Research: Sprinklers – Burnt Offerings

Kemano, British Columbia, was a company town built 50 years ago by Alcan Smelters and Chemicals to support its hydroelectric station. When the station became essentially automated, making the town obsolete, an idea was born. The deserted town was donated to B.C.’s Office of the Fire Commissioner for training and research. This has been known as the Kemano Public Safety Initiative.

As part of this unique opportunity, the National Research Council of Canada was invited to conduct a series of full-scale fire experiments at the town. One series of experiments was designed to evaluate the performance of a cross-linked-polyethylene pipe sprinkler system in an abandoned house that contained furniture to simulate a normal occupancy. The house was a wood-framed bi-level structure with approximately 1,000 square feet per floor (2,000 s.f. total).

The sprinkler system was tested for fires originating in a basement recreation room, in a ground floor bedroom, and in the living room. A residential sprinkler system with quick response heads (rated at 68.3C (155F)) was designed and installed by a commercial residential sprinkler company. The designed flow rate was 50 L/min for the hydraulically remotest sprinkler head and 55 L/min for the hydraulically nearest one. The system was under a static pressure of 552 kPa (80 psi); during sprinkler activation, the main manifold was under a flowing pressure of 241 kPa (35 psi). The system was designed to prevent flashover in the room where fire originated and to allow response time for the fire department.

In all experiments, a single sprinkler head controlled and contained the fire within the original room within one minute of activation. The water spray cooled down the fire compartment. Since the fire was successfully controlled in each experiment, the fire damage was limited to the furniture that was near the ignition source. There was soot deposition as expected, but no structural damage. For experiments with the fire room door closed, temperature and visibility along the egress route were basically not affected by the fire inside that room.

In the recreation room, the plastic pipes and fittings were installed, without protection, on open wood joists. The equipment was exposed to temperatures as high as 140C in two experiments. After exposure to temperatures above their rated temperature of 93C (200F) at 552 kPa (80 psi) for 140 seconds, the pipes and fittings were not visibly damaged, and the sprinkler system was then successfully actuated and controlled the fire.

The effectiveness of heat and smoke detectors was also investigated in the sprinkler experiments. As expected, the heat detectors, which were rated at 57C (135F) and a temperature rise of 8.4C/min (15F/min), were always actuated earlier than the sprinkler. The rate of temperature rise triggered the heat detectors. Smoke detectors, installed on the egress route, were actuated before the sprinkler when the fire room was open to the egress route but after the sprinkler when the fire room door was closed.

Technical data obtained from the Kemano experiments can be used as a basis for further studies on sprinkler systems. Questions can be directed to Dr. Joseph Su at (613) 993-9616, fax (613) 954-0483, or e-mail joseph.su@nrc.ca.

Dr. Joseph Su is a researcher in the Fire Risk Management Program of the National Research Council’s Institute for Research in Construction in Ottawa. George P. Crampton, Don W. Carpenter, Cameron McCartney, and Patrice Leroux are technical officers in the same program.