Description
Thermal insulation test methods approach their lower limits as thermal resistance falls below 0.1 m2⋅K/W. This is the minimum value specified in ASTM C 518 (ASTM International, 2010b) while ASTM C 177 (ASTM International, 2010a) proposes about 0.06 m2⋅K/W. Nevertheless these are the test methods, along with their ISO equivalents, required by Australasian building codes and directed at many products and materials with thermal resistance on the low side of 0.1 m2⋅K/W. Alternatives, such as ASTM E 1530 (ASTM International, 2011), cover much lower resistances but require carefully prepared small specimens and very-high contact pressures and are therefore largely unsuitable for both technical and compliance reasons. For these low resistances, the insulation test methods face large errors because of interface resistance between specimen and the apparatus hot and cold plates. Staying with C 518, the problem can be avoided by using direct measurement of the test specimen surface temperatures, but this is difficult, has its own accuracy issues, and is often impractical for commercial laboratories. This technique is generally used in conjunction with interface materials such as flexible foam between the specimen and the hot and cold plates, to enhance contact and also provide an access path for temperature sensors. The alternative prospect of using these interface materials to ensure good specimen contact has been studied, in conjunction with a simple two-step thermal resistance determination based on the difference between presence and absence of the test specimen.
This article presents results of a study using this difference approach for the measurement of 12 highly conducting materials, including sheets of aluminum, phenolic, HDPE, MgO, bonded rubber and cork granules, PMMA, and compressed wood fiber. For each material, repeated measurements have been performed with four different interface or “buffer” materials: PVC, silicone, EVA, and nitrile. Silicone sponge provides the most uniform results, consistent with a measurably lower hysteresis. The difference technique yielded a lower indicated thermal resistance than direct measurement by between 0.003 and 0.01 m2⋅K/W, with some variation depending on the specimen surface characteristics and to a lesser extent on the choice of buffer. Larger differences were associated with bowed, uneven or roughly surfaced specimens. The difference-technique results have greater variability, but they may be seen as better estimates of the actual specimen resistance, as contact resistance is much lower for soft-surface interfaces. An interface resistance of up to 0.01 m2⋅K/W is large enough to be of significance in many thermal measurements.
Keywords
thermal resistance, measurement, test methods, contact resistance, interface resistance
DOI
10.5703/1288284315544
Included in
Flexible Buffer Materials to Reduce Contact Resistance in Thermal Insulation Measurements
Thermal insulation test methods approach their lower limits as thermal resistance falls below 0.1 m2⋅K/W. This is the minimum value specified in ASTM C 518 (ASTM International, 2010b) while ASTM C 177 (ASTM International, 2010a) proposes about 0.06 m2⋅K/W. Nevertheless these are the test methods, along with their ISO equivalents, required by Australasian building codes and directed at many products and materials with thermal resistance on the low side of 0.1 m2⋅K/W. Alternatives, such as ASTM E 1530 (ASTM International, 2011), cover much lower resistances but require carefully prepared small specimens and very-high contact pressures and are therefore largely unsuitable for both technical and compliance reasons. For these low resistances, the insulation test methods face large errors because of interface resistance between specimen and the apparatus hot and cold plates. Staying with C 518, the problem can be avoided by using direct measurement of the test specimen surface temperatures, but this is difficult, has its own accuracy issues, and is often impractical for commercial laboratories. This technique is generally used in conjunction with interface materials such as flexible foam between the specimen and the hot and cold plates, to enhance contact and also provide an access path for temperature sensors. The alternative prospect of using these interface materials to ensure good specimen contact has been studied, in conjunction with a simple two-step thermal resistance determination based on the difference between presence and absence of the test specimen.
This article presents results of a study using this difference approach for the measurement of 12 highly conducting materials, including sheets of aluminum, phenolic, HDPE, MgO, bonded rubber and cork granules, PMMA, and compressed wood fiber. For each material, repeated measurements have been performed with four different interface or “buffer” materials: PVC, silicone, EVA, and nitrile. Silicone sponge provides the most uniform results, consistent with a measurably lower hysteresis. The difference technique yielded a lower indicated thermal resistance than direct measurement by between 0.003 and 0.01 m2⋅K/W, with some variation depending on the specimen surface characteristics and to a lesser extent on the choice of buffer. Larger differences were associated with bowed, uneven or roughly surfaced specimens. The difference-technique results have greater variability, but they may be seen as better estimates of the actual specimen resistance, as contact resistance is much lower for soft-surface interfaces. An interface resistance of up to 0.01 m2⋅K/W is large enough to be of significance in many thermal measurements.