Effect of Swirl Number and Fuel Type Upon the Combustion Limits in Swirl Combustors

Abstract

Increasing interest in lean fuel premixed swirl combustors has arisen because of reduced NOx emissions. Alternative fuels, including hydrogen-enriched natural gas and by products of process industries such as coke oven gas are now receiving increasing attention. This gives rise to areas of concern including, flashback, temperature levels, blow-off and combustion instability. Flashback with hydrogen containing fuels is of special concern, owing to the high flame speed of hydrogen, to such an extent that diffusion combustion is commonly employed resulting in high NOx emissions. This paper examines the effect of hydrogen containing fuels upon flashback and blow-off in a generic, compact, premixed swirl burner in swirl number regimes representative of those found in practical systems. All results are obtained at atmospheric pressure without air preheat as a precursor to pressurised tests, the burner firing freely into atmosphere for most tests. The swirler has radial tangential inlets firing into a swirl chamber, which then feed into the exhaust. A central fuel injector just extends into the exhaust and is ∼40% of the exhaust diameter, a common industrial size. Four tangential inlets are used for S = 1.47, while nine has been used for S = 1.04 and S = 0.8. Flashback and blow-off are sensitive to the level of swirl, the exhaust configuration and the type of fuel. High swirl numbers, S = 1.47, gave flashback limits with methane considerably worse than those produced at S = 1.04 and S = 0.8, although there were differences in exhaust nozzle configuration. At equivalence ratios ∼1 total mass flow at which flashback occurred (hence velocities) was reduced by a factor of two. Changes in flashback behaviour were especially noticeable when the hydrogen content in fuel blends was > 60% by volume. Blow-off was very much a function of hydrogen content of the fuel and Swirl Number. Best blow-off limits for all fuel blends are obtained at S = 0.8, the worst for S = 1.47. Coke Oven gas (COG) with 65% hydrogen content gave best blow-off limits of the fuels tested, although data was not available for pure hydrogen due to rig limitations.