0
Research Papers: Gas Turbines: Structures and Dynamics

Numerical and Experimental Damping Assessment of a Thin-Walled Friction Damper in the Rotating Setup With High Pressure Turbine Blades

[+] Author and Article Information
J. Szwedowicz

 ABB Turbo Systems Ltd., Bruggerstreet 71a, CH-5401 Baden, Switzerlandjaroslaw.szwedowicz@ch.abb.com

C. Gibert

 Ecole Centrale de Lyon, CNRS/LTDS UMR 5513, 36 Avenue Guy de Collongue, F-69134 Ecully Cedex, Franceclaude.gibert@ec-lyon.fr

T. P. Sommer

 ALSTOM (Switzerland) Ltd., Power Turbo-Systems, Brown Boveri Street 7, CH-5401 Baden, Switzerlandthomas.sommer@power.alstom.com

R. Kellerer

 ALSTOM (Switzerland) Ltd., Power Turbo-Systems, Brown Boveri Street 7, CH-5401 Baden, Switzerlandrudolf.kellerer@power.alstom.com

www.energy.kth.se/adturb2.

J. Eng. Gas Turbines Power 130(1), 012502 (Dec 18, 2007) (10 pages) doi:10.1115/1.2771240 History: Received June 06, 2006; Revised May 29, 2007; Published December 18, 2007

Underplatform friction dampers are possible solutions for minimizing vibrations of rotating turbine blades. Solid dampers, characterized by their compact dimensions, are frequently used in real applications and often appear in patents in different forms. A different type of the friction damper is a thin-walled structure, which has larger dimensions and smaller contact stresses on a wider contact area in relation to the solid damper. The damping performance of a thin-walled damper, mounted under the platforms of two rotating, freestanding high pressure turbine blades, is investigated numerically and experimentally in this paper. The tangential and normal contact stiffness that are crucial parameters in optimal design of any friction damper are determined from three-dimensional finite element computations of the contact behavior of the damper on the platform including friction and centrifugal effects. The computed contact stiffness values are applied to nonlinear dynamic simulations of the analyzed blades coupled by the friction damper of a specified mass. These numerical analyses are performed in the modal frequency domain, which is based on the harmonic balance method for the complex linearization of friction forces. The numerical dynamic results are in good agreement with the measured data of the real mistuned system. In the analyzed excitation range, the numerical performance curve of the thin-walled damper is obtained within the scatter band of the experimental results. For the known friction coefficients and available finite element and harmonic balance tools, the described numerical process confirms its usability in the design process of turbine blades with underplatform dampers.

Copyright © 2008 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Thin-walled damper assembly, (a) thin-walled damper, (b) assembled blade damper system

Grahic Jump Location
Figure 2

Rotating blade setup in vacuum conditions

Grahic Jump Location
Figure 3

Schematic of the piezoelectric shaker

Grahic Jump Location
Figure 4

Instrumentation of the rotating test rig for blades with friction dampers

Grahic Jump Location
Figure 5

Experimental undamped responses of OOP first flexural mode for excitation levels of 1V, 10V, 24V, and 35V at speed no

Grahic Jump Location
Figure 6

Experimental damped responses of OOP first flexural mode for excitation levels of 1V10V, 20V, 30V, and 35V at speed no, where response differences between blades 1 and 2 are due to the asymmetrical contact conditions (Fig. 8)

Grahic Jump Location
Figure 7

Specimen (a), 2D FE model (b) and the comparison of the calculated contact stiffness for friction coefficients of 0.5 and 0.7 with the experimental data (c) measured at the ALSTOM setup for quasistatic tests

Grahic Jump Location
Figure 8

von Mises centrifugal stress (a) and the contact normal stress (b) computed with the coarse parabolic FE model of two blades with the damper

Grahic Jump Location
Figure 9

(a) A 3D parabolic FE model exploiting the symmetry of the damper configuration and (b) the measured contact area compared to ((c) and (d)) the FE contact normal pressure, where σn,O and nO are normalized normal contact stress and rotational speed, respectively

Grahic Jump Location
Figure 10

Scheme of the virtual accelerations acting on the damper for its stiffness assessment in the normal n (a) and tangential t (b) direction, where C is the center gravity of the damper and k means the node in contact

Grahic Jump Location
Figure 11

The determination of the linearized stiffness ki from the nodal deformation ui and reaction force Ri at node in contact for different increments i of the nonlinear FE solution of the contact problem

Grahic Jump Location
Figure 12

The resulting normal contact stiffness kn and global normal stiffness cn of the thin-walled damper in terms of the centrifugal load Fn

Grahic Jump Location
Figure 13

The FE fundamental bending mode shape of interest of the blades with shaker in the rotating setup, where green color denotes blade at standstill

Grahic Jump Location
Figure 14

The experimental damping ratio of the blade in the vacuum condition with the PZT shaker and without damper at 20°C and constant speed nO

Grahic Jump Location
Figure 15

The numerical eigenfrequencies of the rigidly clamped blade and the measured resonance frequencies of pair blades coupled weakly by the disk

Grahic Jump Location
Figure 16

The numerical resonance amplitudes of the tuned blade and experimental resonance amplitudes of two mistuned blades without damper measured at the ECL setup in the vacuum condition at 20°C and speed nO

Grahic Jump Location
Figure 17

(a) FE contact normal stresses on the damper whose the contact area is represented in the nonlinear analysis by a finite number η×ζ of contact points on the rectangular area H×Z considering (b) the technical pressure distribution

Grahic Jump Location
Figure 18

Numerical eigenfrequencies of two tuned blades with sticking underplatform damper for the validation of the damper stiffness properties with respect to the measured resonance frequencies with very low excitation at speed nO and 20°C

Grahic Jump Location
Figure 19

The measured resonance response functions of the mistuned two blades ((a) and (b)) with the damper compared to the numerical results of the two tuned blades with damper for μ=0.1 (c) and μ=0.2 (d), where d means the experimental damping ratio of the blades without damper

Grahic Jump Location
Figure 20

The experimental performance curves of the two mistuned blades with the thin-walled damper compared to the numerical ones of the tuned blades obtained for friction coefficients μ of 0.1 and 0.2

Grahic Jump Location
Figure 21

Illustration of the damping capabilities of the thin-walled damper for the analyzed blades, where “mi” denotes friction coefficient

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

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