Development of a meso-scale liquid-fueled burner for electricity generation through the use of thermoelectric modules
The goal of this research was to design, build and test a small burner and heat exchanger system that could be used as a source of heat for thermoelectric modules (TEMs) for the purpose of generating portable electric power for soldiers in the field. The project was conducted as a subcontract to Marlow Industries Inc. which was under contract from the U.S. Army. The scale of the burner thermal output was to be in the approximate range of 2 kW of heat production and it was to be able to operate on a liquid fuel, specifically JP8. The first burner investigated was a custom burner designed and built at UT. It was tested with various fuel and air delivery systems. Different methods to start it, with the goal of developing an electrical starting system, were also investigated. It was capable of operating at outputs over 1 kW, but was difficult to start reliably and fuel vaporization characteristics were sensitive to operating conditions. Two commercial burners were also studied, each with somewhat different designs. One of those burners, manufactured by MSR, was chosen to be further tested in conjunction with a heat exchanger and thermoelectric modules. The performance of the thermoelectric modules used in this study was determined to be very dependent on an attached resistive load, with a peak power output occurring at approximately 3 ohms. Power output was also determined to increase linearly with increasing temperature difference between the hot and cold sides of the module. Power output followed similar trends as open circuit voltage. The temperatures of the heat exchanger across its width were very uniform, but the accuracy in centering the heat exchanger over the burner could significantly affect temperatures. The time to reach steady state temperatures was relatively insensitive to the length of the heat exchanger. The presence of attached thermoelectric modules reduced the temperature of the heat exchangers and exhaust gas slightly. Reducing the heat exchanger length resulted in higher metal temperatures. Without cooling the cold side of the thermoelectric modules, performance increased while the system was heating up, but then dropped after reaching a peak. Cold side cooling improved thermoelectric performance by increasing its temperature difference. Active cooling with a blower and heat sink provided even better performance than passive cooling using just a heat sink at the expense of a larger parasitic load. The TEMs on the 5 inch long heat exchanger could generate 6.32 W with passive cooling, but active cooling would produce no net power. The 11 inch long heat exchanger could generate 12.8 W with passive cooling, and 16 W net could be generated with active cooling. A heat exchanger efficiency calculation showed that the 16, 11 and 5 inch long heat exchangers were about 94.4%, 93.4%, and 90.7% efficient respectively. This efficiency was defined as the ratio of the heat transferred to the heat exchanger to the heat released in the flame.