= 0.833 (fuel-lean). Sampling of gases was not feasible at other equivalence ratios as extensive formation of particulates blocked the filters and coalescers in the sampling lines.
Sulphur was added to the combustion system as SO2 (gas). Spectrol range (long term stability) cylinders of 2 % vol. SO2 (balance N2) were supplied by BOC Special Gases. In order to obtain the desired increasing amounts of fuel-S (which, in turn, depend upon the mass flow rate of fuel), the necessary volumetric flow rates of SO2 were calculated. Since additional nitrogen is introduced as a balance in the SO2 cylinder, oxygen was also added to preserve the stoichiometry of the combusting mixture. Thus, an algorithm was devised to calculate the exact mass (volumetric) flow rates of sulphur dioxide, nitrogen, oxygen and atomisation air to be introduced in a given combustion system.
The dopant gases were added in the atomisation air stream. The amount of compressed air was progressively reduced to accommodate increasing amounts of dopant gases so that a constant total mass flow rate of 23.2 g/min was maintained. This ensured that no dilution was caused by the addition of dopant gases.
The volumetric flow rates of the gases were controlled by means of Chell Instruments Ltd. "Tele-Hastings" digital mass flowmeters. Monitoring the flowmeters enabled adjustment to the set-point values within ± 5 % deviation. The flowmeters also produced an output of the gas flow rates expressed in standard conditions (ie 0 °C and 1 atm), which was directed to the Orion 3561 D data-logger for recording purposes.
The fuels used in these experiments were the heavy coker gas oil M1, the heavy vacuum gas oils G1 and G2, supplied by Repsol Petróleo S.A., and Orimulsion. This latter fuel was provided by National Power plc. Their specifications are reproduced in "Appendix II". The introduction of other fuels widened the scope of fuel-N and S contents, so that comparative studies of the influence of these variables could be made. Unlike the heavy gas oils used in this thesis, Orimulsion is a [heavy fuel oil:water] emulsion with approximately 30 % water.
Increasing mass flow rates of SO2 gas were added in order to simulate concentrations of fuel-S increasing in steps of 0.5 % by weight, up to a maximum of 2 % wt fuel-S added. For Orimulsion the maximum addition of sulphur was 1.5 % by weight.
Experiments with heavy gas oils were carried out at three equivalence ratios, namely 0.833 (fuel-lean), 1.000 (stoichiometric) and 1.200 (fuel-rich). Experiments with Orimulsion could only be performed at = 0.833 (fuel-lean). Sampling of gases was not feasible at other equivalence ratios as extensive formation of particulates blocked the filters and coalescers in the sampling lines.
The furnace wall temperature was set at 900 °C to ensure that no thermal-NOX was formed. It has been shown in chapter IV that the formation and emission of thermal-NOX is minimum at these experimental conditions. The fuel mass flow rate was kept constant at 3.5 g/min for heavy gas oils and 5 g/min for Orimulsion as this fuel contains 29.8 % of water, which maintained the equivalent fuel oil flow rate of 3.5 g/min.
Sampling of gaseous species was performed at a distance of 500 mm from the atomiser. Previous experimental work showed that no relevant changes in gas species concentrations occurred beyond this distance from the atomiser nozzle at all equivalence ratios. The experimental readings lasted for eight minutes for all gases, except for NO and NO2, which were recorded during alternative periods of four minutes each. The logged values were averaged and standard deviation was calculated, except for nitrogen dioxide, as explained in chapter III. Finally, experimental values were plotted versus the percentage of fuel-sulphur added as SO2 dopant.
Finally, the Thermo-Electron chemiluminescence NOX analyser was tested for possible interference by SO2. No interference occurred.
 Previous |  Table of Contents |  Next |