【Objective】 Measurement of soil heat flux is pivotal to researches on budgeting of surface heat. To measure soil heat flux, there are several ways, one of which is the heat flux plate method, simple and easy to handle, but unpredictable in accuracy.【Method】 In this study, an indoor experiment was carried out in combination with computer simulation to determine errors of the measurement with the soil heat flux plate method and explore its influencing factors. In this paper, with soil deemed as homogenous medium and the heat flux plate as composite material, influences of the composite material on temperature field of and heat flux distribution in the homogenous medium. In addition, analysis was done of applicability of the modificatiton equation Philip put forth in 1961. 【Result】 Results show that when the soil and the plate is quite approximate in heat conductivity measurement error of the method is low and Philip’s equation is relatively low in applicability under transient conditions.【Conclusion】 The study shows that due to difference in thermal conductivity between the heat flux plate and the soil, the heat flow will distort at the edge of the plate, thus leading to occurrence of errors in the measurement. As Philip’s equation simplifies computation in the derivation process, difference is still found between calibrated value and theoretical value when soil temperature varies with time, which indicates that its application is limited in range.