Flutter is an aeroelastic instability, where the vibration of a mode shape of an aerodynamic body such as a blade row is excited by the unsteady aerodynamic forces induced by that vibration. In the absence of sufficient mechanical damping, this positive feedback mechanism causes an exponential growth in vibration amplitude which can quickly lead to component failure due to material fatigue.
The mode shapes of fans and compressors most susceptible to flutter are those with the lowest natural frequencies, typically the fundamental bending (or “flap”, 1F) and torsion mode (1T). These mode shapes generally have different natural frequencies and very little aeroelastic coupling, due to the large mass ratio, and can therefore be considered separately for aeroelastic analysis (unlike in wings, whose flutter stability is largely determined by combined bending-torsion-flutter). However, because of the three-dimensional blade design and the strong annulus contraction, the fundamental bending mode shape of
modern fan and compressor blades usually has a more or less pronounced twist component, meaning that the blade sections move not only perpendicular to their chord (plunge motion) but also rotate about their mid-chord axis (twist motion), as in figure 1. It is known that this twist component adversely affects the aeroelastic stability of the bending mode, which, for attached subsonic flow, is otherwise inherently stable. In addition, the blades in a blade row are coupled mechanically (through the disc) and aerodynamically (through potential flow effects as well as the change in passage area). This leads to vibration patterns with nodal diameters, in which all blades vibrate in the same mode shape, but with a phase-lag called inter-blade phase angle (IBPA), as in figure 2. At appropriate IBPAs, a mode shape which would be aerodynamically damped for an isolated blade, may become unstable.
In this research project, which is conducted in cooperation with the Imperial College London (ICL), the effects of a variable twist component in a bending mode shape on the flutter stability of a linear compressor cascade at different reduced frequencies are to be investigated experimentally (ILR/TU Berlin) and numerically (ICL). The goals of the project are to develop aeroelastic design rules for avoiding flutter and to provide a set of measurement data for reference cases, which can be used to validate aeroelastic simulation codes. For this purpose, a vibrating blade module for the aeroelastic wind tunnel at ILR is being developed and manufactured, which allows implementation of a prescribed blade motion with variable twist component (figure 3).
The aeroelastic stability of the cascade over the full IBPA range can then be assessed employing the aerodynamic influence coefficient (AIC) method, based on the unsteady pressure distribution measured on the surface of the vibrating blade and its neighbours.
Person of contact: Dipl.-Ing. Julian Gambel