Parameters affect the fire resistance performance of floor systems

(1)

Publisher’s version / Version de l'éditeur:

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.

Questions? Contact the NRC Publications Archive team at

PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.

https://publications-cnrc.canada.ca/fra/droits

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

CIB World Building Congress 2004 [Proceedings], pp. 1-10, 2004-05-01

READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.

https://nrc-publications.canada.ca/eng/copyright

NRC Publications Archive Record / Notice des Archives des publications du CNRC :

https://nrc-publications.canada.ca/eng/view/object/?id=01247144-8385-469b-a7c0-71ae73b707bd https://publications-cnrc.canada.ca/fra/voir/objet/?id=01247144-8385-469b-a7c0-71ae73b707bd

NRC Publications Archive

Archives des publications du CNRC

This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at

Parameters affect the fire resistance performance of floor systems

(2)

Parameters affect the fire resistance performance of floor systems

Sultan, M.A.

NRCC-39799

A version of this document is published in / Une version de ce document se trouve dans :

CIB World Building Congress 2004, Toronto, Ont., May 2-7, 2004, pp. 1-10

(3)

Parameters Affect the Fire Resistance Performance of Floor Systems

Mohamed A. Sultan∗

Institute for Research in Construction National Research Council of Canada Ottawa, Ontario, Canada, K1A-0R6

1 ABSTRACT

This paper presents and discusses the results of 28 full-scale fire resistance test assemblies conducted in accordance with the ULC-S101 standard on wood joist, wood-I joist, steel C-joist and wood truss floor systems as part of the collaborative research program on the fire and acoustical performance of floor assemblies. Parameters investigated in this study include the effects of attachment of the gypsum board to resilient channels, insulation type, number of gypsum board layers, joist spacing, resilient channel spacing, sub-floor type and structural load on the fire resistance performance of floor assemblies. Temperatures throughout the systems and floor deflection were measured. Using these results, the effects of the above-mentioned parameters on the fire resistance performance of floor assemblies are discussed.

2 INTRODUCTION

Fire-rated floor assemblies formed with new materials and construction methodologies have been used increasingly in residential buildings. Designers, architects and builders can choose fire rated-floor assemblies from different sources such as the listed assemblies or from the Part 9 Appendix A table of the National Building Code of Canada (NBCC). There are only 15 floor assemblies in Table A-9.10.3.1.B in the latest edition of the NBCC (1995) and the construction industry showed interest in increasing the number of floor assemblies listed in the NBCC table, so that designers, architects, builders, building officials and code users can have a variety of assemblies to choose from. Also, there have been numerous efforts to improve the quality of residential building environments in North America, including efforts to meet public demands for better acoustic isolation. During the same period, new construction materials have been developed and many construction practices and product specifications have changed over the past several years. In response to the changes noted above, the Institute for Research in Construction (IRC), National Research Council of Canada (NRC), in collaboration with industry and government partners, carried out a major research program to measure the fire resistance and acoustic performance of full-scale floor assemblies. Parameters investigated in this study include the effects of attachment of the gypsum board to a framing or to resilient channels, insulation type, number of gypsum board layers, joist spacing, resilient channel spacing, sub-floor type and structural load. The results of the fire resistance study and a study on the acoustical performance were used as the basis for the published update Part 9 Appendix A table of the NBCC that included more than 700 floor assemblies with 200 of them having fire resistance ratings. The list of these assemblies is on the Website (http://www.ccbfc.org/ccbfc/changes/soundfire_E.shtml).

3 DESCRIPTION OF TEST ASSEMBLIES

This study includes twenty-eight floor assemblies, 4.8 m long by 3.9 m wide, that were constructed in accordance with CAN/CSA-A82.31-M91 (1991). Construction details for the assemblies are available in Sultan et al (1998). Four types of framing were used: wood joist, wood-I joist, C-steel joist and wood truss. Details on the joist and resilient channel spacing for each assembly are given in Table 1. Three types of insulation were used: glass, rock fibre and cellulose fibre insulation. For Assemblies with two layers of gypsum board, 12.7 mm thick, (face and base layers) the board joints for both layers were staggered. The construction practice in Canada uses 12. 7 mm thick gypsum board in floor construction for economics and installation reasons.

