0

Manufacturing Materials and Metallurgy

1994;():V005T12A001. doi:10.1115/94-GT-006.
FREE TO VIEW

As a result of normal operating conditions, jet engine parts tend to distort. During repair, many of these parts require some form of selective cutting. Due to the differences between each individual part, automation such as CNC machining becomes complex. The need to map each individual part prior to machining has prohibited repair facilities from introducing sophisticated automation. This paper presents for the first time a novel method that addresses many problems related to selective material removal in Jet Engine repairs. Similar to a terrain-following cruise missile, a laser guided cutter is used to follow a selected datum surface. For example, in the case of Honeycomb Airseals, the thin laser beam penetrates through the honeycomb cell structure to follow the base metal. The method minimizes parent metal removal and under-minimum-wall condition. As a result, life potential of parts is extended, and repair costs are reduced. The ability to track a moving part and modify the cutting path automatically produces one more significant advantage: accurate set-up of a part on a machine is no longer critical. The method has applications in the repair of airseals, frames and cases, and airfoils.

Commentary by Dr. Valentin Fuster
1994;():V005T12A002. doi:10.1115/94-GT-050.
FREE TO VIEW

A research study was undertaken to evaluate whether electron beam welding (EBW) or gas tungsten arc welding (GTAW) could be utilised for repairs to the leading edges of the Turmo IV C compressor blades. These blades are manufactured from Ti-6Al-4V. The study entailed performing a series of welding trials. For the GTAW process a matching filler metal to the parent metal was used whereas for the EBW process, the welds were made autogenously. After metallographic examination of the weld microstructure, mechanical property assessments were undertaken, namely tensile and fatigue tests, the latter being a stringent test to evaluate the performance of the welded joint. The results demonstrated that the EB welds had equivalent properties to the parent metal whereas the GTA welds had poorer fatigue properties due to undesirable microstructure that resulted in the weld zone. The results achieved herein showed that the EBW process would be an appropriate technique for the restoration of these compressor blades.

Commentary by Dr. Valentin Fuster
1994;():V005T12A003. doi:10.1115/94-GT-199.
FREE TO VIEW

This paper describes research and development of 12Cr steel for heavy duty gas trubine disc. The 1300°C class gas turbine developed by Hitachi, Ltd. (H25) is commercially operated for land-based power generator (25MVA) which is realized after five years designing.

The quantitative study of alloying elements, which increase mechanical properties such as creep rupture strength, tensile strength, and toughness, is discussed. The sixteen pieces of real scale disc for 1300°C class gas turbine are examined to evaluate mechanical properties in terms of real operation. The recent study of 12Cr steel disc for 1500°C class gas turbine is also discussed.

Topics: Steel , Gas turbines , Disks
Commentary by Dr. Valentin Fuster
1994;():V005T12A004. doi:10.1115/94-GT-302.
FREE TO VIEW

High Velocity Oxy-Fuel thermal spray systems (HVOF), also known as High Velocity Combustion systems (HVC) are high energy thermal spray combustion processes, producing very hard, high density coatings. These coatings are used in areas where high wear resistance is of particular importance, with metal carbide coatings being typical in gas turbine applications.

Gas turbines use hard face coatings in such areas where vibration is the initial source of the problem. The areas tend to be in the hot end of the gas turbines although certain areas of the cold end are also affected. To date the hard face coating that has been predominately applied in gas turbines particularly in the hot end, is the Praxair (Union Carbide) “D” gun coating. As a result to date, the “D” gun system has had a virtual monopoly with regards to the overhaul/repair of gas turbine components where hard face coating was required.

However new HVOF systems have come on to the market Examples are: CDS, Plasma Technik; Diamond Jet, Metco; Top Gun, UPT U.K. (Miller); Jet Kote, Deloro Stellite; Gun. Metallisation and JP 5000, Hobart Tafa. As a result gas turbine overhaul bases are now in a position to offer a more competitve coating service, producing comparable, and in some cases superior, coatings to those produced by the “D” gun process.

This paper covers, from the Rolls Wood Group point of view, the developments of the HVOF systems, where they appear to be today and how these systems, now allow overhaul bases to offer services not previously available and the ability to develop new coating applications.

Commentary by Dr. Valentin Fuster
1994;():V005T12A005. doi:10.1115/94-GT-305.
FREE TO VIEW

Airfoils of industrial gasturbines are provided with longitudinal cooling holes. The cooling efficiency of these holes may be improved by introducing multiple ribs, so-called turbulators. The Electro Chemical Drilling process is used to drill longitudinal cooling holes with or without turbulators in casted as well as in forged airfoils. Turbulated cooling holes are drilled by varying process parameters during drilling. A FEM model is developed to determine the effect of parameter variations on the shape of the turbulators. The model can also be used to study the effect of fluctuations of parameters on the quality of straight and turbulated holes. The feasibility of the model is demonstrated by comparing calculated and experimental results of a step-wise variation of the cathode voltage.

Commentary by Dr. Valentin Fuster
1994;():V005T12A006. doi:10.1115/94-GT-375.
FREE TO VIEW

The efficiency of steam- and gasturbines is mainly influenced by the geometry and the surface roughness of the turbine blades. Therefore the profile contour of the blades must be machined as accurate as possible. High speed cutting (hsc) offers a lot of advantages for surface finishing of turbine blades. The paper describes the influence of different cutting parameters as well as the importance of tool geometry for the surface quality achievable by high speed milling. Specific requirements for machine tools for high speed milling will be discussed.

Commentary by Dr. Valentin Fuster
1994;():V005T12A007. doi:10.1115/94-GT-399.
FREE TO VIEW

The routine overhaul and repair of gas turbine components requires dimensional restoration of many components. This process is usually accomplished by using various thermal sprays to build up the worn part surface and then machining or grinding the thermal sprays to correct part tolerances and dimensions. These surfaces usually contain numerous holes that generate heavily interrupted surfaces to be machined. The combination of interrupted cuts and a very wear resistant thermal spray causes rapid tool wear on conventional carbide cutting tools. This rapid tool wear also produces poor surface finish, part taper, chipping around the hole edges, and increased tool pressure that could result in lifting or peeling of the sprayed material from the parent metal.

This paper summarizes the results of machining nickel based thermal sprays with polycrystalline cubic boron nitride (PCBN) cutting tools. The tests have shown a minimum 2–5x improvement in surface finish, dimensional control, part taper and up to 10x increase in productivity. The PCBN tools also generated lower cutting forces resulting in reduced stress and higher bond strengths in the part.

This paper presents the data collected and the recommended machining parameters developed under controlled conditions for machining air plasma sprayed Metco 443, Metco 450, Inconel 718, two wire arc applied TAFA 75B, TAFA 73MXC, and high velocity oxygen fuel applied Inconel 718.

Topics: Nickel , Machining , Sprays
Commentary by Dr. Valentin Fuster
1994;():V005T12A008. doi:10.1115/94-GT-426.
FREE TO VIEW

It is generally conceived that a blade surface is flank millable if it can be closely approximated by a ruled surface; otherwise the slow machining process of point milling has to be employed. However, we have now demonstrated that the ruled surface criterion for flank milling is neither necessary nor sufficient Furthermore, many complex arbitrary surfaces typical of our blades in fans, axial compressors, and centrifugal impellers in aviation gas turbines are actually closely flank millable and can be rendered exactly flank millable with one or more passes per surface often without sacrificing, indeed sometimes with gain, in performance.

Commentary by Dr. Valentin Fuster
1994;():V005T12A009. doi:10.1115/94-GT-448.
FREE TO VIEW

Multi-axis laser materials processing systems are having a significant impact on the way turbine engine parts are being cut, drilled, and welded. The success of laser cutting, drilling and welding is based in the ability to concentrate laser energy into a small area and to produce features having narrow heat affected zones.

Reduced tooling expense, fast turnaround, and flexibility for handling design changes and for economical small lot manufacturing are some of the benefits associated with laser processing of turbine engine parts. For example, in replacing hand trimming, laser cutting has increased throughput for trimming a deep drawn gas turbine engine part from 18 pieces per day to 18 pieces in 30 minutes. For another company, laser cutting saved $75,000 in tooling expense for an application involving drilling of 3500, 0.33 mm diameter holes in a 0.36 mm thick turbine engine blade insert. Laser cutting and welding have been key to a major aircraft manufacturer’s implementation of Just-In-Time manufacturing practices.

Recent advances in design and control of laser processing systems have increased the number of applications for laser processing turbine engine parts. New processes have increased the range of applications for which laser processing is qualified. In-process gaging has been used to automate processes that were previously manual and, in doing so, increase the quality of the finished part and productivity of the manufacturing operation.

This paper reviews the laser processes used to cut, drill, and weld turbine engine parts. Some typical applications are presented to illustrate the benefits of laser systems for processing turbine engine parts.

Commentary by Dr. Valentin Fuster
1994;():V005T12A010. doi:10.1115/94-GT-449.
FREE TO VIEW

Gas turbine engine development continues to accelerate, creating more demanding requirements for abradable seal coatings. These coatings are necessary to provide very small clearances between the rotating and stationary parts in order to minimize gap losses and so Increase efficiency. The relatively few abradable coating materials developed over the last 20 years still perform well in many blade tip seal and labyrinth seal applications. However, rising operating temperatures, corrosion and other environmental changes, longer overhaul times and even better tip clearances are dictating the design of new coating materials which requires a strong scientific approach. For example, ways are being Investigated to replace Nickel-Graphite and other flame sprayed coatings being used between 450 and 700°C respectively because of steady state/corrosion/oxidation/erosion and wear problems respectively.

