The integration of heat pipe technology into natural ventilation systems

The aim of this research was to integrate heat pipe technology into natural ventilation air streams to provide passive cooling in regions of hot arid climatic conditions without the use of mechanical assistance. The study used numerical and experimental techniques identified from literature to carry out the research. Design parameters of heat pipes including the working fluid and geometrical arrangement were investigated using Computational Fluid Dynamics (CFD). Flow and thermal behaviour of two-phase heat pipe working fluids including water, ethanol and R134a were investigated and their performance was quantified in terms of heat transfer and the overall effectiveness. The results showed that water has the highest cooling capability for the downstream airflow and its working performance was approximately 24% superior in comparison to refrigerant R134a and 42% higher in relation to ethanol. The analysis further determined that for low-speed hot airstreams, heat pipe working fluid properties play an important role in enhancing heat transfer and that the specific heat capacity of the fluid was the most influential parameter in increasing convective heat transfer by 39%. Subsequent to the working fluid, geometrical arrangements of the heat pipes were studied. Using a fixed physical domain, the findings displayed that the optimum spanwise thickness between the pipes was 50mm (spanwise thickness to pipe diameter ratio of 2.5) while the optimum streamwise distance was 20mm corresponding to the streamwise distance-to-pipe ratio of 1.0. In addition, the periodic time-dependant model determined the thermal response of heat pipes in relation to external temperatures and established the relationship between source temperatures and downstream profiles. The final part of the study validated the CFD findings through full-scale wind tunnel experimentation. Experimental testing was carried out on heat pipes using water and R134a as working fluids at varying spanwise configurations. The error patterns were found to independent of the heat pipe geometry and working fluid. The validation study determined the error range which varied between 0.6% and 18.1% for velocity, 0.7% and 18.8% for pressure and between 0.01% and 2.8% for temperature, showing a good correlation between the CFD and experimental techniques and with previous work found in published literature.