Figure 1: The complex series of issues which comprise Forced Response
Carnegie Mellon University
Jerry H. Griffin, Professor, Mechanical Engineering
Abstract
The GUIde Consortium, a joint government, university, and industrial program to fund research in turbomachinery bladed disk forced response, is described. The development of the Consortium was prompted by the fact that excessive vibration of bladed disks in fans, compressors, and turbines is an industry-wide problem that impacts turbine engine development and reliability. This paper describes the forces behind the development of the Consortium, its structure, including the costs and benefits to each of the participating organizations, and a reflection of the lessons learned from the open interaction between the GUIde members. Finally, the future of the organization is discussed with a look at the evolution of the program to face the changes in technical and financial priorities within the industry.
The GUIde Consortium started in 1991 when a number of companies joined with Carnegie Mellon University and Purdue University to form a partnership that would result in improved technology for dealing with the forced vibration of bladed disks. The decision to form the Consortium was the result of two workshops [Griffin, 1989; Griffin, 1990] that were attended by industry, government, and university representatives, The name "GUIde" was selected for the Consortium. The acronym GUIde stands for Government agencies, Universities, and Industry working together to achieve a specific goal. The word "Guide" also alludes to the idea that by cooperating these organizations can develop an enhanced view of how to achieve their objective and, consequently, serve as guides to each other in obtaining their mutual goal.
The development of the GUIde Consortium on Forced Response of Bladed Disks was prompted by the fact that excessive vibration of bladed disks is an industry-wide problem that impacts engine development, reliability, efficiency, and fleet readiness. A review by the GUIde industrial members revealed that approximately 14% of the unplanned redesigns that occurred during engine development were prompted by excessive vibration. Practically, an average development engine experiences four vibration problems, two of which are considered serious. Serious vibration problems are those that cost on the order of $5M to resolve. Blade vibration problems that develop later in the product cycle may result in significant maintenance costs. For example, one manufacturer reported that vibration led to a failure rate as high as 50 incidents per year in a mature, fielded engine. It has been estimated that an 80% reduction in high cycle fatigue failures would reduce maintenance costs by over $100M per year in the US alone. Excessive vibration also affects engine
|
|
Figure 1: The complex series of issues which comprise Forced Response |
efficiency because the thickness of fan and compressor airfoils are limited by vibration considerations. If thin airfoils could be designed that were more durable, then fan and compressor efficiency could be increased by as much as 2%. Lastly, vibration problems can affect fleet readiness. Blades failing from vibration induced, high cycle fatigue often cause catastrophic engine failures. When this occurs it is sometimes necessary to ground the fleet until corrective action is taken.
The research areas which impact forced response are illustrated in Figure 1. The turbine engine industry maintains complex design system programs which attempt to predict the vibration levels of compressor and turbine bladed disks in order to determine operating regions which might lead to premature blade failure. Technical efforts in these areas are expensive and individual companies are only able to pursue limited programs to improve these design systems. The Consortium allows the turbomachinery companies to leverage their research money approximately 10:1 and attack the issues over a broader front. In fact, if other related research activities that the GUIde members have access to are taken into account, then the leverage on the companies' contributions is closer to 25.1.
The GUIde Consortium provides a structure through which government agencies, universities and industry can contribute to a coordinated research program. The GUIde concept is: The scope of the research program should be focused; technical experts from the government, universities, and industry should jointly set research directions; technology transfer should be enhanced by collaborations instituted through the Consortium; and Consortium-sponsored research should be carefully integrated with government and industrial research programs. An important goal of the Consortium is to surpass the technical capability of foreign owned manufacturers of turbomachinery in producing -durable products. As a result, membership in GUIde is restricted to US companies.
This paper will provide an overview of the GUIde Consortium beginning with an outline of the organizational structure and a description of the various players and their roles. A history of the Consortium is then presented to show the factors leading up to the formation of the group and as a guide to others who might wish to develop similar programs. Finally, the lessons learned from participation in the Consortium are given as well as a look at its evolution and future plans.
