Comparison of Experimental and Theoretical Results for Unsteady Transonic Cascade Flow at Design and Off-Design Conditions

Abstract

This paper presents experimental and theoretical results for turbine cascades performing harmonic oscillations in transonic flow at design and off-design conditions. The experimental investigations were performed in an annular test facility where unsteady blade pressures were measured in two different test cascades, one operating at the nominal inlet flow angle, the other at an incidence angle exceeding the normal value by more than 20 degrees. The corresponding theoretical results were computed with a 2D Euler code which makes use of flux vector splitting in combination with a time-dependent grid generation. The present data were all obtained for tuned bending modes where the blades performed heaving oscillations with the same frequency and amplitude, but with a constant interblade phase angle. For the cascade operating at design conditions, the steady flow was purely subsonic. The other test cascade was run in transonic flow, and a normal shock appeared on the rear part of the blade’s suction surface. It was found that measured unsteady pressure and damping coefficients are well reproduced by the computed results for the first test cascade. In the case of steady off-design flow (the second test cascade), significant differences between experimental and theoretical results are observed.

Viscous and Inviscid Linear/Nonlinear Calculations Versus Quasi 3D Experimental Cascade Data for a New Aeroelastic Turbine Standard Configuration

“…The linear Euler code and the viscous code position the shock too far downstream, but with a better agreement of the viscous results. This was also found by Carstens [1993] and Gruber [1996]. The nonlinear Euler code predicts the shock upstream of the measured position.…”

“…Also single blade vibration data were collected to obtain the influence coefficients. More detailed information of the test facility can be found in Bolcs et al [1983Bolcs et al [ , 1986Bolcs et al [ , 1991Bolcs et al [ , 1993.…”

“…Several unsteady prediction models for flutter at attached flow conditions have appeared in the open literature (and as confidential design methods in the industry) over the last decade (see for example Verdon and Caspar, 1984;Whitehead and Newton 1985;Fransson and Pandolfi, 1986;Smith, 1989;Hall et al, 1989;He, 1989;Whitehead, 1990;Sid6n, 1991;Holmes et al, 1991;Huff et al, 1991;Carstens at al., 1993;Giles et al, 1993;Kahl et al, 1993;Gerolymos et al, 1994;Groth et al, 1996;Gruber et al, 1996). Several of these models show good to excellent agreement against selected test cases of mainly twodimensional nature.…”

“…Some years ago extensive experiments on a two-dimensional section of a vibrating last-stage gas turbine blade (54% span) were performed in the annular test rig at the EPF-Lausanne [Bolcs et al, 1991[Bolcs et al, , 1993. Many experiments have also been performed on the same geometry in the linear test facility at the EPFL [Norryd, 1997;Ott et al, 1998].…”

“…Many experiments have also been performed on the same geometry in the linear test facility at the EPFL [Norryd, 1997;Ott et al, 1998]. This substantial database, especially the one from the annular test facility, has been used by a few researchers for code validation over the years [Bolcs et al, 1991;Carstens et al, 1993;Leyland et al, 1994;Gruber et al, 1996]. However, the blade geometry has not been available to the larger research community till presently.…”