Abstract
Laminar separation induced flow transition has been regarded as one common but complex flow in low pressure turbine (LPT), which is one main source of flow losses of LPT. Lots of studies have demonstrated that the realistic rough blade of LPT favors the reduction of flow losses as the laminar separation can be delayed by increasing the roughness. Subsequently, the flow separation bubble on the suction side can be reduced, and even disappears. The present paper investigates the effects of surface roughness on flow separation of an LPT cascade, T106C. Firstly, a detailed grid-independent study assisted by Richardson extrapolation is presented to illustrate the reliability of the numerical solutions obtained by solving RANS and SST γ -Re˜θ transition model equations. Then a series of roughness with different geometric roughness height (Ra) are imposed on the whole suction side. The effects of roughness on the reduction of separation bubble are illustrated and compared. The results demonstrate that there is one critical Ra, below which the flow losses decrease, while above which the flow losses increase as Ra increases. Finally, different roughness allocations with sand grains imposed on different blade portions of the suction side are studied. Through comparisons, the sensitivities of flow loss reduction to the roughness on different blade portions are evaluated and illustrated. The study paves the way for finding an optimal roughness allocation on the suction side to minimize the flow losses.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call