New plasma and HVOF sprayed coatings have been developed using a systematic approach based on material response to operating conditions, minimizing trial and error. The major steps in the programme were:

1. Selection of constituent materials able to withstand service temperatures up to 325 (AISI-Polyester or Polyimide), 450 (AISI base), 700 (MCrAlY base) and 1100°C (ceramic base) respectively.

2. Powder particle manufacture and coating deposition to guarantee highly reproducible coatings.

3. Coating optimization based on wear tests carried out using a fully instrumented abradability test rig and wear mechanism analysis.

4. An investigation of blade tipping systems for high temperature applications.

This paper discusses the results of plasma sprayed coatings developed for use at 450 and 700°C.

Commentary by Dr. Valentin Fuster
1994;():V005T12A011. doi:10.1115/94-GT-450.
FREE TO VIEW

Technological advances in arc spray have produced an arc spray system that competes very favorably with other thermal spray processes. In the past arc spray was thought of as a process for very large parts that need thick buildups. However, an attachment device has been developed which focuses the pattern and accelerates the particles and is known as the Arc Jet system. The advantages of this device were discussed by Zwetloot, Sampson and Thorpe (1993), Reference 4.

This attachment device coupled with the introduction of metal cored wires that provide the same chemistries as plasma sprayed powders, provides application engineers with a viable economic alternative to existing spray methods.

A comparative evaluation of a standard production plasma spray system was conducted with the arc spray process utilizing the device mentioned above. This evaluation was conducted by an airline company on four major parts coated with nickel aluminum and results show that, for those applications, the arc spray process demonstrated significant benefit. It is expected that other applications may benefit from the use of the arc spray process as well.

Commentary by Dr. Valentin Fuster
1994;():V005T12A012. doi:10.1115/94-GT-475.
FREE TO VIEW

Gas turbine engines are constructed of components with excellent strength and stiffness, a minimum density, a high temperature capability for long times, and at affordable cost. Metallic materials are the centrepiece in fulfilling these requirements. Future gas turbine engines will have to have higher thrust-to-weight ratios, better fuel efficiencies and still lower costs. This will require new and advanced lightweight materials with higher temperature capabilities.

This paper discusses some of the presently applied materials in the fan, compressor and turbine sections of gas turbines, and reviews the material developments that are occurring and will be necessary for the near and long term futures.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster

Ceramics

1994;():V005T13A001. doi:10.1115/94-GT-228.
FREE TO VIEW

This paper addresses the application of ultrasonic methods to damage assessment in ceramics and ceramic matrix composites. It focuses on damage caused by thermal shock and oxidation at elevated temperatures. The damage-induced changes in elastic constants and elastic anisotropy are determined by measuring the velocities of ultrasonic waves in different propagation directions within the sample. Thermal shock damage measurement is performed in ceramic samples of reaction bonded silicon nitride (RBSN) and aluminum oxide. Thermal shock treatment from different temperatures up to 1000°C is applied to produce the microcracks. Both surface and bulk ultrasonic wave methods are used to correlate the change of elastic constants to microstructural degradation and to determine the change in elastic anisotropy induced by microcrack damage. Oxidation damage is studied in silicon carbide fiber/reaction bonded silicon nitride matrix (SCS-6/RBSN) composites. The oxidation is done by exposing the samples in a flowing oxygen environment at elevated temperatures, up to 1400°C, for 100 hours. Significant changes of ultrasonic velocities were observed for composites before and after oxidation. The elastic constants of the composites were determined from the measured velocity data. The Young’s modulus in the fiber direction as obtained from ultrasonic measurements decreases significantly at 600°C but retains its original value at temperatures above 1200°C. This agrees well with the results of destructive tests by other authors. The transverse longitudinal and shear moduli obtained from ultrasonic measurements decrease continually until 1200°C. The results of this work show that the damage-induced anisotropy in both ceramics and ceramic matrix composites can be determined successfully by ultrasonic methods. This suggests the possibility of assessing damage severity using ultrasonic techniques.

Commentary by Dr. Valentin Fuster
1994;():V005T13A002. doi:10.1115/94-GT-309.
FREE TO VIEW

To cope with the increasing demand of electric power, many research and development programs have been performed in the field of electric power industry. Among them, the application of highly thermal resistive ceramics to hot parts of the gas turbines is one of the most promising ways to raise the thermal efficiency of the gas turbine, and several projects have been executed in the U.S.A., Europe and Japan.

Tokyo Electric Power Co., Inc. (TEPCO) also has been conducting a research project to apply ceramic components to hot parts of a 20MW class gas turbine with a turbine inlet temperature of 1300C. In this project. TEPCO and Hitachi have been conducting the cooperative research work to develop a first stage ceramic rotor blade.

After several design modifications, it was decided to select ceramic blades attached directly to a metal rotor disc, and to insert metal pads between the dovetail of the ceramic blade and metal disc to convey the centrifugal force produced by the blade smoothly to the metal disc.

The strength of this ceramic blade has been verified by a series of experiments such as tensile tests, room temperature spin tests, thermal loading tests, and high temperature spin tests using a high temperature gas turbine development unit (HTDU).

In addition, the reliability of the ceramic blade under design and test conditions has been analyzed by a computer program GFICES (Gas turbine - Fine Ceramics Evaluation System) which was developed on the basis of statistical strength theory using two parameter Weibull probability distribution.

These experiments and analyses demonstrate the integrity of the developed ceramic rotor blade.

Commentary by Dr. Valentin Fuster
1994;():V005T13A003. doi:10.1115/94-GT-313.
FREE TO VIEW

A program is being performed under the sponsorship of the United States Department of Energy, Office of Industrial Technology, to improve the performance of stationary gas turbines in cogeneration through the selective replacement of hot section components with ceramic parts. It is envisioned that the successful demonstration of ceramic gas turbine technology, and the systematic incorporation of ceramics in existing and future gas turbines will enable more efficient engine operation, resulting in significant fuel savings, increased output power, and reduced emissions.

The engine selected for the program, the Centaur 50 (formerly named Centaur ‘H’) will be retrofitted with first stage ceramic blades, first stage ceramic nozzles, and a ceramic combustor liner. The engine hot section is being redesigned to adapt the ceramic parts to the existing metallic support structure.

The work in Phase 1 of the program involved concept and preliminary engine and component design, ceramic materials selection, technical and economic evaluation, and concept assessment. A detailed work plan was developed for Phases II and III of the program. The work in Phase II addresses detailed engine and component design, and ceramic specimen and component procurement and testing. Ceramic blades, nozzles, and combustor liners will be tested in subscale rigs and in a gasifier rig which is a modified Centaur 50 engine. The Phase II effort also involves long term testing of ceramics, development of appropriate nondestructive technologies for part evaluation, and component life prediction. Phase III of the program focuses on a 4,000 hour engine test at a cogeneration site.

This paper summarizes the progress on the program through the end of 1993.

Commentary by Dr. Valentin Fuster
1994;():V005T13A004. doi:10.1115/94-GT-420.
FREE TO VIEW

The Ceramic Stationary Gas Turbine (CSGT) Program has utilized the SPSLIFE computer code to evaluate the preliminary design of ceramic components. The CSGT program is being performed under the sponsorship of the United States Department of Energy, Office of Industrial Technology, to improve the performance of stationary gas turbines in cogeneration through the selective replacement of hot section components with ceramic parts. Preliminary design evaluation and life assessment results are presented here for the following components: (1) Stage 1 Turbine Blade, (2) Stage 1 Turbine Nozzle, and (3) Combustor Inner Liner. From the results of the analysis, recommendations are made for improving the life and reliability of the components. All designs were developed in Phase I (preliminary design) of the CSGT program and will be optimized in Phase II (detail design) of the program.

Topics: Design
Commentary by Dr. Valentin Fuster
1994;():V005T13A005. doi:10.1115/94-GT-444.
FREE TO VIEW

A study of crack bridging in long-fiber reinforced ceramic matrix composites (CMC) is the main focus of this paper. The in-situ observation using NDE techniques reveals that 1D SiC/CAS CMCs show only partial fiber bridging behavior while 2D SiC/SiC CMCs possess an excellent bridging mechanism. Laser interferometric technique was successfully applied to measure crack opening displacements (COD) in the notched specimens under four-point flexural loading. The onset of precracking and the effects of fiber bridging were also detected by in-situ optical microscopy and acoustic emission techniques. The applied load-COD relations were shown and on the basis of these results, the bridging stiffness parameter can be determined via the fracture mechanics formulation proposed by Cox and Marshall.

Commentary by Dr. Valentin Fuster
1994;():V005T13A006. doi:10.1115/94-GT-445.
FREE TO VIEW

Initial results of the first step (at Pa = 0.11–0.12 MPa) of an experimental investigation of the basic parameters of full-scale, micro-flame double-zone combustors with flame tubes are presented. This combustion chamber is developed for a 2.5 MW advanced ceramic gas turbine unit. (Sudarev, et al., 1991). This engine, when working at the design operation conditions, has an efficiency range of 41–46%, which is a function of using either intecooling or a heat regeneration scheme. The efficiency is the result of increasing the gas temperature to a maximum turbine inlet temperature of 1250°C and a 2.5 MW pressure ratio of 29. With such high initial parameters of the working media, the problem of nitrogen oxide emissions reduction assumes paramount importance.

