Effects of Labyrinth Seal Variation on Multistage Axial Turbine Flow

Abstract

The need to increase overall turbine efficiency is always a driving force for redesigning a turbine stage. In particular, the labyrinth leakage flows in the endwall regions contribute to an increase of the overall loss generation. In order to asses this mechanism, a detailed study of the effects of labyrinth seal geometry variation on the blade performance is presented. Two different shroud seal geometries have been experimentally investigated in a two stage low speed turbine facility. The seal geometries differ in the size and shape of the re-entry cavity. The baseline seal is designed with a large rectangular re-entry cavity volume in order to dissipate the kinetic energy of the accelerated leakage flow after the seal gap. The re-entry cavity volume of the alternative seal design is reduced in size and a spline shaped contour is added to the endwall using annular inserts. This modification alters the gas path of the leakage jet and changes the incidence angles on the downstream blade rows. The measurements are performed with state of the art pneumatic and fast response pressure probes at various planes within the turbine stage. It is found that the inserts improved the flow profile uniformity at the endwalls. The measurements within the stator passage reveal the origin of the tip passage vortex formation at the blade suction side, already at the inlet to the stator passage. This result does not conform to the classical secondary flow theory, which suggests that the passage vortex migrates from the pressure to the suction side within the stator passage. The origin and formation of the secondary flow passage vortices at rotor hub and stator tip is described in a flow schematic. The generation of streamwise and tangential vorticity at the interaction area of leakage and main flow field also is studied and discussed. The measured overall polytropic turbine efficiency for the second seal configuration, relative to the baseline case, is reduced by 0.3%. The change in the re-entry flow angle of the leakage gas path reduces the negative incidence angle on the rotor hub and increases it at the stator tip leading edge. The secondary flow and mixing loss is reduced at the hub and increased at the tip in the second test case with the smaller cavity volume. Hence, the combination of small clearances and inserts in the re-entry cavities shows no beneficial effect on the overall turbine efficiency.