Characterization of Commercial Thermoelectric Modules for Precision Heat FLux Measurement

TL;DR

This paper explores cost-effective methods for precise heat flux measurement using commercial thermoelectric modules, comparing passive and active modes, and proposing a dual-TEM scheme to improve accuracy and stability.

Contribution

It introduces a dual-TEM measurement scheme that significantly reduces sensitivity uncertainty and demonstrates the calibration and response characteristics of TEMs for heat flux measurement.

Findings

Active mode is ~7 times more sensitive than passive mode.

Dual-TEM scheme reduces sensitivity uncertainty by up to 4 times.

Response time of TEM is approximately 9-11 seconds for heating and cooling.

Abstract

In this article, we present a cost-effective approach to the precision measurement of heat flux using commercial thermoelectric modules (TEMs). Two different methods of measuring heat flux with TEMs are investigated, namely passive mode based on the Seebeck effect and active mode based on the Peltier effect. For both modes, a TEM as a heat fluxmeter is calibrated to show a linear relation between the voltage across the TEM and the heat flow rate from 0 to 100 mW. While both modes exhibit sufficiently high sensitivities suitable for low heat flow measurement, active mode is shown to be $\sim$7 times more sensitive than passive mode. From the speculation on the origin of the measurement uncertainty, we propose a dual TEM scheme by operating the top TEM in passive mode while its bottom temperature maintains constant by the feedback-controlled bottom TEM. The dual-TEM scheme can suppress the sensitivity uncertainty by up to 4 times when compared to the single-TEM passive mode by stabilizing the bottom temperature. The response time of a 1.5 cm $\times$ 1.5 cm TEM is measured to be 8.90 $\pm$ 0.97 seconds for heating and 10.83 $\pm$ 0.65 seconds for cooling, which is slower than commercial heat fluxmeters but still fast enough to measure heat flux with a time resolution on the order of 10 seconds. We believe that the obtained results can facilitate the use of a commercial TEM for heat flux measurement in various thermal experiments