Analytical and experimental analysis of fireplace performance
MetadataShow full item record
A preliminary fireplace design to reduce particulate, carbon monoxide, and unburned hydrocarbon emissions has been fabricated and tested. Using a damper and glass pane, the design solves the problem of uncontrolled air entering the firebox. The pane reduces the inlet area and directs the flow of air towards the charcoal bed. The damper aids in slowing down the inlet air flow and decreasing the amount of excess air in the combustion region. Limiting the supply of inlet air produces a slower flow through the combustion zone resulting in higher temperatures and longer residence times. The aesthetic appearance of the fire has nonetheless been maintained.
The preliminary design was tested using two methods which varied in fuel type, test length, and fuel reloading. Each method followed dilution tunnel particulate sampling EPA method 5G. Carbon monoxide emissions were continuously monitored with an infrared gas analyzer. The average preliminary design particulate emission factor was reduced from baseline testing by 84 percent with method 1 and 86 percent with method 2 testing. The average preliminary design carbon monoxide emission factors for method 1 and 2 testing, were decreased by 57 and 59 percent respectively. A baseline test was defined as the operation of the fireplace as it would arrive from the factory. Safety issues were not addressed in this thesis. Improvements in safety will likely be required and may cause an increase in emissions from the preliminary design.
A two dimensional variable property cartesian coordinate computer model has been written which determines the temperature and velocity distributions in the firebox. Properties of air were used to represent the fluid. The model determined stack flow rates and temperatures in the flue for the preliminary design. These values were then compared to experimental data.
Four calculations were run under various conditions of inlet air, heat source, and baffie temperatures. The heat source temperature ranged from 1000 to 1300 K. The temperatures predicted by the computer model were within 1 to 28 percent of measured values. Stack flow rates determined by the model were 66 to 78 percent less than measured. The model also predicted a recirculation area.
- Masters Theses