Pressure surface near the tip clearance

Effects of tip clearance level on the mass transfer on the pressure surface are shown in the contour and surface plots of fig. 4.16 and fig. 4.17, respectively. Similar to the results for with no tip clearance, the unsteady Taylor-Gortler vortices appear evidently on the pressure surface for all tip clearance levels except the largest one. However, the effect of corner vortices disappear near the tip edge. Instead, the relatively higher mass transfer region near the tip edge after Sp/C > 0.15 is caused by the leakage flow sucked into the tip clearance. This effect of acceleration is so strong that much higher and probably unstable mass transfer rates are induced at the two largest tip clearance level at about Sp/C = 0.15 near the tip edge and extended further away from the tip in the span direction at r/C = 6.90%, which can also be observed from the surface flow visualization in fig. 3.10(d). This strong acceleration perhaps also causes the disappearance of the Taylor-Gortler vortices on the pressure surface at the largest tip clearance level.

Local Sh and Sh/Reli2 at different Sp/C locations starting from the tip in the span direction are plotted in fig. 4.18 and fig. 4.19, respectively, for different tip clearance levels. The trend of mass transfer near the tip on the pressure surface is quite different for the case with no tip clearance: the mass transfer rates have maximum at the tip and decrease gradually in the span direction for all non-zero tip clearances in fig. 4.18. However, the effect of different tip clearance on the mass transfer near the tip is not very strong upstream of Sp/C = 0.09. Between Sp/C = 0.09 and Sp/C = 0.48, the two largest tip clearance levels generate much higher and highly fluctuating mass transfer rates near the tip than the two smaller tip clearance levels. Further downstream of Sp/C = 0.48, the mass transfer rates near the tip increase as tip clearance level becomes larger and is much higher than the case with no tip clearance, though the effect of tip clearance is limited to the region close to the tip edge and generally within Z/C < 0.15. For the normalized Re plots, this trend can also be seen clearly in fig. 4.19. For Sp/C > 0.48, the mass transfer curves tend to collapse to one after Z/C = 0.1 disregarding the effect of the Taylor-Gortler vortices. Generally, the effect of tip clearance is limited to Z/C < 0.1 for the small tip clearance levels.

Spanwise and streamwise averaged Sh on the pressure surface are plotted in fig. 4.20 for different tip clearance levels. From fig. 4.20(a), we can find that the effect of tip clearance is evident only for the two largest tip clearance around Sp/C = 0.2. As suggested earlier, it is caused by the highly fluctuated suction flow at the larger tip clearances. The variation of streamwise averaged Sh vs. Z/C in fig. 4.20(b) shows the effect of tip clearance. For the case without tip clearance, the minimum mass transfer rate around Z/C = 0.05 is induced by the corner vertices (Vpc)t, while the mass transfer rate increases monotonicly toward the tip for all tip clearance cases. For the largest the tip clearance of 6.90%C, the fluctuating Sh, caused by the Taylor-Gortler vortices for smaller tip clearance cases, disappears due to the existence of highly accelerated suction flow on the pressure surface.

At the same tip clearance level of r/C = 0.86%, the effects of mainstream Reynolds number and turbulence intensity on the mass transfer on the pressure surface are displayed in contour and surface plots of fig. 4.21 and fig. 4.22, respectively. Similar to the results without tip clearance, the existence of Taylor-Gortler vortices can be found for the low mainstream turbulence cases while for the high mainstream turbulence case, the effect of unsteady vortices disappears. High mass transfer region near the tip edge can be observed and the mass transfer rates in this region increase with the increase of mainstream Reynolds number. The high mainstream turbulence seems not to greatly affect this relatively high mass transfer region near the tip.

The local mass transfer Sh against Z/C at different curvilinear locations on the blade are shown in fig. 4.23. Near the leading edge (Ss/C < 0.04), the higher the mainstream Reynolds number, the higher mass transfer rate can be obtained. Further downstream, the effect of Reynolds number is not very obvious. However, the high mainstream turbulence level increases the mass transfer rate up 100% at Ss/C = 0.15. The same trend is revealed by normalized Sh plots in fig. 4.24. Mass transfer curves for different Reynolds numbers at low turbulence level collapse further downstream of the leading edge, with the unsmoothness caused by the unsteady Taylor-Gortler vortices. The effect of leakage flow sucked into the tip clearance is limited well within Z/C <0.1 for all case and at most two times as large in Sh as that near mid-span. The effect of mainstream turbulence level is also plotted in fig. 4.25. We can find that at high turbulence level, higher mass transfer rates are obtained with tip clearance than that of without.

Spanwise and streamwise averaged Sh on the pressure surface are shown in fig. 4.26 at the same tip clearance of 0.86%C for different Reynolds numbers and mainstream turbulence intensities. In fig. 4.26(a), we can clearly see that higher mainstream Reynolds number and turbulence intensity cause high mass transfer on the pressure surface in streamwise direction. The same effects can also be found in in fig. 4.26(b) for streamwise averaged Sh vs. Z/C. The high mainstream turbulence also totally eliminates the Taylor-Gortler vortices on the pressure surface.

The present measurements on the effects of tip leakage flow on the mass/heat transfer on the pressure surface is different from the results from Metzger and Rued (1989)'s sink flow measurements, in which higher heat transfer rates (up to 100% near the edge) were found to extend to 30% of the span. In present study, the effect of leakage sink flow on mass transfer is found to limited with 10% of the chord at small clearances.

0 0

Post a comment