An ash monitoring system prototype that consists of a “clean” heat flux sensor and a “dirty” heat flux sensor was developed in this study. The “clean” heat flux sensor was studied numerically and experimentally while the “dirty” heat flux sensor was tested in the experiment. Two different measurement methods were applied on the “clean” sensor, one from the original study and one proposed in the present work. The new method required additional data processing procedures to be able to work in an on-line basis. Among the three data processing procedures developed in this study the central temperature difference procedure was found to be the most reliable one. Numerical results provided valuable information about the heat transfer pattern at the sensing element and also the performance of the sensor at high radiation heat flux levels. A rough calibration of both “clean” and “dirty” heat flux sensors was conducted experimentally. Nevertheless, the experimental results still served as a primary assessment for both sensors. Observed disagreements between the original study and the present one were probably caused by the modification of the sensor structure made in the present study. Sensitivity variation at high heat flux levels did not appear in the numerical results of either measurement method. Low overall sensor temperature change was believed to be the main reason as it meant less change in thermal properties of the sensing disc. Although the results of the study suggested that the proposed oscillation method was less attractive than the original one, it increased the possibility of resolving the surface characteristic variation problem that was considered crucial for the performances of the “clean” heat flux sensor regardless of the measurement method used. / Thesis (Master, Mechanical and Materials Engineering) -- Queen's University, 2009-09-03 16:38:26.48