The objective of the paper is to develop a combustor which would ensure NOx emissions at the design conditions not above 75 mg/Nm3 (at 15% 02) due to application of a double-stage working process of pre-mixture firing. Specific features of fuel burn-up, formation of pollutants at combustion, dependencies of combustor characteristics and upgraded algorithm of combustor loading are also shown.

Commentary by Dr. Valentin Fuster
1994;():V005T13A007. doi:10.1115/94-GT-484.
FREE TO VIEW

The failure probability of notched tensile bars is calculated using the multiaxial Weibull theory. The influence exerted by the stress concentration factor, the stress gradient in the notch root, and the Weibull exponent is analysed.

Commentary by Dr. Valentin Fuster
1994;():V005T13A008. doi:10.1115/94-GT-489.
FREE TO VIEW

Metal matrix composites (MMC) and intermetallic matrix composites (IMC) are materials of complex structure. Nominally defect-free, as-manufactured MMC requires nondestructive evaluation (NDE) for quality assurance and process monitoring purposes. In this work, three NDE techniques — ultrasonics, eddy current, and X-ray radiography — were applied to un-damaged NiFeAI/Wf coupons. Images of the coupons were obtained using the three techniques. The NDE results were compared among themselves, and correlations were sought between these results and microstructural features of the specimen. Consistencies were found among the NDE results and a strong correlation was found between the spatial variation of fiber density and the NDE signals.

Commentary by Dr. Valentin Fuster

Structures and Dynamics

1994;():V005T14A001. doi:10.1115/94-GT-030.
FREE TO VIEW

A subsynchronous vibration problem with a large water injection pump was solved in 1974 by modifying the discharge piping transition to a long taper configuration. This paper describes a quantitative theory based upon the dynamic modeling of the pump rotor and piping system. The mode] successfully duplicates the experimental results of the pump instability and reproduces the subsynchronous vibration described in the original paper.

The analysis results in a log decrement of the system that is dependent upon the time delay of the acoustic pulsation. It shows that the system can be driven unstable for acoustic path lengths that are much less than the quarter wave length. The analysis also shows the log decrement is minimized when the time delay is equivalent to exactly that of an eighth wave length of the resonant frequency of the pump.

Topics: Acoustics , Pumps , Feedback
Commentary by Dr. Valentin Fuster
1994;():V005T14A002. doi:10.1115/94-GT-035.
FREE TO VIEW

The paper describes a synchronous vibration instability problem encountered on a centrifugal compressor with oil lubricated bearings. The problem was solved by modification of the compressor rotor, however the root cause was not completely understood at that time. A possible explanation was based on a theory which suggested differential heating of the bearing journals. It was decided to verify ibis theory by experiments. Therefore a test rotor was designed with identical rotor dynamic characteristics to those of the compressor rotor. In order to fill a gap in the published research on bearing thermohydrodynamics an experimental technique was devised to measure the surface temperature variations around one of the journals of this rotor. The dependence of significant temperature differentials across the journal upon its orbit was confirmed.

Commentary by Dr. Valentin Fuster
1994;():V005T14A003. doi:10.1115/94-GT-038.
FREE TO VIEW

Magnetic bearings represent a new bearing technology which has some advantages over conventional fluid film and rolling element bearings for some applications. The paper describes the basic concepts of magnetic thrust bearing operation involving the magnetic actuator, electronic controls, power amplifier, and sensor. The magnetic actuator is a magnetic circuit which generates attractive forces. These support the rotating shaft While it is often thought that magnetic bearings are highly nonlinear devices, this paper demonstrates that they are linear in both the perturbation flux and current when used in a double acting configuration. Electronic feedback controls are used to stabilize the bearing. Example design parameters are presented for an application to an industrial canned motor pump.

Commentary by Dr. Valentin Fuster
1994;():V005T14A004. doi:10.1115/94-GT-052.
FREE TO VIEW

This paper summarizes the development of hybrid squeeze film dampen (HSFDs) for active control of rotor vibrations. Previously, it was shown both theoretically and experimentally that HSFDs can be used for controlling rotor vibrations (El-Shafei, 1993). This is done by controlling the flow in a squeeze film damper through movable end seals, thus achieving the ability to change the damper from a short damper to a long damper and vice versa. However, the control of the HSFD was manual. In this paper, an automatically controlled circuit is developed for the HSFD, incorporating a pressure control servovalve for controlling the pressure in the scaling chambers. A complete mathematical model of this open-loop system is developed and is implemented on a digital computer. The transient behavior of the system, including the sealing ring dynamics, illustrates that the open-loop system exhibits well behaved, stable, and fast response. In addition it is shown that the HSFD can achieve any amount of damping between the short and long damper modes through the accurate positioning of the sealing rings. The simulation results illustrate that the automatically controlled HSFD can be a very useful device for the active control of rotors. A closed loop control strategy with feedback on rotor speed is also investigated both from the points of view of steady state and transient behaviors. It is shown that this closed loop strategy results in a much improved behavior of the rotor system.

Topics: Bearings , Modeling , Rotors
Commentary by Dr. Valentin Fuster
1994;():V005T14A005. doi:10.1115/94-GT-053.
FREE TO VIEW

An integrated, compact piezo-hydraulic actuator system for active vibration was designed and developed with a primary application for gas turbine aircraft engines. Copper tube was chosen as the transmission line material for ease of assembly. Liquid plastic which meets incompressibility and low viscosity requirements was adjusted to provide optimal actuator performance. Variants of the liquid plastic have been prepared with desired properties between −40°F and 400° F. The effectiveness of this hybrid actuator for active vibration control (AVC) was demonstrated for suppressing critical speed vibration through two critical speeds for various levels of intentionally placed imbalance. A high accuracy closed loop simulation which combines both finite element and state space methods was applied for the closed loop unbalance response simulation with/without AVC. Good correlation between the simulation and test results was achieved.

Commentary by Dr. Valentin Fuster
1994;():V005T14A006. doi:10.1115/94-GT-057.
FREE TO VIEW

The vibration and active control of a flexible rotor system with magnetic bearings is investigated using Hybrid Method (HM) and H control theory with consideration of gyroscopic effect. The hybrid method which combines the merits of finite element method (FEM) and generalized ploynomial expansion method (GPEM) is employed to model the flexible rotor system with small order of plant. The mixed sensitivity problem of H control theory is applied to design the control of system vibration with spillover phenomena for the reduced order plant. The H2 control design is also employed for the comparison to the H design.

The experimental simulation is used to illustrate the effects of control design. It is shown that the H controller design can be very effective to suppress spillover phenomena. In addition, HM control design has robustness to the variation of the parameters of the model. The application of hybrid method (HM) together with H control design is highly recommended for the vibration control of flexible rotor system with magnetic bearings.

Commentary by Dr. Valentin Fuster
1994;():V005T14A007. doi:10.1115/94-GT-060.
FREE TO VIEW

On the basis of characteristics of vibration in the rotor system with spring nonlinearity, a new method for vibration control has been developed. In the method, the spring characteristics of a bearing housing are controlled to be of softening nonlinearity when the rotor supported on it is accelerated and to be of hardening one when it is decelerated. So vibratory amplitudes of the rotor system always vary along the smallest solution curve in the whole operating process. A model of vibration of the rotor system supported on the controllable hearing housing is derived. Its dynamic behaviour is predicted and verified by experiments. Both theoretical and experimental results show that not only vibratory amplitudes and transmitted forces are suppressed significantly but also nonlinear vibration performance of the rotor supported on squeeze film dampers, such as “lock up” at rotor pin-pin critical speeds and asynchronous vibration, can be avoided.

Commentary by Dr. Valentin Fuster
1994;():V005T14A008. doi:10.1115/94-GT-064.
FREE TO VIEW

A neural network based output feedback control system has been developed to optimally suppress rotor vibration caused by the rotor unbalance. The unique feature of this new vibration absorber is its ability to optimally control vibration for a wide range of rotor speeds. Numerical examples dealing with a multi-degree-of-freedom rigid rotor supported by radial magnetic bearings are presented to verify the advantages of this novel neural network based output feedback active vibration absorber.

Commentary by Dr. Valentin Fuster
1994;():V005T14A009. doi:10.1115/94-GT-072.
FREE TO VIEW

In this paper the stability of non-linear misaligned rotor-bearing systems is investigated, using the Lyapunov direct method. A finite element formulation is used to determine the journal bearing pressure distribution. Then the linear and nonlinear stiffness, damping and hybrid (depended on both displacements and velocity) coefficients are calculated A general method of analysis based on Lyapunov’s stability criteria is used to investigate the stability of non-linear misaligned rotor bearing systems. The equations of motion of the rigid rotor on the non-linear bearings are used to find a Lyapunov function using some of the above coefficients, which are depending on L/D ratio and the misalignment angles ψa, ψr. The analytical conditions for the stability or instability of some examined cases are given and some examples for the orbital stability are also demonstrated.

