In this paper, we investigate non-synchronous vibrations (NSV) in turbomachinery, an aeromechanic phenomenon in which rotor blades are driven by a fluid dynamic instability. Unlike flutter, a self-excited vibration in which vibrating rotor blades and the resulting unsteady aerodynamic forces are mutually reinforcing, NSV is primarily a fluid dynamic instability that can cause large amplitude vibrations if the natural frequency of the instability is near the natural frequency of the rotor blade. In this paper, we present both experimental and computational data. Experimental data was obtained from a full size compressor rig where the instrumentation consisted of blade-mounted strain gages and case-mounted unsteady pressure transducers. The computational simulation used a three-dimensional Reynolds averaged Navier-Stokes (RANS) time accurate flow solver. The computational results suggest that the primary flow features of NSV are a coupled suction side vortex shedding and a tip flow instability. The simulation predicts a fluid dynamic instability frequency that is in reasonable agreement with the experimentally measured value.
Fluctuating wall pressure measurements have been made in a separated shock wave/turbulent boundary-layer interaction produced by an unswept compression corner in a Mach 5 flow. Wheeler doublet vortex generators were placed upstream of the corner to study their effect on the fluctuating pressure loads within the interaction. The vortex generators produce significant three dimensionality in an otherwise two-dimensional interaction. They reduce the upstream influence and the length of the region of separation shock motion by 60 % and 64 %, respectively, and they decrease the maximum wall pressure rms near separation by 23% and shift the fluctuations to higher frequencies. The maximum fraction of energy in the resonant frequency band of typical skin panels (i.e., 100-500 Hz) is decreased by 11%. On the corner face, where the maximum loads occur, the maximum wall pressure rms is reduced by 10%, and the maximum fraction of energy in the 100-500 Hz band is reduced by 35%. These changes are due to a fuller boundary-layer profile, a weaker separation shock, increased boundary-layer turbulence causing increased separation shock jitter, and a shorter region of shock motion. Nomenclature Cf= skin friction coefficient / = frequency f c = separation shock zero crossing frequency G(f) = spectral density function LI = intermittent region length P = pressure Reg = Reynolds number based on momentum thickness T = temperature, separation shock period U = velocity X = stream wise distance normalized by 8, Fig. 1 Y = span wise distance normalized by 8, Fig. 1 Z = vertical distance normalized by 8, Fig. 1 ft = fraction of wall pressure variance in the 100-500 Hz band 8 = undisturbed boundary-layer thickness 8* = undisturbed boundary-layer displacement thickness Y = wall pressure signal intermittency 0 = undisturbed boundary-layer momentum thickness O = undisturbed boundary-layer wake strength parameter a -standard deviation Subscripts max = maximum value pw = wall pressure rms = root mean square value t = pitot pressure vg = vortex generator w = wall oo, inf = freestream condition 0 = stagnation quantity
The unsteady aero-dynamics of a single-stage high-pressure turbine blade operating at design corrected conditions has been the subject of a thorough study involving detailed measurements and computations. The experimental configuration consisted of a single-stage high-pressure turbine and the adjacent, downstream, low-pressure turbine nozzle row. All three blade-rows were instrumented at three spanwise locations with flush-mounted, high-frequency response pressure transducers. The rotor was also instrumented with the same transducers on the blade tip and platform and the stationary shroud was instrumented with pressure transducers at specific locations above the rotating blade. Predictions of the time-dependent flow field around the rotor were obtained using MSU-TURBO, a three-dimensional (3D), nonlinear, computational fluid dynamics (CFD) code. Using an isolated blade-row unsteady analysis method, the unsteady surface pressure for the high-pressure turbine rotor due to the upstream high-pressure turbine nozzle was calculated. The predicted unsteady pressure on the rotor surface was compared to the measurements at selected spanwise locations on the blade, in the recessed cavity, and on the shroud. The rig and computational models included a flat and recessed blade tip geometry and were used for the comparisons presented in the paper. Comparisons of the measured and predicted static pressure loading on the blade surface show excellent correlation from both a time-average and time-accurate standpoint. This paper concentrates on the tip and shroud comparisons between the experiments and the predictions and these results also show good correlation with the time-resolved data. These data comparisons provide confidence in the CFD modeling and its ability to capture unsteady flow physics on the blade surface, in the flat and recessed tip regions of the blade, and on the stationary shroud.
Aerodynamic measurements were acquired on a modern single-stage, transonic, high-pressure turbine with the adjacent low-pressure turbine vane row (a typical civilian one and one-half stage turbine rig) to observe the effects of low-pressure turbine vane clocking on overall turbine performance. The turbine rig (loosely referred to in this paper as the stage) was operated at design corrected conditions using the Ohio State University Gas Turbine Laboratory Turbine Test Facility. The research program utilized uncooled hardware in which all three airfoils were heavily instrumented at multiple spans to develop a full clocking dataset. The low-pressure turbine vane row (LPTV) was clocked relative to the high-pressure turbine vane row (HPTV). Various methods were used to evaluate the influence of clocking on the aeroperformance (efficiency) and the aerodynamics (pressure loading) of the LPTV, including time-resolved and time-averaged measurements. A change in overall efficiency of approximately 2–3% due to clocking effects is demonstrated and could be observed using a variety of independent methods. Maximum efficiency is obtained when the time-average surface pressures are highest on the LPTV and the time-resolved surface pressure (both in the time domain and frequency domain) show the least amount of variation. The overall effect is obtained by integrating over the entire airfoil, as the three-dimensional (3D) effects on the LPTV surface are significant. This experimental data set validates several computational research efforts that suggested wake migration is the primary reason for the perceived effectiveness of vane clocking. The suggestion that wake migration is the dominate mechanism in generating the clocking effect is also consistent with anecdotal evidence that fully cooled engine rigs do not see a great deal of clocking effect. This is consistent since the additional disturbances induced by the cooling flows and∕or the combustor make it extremely difficult to find an alignment for the LPTV given the strong 3D nature of modern high-pressure turbine flows.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.