Researching fire performance of construction materials in houses

Volume 16, Number 2, June 2011

Fire testing is performed at an NRC fire research facility

Gaining a better understanding of the performance of construction materials and systems used in houses, and determining their impact on the life safety of occupants under fire conditions, are key areas of research at the NRC Institute for Research in Construction (NRC-IRC).

NRC-IRC researchers recently completed a series of tests to investigate the fire performance of protected floor/ceiling assemblies, and the tenability conditions in a full-scale test facility simulating a single-family house with a basement fire. This is a follow-up study to research on unprotected floor/ceiling assemblies under basement fire scenarios (see Construction Innovation March 2009).

The basement in the test facility had an open stairwell to the main floor. Test floor assemblies were selected from the previous study, constructed with wood I-joists, steel C-joists, metal web wood trusses or solid wood joists over the basement. The test assemblies were protected on the basement side by a regular gypsum board ceiling, residential sprinklers, or a suspended ceiling.

Test assembly Gypsum board ceiling Suspended ceiling Sprinklered
Wood I-joist
Metal web wood truss  
Steel C-joist    
Solid wood joist    

Matrix of full-scale fire experiments

The experiments utilized a relatively severe, fast-growing fire set in the basement to challenge the structural integrity of the floor system above the basement, which provides the normal egress route on the first storey for occupants. The study focused on life safety from the perspective of tenability for occupants on upper storeys, and on the integrity of the floor structure above the basement as egress routes. The table shows a matrix for the full-scale fire experiments.

Some of the key findings from this series of fire experiments include:

  • In the tests conducted on the floor assemblies protected with gypsum board, all fire events exhibited the same chronological sequence: initiation of the fire, activation of smoke alarms, loss of tenable conditions in open areas on upper storeys, and finally structural failure of the test floor assembly above the basement (loss of first-storey egress route).
  • Compared to the experiments conducted previously using the same floor assemblies without gypsum board protection, tenability conditions were similar or improved slightly while the structural performance was improved significantly with gypsum board protection. The time taken to reach structural failure for the protected floor assemblies was much longer than for those with no protection.
  • In the previous study, the time to reach structural failure for the engineered floor assemblies without gypsum board protection was 35-60% shorter than for the solid wood-joist assemblies. With gypsum board protection, all engineered test assemblies had similar structural failure times, matching that of the solid wood joist assembly.
  • The experiment using the test assembly with the suspended ceiling followed the same sequence of fire events. The benefit of the suspended ceiling as a protection measure for the test assembly was marginal compared to the same test assembly without a suspended ceiling.
  • The residential sprinkler systems effectively protected the structural integrity of the test floor assemblies and there was no structural failure or damage. The residential sprinkler systems also kept the conditions tenable in the test house during the experiments. Note that, outside the scope of this study, additional experiments were conducted for sprinkler-protected assemblies under more severe fire conditions. For more information see www.nrc-cnrc.gc.ca/obj/irc/doc/pubs/rr/rr308.pdf.
  • Research on the fire performance of houses includes several phases of studies, with each investigating a specific structural system of single-family houses based on specified fire scenarios. The next phase of research, expected to begin in 2012, will explore the fire performance of wall assemblies for single-family houses.

A full report on this study can be downloaded at www.nrc-cnrc.gc.ca/obj/irc/doc/pubs/rr/rr307.pdf.

For more information

Questions can be directed to Dr. Joseph Su at 613-993-9616 or e-mail joseph.su@nrc-cnrc.gc.ca.