The design of an inverted warm flat roof – where the thermal insulation is above the waterproofing layer – allows for the possibility of rainwater passing through the insulation layer. Moisture can have an effect on both the insulation itself, and on the thermal performance of the roof as a whole. Both must be taken into account as part of inverted roof U-value calculations.

ETAG 031, the document that sets out how inverted roof systems should be assessed, describes procedures for adjusting the thermal conductivity of insulation, and establishing the effect of rainwater cooling if water is present below the insulation layer.

What is design thermal conductivity?

The thermal conductivity of an insulation product as it comes out of the factory is its declared thermal conductivity. Where the lambda is adjusted to account for specific conditions – such as exposure to moisture in an inverted roof – it is known as the design thermal conductivity.

Account is taken of moisture absorbed by diffusion, and additional absorption due to freeze-thaw cycling. The adjustment is calculated by multiplying the declared thermal conductivity by a moisture conversion factor (Fm).

Extruded polystyrene (XPS) insulation is less affected by moisture than expanded polystyrene (EPS). The result is a conversion factor of no more than 1.1 for XPS insulation. Using the performance of Polyfoam Roofboard Extra to illustrate:

  • Declared thermal conductivity: 0.033 W/mK.
  • Design thermal conductivity: 0.034 W/mk (insulation 100mm or thicker).
  • Design thermal conductivity: 0.035 W/mK (insulation less than 100mm).

How is rainwater cooling effect measured?

An allowance must be made for the effect of rainwater absorbing heat energy from the structure and increasing the rate of heat loss from the roof. Three factors are taken into account to calculate it:

  • p, measured in mm/day: the average rate of rainfall during the heating season, based on location-specific data. To achieve an accurate U-value calculation, p should reflect the location of the building, or the closest location with data available.
  • f, a ‘drainage factor’: expressing the percentage of the rainfall (p) that reaches the waterproofing membrane.
  • x, measured in W.day/m².K.mm: the factor for increased heat loss, which is a standard value of 0.040 W.day/m².K.mm.

f and x are typically combined by multiplying them together, so it’s common to see fx values quoted.

Annex C of ETAG 031 describes the test method for establishing a value for f. The test must be on the thinnest insulation board and the thinnest, most permeable ballast. If a product is intended for zero falls roofs then the tested build-up must also be without falls.

A document published by the BBA – Information Bulletin No.4 – quotes the following values for f which may be used without testing (and which are conservative as a result):

  • 0.5 for “roof gardens, green roofs and parking decks with cast concrete finish”.
  • 0.75 for “insulation with rebated joints and an open covering”.
  • 1.0 for “insulation with butt edged joints and an open covering”.

Section of ETAG 031 supports all of this, quoting respective fx values of 0.020, 0.030 and 0.040 for the three categories listed above, which may be used without the need for testing.

ISO 6946 – the international standard that describes the calculation of U-values by the combined method – says that “different types of roof build-up” can enjoy the benefit of a lower fx value where “the effect of the measures are documented in independent reports”.

Usually, this means the use of a water control layer (WCL) – an impermeable (to liquid, but permeable to water vapour) membrane, resistant to rot and UV decay. Following the test described in ETAG 031, a WCL may be so effective that a zero value for f is recorded.

What is the benefit of a water control layer in practice?

BBA Bulletin No.4 – is clear that a water control layer cannot be assumed to be waterproof. A minimum f factor of 2.5% is recommended, resulting in a value for fx of 0.001 (0.025 x 0.040).

This fx value is used in U-value calculations featuring Polyfoam XPS inverted roof solutions that include our Slimline Zero membrane – all independently verified in our BBA certificate. The value is so low that is has almost no impact on the end result.

Both of these moisture correction calculations are widely accepted within the construction industry, but their application is subject to some ongoing debate. The effect of water on insulation, and the application of the moisture correction factor, has been raised by the publication of the 2018 version of BS 6229. You can read more about that debate in this blog post about inverted roof guidance in the revised BS 6229​.

As blue roof constructions become more common, the role of temporary rainwater retention and whether it increases rainwater cooling is also being discussed by industry. We cover that in more detail in this blog post about inverted blue flat roofs.

Published in August 2019.