The approach for achieving the Consortium objectives has been to develop a nationally coordinated research and development
 |
|
Figure 2: Flow diagram of GUIde Consortium |
effort that involves industry, government agencies, and universities. A diagram indicating the current structure of the Consortium is shown as Figure 2. The role of each element is:
Consortium: The GUIde management structure consists of two parts, the center directors and the Steering Committee.
Consortium centers have been established at Carnegie Mellon University (CMU) in Pittsburgh, Pennsylvania and at Purdue University in- West Lafayette, Indiana. The Consortium center director at CMU is Professor J. Griffin and at Purdue University is Professor S. Fleeter. Professors Griffin and Fleeter are responsible for coordinating Consortium research activities in the structural and fluid mechanics areas, respectively, and work together with the Steering Committee in managing the various projects of the Consortium.
Technical guidance to the Consortium is provided by the Steering Committee which consists of one voting member from each of the six participating engine companies and one each from the Air Force and NASA. The members are experts in engine aerodynamic and structural problems and have actively participated in the formation of the Consortium, in setting its objectives, and determining the overall program. The Steering Committee sets the research direction, evaluates proposals, and determines the level of funding which is allocated to each research project.
Air Force Wright Laboratory: The Air Force Aero Propulsion and Power Directorate (APPD) at Wright Patterson Air Force Base participates in the following three ways:
A number of representatives from APPD attend the Steering Committee meetings. Their primary representative is Mr. T. Fecke, who serves on the Steering Committee. Consequently, the Air Force's interests are represented in planning exercises and their research program in forced response is closely integrated into the GUIde plan.
As part of its program of forced response research, the Air Force has a R&D effort which will provide benchmark forced response data for use by all members of the GUIde Consortium. The project, the Augmented Damping of Low Aspect Ratio Fans or ADLARF, is a modem transonic fan design which has been tested at the Compressor Research Facility at Wright-Patterson AFB. This rig test, which has included on-rotor unsteady pressure and strain gauge measurements and flow field laser velocimetry, has been performed with a series of inlet distortion screens Russler et al., 1995]. A unique aspect of this program is that the Air Force has required the prime contractor (General Electric) to make all data on the stage available to other Consortium members. Since the complete blade geometry and flow conditions are Available, the velocity and unsteady pressure measurements and vibratory response data can be used by all members as benchmark data to check their unsteady aerodynamic and structural response codes. As a result, unlike in the past, these multimillion dollar tests will benefit all the companies in the Consortium, not just the prime contractor.
National Aeronautics and Space Administration: NASA/Lewis Research Center (LeRC) actively participates in the GUIde Consortium as follows.
NASA engineers, G. Stefko (structures) and D. Hoyniak (fluids) regularly attend the Steering Committee meetings and determine NASA's vote in matters of policy. Mr. Stefko's group is responsible for structural research projects in forced response and the group that Dr. Hoyniak represents is responsible for unsteady aerodynamic projects associated with forced response. Mr. Stefko and Dr. Hoyniak regularly brief the Steering Committee on the research efforts in forced response at NASA/LeRC. Consequently, the NASA/LeRC efforts in forced response are closely tied to the GUIde effort. For example, NASA/LeRC has research grants with two of the researchers who are also supported by the GUIde Consortium. In these cases, the NASA supported research tends to be more basic in nature, while GUIde funded activities tend to focus on developing those basic research concepts into practical design tools that can be used by industry.
Vibration is also a problem in rocket engines and advanced propulsion systems of interest to NASA. Consequently, like industry, NASA needs to be able to predict blade vibration. NASA/LeRC has an internal research effort, headed by D. Murthy, directed towards establishing a Forced Response Prediction System (FREPS) [Murthy and Morel, 1993]. The projects sponsored by GUIde complement FREPS and will be integrated into the program as the various research results and design tools become available. This work is especially valuable to the Consortium for two reasons: it provides an example of a forced response design system which may be openly examined and discussed by the researchers and all members of the Consortium (the sponsoring companies design systems are proprietary), and, because it is being developed by the government, FREPS will be generally available to industry in the US A previous GUIde meeting at NASA/LeRC has included a workshop which gave the researchers and the Steering Committee a "hands on" appreciation of the FREPS system.
