The South African built environment has been abuzz over the past decade with the release of new energy efficiency management standards, most recently the requirements for Energy Performance Certification for certain buildings. Building owners strive to comply, but what of the old adage; you can’t manage what you can’t measure? One roof insulation manufacturer takes the bull by the horns with some real-life testing. Scott Quarmby describes how he did this using a test house.
This short article looks at the performance of cellulose fibre insulation via empirical testing, as a convenient, efficient and cost-effective way of complying with the National Building Regulations, while also reducing a home or other buildings’ reliance on climate and temperature control measures.
As manufacturer of two of the major cellulose fibre roof insulation brands, Eco-Insulation and Thermguard, we recently undertook testing under real life conditions to back up the generic ‘deemed to comply’ methodology on which R-Values are described in SANS 204:2011. Testing began in summertime, mid-November 2019, ending in February 2020 in an average-sized family home in Gauteng where Climatic Zone 2 (as per SANS 204:2011) applies.
Methodology
Temperature probes were placed in standardised positions throughout a typical three-bedroom home which was fitted with generic bulk blanket insulation. Real time temperature data was recorded on a data logger throughout the day for a period of four weeks. The blankets of insulation were then removed and the ceiling expertly cleaned. Thereafter, recordings were taken for a further week to provide a baseline with no insulation. Finally, the house was fitted with Thermguard cellulose fibre insulation and temperature recordings were taken for a period of eight weeks. In both insulated scenarios, insulation was installed on the minimum deemed-to-satisfy thicknesses for Zone 2 (115mm thickness).
Analysis of results
Under the guidance of a qualified actuary, the three test scenarios were evaluated using an ‘insulation score’. The score quantifies how well the insulation performs in regulating the internal temperature of the home. Prevailing outdoor climatic conditions had also been monitored to ensure an even comparison on days regarded as having similar weather.
As expected, the presence of insulation above the roof led to a much smoother variation in temperature through the day – lower highs and higher lows. The test results provide useful insight to interpreting the actual performance of insulation as opposed to the SANS204 deemed to comply methodology.
Graph shows the temperature profile within the roof space over a 24-hour period – the first probe being placed immediately below the roof tiles and the second probe on top of the ceiling board or below the insulation layer when present.
Significantly, the unaudited comment of the analyst is that the test analysis shows that at a 99% confidence level the cellulose fibre insulation employed is a better insulator than the generic blanket insulation and that both types are significantly better than no insulation at all.
The unaudited conclusions of the independent analyst was that that cellulose fibre was 6.8 times more effective than no insulation, while the generic blanket insulation was 2.8 times more effective than no insulation at all. Put another way, cellulose fibre was 59% more effective as an insulator compared to the insulation blanket material, although both complied with the deemed to comply requirements of SANS204:2011.
The tests seem to point to cellulose fibre being a superior insulation material, even when R values are the same. This brings into question the reliance on R value as the only factor when considering insulation thermal performance. Potentially, another factor to consider is the more efficient method of application of cellulose fibre into the ceiling where it flows under pneumatic pressure into every nook and cranny, creating a seamless fill above the ceiling.
The graph displays the set of results in the insulated scenario where the ambient temperatures were similar for a 24-hour period. Note at peak period of heat, a differential of at least 2 degrees Celsius was observed between the two materials.
