b'Figure 3: 1.2- by 1.2-m (4- by 4-ft.) GHB.Figure 2: Typical insulation fastener densities in the design of roofing assemblies (FM Global).with all the fasteners in the study. Figure 5 shows the insulation and fastener layout, depicting the insula-tion thickness and fastener lengths.In each design category, the testing started with quantifying the thermal performance of the assemblyFigure 4: Fastener density.without fasteners and followed by installing fasteners as per the fastener density and determining the respective effective R-value. The measured effective R-value of each fastener configuration was compared relative to the respective assembly without fasteners, and the percentage decrease in the effective R-value from thermal bridging was determined. Following this approach, the measured data obtained from the testing of each roof configuration, with each insulation type in each design category, were determined and grouped based on fastener density. These data were fitted with a polynomial curve trend line to obtain the generalized curves as shown in Figure 6, which highlights the relationship between fastener density and effective R-value of the roof assembly. The data show that as the number of fasteners per unit area increased, the conductive heat flow increased, lowering the overall thermal performance of the roof assembly. The decrease in effec-tive thermal resistance with fasteners was found to be consistent, irrespective of insulation type having an equivalent thermal resistance. Although the maximum fastener density tested in this study was 6.73 fasteners/m 2(0.63 fasteners/ft 2 ), the trend line was extrapolated to 8 fasteners/m 2 , as shown in Figure 6. The trend of generalized curves indicated that the thermal resistance of insulation influences the thermal bridging effects. For example, with the same fastener density, thermal bridging losses are higher in R-36 assemblies compared to R-31 and R-26. Similarly, the thermal losses are higher in R-31 relative to R-26 assemblies. The R-26 and R-31 assemblies, tested with the fastener densities ranging from 2.69 to 6.73 fasteners/m 2(0.25 to 0.63 fasteners/ft. 2 ), showed an average of five to 12 per cent decrease in the effective R-value, and the R-36 assemblies measured 6.4 to 13.3 per cent decrease in the effective R-value.Figure 5: Insulation and fastener layout (FL = fastener length).The developed fastener impact factor curves will enable a designer to predict thermal losses for different fastener densities followed in the industry and design roof assemblies accommodating the resulting losses.ZONE 1 2 3 4 5 6 7 8Effective Thermal Resistance 21 26 26 31 31 31 36 36Fasteners/ft2 Fasteners/m2 % Decrease in Effective Thermal ResistanceFigure 6b: Fastener impact factors.0.250 2.69 - 4.4 4.4 5.3 5.3 5.3 6.4 6.40.375 4.04 - 6.5 6.5 7.7 7.7 7.7 9.1 9.10.625 6.73 - 10.5 10.5 11.7 11.7 11.7 13.3 13.3Figure 6a : Fastener impact factors.ORN THE ONLY SOURCE FOR PROFESSIONAL ICI ROOFING CONTRACTORS IN ONTARIO ONTARIO ROOFING NEWSISSUE 1 202023'