Dr. Hoyniak works in the Turbomachinery Technology Group (Propulsion Systems Division) at NASA/LeRC that is responsible for developing steady and unsteady aerodynamic codes for use in propulsion system analysis. In addition to serving as a working member of the Steering Committee, Dr. Hoyniak has given presentations to the Consortium members that provide an overview of the analytical and experimental activities in aerodynamics at NASA/LeRC. One of the current research projects funded through the Consortium is a joint research project with NASA/LeRC on unsteady aerodynamics.
Of special interest is a cooperative effort between Pratt & Whitney and NASA/LeRC to generate benchmark unsteady aerodynamic database for airfoils representative of advanced fan configurations. The effort is executed under a "Space Act Agreement" in which Pratt & Whitney has designed, fabricated, and instrumented a set of airfoils to be tested in NASA's Oscillating Cascade test facility. The experiment will provide a valuable database for transonic stall flows at a wide range of incidence angles. Pratt & Whitney has agreed to share this data with the Consortium at no cost.
Universities and Research Institutions: Six research projects are supported through GUIde. Researchers in the areas of unsteady aerodynamics and structural response were asked to propose research to address the development of analytical models, supporting experimental data, design tools, and design strategies for controlling forced response. The relative merits of the proposed research projects were determined by the Steering Committee at a two day evaluation meeting and an integrated research program was developed that included six of these projects. Two of the projects were initiated in 1991 using funds provided completely by industry, while the other four began in 1992 with funding provided through an Air Force contract. Five of these projects are scheduled to last approximately four years. The projects and their respective principal investigators are listed below:
Project 1:Design Tools for Predicting the Forced Response of Mistuned Bladed Disks. C. Pierre and J. MacBain, University of Michigan.
Project 2: Tip Mode Vibration in Low Aspect Ratio Blading. J. Griffin, Carnegie Mellon University.
Project 3: Development of a Computer Aided Environment for the Design of Friction Dampers and Shrouds. C. Menq, Ohio State University.
Project 4: Numerical Unsteady Aerodynamic Simulator for Blade Forced Response Phenomena. J. Verdon, United Technologies Re search Center.
Project 5: High Mach Number Wake Forcing Functions. S. Fleeter, Purdue University.
Project 6: Forced Response Unsteady Aerodynamics of Impeller Blades. S. Fleeter, Purdue University
Carnegie Mellon University serves as the contracting agency for each of the research programs and manages these contracts as described below. Researchers are free to publish the results of their research efforts and are not restricted from distributing the products of their Consortium contracts (primarily computer programs and experimental data). However, the Air Force is not allowed to disseminate the products of the four contracts it funds to companies outside the GUIde membership for a period of two years following the expiration of the GUIde contract (currently 1996). Maintaining ownership of the products of their research allows the researchers to charge a lower fee and allows the Consortium to fund a larger number of projects. Through the required annual review of the research efforts, there is increased interaction between the industrial representatives and the researchers which helps to guide the research towards real industry needs and speed the transition of developing technology into the company design systems. Researchers are also encouraged to submit draft copies of externally published technical papers to the members of the Steering Committee for review prior to publication. This review is simply to ensure that there is no release of company proprietary information.
Industry: The industrial members of GUIde support its activities in a number of important ways.
Since its inception, the industrial participants have played a key role in organizing the Consortium, setting objectives, and determining the overall program of the GUIde Consortium. A key benefit of the organization is the involvement of those people who are actually responsible for integrating the technology developed through the Consortium into their individual company design systems.
The industrial members also participate in the formal, annual review of all research projects. In addition, company representatives serve on smaller subcommittees that monitor each individual research project to insure their relevance through use of test cases of real hardware and comparisons with actual component test data.