∗_{Dr. Mohamed A. Sultan is a Senior Researcher and Group Leader of the fire resistant}

Subprogram of the Fire Risk Management Program, Institute for Research in Construction, National Research Council of Canada, Ottawa, ON, Canada K1A-0R6

(4)

4 Instrumentation AND TEST CONDITIONS

Thermocouples were used to measure temperatures inside the floor cavity and on each gypsum board surface. Details on the locations of the thermocouples can be found in Sultan et al. (1998). The temperature on the unexposed surface for all assemblies was measured at 9 locations in accordance with CAN/ULC-S101-M89 (1989). The loads used in this study are given in Table 1. Tests were conducted in accordance with CAN/ULC-S101-M89 that is similar to ASTM E119 (1988). As the fire resistance tests are expensive and time consuming process, no repeat test was conducted during this series, however, in

other test series, the test result variability of same construction assemblies was ± 4% (systematic test

error).

5 RESULTS AND DISCUSSION

The results of the 28 full-scale fire resistance floor tests are summarised in Table 1. The effects of different parameters on the fire resistance of floor assemblies are discussed below.

Table 1 Assembly Parameters and Fire Resistance Test Results

Assembly Joist Ceiling Finish Sub-Floor Cavity Insulation Resilient Channels Load Time to Type of Number Type Depth Spacing Type Thick Layers Type Thick Layers Type Thick Location Orient. Spacing (N/m²) Failure Failure

(mm) (mm) (mm) (mm) (mm) (mm) (min)

