.


Wind Loads on Houses: Destructive Model Testing of a Residential Gable Roofed House (March 24, 2004) (PDF)







The Institute for Catastrophic Loss Reduction

Toronto Office
20 Richmond Street East, Suite 210, Toronto, Ontario M5C 2R9
Tel: (416) 364-8677
Fax: (416) 364-5889

London Office
Boundary Layer Wind Tunnel Laboratory
Western University
1151 Richmond Street, London, Canada N6A 5B9
Tel: (519) 661-3234
Fax: (519) 661-4273