Determining Heat Transfer and Pressure Drop of Inline and Semi-staggered Tube Banks in Cross-flow

Abstract

Flue gas air preheater is a piece of equipment used to heat combustion air in steam gen-erators that use the combustion of fuels as heat source. Modern steam boilers use ad-vanced Rankine cycle and the use of flue gas air preheater as part of the process in-creases thermal efficiency of the boiler. Flue gas air preheaters are typically recuperative cross-flow tube banks, in which the flows are separated by the tube walls. The heat transfers from flue gas to air by the means of convection, conduction and radiation. Convective heat transfer occurs from tube inside surface to internal flow and from ex-ternal flow to tube outside surface along with radiative heat transfer. Conduction occurs through the tube wall. These heat transfer modes form a thermal circuit. There have been numerous studies and publications on heat transfer and pressure drop of in-line arranged and fully staggered tube banks in cross-flow, but hardly any regarding semi-staggered tube banks. For that reason experimental work was done to determine these attributes for a semi-staggered tube bank and with preset tube longi-tudal and transversal tube pitches. For comparison the tests were done for otherwise geometrically matching in-line tube bank. Testing system was designed and built to Valmet R&D center in Messukylä, Tampere. Although normally air preheaters operate with air and flue gas, different ap-proach was used in the test system. The system comprised of separate water and air sys-tems, of which water system was used as heat source for the exchanger and was chosen to be tube bank internal flow. Air was forced through the tube bank as external flow. The flows were measured along with air temperatures before and after tube bank. A heat transfer model based on literature references was produced for measurement data pro-cessing as heat transfer couldn’t be measured directly. As tube bank external flow con-vective heat transfer was of interest, the heat transfer model was designed to return it as a result. Pressure drop across tube bank was measured directly. The results showed that the measurements used as input for the heat transfer model were successful and the heat transfer model worked as test run repeatability was on good level. Pressure drop measurements were not successful as repeatability was not on a satisfactory level, but yielded comparable results between the two tested tube banks. External flow convective heat transfer coefficient was higher for in-line tube bank in low Reynolds number region, but lower than for semi-staggered one in high Reynolds number region. Results for pressure drop across tube bank indicated opposite behavior.