Newest Issue

Research Papers

J. Turbomach. 2016;139(2):021006-021006-11. doi:10.1115/1.4034480.

Previous work has shown that low-stagger contouring near the endwall of a nominally high-lift and high-stagger angle front-loaded low-pressure turbine (LPT) airfoil is successful in reducing endwall loss by limiting the development and migration of low momentum fluid associated with secondary flow structures. The design modification that leads to loss reduction in that study was determined from an intuitive approach based on the premise that reducing flow separation near the endwall will lead to reduced loss production. Those authors also relied heavily upon Reynolds-averaged Navier–Stokes (RANS) based computational tools. Due to uncertainties inherent in computational fluid dynamics (CFD) predictions, there is little confidence that the authors actually achieved true minimum loss. Despite recent advances in computing capability, turbulence modeling remains a shortcoming of modern design tools. As a contribution to overcoming this problem, this paper offers a three-dimensional (3D) view of the developing mean flow, total pressure, and turbulence fields that gave rise to the loss reduction of the airfoil mentioned above. Experiments are conducted in a linear cascade with aspect ratio of 3.5 and Re = 100,000. The results are derived from stereoscopic particle image velocimetry (PIV) and total pressure measurements inside the passage. Overall, the loss reduction correlates strongly with reduced turbulence production. The aim of this paper is to provide readers with a realistic view of mean flow and turbulence development that include all the components of the Reynolds stress tensor to assess, at least qualitatively, the validity of high fidelity computational tools used to calculate turbine flows.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2016;139(2):021007-021007-10. doi:10.1115/1.4034684.

The secondary flow increases the loss and changes the flow incidence in the downstream blade row. To prevent hot gases from entering disk cavities, purge flows are injected into the mainstream in a real aero-engine. The interaction between purge flows and the mainstream usually induces aerodynamic losses. The endwall loss is also affected by shedding wakes and secondary flow from upstream rows. Using a series of eddy-resolving simulations, this paper aims to improve the understanding of the interaction between purge flows, incoming secondary flows along with shedding wakes, and mainstream flows on the endwall within a stator passage. It is found that for a blade with an aspect ratio of 2.2, a purge flow with a 1% leakage rate increases loss generation within the blade passage by around 10%. The incoming wakes and secondary flows increase the loss generation further by around 20%. The purge flow pushes the passage vortex further away from the endwall and increases the exit flow angle deviation. However, the maximum exit flow angle deviation is reduced after introducing incoming wakes and secondary flows. The loss generation rate is calculated using the mean flow kinetic energy equation. Two regions with high loss generation rate are identified within the blade passage: the corner region and the region where passage vortex interacts with the boundary layer on the suction surface. Loss generation rate increases dramatically after the separated boundary layer transitions. Since the endwall flow energizes the boundary layer and triggers earlier transition on the suction surface, the loss generation rate close to the endwall at the trailing edge (TE) is suppressed.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2016;139(2):021008-021008-9. doi:10.1115/1.4034556.

For film cooling of combustor linings and turbine blades, it is critical to be able to accurately model jets-in-crossflow. Current Reynolds-averaged Navier–Stokes (RANS) models often give unsatisfactory predictions in these flows, due in large part to model form error, which cannot be resolved through calibration or tuning of model coefficients. The Boussinesq hypothesis, upon which most two-equation RANS models rely, posits the existence of a non-negative scalar eddy viscosity, which gives a linear relation between the Reynolds stresses and the mean strain rate. This model is rigorously analyzed in the context of a jet-in-crossflow using the high-fidelity large eddy simulation data of Ruiz et al. (2015, “Flow Topologies and Turbulence Scales in a Jet-in-Cross-Flow,” Phys. Fluids, 27(4), p. 045101), as well as RANS k–ϵ results for the same flow. It is shown that the RANS models fail to accurately represent the Reynolds stress anisotropy in the injection hole, along the wall, and on the lee side of the jet. Machine learning methods are developed to provide improved predictions of the Reynolds stress anisotropy in this flow.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2016;139(2):021009-021009-10. doi:10.1115/1.4034685.

Three-dimensional design is central to all modern compressor design systems, but many of these methods still rely on a two-dimensional and sectional view of aerodynamics at their core. This paper argues that this view fundamentally limits design by not considering the effect, on separation and loss, of the pressure gradient on the surface of the blade perpendicular to the meridional direction, here known as the transverse pressure gradient. The first part of the paper details how altering the transverse pressure gradient, by changing a blade's 3D stacking, switches the way in which the blade aerodynamically “fails,” from a open corner separation to a trailing edge separation. It also shows how the transverse pressure gradient significantly changes the blade profile loss. In the second part, the effect of the transverse pressure gradient on the uncertainty inherent in the compressor design space is investigated. It is shown that as blade pitch–chord ratio is raised and the amount of 3D stacking is lowered, the uncertainty of predicting a compressor's operating range is significantly raised. By increasing 3D stacking and the strength of the transverse pressure gradient, it is shown that this uncertainty can be significantly reduced.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In