This article discusses the use of friction damper technology in turbine blades. In 1980, Jerry Griffin published an integrated approach for the underplatform friction damper design, utilizing centrifugal loading. The idea was to apply an individual metal piece, which is pressed by the centrifugal load against the platforms of vibrating turbine blades. The dissipation energy is then produced by friction sliding between the vibrating platforms and the pressed damper. Griffin’s findings have opened up friction damping technology, which is now commonly utilized by many Original Equipment Manufacturers in gas and steam turbines. Every year, new publications show the development of sophisticated interdisciplinary knowledge for predicting the nonlinear blade dynamic behavior in the most reliable manner. Friction dampers reduce resonance amplitudes several times with respect to that for sticking contact condition. But they only act efficiently in a narrow frequency range for the resonance of interest. Therefore, other technologies are continuously being developed, based for instance on piezomaterials, which can extend the allowable limits of High Cyclic Fatigue for the conventional blade alloys.
From the beginning of time friction has accompanied mankind, who has empirically applied its nature to create fire for roasting and safety. A little later mankind invented the wheel to reduce frictional drag when carrying goods. Up to date the demand on frictionless motion in rotating machines is an engineering challenge, providing magnetic bearings as a solution. On the other hand, frictional drag is readily utilized for friction clutches, breaks, welding, and friction dampers.
At the end of the 15th century, Leonardo da Vinci had experimentally shown the proportional relationship between frictional drag to external loading as well the independency of friction drag to the contact area. 150 years later, da Vinci's unpublished findings were confirmed by two Amontons's laws (1699), which were verified mathematically as non-linear friction forces by de Coulomb in 1781. Finally, in 1950 a physical interpretation of these laws were explained by Bowden and Tabor, who proved that the true contact area is determined by the asperities at the contact surfaces whose higher numbers come into contact by enlarging the normal force.
In the lifetime assessment, the alternating stress against the mean stress determines the fatigue damage of a mechanical component, as determined with the Haigh's Mean Stress Diagram. Reduction of the alternating stress has a direct impact on the lifetime extension, especially for engines operating with variable rotational speed. With respect to turbine blades, one approach is to decrease flow excitations, which is a challenge since higher demands on turbine performance can be achieved with 3D blading technology. Now blade excitation mechanisms, which are traditionally empirically defined, are being intensively investigated with a growing number of different Fluid-Structure Interaction projects. Another alternative is to increase damping capabilities based on internal material dissipation, aero-damping and frictional dissipation at the blade interfaces. Although material damping increases for higher alternating stresses, its overall influence on suppressing vibration can be neglected in the design process. High aero-damping basically means bigger imperfection in the aerodynamic design of the turbine profile, so that this phenomenon is any key driver for reducing the alternating stress. Then, there is nothing left besides an increase in the frictional dissipation at the blade interfaces.
The first analytical analysis of forced vibrations with friction sliding was performed by den Hartog (1931), who approximated the non-linear friction Coulomb force as an equivalent viscous damping. Earliest theoretical investigations of blade vibrations with friction sliding appeared in the 70s. The major motivation of those works was based upon old design practice, which allowed for adding additional lacing wires or zigzag bolts between blades for reducing resonant amplitudes to an acceptable level. Earles and Williams (1972) extended Hartog's formulation by developing a new linearization concept of the non-linear friction force on a rigid contact (Fig. 1a). For dynamic modeling of tuned and mistuned assemblies, similar models for a single contact point representing a rigid contact coupling were applied by different researchers (Muszynska and Jones, 1978).
Already then, den Hartog's measurement demonstrated hysteresis relationships between the tangential force and tangential contact displacement (Fig. 1c-d), caused by local elastic deformations between contacting bodies. The frictional hysteresis loop can be represented by the resulting elastic stiffness CT for the sticking contact condition (Fig. 1), friction coefficient μ, normal contact load FN and the relative contact oscillation uT, which depends on the excitation load. Concerning periodic vibrations, non-linear friction forces are defined by applying the Harmonic Balance Method either with a one-term or with multi-term harmonics (Fig. 1). The reliability of the numerical solution increases by applying more Fourier harmonics in the linearization of the non-linear friction hysteresis loop. In this modeling, the friction coefficient is a physical input depending on the material pair in contact, and the contact stiffness can be either measured experimentally or computed with the Finite Element Method (FEM).