1 WJ* 235 406 X 12.7 1 Ply 15.5 1 *** *** *** Per 406 3838 30 Flame

2 WJ 235 406 X 12.7 1 Ply 15.5 1 *** *** *** Per 406 3838 45 Flame

3 WJ 235 406 X 12.7 1 Ply 15.5 1 G1 90 B Per 406 3828 36 Flame

4 WJ 235 406 X 12.7 1 Ply 15.5 1 M1 90 B Per 406 3838 60 Flame

5 WJ 235 406 X 12.7 1 Ply 15.5 1 C1 93 T Per 406 3838 59 Flame

6 WJ 235 406 X 12.7 2 Ply 15.5 1 *** *** *** Per 406 3838 80 Flame

7 WJ 235 406 X 12.7 2 Ply 15.5 1 G1 90 B Per 406 3838 67 Flame

8 WJ 235 406 X 12.7 2 Ply 15.5 1 M1 90 B Per 406 3838 72 Flame

9 WJ 235 406 X 12.7 2 Ply 15.5 1 C1 125 T Per 406 3838 74 S/F

10 WIJ 240 406 X 12.7 1 Ply 15.5 1 *** *** *** Per 406 4605 42 S/F

11 WIJ 240 406 X 12.7 1 Ply 15.5 1 M1 90 B Per 406 4653 46 S/F

12 WIJ 240 406 X 12.7 1 Ply 15.5 1 C1 90 T Per 406 4654 52 S/F

13 WIJ 240 406 X 12.7 2 Ply 15.5 1 *** *** *** Per 406 3934 72 S/F

14 WIJ 240 406 X 12.7 2 Ply 15.5 1 G1 90 B Per 406 3982 64 S/F

15 WIJ 240 406 X 12.7 2 Ply 15.5 1 M1 90 B Per 406 3982 77 S/F

16 WIJ 240 610 X 12.7 2 Ply 19 1 G1 90 B Per 406 2590 74 S/F

17 WIJ 240 610 X 12.7 2 Ply 19 1 G1 90 B Per 610 3120 65 S/F

18 SJ 203 406 X 12.7 2 Ply 15.5 1 *** *** *** Per 406 2950 74 S/F 19 SJ 203 406 X 12.7 2 Ply 15.5 1 G1 90 B Per 406 2950 68 S/F 20 SJ 203 610 X 12.7 2 Ply 19 1 G1 90 B Per 406 1799 69 S/F 21 SJ 203 406 X 12.7 2 Ply/Con 15.5/38 2 G1 90 B Per 406 1915 60 S/F 22 WJ 235 406 X 12.7 2 Ply 15.5 1 *** *** *** Per 406 5075 69 S/F 23 WJ 235 406 X 12.7 2 Ply 15.5 1 G1 90 B Per 406 5075 65 S/F 24 WT 305 406 X 12.7 2 Ply 15.5 1 *** *** *** Per 406 5602 69 S/F 25 WT 305 406 X 12.7 2 ply 15.5 1 G1 89 B Per 406 5602 65 S/F 26 WT 305 406 X 12.7 2 Ply/Con 15.5/38 2 G1 89 B Per 406 5123 72 S/F 27 WT 305 610 X 12.7 2 Ply 19 1 G1 89 B Per 406 3783 68 S/F 28 WT 406 610 X 12.7 2 Ply 19 1 G1 89 B Per 610 3783 55 S/F S/F - Structural Failure and Flame Penetration G1 - Glass Fibre Insulation *** - Null Value SJ - Steel Joist

Per - Perpendicular to Joists M1 - Rock Fibre Insulation T - Top Con - Concrete

X - Type X WIJ - Wood-I Joist B - Bottom WJ* - Gypsum Board Screw Spacing from Boa C1 - Cellulose Fibre Insulation (wet spray) Ply - Plywood WJ - Wood Joist WT - Wood Truss

Flame - Flame penetration through the subfloor S/F - Structural failure due to excessive deflection and flame penetration from T&G subfloor joint

5.1 Effect of Attachment of the Gypsum Board

Floor Assembly Nos. 1 and 2 with wood joists were tested to investigate the effect of the gypsum board screw spacing (10 mm and 38 mm) from board edges with a single layer of gypsum board ceiling finish attached to resilient channels. Assembly No. 1, with screws at 10 mm, provided 30 min of fire resistance while Assembly No. 2, with screws at 38 mm, provided 45 min. These results showed that by moving the screws away from the board edges (from 10 mm to 38 mm), the fire resistance increased by 50%. This can be explained below.

The results shown in Figure 1 (a and b) indicate that the temperature began to increase on the gypsum board facing the floor cavity approximately 2 min after the test was started in both assemblies.

There was a gradual increase in temperature to 125°C at 18 min in the assembly with screws at 10 mm

and at 15 min in the assembly with screws at 38 mm and at that time, the water in the board was driven off and the gypsum board core became dry; consequently, the board edges started to shrink, especially at the vicinity of the screw heads due to rapid heat transfer by the steel screws to the gypsum core around

(5)

3 them. The board edges peeled away from the screw heads much faster in the assembly with screws at10 mm, as they were located much closer to the edges, than in the assembly with screws at 38 mm. Thus, the sub-floor and joist sides were fully exposed to the furnace heat much earlier in the former assembly. This accelerated the burning of the joists and sub-floor and caused the assembly with screws at 10 mm from the board edges to fail earlier.

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T em p er atu re ( °C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp

Figure 1. Temperature Distributions for Floor Assemblies with Screw spacing at 10 mm and 38 mm from Board Edges Average Values

(a) Screw spacing at 10 mm

Furnace Temp. - Temperature-Time Relationship

SL/Cav. - Average Surface Temperature of Gypsum Board Facing Floor Cavity SF/Cav. - Average Surface Temperature of Sub-Floor Facing Floor Cavity UnExp. - Average Temperature on the Unexposed Surface

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T em p er atu re ( °C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp (b) Screw spacing at 38 mm Average Values

Assembly No. 1 Assembly No. 2

5.2 Effect of Insulation Installation and Types

Assembly Nos. 2 to 5 (wood joist with 1 layer of gypsum board), Nos. 6 to 9 (wood joist with 2 layers of gypsum board), Nos. 10 to 12 joist with 1 layer gypsum board), Nos. 13 to 15 (wood-I-joist with 2 layers of gypsum board) and Nos. 18, 19 (C-steel (wood-I-joist with 2 layers of gypsum board) and Nos. 24 and 25 (wood truss with 2 layers of gypsum board) were tested to investigate the effect of insulation type (glass, rock and cellulose fibre) and insulation installation on the fire resistance.

