Thermoelectric Energy Harvesting from Transient Ambient Temperature Gradients

Abstract

We examine a thermoelectric harvester that converts electrical energy from the naturally occurring temperature difference between ambient air and large thermal storage capacitors such as building walls or the soil. For maximum power output, the harvester design is implemented in two steps: source matching of the thermal and electrical interfaces to the energy source (system level) followed by load matching of the generator to these interfaces (subsystem level). Therefore, we measure thermal source properties such as the temperature difference, the air velocity, and the cutoff frequency in two application scenarios (road tunnel and office building). We extend a stationary model of the harvester into the time domain to account for transient behavior of the source. Based on the model and the source measurements, we perform the source and load matching. The resulting harvester consists of a pin fin heat sink with a thermal resistance of 6.2 K/W and a cutoff frequency 2.5 times greater than that of the source, a thermoelectric generator, and a DC/DC step-up converter starting at a total temperature difference of only ΔT = 1.2 K. In a final road tunnel field test, this optimized harvester converts 70 mJ of electrical energy per day without any direct solar irradiation. The energy provided by the harvester enables 415 data transmissions from a wireless sensor node per day.