The centrifugal loading acting on the blade induces high contact stresses in the rotor groove, which generate sticking contact conditions. Also, shroud or winglet connections require the reliable sticking contact condition, which integrates all rotating airfoils into the disc assembly in service. In 1980, Jerry Griffin published an integrated approach for the under-platform friction damper design, utilizing this centrifugal loading. His idea was to apply an individual metal piece, which is pressed by the centrifugal load against the platforms of vibrating turbine blades (Fig. 2). The dissipation energy is then produced by friction sliding between the vibrating platforms and the pressed damper. For a real gas turbine blade, his experimental data confirmed well the numerical results, which were calculated with the macro-slip approach (Fig. 1b). Griffin identified that the tangential contact stiffness CT is a key parameter in the damper optimization (Fig. 2).
Besides a good understanding of blade mechanics, the design process for the friction airfoil damper requires proper numerical tools solving non-linear dynamic equations, as well the measured frictional coefficients and contact stiffness at different temperatures. Therefore, several universities have followed Griffin's idea and developed numerical tools considering micro-slip, roughness of the contact surface, 3-dimensional motion of the damper including sliding, sticking and open contact modes on different forms of dampers as well shrouds, winglet, zigzag and wire couplings. Until now, these methodologies have not been implemented in commercial FEM suits. Close collaborations have been ongoing between different OEM (Original Equipment Manufacturer) and academia to assure further progress of this technology.
Griffin's findings have opened up friction damping technology, which is now commonly utilized by many OEMs in gas and steam turbines. A huge number of research papers and patents have been published during the last 30 years. Every year, new publications show the development of sophisticated inter-disciplinary knowledge for predicting the non-linear blade dynamic behavior in the most reliable manner. Friction dampers reduce resonance amplitudes several times with respect to that for sticking contact condition. But they only act efficiently in a narrow frequency range for the resonance of interest. Therefore, other technologies are continuously being developed, based for instance on piezo-materials, which can extend the allowable limits of High Cyclic Fatigue for the conventional blade alloys.
Gas turbine users to meet in houston, oct 4-7, 2010
Focused on the daily operational and managerial challenges faced by turbomachinery users, the ASME Gas Turbine Users Symposium (GTUS) is a unique forum for current, practical and interactive exchange among gas turbine users, manufacturers and consultants.
The symposium offers users a great opportunity to learn about the most recent industry best practices, new technology and developments arranged in a series of discussions, panels and tutorials, an ideal setting for the exchange of information with other industry practitioners.
Registrants may expect to network and exchange the latest in technology and best practices with experts from across the spectrum of gas turbine design, application and field proficiency. Program topics emphasize operations and maintenance, as well as gas turbine design, advances and environmental issues.
Preceding the symposium, IGTI will also present three, one-day workshops offering in-depth, fundamental training.
In addition, GTUS will again co-locate with Texas A&M's Turbomachinery Symposium, to be held Oct. 4-7, 2010, at the George R. Brown Convention Center in Houston. Thus, GTUS registration not only provides admission to all GTUS sessions but also includes free access to the 39th Turbomachinery Symposium exhibit floor, complimentary lunches and evening meals, gas turbine users networking, and an invitation to the Turbomachinery welcome address. For a nominal upgrade fee, delegates can be registered for both GTUS and the 39th Turbomachinery Symposium.
Visit www.asmeconferences.org/gtus10 for more details and the latest updates on GTUS 2010.
Why YOU Should Attend GTUS 2010
39th Turbomachinery Symposium Exhibition
High Quality Presentations from Experts
In this challenging economy, get the most for your training/education/conference budget by attending IGTI training workshops, the ASME Gas Turbine Users Symposium and the 39th Turbomachinery Symposium, all together in Houston, October 4-7, 2010.