Technical Series 96-212
When the CSA standards for "Gypsum Board Building Materials and Products", CAN/CSA-A82.27 was revised in 1991, the mass per unit area requirement for gypsum board products was removed. About the same time, changes to the 1990 edition of the National Building Code of Canada increased the sound transmission ratings between dwellings. Concerns regarding the impact of these changes on the fire resistance of insulated and non-insulated gypsum board protected wall assemblies prompted this joint research project between the Institute for Research in Construction at the National Research Council, Canada Mortgage and Housing Corporation and seven industry partners.
Forty-eight small-scale assemblies, 914 mm high by 914 mm wide were tested in a small-scale propane-fired vertical test furnace with an 810 mm by 810 mm opening. The samples were sealed at the edges against the furnace with ceramic fibre blanket. The tests were carried out by exposing the assemblies to heat. Thermocouples placed within the furnace were used to control the furnace temperature to match as closely as possible the CAN/ULC-S101-M89 standard temperature-time curve. Thermocouples were also used to monitor the temperature of the assembly. Depending on the assembly, between 12 and 30 thermocouples were placed through the assembly. Figure 1 provides an indication of the thermocouples for one of the test assemblies. Nine thermocouples in a 3 x 3 grid pattern were placed on the unexposed surface of the gypsum board. The thermocouples in the corners of the grid and in the middle were placed under insulated pads. The furnace and wall assembly temperatures were recorded at one-minute intervals. An assembly was considered to have failed if a single point thermocouple temperature on the unexposed face rose above 180°C, or if the average of the five thermocouples under the insulated pads rose 140°C above the ambient temperature of approximately 22°C, or if there was passage of flame or gases hot enough to ignite cotton waste.
Figure 1. Example of Thermocouple Locations
Five different gypsum boards were included in the testing:
The gypsum board was mounted on 90 mm steel studs at 600 mm spacing (one test used 65 mm steel studs) or 89 mm wood studs at 400 mm spacing (one test used 65 mm wood studs). Either one or two layers of gypsum were mounted on both sides of the studs. In nine of the assemblies, resilient channels were used. The cavity insulation and its thickness was varied. Three insulation types were used:
The assemblies tested and the time to failure for each assembly are summarized in Table 1.
The gypsum board orientation had a minor effect on the fire resistance. An assembly with a vertical joint backed by a steel stud provided a slightly better fire resistance than an assembly with a horizontal joint orientation.
The use of resilient channels on either the exposed side or the unexposed side had no effect on the fire resistance of the assembly. A similar result was found when resilient channels were used on both sides.
The effects of insulation varied depending on the layers of gypsum board used on each side of the assembly. To aid in reporting, the he reporting convention used is to list the layers of gypsum board on the exposed side followed by the layers of gypsum board on the unexposed side. For example, a 1x2 assembly indicates that there was one layer on the exposed side and 2 layers on the unexposed side.
Glass Fibre: In most cases, the addition of glass fibre insulation had no effect on the fire resistance of the assemblies compared to a non-insulated assembly. When 2 layers of 12.7 mm thick Type X gypsum board were applied to both sides of the studs there was a small increase in the fire resistance of 8%. When 2 layers of 12.7 mm thick lightweight gypsum board were used on both sides, the fire resistance was actually worse than for a non-insulated cavity. With small-scale tests, failure is predominantly due to heat transfer through the gypsum board layers. With the glass fibre insulation in the cavity, there is a more rapid temperature increase in the gypsum board on the exposed side which causes premature failure/splitting of the gypsum board, exposing the cavity to direct flame earlier in the test, resulting in an earlier failure.
Mineral Fibre Insulation: In all but one case, the mineral fibre insulation improved the fire resistance compared to an uninsulated cavity, from 20% for 2 layers of 12.7 mm thick regular lightweight gypsum on both sides, to 50% for 1 layer of 12.7 mm thick Type X gypsum board on both sides. In an asymmetrical installation of 15.9 mm thick Type X gypsum board (1x2 arrangement, meaning one layer on the exposed face and 2 layers on the unexposed face), the gypsum board split prematurely. However, the mineral fibre batts remained intact in the cavity for some time, providing a fire resistance to the unexposed gypsum more or less equal to the fire resistance lost due to the early splitting of the gypsum board on the exposed side.
Cellulosic Fibre Insulation (Dry Blown): In all but one case, the cellulosic fibre insulation improved the fire resistance compared to an uninsulated cavity, from 22% for 2 layers of 12.7 mm thick Type X gypsum board both sides, to 56% for one layer 12.7 mm thick Type X gypsum board on the exposed face and 2 layers on the unexposed face. In the case of the 15.9 mm thick Type X gypsum board, 1x2, the fire resistance with the cellulose was 17% worse than an uninsulated cavity. It was observed that during the test when a piece of gypsum board cracked and fell, the cellulose was completely consumed leaving not protection for the gypsum board on the unexposed side.
Mineral Fibre Insulation: The thickness of mineral fibre insulation (40 mm and 90 mm) did not affect the fire resistance of an assembly with one layer Type X gypsum board exposed and 2 layers unexposed. However, the 90 mm thick mineral fibre insulation provided an 18% improvement over the 40 mm thick insulation in the assembly that had 2 layers Type X gypsum board on both sides.
Cellulosic Fibre Insulation (Wet Sprayed): The comparison of 40 mm cellulosic and 90 mm cellulosic was carried out only for one layer 12.7 mm thick Type X gypsum board exposed and 2 layers 12.7 mm thick Type X gypsum board unexposed. The 40 mm thick cellulosic fibre insulation provided 14% better fire resistance.
The assemblies (non-load-bearing) constructed with wood studs were found to have a 7 to 12% better fire resistance than comparable assemblies with steel studs.
A reduction in mass/unit area of 7.82 kg/m2 to 7.35 kg/m2 for a double layer gypsum board assembly resulted in a 21% poorer fire resistance.
The presence of glass fiber in lightweight gypsum board did not affect the fire resistance of the assemblies.
In an assembly with one layer of 12.7 mm thick Type X gypsum board exposed and 2 layers 12.7 mm thick Type X gypsum board unexposed, the dry blown cellulose fiber provided a 41% better fire resistance than the assembly with the wet spray.
Varying the depth of both steel studs and wood studs did not affect the fire resistance.
The study revealed that the mass per unit area of the gypsum board does have an effect on the fire resistance performance of the assemblies. Since the requirement for mass per unit area has been removed from the CAN/CSA-A82.27 Standard, "Gypsum Board Building Materials and Products", designers may have to specify morethan just the thickness of gypsum board for fire-rated assemblies. The use of resilient channels for sound control does not affect the fire resistance performance. The use of fibreglass insulation could have a negative effect, so when fibreglass is used for sound control in a fire-rated wall assembly, it should be used with care. Mineral fibre insulation may prove to be a good alternative to fibreglass as it provides better fire resistance performance.
Project Manager: Jacques Rousseau
Research Report: Results of Fire Resistance Tests on Snall-Scale, Insulated and Non-Insulated, Gypsum Board Protected Wall Assemblies
Research Consultant: National Reseach Council Canada, National Fire Laboratory
full report on this research project is available
from the Canadian Housing Information Centre.
The information in this publication represents the latest knowledge available to CMHC at the time of publication, and has been thoroughly reviewed by experts in the housing field. CMHC, however, assumes no liability for any damage, injury, expense or loss that may result from use of this information.
Table 1. Description of Assemblies Tested and Fire Test Results
©1999 CMHC-SCHL. All rights reserved.