Funding for the Consortium comes from a number of sources as depicted in Figure 3; funding for the four year life of the Consortium totals nearly $2.5M. Roughly 40% of the overall Consortium funding is
 |
|
Figure
3: Funding sources for the GUIde Consortium.
|
provided by the Air Force through a contract with the GUIde Consortium entitled Forced Response Consortium Initiative. This contract, along with a $50K per year contribution by NASA/LeRC, is used to fund four of the research projects that are being conducted at universities and research laboratories, Each industrial member company contributes $50K per year the total of which represents approximately half the overall GUIde funding. These contributions are used to fund centers at CMU and Purdue, to fund two of the six research projects, and to cost share the Air Force contract activities. The final contribution to the funding is the cost share by CMU.
Carnegie Mellon University is the legal entity that forms contractual agreements with the industrial partners, governmental funding agencies, and other participating research institutions. The privileges and responsibilities of the companies that join the Consortium have been clearly defined in an agreement with CMU. The center at Purdue University has been established through a contract with CMU and other institutions that are given research contracts by the Consortium become subcontractors to CMU. In addition,CMU has also been responsible for generating and submitting funding proposals to various government agencies on behalf of the Consortium.
This section presents the history of the Consortium as a potential model for other organizations interested in developing similar programs. In the late 1980's, R. Kielb, while at NASA/LeRC, wanted to organize a consortium of universities and NASA to address research issues in the area of structural response. He discussed the idea with T. Fecke at APPD to determine the Air Force's level of interest in the project and the possibility of funding. At this time there had been a reduction in Air Force funding for forced response while the turbomachinery industry faced a rising level of vibration problems in demonstrator and production engines. In order for the Air Force's APPI) to support the project, it needed to be tied to the Integrated High Performance Turbine Engine Technology (IHPTET) program which meant that it must include industrial participation.
During the same time, the second author found that while there was a general need for work in the area, there was insufficient funding available to support the work on an individual company basis. The common research needs and the restrictive funding prospects led to the idea of pooling resources between companies and pursuing -a combined research path. While on a visit to Wright-Patterson AFB, J. Griffin and T. Fecke discussed the consortium idea where it became clear that such an organization should include elements from government, industry, and the universities. Additionally, it became clear that any research into the forced response issue must also include the definition of the aerodynamic forcing functions. S. Fleeter at Purdue University was asked to be the focal point in this area.
A workshop [Griffin, 1989] was held in November, 1989 to bring together the interested parties in industry, NASA, and the Air Force to determine the research needs in the area of bladed disk response and the potential benefits of a consortium in this area. This meeting included representatives from General Electric, Pratt & Whitney, Allison, Allied Signal, Textron-Lycoming, Teledyne, Williams International, Westinghouse Electric, Rocketdyne, and Solar Turbine, some of whom decided not to participate in the final GUIde Consortium effort. Once the need had been established, a second workshop (Griffin, 1990] was held in March, 1990 to define the consortium structure and to define the Research Agreement between CMU and the industrial partners. In defining the financial structure of the Consortium,
one difficulty which arose was the inability of the industrial members to commit funding for more than a single year at a time. The agreement was therefore written which allowed any industrial company to drop out of the Consortium at any time but which would cause forfeiture of any rights to the GUIde products. This second workshop also included presentations by some of the researchers interested in proposing efforts for funding by the Consortium. These presentations gave the company representatives examples of some of the research available to GUIde members, giving them specific cases of potential technical improvements which they could relate to upper-level managers. This would allow management to properly assess the payoff of the Consortium to their respective companies. The final list of participating companies includes: Allied-Signal, Allison, General Electric, Pratt & Whitney, Textron-Lycoming, and Westinghouse Electric.
