Guarded hot plate method (GHP method) is the most popular way of measuring thermal conductivity of thermal insulation. However, there are large differences among the thermal conductivities measured by different apparatuses which are made in different institutions in the temperature range above 100 °C. It is considered the reason that we have no standard material for measurement of thermal conductivity. The standard material is only glass wool which are produced by NIST (National Institute of Standards and Technology) in a temperature range from about 7 to 67 °C. Then, I have researched how far the influence of the difference in measurement methods and apparatuses extends. Furthermore, I have investigated how to measure accurately thermal conductivity in the temperature range which there are few reference materials. In this study, I proposed to increase the accuracy of thermal conductivity of thermal insulation by comparing with values obtained using different methods. I investigated the practicality of the comparison of the results obtained by different methods, which are the GHP, cyclic heat, transient hot wire, and Hot Disk methods, for the accurate measurement at the temperature range from −170 to 1300 °C. First, I developed the three types of measurement apparatuses; one can measure thermal conductivity by using both the GHP and cyclic heat methods in the temperature range from 100 to 1300 °C, the second can measure using both the cyclic heat and transient hot wire methods in the temperature range from 100 to 1000 °C, and the third can measure using both the cyclic heat, transient hot wire, and Hot Disk methods in the temperature range from −170 to 25 °C. Next, I measured thermal conductivities of various thermal insulations using these apparatuses. In the temperature range above 100 °C, the results obtained by using the GHP and cyclic heat methods agree with each other within ± 10% deviations. In like manner, the results obtained by using the cyclic heat and transient hot wire methods agree with each other within ± 10% deviations. Furthermore, in the temperature range from −120 °C to 25°C, the results obtained by using the cyclic heat, transient hot wire, and Hot Disk methods agree with each other within ± 10% deviations. Therefore, it is thought that to compare the thermal conductivities obtained by the different measurement methods will be practical for improvement of the accuracy measurement at the temperature range in the absence of reference materials.
