Date on Master's Thesis/Doctoral Dissertation

6-2009

Document Type

Master's Thesis

Degree Name

M. Eng.

Department

Mechanical Engineering

Committee Chair

Sharp, Michael Keith

Subject

Heat pipes; Solar heating--Passive systems; Solar heating--Research; Solar energy--Research

Abstract

The one dimensional heat transfer, thermal diode effect of heat pipes makes them ideal for passive solar applications. Gains in a heat pipe passive solar wall are not lost during cloud cover or periods of low irradiation. An experimental model was built to test the performance of a heat pipe passive solar wall in a laboratory setting. Experimental variations included varying fluid fill levels and the addition of insulation along the adiabatic section of the heat pipe. Filling the heat pipe to 120% volume of the evaporator section and insulating the adiabatic section achieved efficiencies of 85%. The average rate of water tank temperature change for 80%, 100%, 120%, and 140% fill levels was 1.153°C/hr, 1.195°C/hr, 1.227°C/hr, and 1.203°C/hr, respectively. The addition of fins along the condenser of the heat pipe did not significantly enhance the overall performance. A computer model was constructed to simulate the performance of direct gain, indirect gain, and integrated heat pipe passive solar systems in different climates. The locations selected include: Louisville, Kentucky; Albuquerque, New Mexico; Madison, Wisconsin; and Rock Springs, Wyoming. Integrated heat pipe systems performed better than their direct and indirect gain counterparts in all climates. The water wall indirect gain system outperformed concrete wall systems in all climates. The lowest solar fractions were achieved by the direct gain system in all climates except Albuquerque, New Mexico (sunny and cool). The mild temperatures and cloudiness of Louisville, Kentucky provided an excellent climate for passive solar systems. Passive solar systems in Louisville (cloudy and cool) performed better than Madison, Wisconsin (cloudy and cold) and Rock Springs Wyoming (sunny and cold). Integrated heat pipe wall design variations were analyzed to achieve the best overall performance. These variations included: glazing thickness, extinction coefficient, absorber plate and heat pipe material, selective surface, insulation thickness, and water tank size. Decreasing the thickness of the glazing by 7.15 mm improved the solar fraction by 2.69%. Changing from a greenish cast glass to a low iron glass reduces the extinction coefficient and improves solar fraction by 8.82%. Switching absorber plate and heat pipe material from copper to aluminum reduces cost and decrease the solar fraction by 2.23%. Having the absorber plate electroplated with a black chrome surface opposed to painting the collector with flat black paint increase the solar fraction from 19.28% to 48.36%. Changing the insulation thickness around the collector improves the overall losses of the collector. Raising the R-value by a factor of 5 improved the solar fraction by 0.12%. Increased water tank size improves solar fraction but increases the units weight. Validation of the computer model was made by simulating the laboratory experiments and comparing the data. Temperatures across the system were matched by adjusting the calculated conductances. The computer simulation was able to graphically match the measured temperatures of the experiment. Conduction between the evaporator and condenser sections of the heat pipe was only 6% of the calculated value. An economic assessment was then performed to analyze the overall cost of the unit, deliver design recommendations, optimize cost, complete life cycle costing and calculate payback period. The most cost effective design would incorporate an aluminum absorber plate with four heat pipes. The solar glazing would have one cover and be made with low iron glass (3/32nd inch thick). For the city of Louisville with a load to collector ratio of 10 W/m2K the total cost of the unit would be $1825.46 for a collector area of 2.43 m2 (26.12 ft2). The solar fraction provided by this optimized unit would be 42.67% of the heating load.

Share

COinS