In late 1990 a Request For Proposals (RFP) was distributed to a number of nationally known researchers in the areas of unsteady aerodynamics and structural response. They were asked to propose research that addressed the development of analytical models, supporting experimental data, design tools and design strategies for controlling forced response. In response to the RFP, the Consortium received eighteen proposals which were evaluated by the Steering Committee. The relative merits of the proposed research were determined by the Committee at a two day evaluation meeting in early 1991 and an integrated research program was developed that included six essential research projects. Of these projects, two were initiated in 1991 using funds provided completely by the industrial dues. The other four projects were incorporated into a proposal to the Air Force Aero Propulsion and Power Directorate entitled the Forced Response Consortium Initiative. These four projects were selected based on both financial considerations and in an effort to tailor the proposal toward specific Air Force needs. This Consortium proposal was selected for funding by the Air Force and the kick-off meeting for the contract and the final four projects was held in August, 1992.
Since the initiation of the Air Force contract the Consortium has met twice each year. Reviews of the funded research efforts are held on an annual basis and meetings of the. Steering Committee are held during these reviews as well as at the six-month intervals between reviews. Planning for the follow-on GUIde I I Consortium is currently underway with this effort discussed more fully in a later section of this paper.
Technology transfer in the GUIde Consortium occurs through regular processes such as the annual views and periodic progress reports on each research activity. In addition, subcommittees follow each project and closely interact with the researchers. Subcommittees members provide researchers with data on specific examples of the types of problems that need to be solved. For example, they have supplied the detailed geometry, analytical and test data on tip modes in an advanced low aspect ratio fan blade design, on mistuning in a commercial fan stage, and on bench tests of friction dampers. This information is provided in order to clarify the objectives of the research and also to give specific test cases that the researchers can analyze in order to validate the new tools that they develop. Initial versions of several computer programs for computing mistuning, friction damping, shrouded stage response, and unsteady aerodynamic response have been developed and distributed to members. Industry is in the process of assessing these codes by using them to analyze a variety of applications such as ATEGG and JTAGG components, wide chord fans, and blisk designs. These experiences, in turn, provide feedback to the researchers.
In addition, the GUIde meetings provide a conduit for the transfer of additional related information. For example, the Pratt & Whitney/NASA LeRC oscillating cascade data that will be distributed to GUIde members will provide an enhanced empirical basis for predicting stall flutter and aerodynamic damping under transonic flow and high loading conditions. The Air Force ADLARF test on a modern transonic fan design was fully instrumented for both vibration and flow measurements and the resulting data are in the process of being distributed to GUIde members. Consequently, because of the Consortium, improved communication and technology transfer are taking place between the industrial, government, and university members.
There have been a number of important lessons which may be taken from the association of the Consortium members during the first two years of its operation. The first is that an organization of this type must include the people who will actually use the technology. Each of the industrial member participants are responsible for their individual company's forced response design system. This allows them to better guide the research efforts and give the appropriate feedback to the researchers. With this guidance, the researchers provide a product which is better tailored to specific industry problems, providing enhanced technology transfer. The difficulty with this lower managerial level technical membership is one of visibility. Advertisement of Consortium accomplishments within each company is an important issue which must be addressed in order to answer the "what have you done for me lately" question often asked by upper-level managers.
An organization of this type does require a level of cooperation which is somewhat unusual in the development of turbomachinery technology. The engine company representatives must be willing to cooperate with the individuals against whom they normally compete, but are rewarded with research over a much broader range of topics. Researchers must accept a far greater level of interaction and guidance from their funding source, but with the knowledge that they are doing work that will be utilized by a larger audience. Government entities must accept a smaller, more equal role in decisions relative to research directions, but see the research benefit the industry as a whole which improves the technical edge against foreign competition.
The contract for the GUIde Consortium is due to expire in mid-1996 when all six research projects will have been completed. In response to the success of the consortium concept, a follow-on proposal for GUIde 11 is currently being prepared for submission for Air Force funding. This next generation of GUIde will include a number of changes which reflect the changing interests of the member organizations and the realities of current corporate internal research funding policies.