Floor Assemblies with Wood Joists – The results shown in Figure 2 (a to d) and test observations, for insulated and non-insulated assemblies with wood joist and 1 layer of gypsum board (Assembly Nos. 2 to

5), indicate that the temperature increase was substantially higher (900°C) at the gypsum board on the

cavity side (SL/Cav) for insulated assemblies than on a non-insulated assembly (450°C). As a result of

the substantial increase in board temperature (at 27 min for assemblies with either glass or rock fibre insulation and at 35 min for assembly with cellulose fibre insulation), the board cracked and fell off, unlike the non-insulated assembly. However, in the assembly with glass fibre insulation, the fibre melted when it was exposed to furnace heat in about 2 to 3 min after the gypsum board had fallen off; consequently, the sub-floor and joist sides were exposed to the furnace heat and the glass fibre was unable to compensate for the earlier failure of the gypsum board and provided a negative effect in the fire resistance. The rock and cellulose fibre remained in place after the gypsum board fell off and was able to compensate for the early failure of the gypsum board as well as protect the joist and sub-floor from furnace heat and thus, both rock and cellulose fibre insulation provided a positive effect on fire resistance. The fire resistance results of Assembly Nos. 2 to 5 given in Table 1 showed that, compared to a non-insulated assembly, which provided a 45 min fire resistance, the installation of the glass fibre reduced the fire resistance by 20% while the rock and cellulose fibre increased the fire resistance by 33% and 31%, respectively. The results shown in Figure 3 (a to d) for Assembly Nos. 6 to 9 (tested with 75% design load) with 2

layers of gypsum board indicate that the temperature increase was also substantially higher (900°C) at

the gypsum board on the cavity side for insulated assemblies than a non-insulated assembly (500°C). As

a result of the substantial increase in board temperature, the board cracked and fell off at approximately 60 min, unlike the non-insulated assembly at 75 min. In an insulated assembly with 2 layers of gypsum board, the exposure time for indirect furnace heat (conduction through the board) was much longer than in an insulated assembly with 1 layer of gypsum board. The deteriorated glass, rock and cellulose fibre insulations were unable to compensate for the earlier failure of the gypsum board and thus, all insulation provided a negative effect on fire resistance. The fire resistance results for Assembly Nos. 6 to 9 given in

(6)

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T e m p eratu re ( °C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp

Figure 2. Temperature Distributions in Insulated and Non-Insulated Floor Assemblies with Solid Wood Joists and a Single Layer of Gypsum Board

Average Values (a) No insulation Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T em p er at u re ( °C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp

(b) Glass fibre insulation

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T em p er at u re ( °C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp

(c) Rock fibre insulation

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T em p er at u re ( °C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp

(d) Cellulose fibre insulation Average Values

Average Values

Figure 3. Temperature Distributions in Insulated and Non-Insulated Floor Assemblies with Solid Wood Joists and a Double Layer of Gypsum Board

Average Values

(a) No insulation (b) Glass fibre insulation

(c) Rock fibre insulation (d) Cellulose fibre insulation Average Values

Average Values

BL/Cav. - Average Surface Temperature of Gypsum Board Facing Floor Cavity SF/Cav. - Average Surface Temperature of Sub-Floor Facing Floor Cavity UnExp. - Average Temperature on the Unexposed Surface

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Temperat ure (° C) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Te mp era ture ( °C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Te mp er a tur e ( °C) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Te mp er a tur e ( °C) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp

(7)

