| Abstract |
The deposition on a superheater and economizer may come from loose ash particles adhering to their tube surfaces. It causes a fouling thermal resistance, an increase in the flow pressure drop, and hence a higher operating cost. An active approach such as using a high-grade fuel can diminish the effect. How- ever, it is costly. The deposition may be predicted using sophisticated computational fluid dynamics soft- ware. Alternatively, for the asymptotic mode of deposition, it might be predicted by the lumped models proposed by various researchers. However, they can be used to predict only the saturated time and the overall saturated thermal resistance. In this work, a novel passive approach is proposed. Isosceles triangle cross section wedges were installed parallel to the tubes to increase the wall shear stress due to the flow (t) and hence decrease the deposition. In addition, by modeling the tube bundle as a lump, a mathemati- cal formulation for predicting the deposition behavior of the tube bundle was developed. The experiment was carried out on an inline 6 rows –3 columns tube bundle with constant air velocity and ash feed rate. Four isosceles triangular wedges with the same height but with different base ( b) to height ( e ) ratio or aspect ratio ( α), with α= 0.275, 0.550, 0.758 and 1 being used. Three wedge positions from the tube bundle inlet in the direction of the flow ( x ) to the tube bundle length ( L ) ratios ( x/L ) are considered: at the middle between the second row and the third row ( x/ L = 1 / 3 ), at the middle between the fourth row and the fifth row ( x/ L = 2 / 3 ), and at the tube bundle outlet ( x/ L = 1 ). The regression approach was used to analyze the overall thermal resistance obtained from the experiment, according to the obtained mathematical relation. The saturated time and the overall saturated thermal resistance obtained from the analysis were used to calculate the deposition rate on the tubes ( f d ), the removal rate of the deposition, deposition growth rate, coefficient of the deposition ( k d ) and coefficient of the removal ( k r ). The result showed that the overall saturated thermal resistance and deposition were inversely proportional to t. f d , the removal rate, the deposition growth rate, k d and k r increase with the wedge distance. Except for f d and k d , those parameters were sensitive to a . The tube bundle with wedges has higher values for those parameters. For the bundle with wedges, those parameters were inversely proportional to t . The product of k r and t is the governing parameter of the saturated time |
Publication
Journal Publication
ไทย