OCR Text |
Show temperature was normally controlled to be 1123 K. The static gas pressure differential between the bottom (825 m m above the distributor) and top (5948 m m above the distributor) of the riser was closely monitored and used as the indication of the circulation stability. Most of the experiments reported in this paper were conducted with a fixed total air supply and distribution, as shown in Table 2. Stoichiometry in the riser was changed by coal feed rate only so that the gas velocity above the secondary injection ports was kept at about 5 m/s at 1123 K for all experiments. Other basic operating conditions are also shown in Table 2. Table 1 Analysis of coals (wt% as analysed) Proximate Ultimate (UK) (Colombia) (UK) (Colombia) Moisture Ash Volatile matter Fixed carbon* 3.7 3.5 37.9 54.9 3.5 3.8 33.8 57.9 C H 0* N S 73.1 4.8 12.3 1.3 1.3 76.7 5.1 8.8 1.5 0.7 Gross calorific value (MJ/kg) 28.92 30.063 by difference Table 2 Experimental conditions Primary air flow rate Secondary air flow rate Tertiary air flow rate E H E air flow rate Exhaust 0 2 (prior to afterburning) Bed temperature Pressure differential (bottom - top) Coal feed rate Thermal input from coal Air/fuel stoichiometry of the combustor (excluding E H E & afterburning) (m3/min at STP) (mVmin at STP) (m3/min at STP) (m3/min at STP) (%) (K) (mm H20) (kg/h) (kW) - 1.328 0.200 0.200 0.160 3.0-6.0 1123±20 60 11.3- 13.5 90-110 1.04- 1.24 Concentrations of gas species, on a dry basis, in the exhaust of the secondary cyclone were continually measured by on-line analysers. 0 2 was measured by a paramagnetic analyser (Sybron Taylor-Servomex OA570), C O and N 2 0 were measured by NDIR analysers (Hartmann & Braun, Uras 3 G), N O and N O x were measured by a chemiluminescent analyser (Analysis Automation Ltd. Model 443), S 0 2 was measured by an infrared analyser (Analytical Development Company). All analysers were calibrated frequently, at least daily, using B O C standard calibration gases, to minimise analytical errors. T w o probes were used to extract exhaust gases to sampling lines. A stainless steel hot-probe (4 m m id) supplied gas to the S 0 2 analyser via an electrically heated hot gas sampling line (393 K), two dryers (Magnesium perchlorate) and two filters. A water-cooled stainless steel probe (4 m m id) was used to extract gases for the other gas analysers which were connected to a separated sampling line consisting of water traps, condensers, filters and pumps. Where possible, Teflon tubes were always used in the sampling lines to avoid adsorption and corrosion. Periodically, exhaust gases were also sampled with Teflon bags for further analysis of N 2 0 using a gas chromatograph (GC) equipped with an electron-capture detector. In this case, exhaust gases were extracted through a IN Na2C03 solution to remove S 0 2 and anhydrous CaS04 to |