5 Table 1 showed that, compared to a non-insulated assembly, which provided an 80 min fire resistance, the insulation reduced the fire resistance by 16% with glass fibre, by 10% with rock fibre and by 7.5% with cellulose fibre. The failure mode for these assemblies was flame penetration through the sub-floor. Floor Assemblies with Wood-I Joists - The results shown in Figure 4 (a to c) for assemblies with 1 layer of

gypsum board indicate that the temperature increase was substantially higher (900°C) at the gypsum

board on the cavity side (SL/Cav) for insulated assemblies than a non-insulated assembly (450°C). As a

result of the substantial increase in board temperature, the board cracked and fell off. However, the rock and cellulose fibre remained in place after the gypsum board fell off and was able to compensate for the early failure of the gypsum board as well as protect the joist and sub-floor from furnace heat and thus, both rock and cellulose fibre insulation provided a positive effect on fire resistance. The fire resistance results for Assembly Nos. 10 to 12 given in Table 1 showed that, the insulation increased the fire resistance by 10% in the assembly with rock fibre and by 24% in the assembly with cellulose fibre compared to an assembly with no insulation in the floor cavity. In assemblies with wood-I-joists, the rock and cellulose fibre, like in the assembly with wood joists, provided a positive effect on fire resistance.

Figure 4. Temperature Distributions in Insulated and Non-Insulated Floor Assemblies with Wood-I-Joists and a Single Layer of Gypsum Board

Average Values

(a) No insulation (b) Rock fibre insulation

(c) Cellulose fibre insulation

Average Values

Assembly No. 12 Furnace Temp. - Temperature-Time Relationship

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T e m pe rat ur e ( °C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Tem pera tu re (° C ) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Tempe rature ( °C) 0 200 400 600 800 1000 Furnace Temp SL/Cav SF/Cav UnExp

The results shown in Figure 5 (a to c), for Assembly Nos. 13 to 15 (tested with 100% design load) with 2

layers of gypsum board indicate that the temperature increase was substantially higher (900°C) at the

gypsum board on the cavity side for insulated assemblies than a non-insulated assembly (400°C). As a

result of the substantial increase in board temperature, the board cracked and fell off at approximately 61 min for the assembly with glass fibre, 70 min for the assembly with rock fibre and non-insulated assembly. However, in the assembly with glass fibre insulation, the fibre melted when exposed to furnace heat in about 3 min; consequently, the sub-floor and joist sides were exposed to the furnace heat and glass fibre was unable to compensate for the earlier failure of the gypsum board and provided a negative effect in the fire resistance while the rock fibre remained in place after the gypsum board fell off and was able to compensate for the early failure of the gypsum board as well as protect the joist and sub-floor from furnace heat and thus, the rock fibre insulation provided a positive effect on fire resistance. The fire

(8)

resistance results for Assembly Nos. 13 to 15 given in Table 1, showed that the insulation reduced the fire resistance by 7% in the assembly with glass fibre and increased the fire resistance by 7% in the assembly with rock fibre compared to an assembly with no insulation. The failure mode for these assemblies was structural failure as the wood-I joists were unable to carry the applied load.

Figure 5. Temperature Distributions in Insulated and Non-Insulated Floor Assemblies with Wood-I-Joists and a Double Layer of Gypsum Board

Average Values

(c) Rock fibre insulation

Average Values

Assembly No. 15 Furnace Temp. - Temperature-Time Relationship

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 Tem per atu re ( °C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T em p er at u re ( °C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T e m p er at u re (° C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp

Floor Assemblies with C-Steel Joists – The results shown in Figure 6 (a and b) for Assembly Nos. 18 and

19 indicate that the temperature increase was substantially higher (900°C) at the gypsum board on the

cavity side (SL/Cav) for insulated assemblies than a non-insulated assembly (400°C). As a result of the

substantial increase in board temperature, the board cracked and fell off at approximately 62 min for the assembly with glass fibre, 71 min for non-insulated assembly. However, in the assembly with glass fibre insulation, the fibre melted when it was exposed to furnace heat in about 3 min; consequently, the sub-floor and joist sides were exposed to the furnace heat and glass fibre was unable to compensate for the earlier failure of the gypsum board and provided a negative effect in the fire. The fire resistance results for Assemblies No. 18 and 19 given in Table 1, showed that the installation of glass fibre insulation in the floor cavity of the assembly with a double-layer gypsum board reduced the fire resistance by 8% compared to an assembly with no insulation in the floor cavity.