Commentary by Dr. Valentin Fuster
1994;():V005T14A010. doi:10.1115/94-GT-077.
FREE TO VIEW

Linear vibrations are studied for a straight uniform finite beam element of general orientation spinning at a constant angular speed about a fixed axis in the inertial space. The gyroscopic and circulatory matrices and also the geometric stiffness matrix of the beam element are presented. The effect of the centrifugal static axial load on the bending and torsional dynamic stiffnesses is thereby accounted for. The Rayleigh/Timoshenko/Saint-Venant theory is applied, and polynomial shape functions are used in the construction of the deformation fields. Non-zero off-diagonal elements in the gyroscopic and circulatory matrices indicate coupled bending/shearing/torsional/tensional free and forced modes of a generally oriented spinning beam. Two numerical examples demonstrate the use and performance of the beam element.

Commentary by Dr. Valentin Fuster
1994;():V005T14A011. doi:10.1115/94-GT-078.
FREE TO VIEW

A nonlinear approach based on the method of averaging has been developed to study unstable lateral vibrations of rotors in high-speed turbomachinery. The method makes use of an extended concept of bearing dynamic coefficients which are defined by applying small perturbations to the dynamic equilibrium positions. It has been applied to determine the stability threshold of flexible rotors supported on short and tilting-pad journal bearings respectively. Stability performance of systems supported on short journal bearings is improved by the presence of mass unbalance. However, for flexible rotors supported on tilting-pad journal bearings, increased unbalance can lead to lower rotordynamic stability margins. This has the significant implication that stability estimates from linear analysis are not always conservative. The present method provides a computationally more efficient way to better understand the nonlinear vibration behaviour of high-speed turbomachinery.

Topics: Turbomachinery
Commentary by Dr. Valentin Fuster
1994;():V005T14A012. doi:10.1115/94-GT-081.
FREE TO VIEW

The dynamic behavior of the rotating machinery supported by the hydrodynamic journal bearings is significantly influenced by the dynamic characteristics of the oil-film. In the present work an efficient identification method is used to identify the stiffness and damping coefficients of the tilting pad and cylindrical journal bearings of flexible rotor-bearing system. The method uses FRFs (Frequency Response Functions) obtained by the measurements and the finite element method. The accuracy and feasibility of the method were tested and demonstrated by theoretical simulation. The possible effects of oil-film inertia is also verified by the theoretical simulation. The method can be further extended to identify twelve linearized oil-film coefficients.

Topics: Bearings
Commentary by Dr. Valentin Fuster
1994;():V005T14A013. doi:10.1115/94-GT-092.
FREE TO VIEW

A new technique for measuring and actively controlling dynamic bearing loads in rotating machinery is presented. Bearing loads are estimated using the Deflection-Coefficient Method, a technique which does not rely on a full system model, and which applies commonly-used shaft-deflection measurement equipment to obtain estimates of bearing loads. The estimated bearing load is used as an error signal in an adaptive feedforward disturbance rejection controller. The result is a control system which can selectively minimize dynamic bearing loads in real time in rotating machinery systems. The method is applied to a numerical model of a typical rotating machinery system to suppress dynamic reaction forces at bearing supports.

Topics: Machinery , Stress , Bearings
Commentary by Dr. Valentin Fuster
1994;():V005T14A014. doi:10.1115/94-GT-093.
FREE TO VIEW

This paper presents an analysis of the dynamic response of the high stiffness gas journal bearing to a harmonic load. The dependence between the frequency of the load, and the amplitude and phase angle of the shaft vibration are estimated from numerical simulation. Results of this simulation enable an attempt of identification of the high stiffness bearing as a linear system with two (or in simplification - one) degrees of freedom, to formulate simple dynamic models of the bearing. Stiffness and damping coefficients are calculated and compared with the simple gas journal bearing. The comparison suggests, that the investigated gas bearing retains high stiffness not only for slow changes of the load which appear in its application to grinding machines, but even when the frequency of the load is equal to several rad/s.

Commentary by Dr. Valentin Fuster
1994;():V005T14A015. doi:10.1115/94-GT-100.
FREE TO VIEW

This paper presents the results of an investigation into the dynamic structural properties of self-acting compliant foil journal bearings. A test facility with a journal supported by a compliant foil journal bearing was built. The nonrotating journal was driven by two shakers which were used to simulate the dynamic forces acting on the bump foil strips. The structural stiffness and equivalent viscous damping coefficients are calculated based on the experimental measurements for a wide range of operating conditions. The results are compared to the analytical predictions obtained by a theoretical model developed earlier, and the effects of frequency are investigated. Both theoretical and experimental results show that an increase in the excitation frequency decreases the direct damping term. The effect of frequency on the cross-coupling terms is much less than on the direct terms.

Commentary by Dr. Valentin Fuster
1994;():V005T14A016. doi:10.1115/94-GT-102.
FREE TO VIEW

Equivalent radial stiffness and damping coefficients of ball bearing joints are identified from a bearing test rig through measured linear frequency domain transfer functions. Common least square techniques are used to curve fit analytical transfer functions of a MDOF-mechanical model of the test rig to the measured transfer functions. Both, the damping caused within the dry Hertzian contacts between the balls and the raceways and the damping within the bearing to housing interface are determined separately. A substantial interface damping is observed that may explain significant deviations between theoretical and experimental results stated in the literature.

Commentary by Dr. Valentin Fuster
1994;():V005T14A017. doi:10.1115/94-GT-103.
FREE TO VIEW

There is a dramatic change in the shape of the semi-elliptical cracks growing due to combined cyclic bending and tensile loading of plates. The growing cracks change their shape such that they follow preferred propagation patterns (PPPs). It is observed that computation of strain energy release rate (G) in surface cracks plays an important role in determining the preferred propagation pattern (PPP). Numerical integration techniques have been employed to compute total strain energy (TSE) release rate. The variation of TSE with crack depth (a/t) for different bending ratios (Rb) is presented and discussed. A program in BASIC is written to directly simulate the propagation of crack on the screen. It is shown that for a given cyclic loading field, the PPP represents an upper limit on the aspect ratio of any surface crack growing due to this cyclic loading.

Topics: Surface cracks
Commentary by Dr. Valentin Fuster
1994;():V005T14A018. doi:10.1115/94-GT-107.
FREE TO VIEW

A method for dynamic analysis of flexible bladed-disk/shaft coupled systems is presented in this paper. Being independant substructures first, the rigid-disk/shaft and each of the bladed-disk assemblies are analyzed separately in a centrifugal force field by means of the finite element method. Then through a modal synthesis approach the equation of motion for the integral system is derived. In the vibration analysis of the rotating bladed-disk substructure, the geometrically nonlinear deformation is taken into account and the rotationally periodic symmetry is utilized to condense the degrees of freedom into one sector. The final equation of motion for the coupled system involves the degrees of freedom of the shaft and those of only one sector of each of the bladed-disks, thereby reducing the computer storage. Some computational and experimental results are given.

Commentary by Dr. Valentin Fuster
1994;():V005T14A019. doi:10.1115/94-GT-108.
FREE TO VIEW

A numerical study of the dynamical response and the life estimation of a gas turbine blade which is subjected to transient nozzle excitation is presented. The mechanical and mathematical model for the blade, the exciting unsteady aerodynamic forces and the life estimation techniques are described and the solution procedure and its realization in a computer code is discussed. For an axial gas turbine compressor blade subjected to unsteady lift and drag forces during a run-up and run-down process numerical results are presented and the relation between the damping ratios, the speed of the run-up/down and the estimated fatigue life is discussed.

Commentary by Dr. Valentin Fuster
1994;():V005T14A020. doi:10.1115/94-GT-109.
FREE TO VIEW

This paper proposes an improved squeeze film damper which will prevent the bistable operation associated with conventional squeeze film dampers at large unbalances and/or at small bearing parameters. It consists of a conventional squeeze film damper with a flexibly supported outer ring. This secondary flexible support is considered to be massless, and to have a constant stiffness and damping. The effectiveness of this damper in preventing bistable operation is investigated over a wide range of operating conditions for a rigid rotor supported on a centrally preloaded squeeze film damper. It is shown that depending on relevant parameters such as the stiffness ratio between the secondary support and the retaining spring, the damping coefficient of the support, and the mass ratio between the damper outer ring and the rotor, this proposed damper is very effective in preventing bistable operation even for high unbalance conditions.

Commentary by Dr. Valentin Fuster
1994;():V005T14A021. doi:10.1115/94-GT-110.
FREE TO VIEW

Modelling and analysis of a rotor-bearing system with a new type of active oil bearing are presented. The active bearing basically consists of a flexible sleeve and a pressure chamber. The deformation of the sleeve can be controlled by the chamber pressure during the operation, and so can the pressure distribution of the oil film.

Finite Element Methods (FEMs) and the Guyan condensation technique were utilised to create mathematical models for both the rotor and the flexible sleeve. The hydrodynamic pressure distribution of the oil film, for the instantaneous positions and velocities of the flexible sleeve and rotor, was approximated by Reynolds equation. The influence of the chamber pressure on the stability of the rotor system was investigated by numerical simulation based on the nonlinear model. The results showed that the stability of the rotor-bearing system can be improved effectively by implementation of the active bearing.

Commentary by Dr. Valentin Fuster
1994;():V005T14A022. doi:10.1115/94-GT-111.
FREE TO VIEW

In turbomachinery application Active Magnetic Bearings (AMB) have become an interesting alternative to conventional bearings. They not only offer an excellent solution to suspend the rotor, but also allow for sophisticated real-time data acquisition and monitoring with no additional hardware within the machine.

