This study presents a simple, inexpensive and new method for the thermal conductivity measurement of low thermal conductivity materials using thermoelectric technology. The proposed system consists of a thermoelectric cooler, a thermoelectric generator and two heat sinks cooled with two fans. To measure the thermal conductivity, a sample is placed between a thermoelectric cooler and a thermoelectric generator. The heat generated on the hot side of the thermoelectric cooler passes through the sample and enters the thermoelectric generator makes a temperature difference between its two sides. The temperature difference produces an electrical voltage which is a function of the thermal conductivity, thickness of the sample and the power consumption of the thermoelectric cooler. As a result, a mathematical modeling was used to obtain a nonlinear function for estimation of the thermal conductivity as a function of produced electrical voltage, thickness of the sample, electrical voltage and electrical current of the thermoelectric cooler. The proposed method evaluated by testing several materials with known thermal conductivity (0.02 W.m−1. K−1 - 0.05 W.m−1. K−1) and the results showed a good agreement (maximum error of 10%) between the reference values and the experimental data. The measuring time of the system is considerably short and it only took 12 min. Repeatability of the measurement method for different materials showed that the rate of change is between 0.5% - 6%. A linear correlation for thermal conductivity estimation of Polystyrene with density of 8.0 Kg.m−3 to 25 Kg.m−3 has been proposed with a maximum error of 7%. Another linear correlation has been proposed for thermal conductivity estimation of other materials (0.02 W. m−1. K−1 to 0.05 W. m−1. K−1) with a maximum error of 10%.