Floor Assemblies with Wood Truss – The results shown in Figure 7 (a and b) for Assembly Nos. 24 (no insulation) and 25 (glass fibre insulation) indicate that the temperature increase was substantially higher

(900°C) at the gypsum board on the cavity side for insulated assemblies than a non-insulated assembly

(450°C). As a result of the substantial increase in board temperature, the board cracked and fell off at

approximately 62 min for the assembly with glass fibre, 66 min for the non-insulated assembly. However, in the assembly with glass fibre insulation, the fibre melted when it was exposed to furnace heat in about 3 min; consequently, the sub-floor and joist sides were exposed to the furnace heat and glass fibre was unable to compensate for the earlier failure of the gypsum board and provided a negative effect in the fire. Resistance The fire resistance results for Assembly Nos. 24 and 25 given in Table 1, showed that the installation of glass fibre insulation in the floor cavity of the assembly with wood truss and a double-layer

(9)

7 gypsum board reduced the fire resistance by 6% compared to an assembly with no insulation in the floor cavity.

Figure 6. Temperature Distributions in Insulated and Non-Insulated Floor Assemblies with C-Steel Joists and a Double Layer of Gypsum Board

Average Values

Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T e m p eratu re (° C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T e m p eratu re (° C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T e m p er at u re ( °C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp

Figure 7. Temperature Distributions in Insulated and Non-Insulated Floor Assemblies with Wood Truss and a Double Layer of Gypsum Board Assembly No. 24 Average Values (a) No insulation Time (min) 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 T e m p er at u re ( °C ) 0 200 400 600 800 1000 Furnace Temp BL/Cav SF/Cav UnExp Average Values (b) Glass fibre insulation

Assembly No. 25

5.3 Effect of Number of Gypsum Board Layers

Floor Assemblies with Wood Joists - Floor Assembly Nos. 2 and 6 were tested to investigate the effect of the number of gypsum board layers on the fire resistance of non-insulated wood joist assemblies. The fire resistance of these assemblies given in Table 1, showed that the assembly with a double-layer of 12.7 mm gypsum board ceiling finish provided an increase in fire resistance of 78% compared to an assembly with a single layer of 12.7 mm gypsum board.

Floor Assemblies with Wood-I Joists - Floor Assembly Nos. 10 and 13 were tested to investigate the effect of the number of gypsum board layers on the fire resistance of non-insulated wood-I-joist assemblies. The results given in Table 1, showed that the assembly with a double-layer gypsum board ceiling finish provided an increase in the fire resistance of 71% compared to an assembly with a single layer of 12.7 mm gypsum board. For the assemblies above, the more layers of 12.7 mm gypsum board used, the more protection was provided to the solid wood or wood-I joists and thus, the better the fire resistance.

(10)

5.4 Effect of Joist or Truss Spacing

Floor Assemblies with Wood-I Joists - Floor Assembly Nos. 14 and 16 were tested to investigate the effect of joist spacing (406 mm o.c. and 610 mm o.c.) on the fire resistance of wood-I-joist floor assemblies with a double layer gypsum board ceiling finish and glass fibre insulation in the floor cavity. Assembly No. 14 (with 406 mm o.c. joist spacing) provided 64 min of fire resistance while Assembly No. 16 (with 610 mm o.c. joist spacing) provided 74 min. These results need to be confirmed with repeat tests because as seen blow for other floor frames, the effect of joist spacing was insignificant.

Floor Assemblies with C-Steel Joists - Floor Assembly Nos. 19 and 20 were tested to investigate the effect of joist spacing (406 mm o.c. and 610 mm o.c.) on the fire resistance performance of steel joist floor assemblies with a double layer gypsum board ceiling finish and glass fibre insulation in the floor cavity. Assembly No. 20 (with 406 mm o.c. joist spacing) provided 68 min, while Assembly No. 21 (with 610 mm o.c. spacing) provided 69 min. The difference in the fire resistance is within the systematic error of the test procedure. These results showed that, unlike the case with wood-I-joist assemblies, the effect of steel joist spacing on fire resistance is insignificant. For these assemblies, the joist spacing did not play a role in the fire resistance due to the heat transfer from the steel joist edges to the web.

