Session: 32-05 High-Speed LPT

Paper Number: 151856

URANS Investigation of Unsteady Wakes and Purged Hub Cavity Impact on the Aerodynamics of a High Speed Low-Pressure Turbine Cascade 

The low-pressure turbine (LPT) is a critical component in aeroengine applications, as it drives the fan, which, in a modern high-bypass engine, accounts for up to 80% of the total thrust. In modern LPT design, the trend is to increase blade loading, and in geared engine configurations, higher spool rotational velocities are achieved since the LPT is not constrained by the lower speed of the fan. Consequently, LPTs operate at transonic Mach numbers (approximately 0.9) and low Reynolds numbers (on the order of 10^4 to 10^5), where compressibility effects and potential boundary layer transition have a significant impact on the aerodynamic performance of the blades. Accurate numerical prediction of the aerodynamics in these configurations is further complicated by the periodic impingement of wakes from upstream blade rows and the presence of purged inter-disk cavities.
To the authors’ knowledge, this study represents the first to numerically investigate those complex effects on the secondary flows and blade loading of the SPLEEN linear cascade configuration with wake generators and hub cavity. Available experimental investigations were performed by the Von Karman Institute under engine-representative operating conditions. The experimental apparatus is modular, allowing for the installation of a rotating disk equipped with bars to replicate the periodic wakes generated by upstream blade rows. Additionally, a hub cavity may be introduced to account for the interaction between the purge flow and the mainstream flow.
The nominal operating condition under investigation corresponds to a Mach number of 0.9 and a Reynolds number of 70k. The cylindrical bars rotate at a frequency of 5280 Hz and, when the cavity is installed, the purge flow is approximately 1.2% of the mainstream mass flow.
Unsteady Reynolds-Averaged Navier-Stokes (URANS) simulations were performed using elsA, an in-house ONERA-Safran CFD solver. The computational fluid domain extends over one pitch of the linear cascade and includes both the rotor (wake generator) and the stator (turbine blade), with a 1 to 1 periodicity between the two. The numerical results were compared to steady and unsteady measurements provided by the Von Karman Institute, upstream and downstream of the linear cascade, as well as measurements on the blade surface. Two turbulence models were assessed: the k-ω Menter SST, an eddy viscosity reference model, and the SSG/LRR-ω, a Reynolds Stress Model. Previous studies have demonstrated that the latter model provides results in closer agreement with experiments.
The numerical simulations successfully captured the shock wave around the rotating bars, in line with previous Large Eddy Simulations (LES) performed on a 2D slice of the cascade and with experimental visualizations of the flow around cylinders at transonic Mach numbers. The pitch-wise distribution of total pressure downstream of the rotor closely matches experimental measurements, although slight discrepancies in the axial velocity and static pressure distribution were observed. The pitch-wise evolution of total pressure losses at multiple span-wise locations downstream of the cascade also aligns well with the experimental data. Fourier transforms of the pressure fluctuation signals upstream of the cascade and on the blade surface indicate that the simulations accurately capture the bar-passing frequency, though the amplitude of the fluctuations is overestimated.

Presenting Author: Alessandro Schenatti Safran Helicopter Engines

Presenting Author Biography: Alessandro Schenatti is a PhD candidate at Safran Helicopter Engines and ONERA Meudon. The PhD thesis project focuses on the characterisation of the unsteady aerodynamics excitations on turbine blades with particular consideration to the mutual interactions between the peripheral components (rim seal cavities) and the main duct flow, at engine representative conditions of low-pressure and high-pressure turbines. The flow field and frequency content analyses are performed on URANS numerical simulations.
He has receveid the double degree in Aeronautical Engineering from Politecnico di Milano and ISAE Supaero and a research master in fluid dynamics, energy and transfers at ISAE Supaero.

Authors:

Alessandro Schenatti Safran Helicopter Engines
Thomas Bontemps DAAA, ONERA, Institut Polytechnique de Paris
Luis Bernardos DAAA, ONERA, Institut Polytechnique de Paris
Emma Croner Safran Helicopter Engines
Sébastien Le-Guyader Safran Helicopter Engines
Sergio Lavagnoli von Karman Institute for Fluid Dynamics
Emmanuel Laroche DMPE, ONERA, Université de Toulouse