The present paper describes a digital controller environment for active magnetic bearing systems. The controller computer consists of a digital signal processor board with a fast parallel link to a PC. Special software has been developed to directly interfaces the control program to the software package MATLAB1 which is used on the host computer for data processing, identification and controller layout. These facilities are used to achieve time and frequency domain measurements for fast on-site tuning and data monitoring. Different examples are presented to demonstrate these powerful tools. Possible procedure for future system identification are discussed.

Commentary by Dr. Valentin Fuster
1994;():V005T14A023. doi:10.1115/94-GT-112.
FREE TO VIEW

Over the past seven years, an extensive hybrid bearing development program has been conducted at Textron Lycoming. This paper will report the details of testing and the extraordinary results which can be obtained with silicon nitride balls as applied in hybrid bearings on gas turbine engines. This paper describes the analytically predicted advantages which low mass silicon nitride balls offer at speeds over 2.0MDN. Rig testing comparing hybrid bearings to standard bearings is reported. Testing included heat generation evaluation which showed that hybrid bearings generate an average of 40% less heat than standard bearings.

Rig simulation of the AGT1500 mission duty cycle demonstrated that the hybrid silicon nitride bearing system is robust enough to handle the most severe operating conditions. Testing under severe slipping/skidding conditions demonstrated good resistance to skid failure. Under conditions selected to produce high wear, no wear was induced in a hybrid bearing while severe wear was induced in the M50 steel bearing. These preliminary successes lead to active engine testing on the AGT1500 and a new test program to demonstrate operation at 4.0 MDN.

As a result of these programs Textron Lycoming now considers hybrid ceramic bearings as a viable design to be used in high speed development applications. This paper provides design detail and test data covering the work outlined above.

Commentary by Dr. Valentin Fuster
1994;():V005T14A024. doi:10.1115/94-GT-113.
FREE TO VIEW

The influence of fluid dynamic forces in a fluid-lubricated bearing on the imbalance response of a flexible rotor is examined with respect to effects on polar-plot balancing. The results of an experiment clearly demonstrate a journal phase lag at the center of the bearing which approaches 90° relative to the response of a mid-span massive disk over a range of rotor speeds from well below to well above the first lateral bending mode natural frequency of the rotor. The experimental phase lag is nearly constant over the entire range of rotor speeds. The results of this experiment are found to be in good agreement with results from both an analytical 2-degree-of-freedom model and a finite element analysis. The analytical model reveals that the severity of the phase lag effect is related to the ratio of the Quadrature Dynamic and Direct Dynamic stiffnesses acting at the journal. Calibration-weight balancing is found to be unaffected by the phase lag problem. The authors conclude that, - unless this phase lag is taken into account, polar-plot balancing may be difficult when vibration data from a relatively flexible rotor are obtained very close to a fluid-lubricated bearing.

Topics: Fluids , Bearings
Commentary by Dr. Valentin Fuster
1994;():V005T14A025. doi:10.1115/94-GT-127.
FREE TO VIEW

Nonsynchronous perturbation techniques developed over the last few years have proven to be a very powerful tool for parameter identification of rotating systems. In order to obtain interpretable results, some limitations have had to be imposed on the analytical models and rotor systems used in the process, such as applied forces, and measurement transducers had to be located near major masses on the rotor. In the laboratory environment these limitations can usually be accommodated, but not always, while in the field, compliance is almost impossible. This paper explores the use of finite element modeling, measured vibration response data, and optimization techniques to extend the applicability of parameter identification through nonsynchronous perturbation techniques to those systems in which the perturbation forces and/or resultant response measurements cannot be conveniently located at the mass centers. In this technique, a finite element model of the rotor system is constructed, and all known information about the system parameters input to a computer program designed to operate on a personal computer. The program then computes the theoretical response of the system, by processing user-supplied initial conditions for the unknown system parameters, and compares these results with the vibration responses measured on the machine, and collected by the data acquisition system. The unknown parameters are then modified using local convergence optimization techniques until the error between the theoretical and measured responses is acceptably small. If the system parameters under investigation coincide with the unknown parameters in the computer program, they are identified in the process.

This technique was applied to an experimental rotor system constructed such that the parameters under investigation, the direct and quadrature dynamic stiffness components for a plain oil-lubricated journal bearing operating at low eccentricity, could be determined by both this technique and conventional unbalance force testing. The results of the nonsynchronous tests are presented in the paper.

Topics: Optimization
Commentary by Dr. Valentin Fuster
1994;():V005T14A026. doi:10.1115/94-GT-128.
FREE TO VIEW

Usually while modelling rotor-bearing systems the bearings are treated as point supports. In the present paper, using the finite element technique, the unbalance response of rotors is studied by considering distributed bearing stiffness and damping. The bearing stiffness and damping terms are derived by the principle of virtual work. Unbalance responses of rotors with bearing distributed effects are compared with the model using point supports and for different supports Viz., cylindrical journal bearings, tilting pad journal bearings, offset and three lobe journal bearings.

Commentary by Dr. Valentin Fuster
1994;():V005T14A027. doi:10.1115/94-GT-129.
FREE TO VIEW

It is argued in this paper that errors in rotor balancing practice are often due to the difference between the angular location of vibration transducers and that of the vibration modes of the rotor.

A methodology is presented for uncoupling of the vibration modes of an unbalanced rotor. This leads to the theoretical elimination of the errors in determining the phase angle of the resonant response of the rotor and hence in the angular location of the balancing weights.

Differences between theory and reality are here briefly discussed although field application methodologies and experiences will be the subject of another paper.

Commentary by Dr. Valentin Fuster
1994;():V005T14A028. doi:10.1115/94-GT-203.
FREE TO VIEW

The vibration of grouped blades on a flexible disk cannot be analyzed using procedures developed for individual, uncoupled groups. The bladed-disk system must be considered either in whole, or as a structure composed of cyclically-symmetric sectors. The latter approach, while being computationally more economical, is more complicated to implement because both the structural system and the applied forces must be transformed to a system based on the geometry of a single sector. However, the sector transformation permits a natural ordering of the modes that cannot be obtained from a complete system model. This paper describes the development of the sector model, and the associated transformation of the applied harmonic forces typically prevalent in turbomachinery. The displacements are expressed in a series of the natural modes of a transformed sector, and expressions are developed for maximum displacements of the complete structure. This approach leads to an easier interpretation of the analytical results and an improved physical understanding of the response. For example, it is shown that a single harmonic, or engine-order excitation, can cause response in only a restricted subset of modes.

Topics: Blades
Commentary by Dr. Valentin Fuster
1994;():V005T14A029. doi:10.1115/94-GT-204.
FREE TO VIEW

The vibration of grouped blades on a flexible disk should, for purposes of economy and clarity of modal identification, be analyzed using procedures developed for cyclically-symmetric structures. In this paper, a numerical model, based on the theory of cyclically-symmetric structures, is applied to the vibration analysis, and in particular, the harmonic response, of a flexible disk supporting a number of groups, or packets, of turbine blades. Results are presented to show variations in the modal participation factors as a function of such parameters as disk flexibility, blade density, and the total number of assembled groups. It is also shown that many characteristics of the system spectra of natural frequencies are strongly dependent on the number of blade groups.

Topics: Blades
Commentary by Dr. Valentin Fuster
1994;():V005T14A030. doi:10.1115/94-GT-206.
FREE TO VIEW

Unsteady, separated, high Reynolds number flow over an airfoil undergoing oscillatory motion is investigated numerically. The compressible form of the Reynolds-averaged governing equations is solved using a high-order, upwind biased numerical scheme. The turbulent flow region is computed using a one-equation turbulence model. The computed results show that the key to the accurate prediction of the unsteady loads at stall flutter conditions is the modeling of the transitional flow region at the leading edge. A simplified criterion for the transition onset is used. The transitional flow region is computed with a modified form of the turbulence model. The computed solution, where the transitional flow region is included, shows that the small laminar/transitional separation bubble forming during the pitch-up motion has a decisive effect on the near wall flow and the development of the unsteady loads. Detailed comparisons of computed fully turbulent and transitional flow solutions with experimental data are presented.

Commentary by Dr. Valentin Fuster
1994;():V005T14A031. doi:10.1115/94-GT-270.
FREE TO VIEW

A time-marching approach is adopted in developing a thermal/structural program with linked flow-solid modeling capability. The Blade Life Analysis & Design Evaluation for Combustion Turbines (BLADE-CT) program analyzes gas turbine blade thermal-mechanical stress and natural frequencies under the boundary conditions which result from the gas flow and the cooling/barrier flow within a given turbine stage. Using the finite element method, the blade temperatures obtained from transient/steady-state thermal solutions can be utilized to compute thermal stresses and dynamic stresses under operating conditions for assessing thermal-mechanical fatigue damage in combustion turbine blades. A customized and automated mesh generation routine is developed to model cooled (spanwise multihole configurations) and solid gas turbine blades. By coupling the NASA flow programs, PCPANEL (potential flow), STAN5 (heat transfer boundary layer), and CPF (coolant passage flow) as part of an automated flow-structural analysis approach, a more efficient and accurate thermal and thermal stress calculation can be achieved. The calculated blade temperatures can be also applied for the frequency analysis to account for temperature effects. The coupled fluid-structure interaction program approach for thermal-mechanical analysis and an example of a spanwise cooled blade steady state analysis are presented.