Floor Assemblies with Wood Truss - Floor Assembly Nos. 25 and 27 were tested to investigate the effect of joist spacing (406 mm o.c. and 610 mm o.c.) on the fire resistance of wood truss floor assemblies with a double layer gypsum board ceiling finish and glass fibre insulation in the floor cavity. Assembly No. 25 (with 406 mm o.c. joist spacing) provided 65 min of fire resistance while Assembly No. 27 (with 610 mm o.c. joist spacing) provided 68 min. The difference in the fire resistance is within the systematic error of the test procedure. In a floor assembly exposed to furnace heat with wood truss framing, irrespective of truss spacing, the heat from the gypsum board surface facing floor cavity can move freely from cavity to cavity. These results showed that, also unlike the case with wood-I-joist assemblies, the effect of wood truss spacing on fire resistance is insignificant.

5.5 Effect of Resilient Channel Spacing

Floor Assemblies with Wood-I Joist - Floor Assemblies 16 and 17 were tested to investigate the effects of the resilient channel spacing (406 mm o.c. and 610 mm o.c.) on the fire resistance of wood-I-joist floor assemblies with a double layer gypsum board ceiling finish and glass fibre insulation in the floor cavity. Assembly No. 16 (with 406 mm o.c. channel spacing) provided 74 of min fire resistance while Assembly 17 (with 610 mm o.c. channel spacing) provided 65 min. The assembly with the wider resilient channel spacing provided less fire resistance due in part to the lesser number of fasteners for the gypsum board. The more screws the assembly has, the better chance for the gypsum board to remain in place and protect the frame and thus, the better the fire resistance.

Floor Assemblies with Wood Truss - Floor Assembly Nos. 27 and 28 were tested to investigate the effect of resilient channel spacing (406 mm o.c. and 610 mm o.c.) on the fire resistance of wood truss floor assemblies with a double layer gypsum board ceiling finish and glass fibre insulation in the floor cavity. Assembly No. 27 (with 406 mm o.c. resilient channel spacing) provided 68 min of fire resistance while Assembly No. 30 (with 610 mm o.c. joist spacing) provided 55 min. This was the same as assemblies with wood-I joist, the assembly with the wider resilient channel spacing provided less fire resistance due in part to the lesser number of fasteners for the gypsum board. The more screws the assembly has, the better chance for the gypsum board to remain in place and protect the frame and thus, the better the fire resistance.

5.6 Effect of Sub-floor Type

Floor Assemblies with Wood Truss - Assembly Nos. 25 and 26 were conducted to investigate the effect of adding concrete topping above the plywood sub-floor in assemblies with wood truss. Assembly 25 with no concrete topping provided 65 min of fire resistance and Assembly 26 with concrete topping provided 72 min. Unlike the assembly with steel joist below, the gypsum board base layer fell into the furnace at more or less the same time (56 min). Having a concrete topping has prevented the flame spread from the plywood subfloor T and G joints until the concrete cracked and flame spread from it longer than in assembly with no topping above the plywood subfloor.

Floor Assemblies with Steel Joist - Assemblies 20 and 21 were conducted to investigate the effect of adding concrete topping above the plywood sub-floor on assemblies with steel joist. Assembly 20 with no

(11)

9 concrete topping provided 68 min of fire resistance and Assembly 21 with concrete topping provided 60 min. The gypsum board base layer fell into the furnace much earlier (at 52 min) in the assembly with concrete topping than in the assembly without topping (at 61 min). The results of the assembly with concrete topping need to be repeated to confirm whether the fall of the gypsum board is due to a defect in the board or the installation of concrete topping.

5.7 Effect of Load

Assembly Nos. 6, 7, 22 and 23 were conducted to determine the effect of load on the fire resistance. Assembly No. 6 (no insulation) and No. 7 (with insulation) were tested with 75% design load and Assembly No. 22 (no insulation) and No. 23 (with insulation) were tested with 100% design load. Assembly No. 6 provided 80 min and Assembly No. 22 provided 69 min fire resistance. Also, Assembly No. 7 provided 67 min and Assembly No. 23 provided 65 min fire resistance. These results showed that when the load increased by 25%, the fire resistance decreased by 14% for the non-insulated assembly and by 3% in the insulated assembly.

