BRANZ recently developed a procedure to determine the ‘real life’ thermal performance of reflective foil, bulk insulation or a combination of both. Some interesting results came from the study.
Measuring the thermal performance of a foil floor insulation system depends on both the construction details and parameters like ventilation and the condition of the foil surface. It is usually impractical, or even impossible, to construct a representative test system that would enable reliable laboratory measurements (in an apparatus like a guard hot box).
It is possible to use a laboratory-based measurement to confirm theoretical predictions for ideal systems where the foil is new and there is no ventilation or moisture to affect the reflectiveness of the foil surface. In practice, the thermal performance of foil systems is best measured in-situ using heat flux transducers and data acquisition in a way that effectively averages the large daily variation in heat flow over a period of a week or more.
The technique used here has its basis in a number of previous BRANZ projects.
Use of heat flux transducers
BRANZ has been using heat flux transducers for field measurements of heat flow since the 1970s. Large panel transducers were used to investigate the thermal performance of slab-on-grade floors and for a survey of the thermal performance of houses in the early 1980s that measured the R-value of walls, floors and ceilings.
Whilst these heat flux transducers have been an important part of a number of research projects, their use has largely been restricted to scientists and technicians from BRANZ. They have mainly been used in the winter months when there is sufficient heat flow through building envelopes or in specialist applications such as cool stores.
|Foil draped||Foil along bottom of joists||Rigid fibrous polyester|
|Test period (days)||8||8||20||20||16|
|Temperature difference between upper surface of the floor (including carpet) and the air underneath the floor (°C)||12–18||0–7||12–22||-1–7||18–25|
|Average temperature difference (°C)||14||3||16||3||22|
|Heat flow (W/m2)||13–14||2–5||12.5–15||1–5||9.5–11.5|
|Accumulative R-value (m2 K/W)||1.10||1.05||1.25||1.55||2.05|
|R-value range assuming 10% uncertainty in the measurement||1.0–1.2||0.95–1.15||1.1–1.4||1.4–1.7||1.85–2.25|
1980s draped foil studies
The 1980s survey found that the actual R-values of suspended timber floors insulated with draped foil were less (about 1.3 m2°C/W) than the theoretical values (2.6 m2°C/W). Simplifying assumptions must therefore be applied to the theoretical values because of the complexity of the radiant heat exchange processes and the generally dynamic nature of the heat flow.
The actual R-value is determined from the accumulative sum of temperature difference and accumulative sum of heat flow.
The 1980s field survey floors all had draped foil, whereas retrofitting existing homes involves retrofitting foil insulation by fixing to the bottom of floor joists.
New foil study with heating box
Two goals of the recent project were to simplify the use of the equipment and to develop a ‘heating box’ to enable measurements over a wider range of ambient conditions. The heating box consists of an insulated box designed to surround the heat flux transducer panel and provide a higher constant air temperature than the surrounding room.
The thermal performance of a suspended timber floor insulated with foil was investigated for two scenarios: when foil is draped over the floor joists and when foil is installed along the bottom of the joists as in a retrofit situation. For the draped foil example, the aim was to measure the performance of foil that had been installed 15 years ago so that the effects of contamination and/or corrosion of the foil surface could be included in the measurement.
Table 1 summarises the results for the six sets of measurements performed at two locations in one house. Location 1 was in a spare bedroom that was heated only indirectly, hence the use of the heating box to boost the temperature difference. Location 2 was in the lounge. Although the lounge had a solid fuel heater, the temperature at floor level was only a few degrees above the temperature under the floor.
At least 10°C difference needed for tests
Previous studies have estimated a 10% uncertainty in determining thermal resistance using the heat flux transducers, based on an assumption that the average temperature difference is at least 10°C. If the temperature difference is less than 10°C, the method becomes less reliable and repeat measurements are needed. This 10% margin includes calibration errors and uncertainties associated with installation and in-use conditions, and is an acceptable margin of error.
For the measurements at location 2 where there was no heating box, the average temperature difference was only 3°C. The analysis results in an R-value with a much higher margin of error.
Retrofitted and draped foil perform equally
The improved equipment proved its worth by enabling more reliable and quicker in-situ measurements, offering the opportunity to carry out in-situ testing.
As expected, the older foil in the draped foil example has reduced thermal performance compared with new foil.
The test conducted confirmed that retrofitted foil installed to the underside of the joists should be expected to have similar thermal performance to draped foil. Like draped foil, however, its thermal performance should be expected to decrease with age as the foil contaminates and loses its shiny surface. It is reasonable to assume that the performance in another 15 years would be similar to the current performance for the draped foil, as it shouldn’t degrade any further.
For the rigid fibrous polyester insulation, the measured performance is much closer, but slightly above what is predicted from a theoretical calculation.
Longer durability or higher R-value
The study indicates if longer-term thermal performance is needed, for example, 50-year durability requirements of the New Zealand Building Code, then bulk insulation materials such as fibrous insulation or expanded foam should be used instead.
Likewise, if R-values above the standard R1.3 provided by draped or retrofitted foil are needed, then the only solution is to either use multiple layers of foil or bulk insulation products.
Product R-value better for bulk insulation
It is possible to measure the in-situ thermal performance for bulk insulation materials, but because of the higher measurement uncertainty associated with the method, it is better to use the traditional heat flow meter apparatus to measure the product R-value.
The in-situ measurement equipment has the ability to investigate installation quality and to measure systems that contain air spaces and reflective surfaces. Since each result is unique to the particular construction that is tested, it is only through measurements across multiple locations that an average actual performance can be established.
This research was funded by the Building Research Levy.
Articles are correct at the time of publication but may have since become outdated.