Commentary by Dr. Valentin Fuster
1994;():V005T14A032. doi:10.1115/94-GT-271.
FREE TO VIEW

Model testing has long been used to validate, augment, or replace structural analysis. Choice of the extent of model usage depends on the time, budget, and skills available, and upon the complexity of the structure and the consequences of the results. In particular, plastic models have seen significant usage in understanding the static and dynamic behavior of many different industrial structures. Today’s ability to generate scaled plastic models directly from 3D CAD geometry models via stereolithography (SLA) has opened a new chapter on model testing. SLA models are easily size-scaled and quickly produced, and some SLA polymers have mechanical properties that are well suited for dynamic testing. The use of SLA models makes possible early testing and accelerated test development work that can reduce costs and shrink the product development cycle. This paper describes the elements of using SLA plastic models for dynamic testing, including size and frequency scaling. Examples of SLA applications at Textron Lycoming for rap/shaker testing and for holography are discussed.

Commentary by Dr. Valentin Fuster
1994;():V005T14A033. doi:10.1115/94-GT-272.
FREE TO VIEW

At present, when numerical methods are used to analyze the dynamic characteristics of blades with densely distributed small holes, it is difficult to consider the influence of small holes accurately. A new method is presented regarding air-cooled turbine rotor blades of gas turbine engines. To investigate the topic, a series of experimental rules are simulated effectively by theoretical formulae. The method is developed by combining the experimental results and finite element analysis. To include the effects of small holes on dynamic behaviors of blades, the equivalent concept is employed. In a generalized eigen-equation, the modified stiffness matrix and mass matrix are used to perform an accurate modal analysis of the blades. When making comparison between the analytical results obtained by using the method of this paper and the experimental data, a good agreement is seen. Accuracy, reliability, and practicality of the method presented in this paper are verified.

Topics: Blades
Commentary by Dr. Valentin Fuster
1994;():V005T14A034. doi:10.1115/94-GT-273.
FREE TO VIEW

The untwist of rotating blades in turbomachines treated so far in the literatare simply as a pure elasticity problem is generalized and formulated rigorously as a problem of aeroelasticity by variational principles (VPs) and generalized VP (GVP). It takes into account not only the centrifugal force, but also the aeroelastic interaction between blades and the flow as well as the elastic distortion of the cross section shape of blades, assuming the material to be linearly elastic but nonisotropic. Thus, a new rigorous theoretical basis for the finite element analysis of blade untwist in turbomachine design is provided.

Topics: Rotating blades
Commentary by Dr. Valentin Fuster
1994;():V005T14A035. doi:10.1115/94-GT-274.
FREE TO VIEW

A method for the optimization of stator blade stagger angles and of strut circumferential position in order to shield a rotor cascade from downstream strut pressure disturbances is presented. Potential flow interaction between rotor, stator and strut rows is analyzed by a boundary integral method for unsteady incompressible 2-D flows, based on singularity superposition over every blade profile. The flow field is solved in terms of velocity, and the pressure field is computed using the unsteady Bernoulli equation. An optimization technique based on a constrained minimization problem is applied; objective functions related to pressure coefficient and lift coefficients are considered. Force coefficients in the tangential and axial directions, and the momentum coefficient acting on the blades are calculated by integrating the pressure distribution over the blade profile. Results of the stagger angle and strut position optimization are presented and discussed. An analysis in the frequency domain is also performed.

Commentary by Dr. Valentin Fuster
1994;():V005T14A036. doi:10.1115/94-GT-291.
FREE TO VIEW

A novel technique for computing unsteady flows about turbomachinery cascades is presented. Starting with a frequency domain CFD description of unsteady aerodynamic flows, we form a large, sparse, generalized, non-Hermitian eigenvalue problem which describes the natural modes and frequencies of fluid motion about the cascade. We compute the dominant left and right eigenmodes and corresponding eigenfrequencies using a Lanczos algorithm. Then, using just a few of the resulting eigenmodes, we construct a reduced order model of the unsteady flow field. With this model, one can rapidly and accurately predict the unsteady aerodynamic loads acting on the cascade over a wide range of reduced frequencies and arbitrary modes of vibration. Moreover, the eigenmode information provides insights into the physics of unsteady flows. Finally we note that the form of the reduced order model is well suited for use in active control of aeroelastic and aeroacoustic phenomena.

Commentary by Dr. Valentin Fuster
1994;():V005T14A037. doi:10.1115/94-GT-293.
FREE TO VIEW

A numerical method has been developed to predict the forced response of bladed disks due to a wake excitation from upstream blade rows. The structure is modelled by a 3D finite element mesh of a bladed disk segment. Using cyclic symmetry, this model provides a modal base for the rotating structure. The aerodynamic damping of the vibratory modes and the excitation pressures on the blades due to the propagation of upstream flow defects are computed separately using the same 3D unsteady Euler analysis software. A modal response solution of the aeromechanical system is then performed.

This analytical methodology has been used to study the forced response of an experimental high pressure compressor blisk. The results are analysed and compared with actual rig tests.

Commentary by Dr. Valentin Fuster
1994;():V005T14A038. doi:10.1115/94-GT-294.
FREE TO VIEW

The purpose of this work is to validate a time-nonlinear three-dimensional Euler solver for vibrating cascades aerodynamics by comparison with available theoretical semi-analytical results from flat-plate cascades. First the method is validated with respect to the purely two-dimensional theory of Verdon (for supersonic flow) by computing two-dimensional vibration (spanwise constant) in linear three-dimensional cascades. Then the method is validated by comparison with the theoretical results of Namba and the computational results of He and Denton, for subsonic flow in a linear three-dimensional cascade with three-dimensional vibratory mode. Finally the method is compared with results of Chi from two subsonic rotating annular cascades of helicoïdal flat-plates. Quite satisfactory agreement is obtained for all the cases studied. A first code-to-code comparison is also presented.

Commentary by Dr. Valentin Fuster
1994;():V005T14A039. doi:10.1115/94-GT-297.
FREE TO VIEW

This paper presents the single objective optimization and the multi-objective optimization for a flexible rotor system with magnetic bearings. The weight of rotor shaft and the transmitted forces at the magnetic bearings are minimized either individually or simultaneously under the constraints on the critical speeds and the control currents of magnetic bearings. The design variables are the cross-sectional area of the shaft, the bias currents of magnetic bearings, and the positions of the disk and the magnetic bearings. The dynamic characteristics are analyzed using the generalized polynomial expansion method and the sensitivity analysis is also studied. For single objective optimization, the method of feasible directions (MFD) is applied. For multi-objective optimization, the methods including the weighting method (WM), goal programming method (GPM), and the fuzzy method (FM) are employed. It is found that the system design can be significantly affected by the choices of the bias currents of magnetic bearings, the position of the disk with unbalance and the magnetic bearings. The results also show that a better compromized design can always be obtained for multi-objective optimization.

Commentary by Dr. Valentin Fuster
1994;():V005T14A040. doi:10.1115/94-GT-298.
FREE TO VIEW

An aircraft crash in the Netherlands was caused by disintegration of a jet engine. Fractography showed that the chain of events started with stress corrosion cracking (SCC) of a pin attached to a lever arm of the compressor variable vane system. Such a lever arm-pin assembly costs only a few dollars. Investigation of hundreds of pins from the accident and a number of identical engines revealed that this was not an isolated case. Many pins exhibited various amounts of SCC. The failed pin in the accident engine happened to be the first fractured one.

SCC requires the simultaneous presence of tensile stress, a corrosive environment and a susceptible material. In this case the stress was a residual stress arising from the production method. There was a clear correlation between the presence of salt deposits on the levers and SCC of the pins. It was shown that these deposits were able to reach the internal space between the pin and lever arm, thereby initiating SCC in this space. The corrosive environment in Western Europe explains why the problem manifested itself in the Netherlands at a relatively early stage in engine life. The main point is, however, that the manufacturer selected an SCC-prone material in the design stage. The solution has been to change the pin material.

Commentary by Dr. Valentin Fuster
1994;():V005T14A041. doi:10.1115/94-GT-341.
FREE TO VIEW

This paper describes the results of a study into the dynamic behavior of a magnetic bearing system. The research focuses attention on the influence of nonlinearities on the forced response of a two-degree-of-freedom rotating mass suspended by magnetic bearings and subject to rotating unbalance and feedback control. Geometric coupling between the degrees of freedom leads to a pair of nonlinear ordinary differential equations which are then solved using both numerical simulation and approximate analytical techniques. The system exhibits a variety of interesting and somewhat unexpected phenomena including various amplitude driven bifurcational events, sensitivity to initial conditions and the complete loss of stability associated with the escape from the potential well in which the system can be thought to be oscillating. An approximate criterion to avoid this last possibility is developed based on concepts of limiting the response of the system. The present paper may be considered as an extension to an earlier study by the same authors which described the practical context of the work, free vibration, control aspects and derivation of the mathematical model.

