Straw in steady state conditions is an unexceptional insulator in the context of insulating materials that are typically considered when designing a building envelope.
|Insulation Material||Typical Thermal Conductivity Values, W/mK||Thickness required in a frame construction to provide U = 0.2 W/m2K|
|Hemp-lime||0.06 - 0.09||450mm|
|Straw||0.05 - 0.07||350mm|
|Wood/Hemp/Wool Fibre||0.038 - 0.040||200mm|
|Mineral Wool||0.032 - 0.040||175mm|
|Rigid Foam||0.022 - 0.030||150mm|
Note: of the materials given in the table above straw is the only one that can also be used as the principle structural element.
Note: U = 0.2 W/m2K is a steady state target proposed in case studies for the 2013 Part L draft, the new regulations do allow dynamic methods of analysis where natural fibre insulation materials perform significantly better.
However, the first point to note is that in common with all natural fibre insulation, straw performs better under dynamic conditions. So if a typical day/night cycle is considered better in-situ performance will be recorded than would be predicted from a typical U value calculation.
The next point to consider is that straw is used in bale form and typically laid flat to maximise stability during construction. So more straw is put into the building to fascilitate the construction sequence than would be needed to provide adequate thermal performance.
Finally straw is low density and the higher density renders are thin. So thermal mass and thermal lag time are less than for a traditional masonry construction. However natural fibres interact with moisture in the air which causes a 'virtual' thermal mass effect as water vapour changes phase, this has been more thoroughly investigated in hemp-lime than straw, but is still not fully understood.
A carefully constructed straw-bale building has excellent thermal performance because of its combination of the bales' high insulative value and the thermal mass provided by the interior's thick plaster coating. (Read the section on thermal mass for more on the advantages of a high mass construction.)
- A good starting point is a discussion of what R-value is, and what it is not. It is not an absolute measure of how energy efficient your building is. It is not even a perfect way of predicting the wall’s contribution to thermal comfort. It is one piece of information about the wall that, with other information, can enable you to estimate the heat loss and heat gain through the walls. R-value is the inverse of U-factor (R = 1 / U). U-factor is a measure of thermal conductance, or how easily a material (or system) allows heat to pass through it. This is how U-factor is defined (in the U.S.): the number of British thermal units that pass through one square foot of a material (or system) per hour with a one degree Fahrenheit temperature difference between the two sides of the material. Mathematically:
Btu = British thermal units, a = area in square feet, F = temperature fahrenheit
- In most other countries U-factor is defined in terms of Watts per square meter per Kelvin [W/(m2*K)]. To convert metric (SI) U-factors to inch-pound (IP) U-factors multiply by 0.1761; to convert the other way, simply divide by 0.1761. To convert IP R-values to metric R-values, multiply by 5.6783.
- When a laboratory tests a material (or system) to determine its thermal conductance or resistance, they determine the heat flow from one side to the other on the basis of measured surface temperatures and heat energy required on the warm side of the wall to maintain a steady heat flow. This provides the U-factor, which is then converted to R-value for some purposes. (Nehemiah Stone, 2003)
(the following comments are imperial R value) The theoretical R-value (thermal resistance) for a 16.5 inch (420 mm) straw bale was calculated by Joseph McCabe as 52 (RSI-9.2). This is compared with a theoretical R-value for 3.5 inch (90 mm) of fibreglass (the conventional insulation material used in home construction) of 13 (RSI-2.3). This means fibreglass has an R-value of about 3.7 per inch (RSI-0.26 per centimeter) and straw bales have about 3.2 per inch (RSI-0.22 per centimeter).
Some lab tests of straw-bale assemblies have found significantly lower R-values in practice. However, the more conservative of these results still suggests an R-value of 28 for an 18" wall , which is a significant improvement over the R-14 of an energy-efficient insulated 2x6 wall. Straw-bale experts suggest that it is possible to approach theoretical R-values by giving more attention to detailing, but this has never been documented.
- Tests have shown a range of values from R-17 (for an 18” bale wall) to R-55 (for a 23” bale). Analysis at Oak Ridge National Lab, among other places, has shown that R-values for insulation materials used in “standard” walls are generally much higher than the R-value for the wall as an assembly of disparate materials. Joe McCabe recently postulated that the same phenomenon could account for the difference between the high values from his testing of bales and the lower values obtained in the 1998 Oak Ridge test of a straw bale wall system. While it is possible that the relatively low densities where bales abut each other might contribute to greater heat loss than would be measured through an individual bale, it is unlikely that this would account for the entire difference. This difference between bales and bale walls is similar to the difference between standard insulation and what is found in stud framed walls (insulation voids, thermal bridges, uninsulated headers, and other faults).
- It is noteworthy that all tests of straw bale wall systems prior to the Oak Ridge test in 1998 had potentially significant shortcomings and should not be considered particularly reliable. The last Oak Ridge test had no identified deficiencies and is considered by most to be an accurate determination of the thermal resistance of straw bale walls. ORNL determined the R-value to be R-27.5 (or R-1.45/inch), or R-33 for three string (23”) bale wall systems. Shaving a bit off the top just for conservatism's sake, the California Energy Commision officially regards a plastered straw bale wall to have an R-value of 30.