Hygrothermal Performance of Protected Membrane Roof Systems: Part 2: Experimental Verification

In a previous paper by the same author [1], a simplified analysis of the heat and moisture response of protected membrane roofs (PMR) was discussed. The analysis included a ΔU-correction to the thermal transmission coefficient U of the equivalent, conventional roof to account for the effects of percolation, evapora tion, and butt joint losses between the insulation boards. This second paper presents experimental results.

The hygrothermal performances of five Protected Membrane Roof systems (PMR) were monitored to a varying degree of completeness for a number of years in the period between 1981-1992. The systems included extruded polystyrene boards, mineral fibre boards, a composite foam-concrete board manufactured with expanded polystyrene, and a composite foam-concrete board manufactured with extruded polystyrene.

The monitoring indicated that the heat and moisture response of a protected mem brane roof is substantially reduced when ponding water is present either below or above the insulation. The reduction increases when the vapour diffusion resistance of the insulation material is low. Both conclusions supported the ΔU-correction, derived from the simple model. They have prompted changes in the design of PMRs and modifications to the corresponding design rules.