Commentary by Dr. Valentin Fuster
1994;():V005T14A042. doi:10.1115/94-GT-347.
FREE TO VIEW

An aeroelastic stability analysis of a cascade of engine blades coupled only through aerodynamica is developed. The unique feature of the analysis is the direct use of unsteady aerodynamic pressures, rather than lifts and moments, in calculating the susceptibility of a cascade to flutter. The approach developed here is realistic and relevant for analysis of low aspect ratio blades. However, in the calculations presented in this paper, the surface is assumed to be divided into equal elemental areas. The formulation leads to a complex eigenvalue problem, the solution of which determines the susceptibility of the cascade to flutter. The eigenvalues of an assembly of alternately mistuned blades, operating at high reduced frequencies, appear to be very sensitive to the level of mistuned frequencies. The locus of eigenvalues shows a strong tendency to split even for very small percentage differences between the frequencies of the two sets of blades. Further, blades with identical frequencies, but alternately mistuned mode shapes, operating at high reduced frequencies show a tendency towards instability.

Commentary by Dr. Valentin Fuster
1994;():V005T14A043. doi:10.1115/94-GT-355.
FREE TO VIEW

The dynamic behavior of a geared rotor bearing systems with coupled torsional-lateral motion is investigated using a Hybrid Method which combines the merits of the Finite Element Method and the Generalized Polynomial Expansion Method. The effects of the stiffness and damping of the bearings and the gear mesh, the mesh angle and the location of the gear mesh are also studied. Natural whirl speed analyses and steady state responses due to mass unbalance and transmission error excitation are presented. The accuracy and the efficiency of the method are demonstrated. The results show that the torsional-lateral coupled modes of the system are strongly affected by the stiffness, damping, and mesh angle of the gear mesh. The steady state response amplitudes due to transmission error at tooth passing frequency are high. The application of hybrid method to the geared rotor-bearing system leads to improved computational efficiency compared to the finite element method without lose of accuracy.

Topics: Bearings , Rotors
Commentary by Dr. Valentin Fuster
1994;():V005T14A044. doi:10.1115/94-GT-379.
FREE TO VIEW

In this paper, we apply damage tolerance analysis to a certain compressor blade by finite element and fracture mechanics methods. The blade is analysed by using the isoparametric finite element with 20 nodes, its stress and displacement fields are determined, and the critical position where the crack probably initiates is located. Then the value of J–integral with different crack sizes in the critical position are computed using the 3-D J-integral program. On the basis of the fracture toughness of blade material, Je, the blade extension limit δe, and the deviation of natural frequency, Δfe, the failure criterion of the blade is discussed and the critical crack size, ae, is determined. Then according to the crack growth rate obtained from the testing, the residual life of the blade with different crack sizes is evaluated. The maximum allowable crack size, a0*, is determined using the condition that the crack would not grow to critical crack size during double overhaul periods. This value can be used to design the structure of compressor blade or to establish the criterion to judge whether a compressor blade with a crack can be used. In this way the safety and reliability of airplanes is maintained and the costs are reduced. So the damage tolerance analysis of compressor blades is practical and economical.

Commentary by Dr. Valentin Fuster
1994;():V005T14A045. doi:10.1115/94-GT-383.
FREE TO VIEW

This paper deals with the most relevant experimental results regarding turbine blade dynamics acquired from several designs: from both military and commercial aeroengines to industrial gas turbines. On this basis, a common design methodology has been validated for the different applications.

Campbell and Goodman diagrams are the basic tools used to verify resonances and vibratory stress margins. In addition to that, a frequency response type of calculation was developed, nevertheless facing the problem of exciting force and damping assumption. High emphasis is given to evaluating the damping characteristics of the blade: in-house programs have been validated through dedicated tests of shroud optimization and platform damper design.

Experimental tests are in any case necessary when a new blade is designed; however the confidence in the design tools gained by their experimental validation is likely to give better chances to have a good new design, shortening the relative development phase.

Commentary by Dr. Valentin Fuster
1994;():V005T14A046. doi:10.1115/94-GT-398.
FREE TO VIEW

To assess correctly the effects of transient vibration in a system with imbalance care is required in modelling the system. This is particularly true in cases of extreme imbalance e.g. a blade-off simulation in turbo-machinery. Generally, however, the imbalance is modelled as a simple mrΩ2 term applied when the blade is released but this does not include all possible terms. This paper presents the detailed equations of motion of a flexible rotor system with distributed imbalance. The equations are presented in a rotating coordinate system. The modelling includes coupling between the torsional, lateral and axial motions. A simpler model of a two disk system is then presented in fixed coordinates. The disks which can move laterally am connected by a massless shaft which has both lateral and torsional stiffness giving the system six degrees of freedom. An analysis is presented showing that the model is the same as the conventional model for steady state circular orbits. Results from a simplified blade-off simulation are then presented and compared to the standard mrΩ2 model. The conclusion drawn from these simulations is that the additional terms should be included for high angular acceleration transient problems.

Topics: Rotors
Commentary by Dr. Valentin Fuster
1994;():V005T14A047. doi:10.1115/94-GT-410.
FREE TO VIEW

A new method for quantitatively predicting the interaction between the unsteady aerodynamics and structural dynamics was developed. The unsteady flow with large scale separation is simulated numerically with the developed Discrete Vortex Method. In this calculation process, the instantaneous unsteady force and moment of force acting on the blades can be obtained at each time step. On other hand, the cascade is considered as an elasticity system with damping including the effects of the interblade phase angle. The blades are excited to vibrate by the unsteady force and moment. To deal with the above two respects, the resulting code consists of two basic subprograms which exchange their data between each other at every computational time step and the non-linear problem is then solved in a time-marching discrete continuation fashion.

A series of numerical tests are presented including the parameters studies, such as: effects of the damping, incidence etc. For each case, a large number of computational steps is intended to provide sufficient information about the non-linear behavior of the stall flutter. The applicability of the method is demonstrated numerically.

Commentary by Dr. Valentin Fuster
1994;():V005T14A048. doi:10.1115/94-GT-476.
FREE TO VIEW

The rise of the unsteady pressure magnitude along the surface of a cascade blade in unsteady transonic flow is examined. It is shown that a similar rise in the unsteady pressure may occur for high subsonic flows where the mean flow is near sonic condition. For a subsonic cascade this unsteady pressure bulge is found to be associated with the cut-on of a new acoustic mode in the upstream direction. The level of the pressure bulge is significantly reduced as a downstream propagating mode cuts on. It is therefore proposed that this phenomenon is the result of the blockage of upstream propagating acoustic waves by the transonic mean flow. A transonic convergent-divergent nozzle is used as a model for investigating the acoustic blockage effect. Analytical and numerical computations using unsteady nonlinear Euler equations are then carried out to analyze and quantify the upstream and dowstream propagation of acoustic disturbances in the nozzle. The results confirm the sharp rise in the pressure of the upstream propagating disturbances at the nozzle throat as a result of the acoustic blockage.

Commentary by Dr. Valentin Fuster

Controls, Diagnostics and Instrumentation

1994;():V005T15A001. doi:10.1115/94-GT-003.
FREE TO VIEW

Current generation mechanical diagnostic equipment is designed to identify individual events or trends in the output of sensors mounted on a mechanical component, subsystem, or system. Such equipment can provide a useful indication that a failure condition may be developing, but it cannot provide reliable predictions of the remaining safe or operational life. Typically, these diagnostic systems simply compare the output of individual sensors against a priori thresholds to establish a measure of the system’s health. Two problems result from this approach: (1) there is no advantage taken of possible synergy among the sensors, i.e., the determination of health is one dimensional; and (2) the diagnosis provides only a statement regarding the current equipment health, but docs not provide a prediction of the time remaining to failure. This often leads to an operational environment in which diagnostic equipment outputs arc either ignored because of frequent false alarms or frequent (and costly) time-based preventive maintenance is performed to avoid hazardous failures. This paper describes a new approach to the development of a more robust diagnosis and prognostic capability. It is based on the fusion of sensor-based and model-based information. Sensor-based information is the essence of current diagnostic systems. Model-based information combines dynamic models of individual mechanical components with micromechanical models of relevant failure mechanisms, such as fracture and crack propagation. These micromechanical models must account for initial flaw size distribution and other microstructural parameters describing initial component condition.

A specific application of this approach is addressed, the diagnosis of mechanical failure in meshing gears. Four specific issues are considered: (a) how to couple a validated numerical simulation of gear transmission error (due to tooth spacing irregularity, contour irregularity, or material inhomogeneity) with physically and empirically-based descriptions of fatigue crack growth to predict a failure precursor signature at the component level; (b) how to predict the manifestation of this signature at the subsystem or system level where sensors are located; (c) how to fuse this model-based information with the corresponding sensor-based information to predict remaining safe or operational life of a gear; and (d) issues associated with extending this methodology to bearings and other rotating machinery components.

Commentary by Dr. Valentin Fuster
1994;():V005T15A002. doi:10.1115/94-GT-023.
FREE TO VIEW

The advent of a new generation of transient rotating turbine simulation facilities, where engine values of Reynolds and Mach number are matched simultaneously together with the relevant rotational parameters for dimensional similitude (Dunn et al [1988], Epstein et al [1984]. Ainsworth et al [1988]), has provided the stimulus for developing improved instrumentation for investigating the aerodynamic flows in these stages.

Much useful work has been conducted in the past using hot-wire and laser anemometers. However, hot-wire anemometers are prone to breakage in the high pressure flows required for correct Reynolds numbers, Furthermore some laser techniques require a longer runtime than these transient facilites permit, and generally yield velocity information only, giving no data on loss production.