6 Conclusions

In this paper, the results of twenty-eight full-scale fire resistance tests conducted to investigate various parameters on the fire resistance of lightweight frame floor assemblies are presented. Based on the results of the tests, the following conclusions can be drawn:

1. An assembly with screws located further away from board edges (38 mm versus 10 mm) provides higher fire resistance.

2. In an assembly with wood joists and a single-layer gypsum board ceiling finish, the glass fibre reduced the fire resistance while the rock and cellulose fibre increased the fire resistance compared to an non-insulated assembly. In assemblies with a double-layer of 12.7 mm gypsum board finish, the glass, rock and cellulose fibre all reduced the fire resistance compared to an non-insulated assembly.

3. For a floor assembly with wood-I-joists and single layer of 12.7 mm gypsum board ceiling finish, the rock and cellulose fibre insulation increased the fire resistance compared to an non-insulated assembly. In assemblies with a double layer of 12.7 mm gypsum board finish, the glass fibre reduced the fire resistance while rock fibre increased the fire resistance compared to an non-insulated assembly.

4. For a floor assembly with wood truss and double layer of 12.7 mm gypsum board finish, the glass fibre reduced the fire resistance compared to an non-insulated assembly.

5. Assemblies with two layers of 12.7 mm gypsum board with staggered joints provided a significant increase in the fire resistance compared to an assembly with one layer of 12.7 mm gypsum board. 6. For a floor assembly with either wood-I joist or wood truss and glass fibre insulation as well as a

double layer of 12.7 mm gypsum board, the effects of resilient channel spacing (406 mm o.c. and 610 mm o.c.) was significant.

7. For floor assembly with either C-steel joists or wood truss and a double-layer of 12.7 mm gypsum board ceiling finish, glass fibre insulation reduced the fire resistance compared to a non-insulated assembly and the effect of joist or truss spacing (406 mm o.c. and 610 mm o.c.) was insignificant. 8. Adding concrete topping to plywood sub-floor increased the fire resistance for a wood truss assembly

with two layers of gypsum board.

(12)

7 ACKNOWLEDGMENT

The author wishes to thank the NRC/IRC staff responsible for conducting the fire resistance tests. These include John Latour, Jocelyn Henrie, Patrice Leroux, Rock Monette and Yves Seguin.

The National Research Council of Canada wishes to acknowledge the technical and financial contributions of the following research partners:

• Boise Cascade (USA)

• Canada Mortgage and Housing Corporation

• Canadian Home Builders Association

• Canadian Portland Cement Association

• Canadian Sheet Steel Building Association

• Canadian Wood Council

• Cellulose Insulation Manufacturers

Association of Canada

• Cellulose Insulation Manufacturers

Association (USA)

• Forinteck Canada Corporation

• Gypsum Association (USA)

• Gypsum Manufacturers of Canada

• Louisiana-Pacific Corporation (USA)

• Nascor Inc.

• Ontario Ministry of Municipal Affairs and

Housing

• Owens-Corning Canada

• Owens-Corning (USA)

• Roxul Inc.

• Trus Joist MacMaillan (USA)

• Willamate Industries (USA) Johns Manville

Corporation (USA)

• Truss Plate Institute of Canada

• Truss Plate Institute (USA)

8 REFERENCES

National Building Code of Canada (Part 9), National Research Council of Canada, Ottawa, ON, 1995. CAN/CSA-A82.31-M91, Gypsum Board Application: Rexdale, Ontario, 1991 Canadian Standards Association.

Sultan, M.A., Seguin, Y.P. Leroux P., Results of Fire Resistance Tests on Full-Scale Floor Assemblies, Internal Report No. 764, Institute for Research in Construction, National Research Council of Canada, Ottawa, ON, 1998.

CAN/ULC-S101-M89, Standard Methods of Fire Endurance Tests of Building Construction and Materials. Underwriters' Laboratories of Canada, Scarborough, ON, 1989.

ASTM E119-88, Standard Test Method for Fire Tests of Building Construction and Materials, ASTM, West Conshohocken, PA, 1988.