The first change represents an expansion of the scope of the research to include the broader area of blade durability. Specifically, the research areas for GUIde 11 are: unsteady aerodynamics including blade flutter, vibration control through damping, nonintrusive measurement, probabilistic HCF, and optimization of aerodynamics and structural response. This change is in response to the movement of industrial and military focus away from strictly performance based criteria to include the larger issues of engine cost, maintainability, and life.
The reduction in the available pool of discretionary industrial internal research money has also caused a change to the industrial dues structure. Instead of a company contribution of $50K per year to the Consortium, membership will require dues of $l0K per year. The companies are to spend another $40K per year to use the technology developed under GUIde I in current engine development programs and report their experiences back to the Consortium. While reducing the direct outlay of precious research money, this new structure will greatly enhance the transfer of the GUIde I technology into the company design systems as well as advertise the benefits of Consortium membership to upper-level managers. Companies which are currently outside of the Consortium are invited to participate, but will require either a "buy-in" of GUIde I dues or payment of $50K per year for the first two years of GUIde 11 at which point all GUIde I products will be available and the split fee schedule may be used.
The reduced pot of industrial funds may force the Consortium to scale back the number of research projects which can be funded unless additional sources of funding can be found. It is hoped, however, that the expansion of the scope of the research from forced response to blade durability issues will attract a wider range of users and hence a larger resource pool. Efforts are currently underway to solicit funding from other agencies including the Navy, FAA, DoE, as well as other organizations within the Air Force. Additional funding may also become available through the NASA Advanced Subsonic Technology (AST) program.
The GUIde Consortium on Forced Response of Blade Disks was formed as a partnership between government, universities, and industry to address the industry-wide problem of excessive vibration in turbomachinery bladed disks. Currently, six turbine engine manufacturers, the Air Force, and NASA fund the effort which sponsors six research projects equally divided between structural response and aerodynamic issues. This paper has presented the structure of the Consortium, the roles of each of its member organizations, and the costs and benefits of their participation. The Consortium, now entering its third year of operation, has shown that companies from a highly competitive industry can successfully work together to achieve far more than would be possible individually. Additionally, the Consortium has proven to be far more cost effective for the government in the dissemination of experimental test data. Details of programs, such as the Air Force's ADLARF or NASA's oscillating cascade tests, reach a far larger audience and, consequently, are of real use to more than a single contractor.
For others interested in forming a similar organization, it should be understood that there is a large effort required to develop this type of program. All participants must be willing to cooperate with people against whom they normally compete. However, because of the substantial benefits available through a program such as the GUIde Consortium, it is important that we make the effort. Ultimately, if we are unable to cooperate, then as a nation we will all lose.
The authors would like to acknowledge the important contributions of the other members of GUIde in the establishment of and participation in the Consortium: S. Fleeter at Purdue University, T. Fecke with the Air Force at Wright-Patterson AFB, G. Stefko and D. Hoyniak at NASA/LeRC, Y. EL-Aini and G. Hilbert at Pratt & Whitney, R. Sharma at Pratt & Whitney, Canada, R. Kielb at General Electric Aircraft Engines, R. Stockton at Textron-Lycoming, D. Burns and W. Dalton at Allison Engines, J. Adams at Allied Signal, and A. Partington and M. Brown at Westinghouse Electric.
Griffin, J.H., 1989, "GUIde Consortium on Research in Forced Response," in Proceedings of the First CMU/Purdue/Air Force Workshop, Carnegie Mellon University, Pittsburgh, PA.
Griffin, J.H., 1990, "GUIde Consortium on Research in Forced Response," in Proceedings of the Second CMU/Purdue/Air Force Workshop, Carnegie Mellon University, Pittsburgh, PA.
Murthy, D.V. and Morel, M.R., 1993, "Turbine Blade Forced Response Prediction Using FREPS," SAE Paper 931373.
Russler, P., Rabe, D., Cybyk, B. and Hah, C., 1995, "Tip Flow Fields in a Low Aspect Ratio Transonic Compressor," Paper to be presented at the 1995 ASME Turbo Expo, Houston, TX.