Advances in semiconductor aerodynamic probes are beginning to fulfil this perceived need. This paper describes advances made in the design, construction, and testing of two and three dimensional fast response aerodynamic probes, where semiconductor pressure sensors are mounted directly on the surface of the probes, using techniques which have previously been successfully used on the surface of rotor blades (Ainsworth, Dietz and Nunn [1991]). These are to be used to measure Mach number and flow direction in compressible unsteady flow regimes.

In the first section, a brief review is made of the sensor and associated technology which has been developed to permit a flexible design of fast response aerodynamic probe. Following this, an extensive programme of testing large scale aerodynamic models of candidate geometries for suitable semiconductor scale probes is described, and the results of these discussed. The conclusions of these experiments, conducted for turbine representative mean and unsteady flows, yielded new information for optimising the design of the small scale semiconductor probes, in terms of probe geometry, sensor placement, and aerodynamic performance.

Details are given of a range of wedge and pyramid semiconductor probes constructed, and the procedures used in calibrating and making measurements with them. Differences in performance are discussed, allowing the experimenter to choose an appropriate probe for the particular measurement required.

Finally, the application of prototype semiconductor probes in a transient rotor experiment at HP turbine representative conditions is described, and the data so obtained is compared with (PD solutions of the unsteady viscous flow-field.

Commentary by Dr. Valentin Fuster
1994;():V005T15A003. doi:10.1115/94-GT-026.
FREE TO VIEW

A system for fast-response probe measurements in turbomachine flows has been developed and tested. The system has been designed for 40 kHz bandwidth and used with various in-house built probes accommodating up to 4 piezoresistive pressure transducers. The present generation of probes works accurately up to several bar pressure and 120°C temperature. The probes were found to be quite robust.

The use of miniature pressure transducers placed in the head of probes showed that a precise packaging technique and a careful compensation of errors can considerably improve the accuracy of the pressure measurement.

Methods for aerodynamic probe calibration and off-line data evaluation are briefly presented. These aimed, i.e. in the case of a 4-hole probe, at measuring the velocity fluctuations as characterised by yaw, pitch, total pressure and static pressure and at deriving mean values and spectral or turbulence parameters.

Applications of the measuring system to turbomachinery flow in a radial compressor and to a turbulent pipe flow demonstrate the performance of the measuring system.

Commentary by Dr. Valentin Fuster
1994;():V005T15A004. doi:10.1115/94-GT-027.
FREE TO VIEW

The influence of the probe size and geometry on the quality of fast-response measurements in turbomachines has been experimentally investigated.

For investigations in the static domain (time independent flows) probes were calibrated in two continuously operating wind tunnels in the range 0.2< Ma < 1.2. For dynamic calibrations in time variant flows model experiments in water (0.025 < k < 0.4, reduced frequency) were performed.

Aerodynamic characteristics were determined for a great number of probe geometries such as circular cylinders and wedge-type probes with varied apex angles, locations of the sensing holes and leading edge shapes. The experiments comprised investigations in tolerance ranges for prismatic total pressure probes, yaw angle sensitivity, yaw angle and Mach number effects on calibration and influence of dynamic yaw angle fluctuation on probe characteristics. As a result of the experiments errors due to static and dynamic aerodynamic effects could be quantified.

The majority of the errors arising during measurements in turbomachines can be directly related to the probe size. An important number of these errors are systematic and can be analytically modelled and hence their influence corrected.

In fluctuating flows the most severe measurement errors, which often may exceed the quantity of interest, are due to dynamic stall effects. This phenomenon, which is of transient nature and cannot be corrected, is typical for sharp wedge probes but is not present with circular cylinders or the effects being much smaller with very blunt wedges.

Topics: Design , Probes
Commentary by Dr. Valentin Fuster
1994;():V005T15A005. doi:10.1115/94-GT-034.
FREE TO VIEW

This paper describes the design, operation, construction and demonstration of a new type of high bandwidth unsteady temperature sensor based on fibre optics, and capable of operating in a high speed multi-stage research compressor with flow representative of jet engine conditions. The sensing element is an optical coating of zinc selenide deposited on the end of an optical fibre. During evaluation in aerodynamic testing, a 1 K gas temperature resolution was demonstrated at 9.6 kHz and an upper bandwidth limit of 36 kHz achieved.

Commentary by Dr. Valentin Fuster
1994;():V005T15A006. doi:10.1115/94-GT-039.
FREE TO VIEW

A scanning radiation thermometry technique for determining temperature distributions in gas turbines is presented. The system, an enhancement of an earlier work, can be used by operators even without special training, since the temperature distribution is measured and corrected in terms of the error due to the reflected radiation only on the basis of the turbine’s known geometry and the physical properties of the materials. In the proposed model, the surface-exitent radiances are directly acquired via 360-deg scans. Experimental testing was performed on a static turbine-blading model. Since the angle factors emerged as a notable influence on the accuracy of the model, two angle factor calculation methods, selected for suitability from a literature survey, are exhaustively investigated and their selection criteria, defined.

Commentary by Dr. Valentin Fuster
1994;():V005T15A007. doi:10.1115/94-GT-040.
FREE TO VIEW

It is difficult to make a reliable measurement of running clearance in the hostile environment over the blading of a modern gas turbine. When engine manufacturers require the measurement to be made over every blade during live engine tests, system reliability, ruggedness and ease of operation are of primary importance.

This paper describes a tip clearance measurement system that can measure clearance over every blade around a rotor. The measurement system concept is presented, and the system design described in detail. Commissioning of the measurement system on a compressor test facility, and the results obtained are discussed. An analysis of system performance during the commissioning trials concludes the paper.

Commentary by Dr. Valentin Fuster
1994;():V005T15A008. doi:10.1115/94-GT-058.
FREE TO VIEW

The pursuit of reliable engine performance methods is as old as the gas turbine itself. Contemporary methods of engine fault isolation and assessment have centered around the use of modem estimation algorithms similar in nature to the Kalman filter and its derivatives. Those who are familar with such programs will offer the following quip, (apparently known to many but attributed to none), that the Kalman filter works well when you already know the answer, in other words, when the a-priori estimates are sufficiently close to the true values such as to produce small measurement residuals. Unfortunately, the ubiquity of measurement error ensures that this is generally not the case. A pivotal requirement for successful diagnostics becomes the ability to detect and assess the measurement error existing in raw engine data in an effort to mitigate its effect on potential engine fault mis-interpretation. The purpose of this paper is to outline the mathematical basis for the techniques in present use and to offer a generalization to a large measurement error compensation method which has been utilized effectively for over a decade.

Commentary by Dr. Valentin Fuster
1994;():V005T15A009. doi:10.1115/94-GT-116.
FREE TO VIEW

Conventional 3-hole wedge probes fail to measure the correct static pressure when operating in close proximity to a wall or boundary through which the probe is inserted. The free stream pressure near the outer wall of a turbomachine annulus may be over indicated by upto 20% of the dynamic head.

This paper reports a factorial experiment in which the effects of probe wedge angle, stem length, stem shape, flow yaw and pitch angles on this so-called ‘wall proximity effect’ are quantified at representative flow Mach numbers and turbulence intensities. For a given wedge angle, the probe stem length and Mach number are shown to be statistically the most significant of the tested variables. Wall proximity effect is also shown to be influenced by the probe pitch angle, but is largely independent of yaw angle and free stream turbulence intensity.

Commentary by Dr. Valentin Fuster
1994;():V005T15A010. doi:10.1115/94-GT-117.
FREE TO VIEW

Several tests were conducted on a GE Frame 7 gas turbine to determine its dynamic characteristics. The objective is to obtain a model that can be used for controller design. The tests consisted of adding sequences of square waves to the two inputs — the fuel reference and the inlet guide vane angle reference — and recording the inputs and the outputs. This method of exciting the system afforded us with a way of separating the data sets into two categories, the first, in which the fuel reference was changed, and the second, in which the guide vane angle reference was changed. Least-squares system identification techniques were used to obtain linear models using a selection criterion that was a measure of how well a model fit both the sets of data. This brought in a measure of robustness to the models thus making them ideal for use in controller design. This paper summarizes the results from these tests, contains plots that show how well the linear models are able to fit the recorded data, and finally, provides some recommendations for others doing similar work.

Commentary by Dr. Valentin Fuster
1994;():V005T15A011. doi:10.1115/94-GT-130.
FREE TO VIEW

In keeping with the general industry trend of applying Full Authority Digital Electronic Control (FADEC) technology to small gas turbine engines, Textron Lycoming and Chandler Evans Division of Coltec Industries have developed and qualified a single channel control system for use on the Textron Lycoming LF507-1F turbofan engine. The LF507-1F is the world’s smallest FADEC-equipped airline turbofan engine and is the only FADEC-equipped turbofan developed and certified for the regional jetliner market. The application for this powerplant, the four-engine AVRO International Aerospace RJ Avroliner series of aircraft, began airline service in April of 1993.

The FADEC employs modern control algorithms to achieve surge-free operation over the flight envelope while providing rapid transient performance and crisp handling qualities. The control interfaces with the aircraft via an ARINC 429 data link to control each engine automatically to the desired power setting with or without N1 synchronization. A simple hydromechanical backup control provides full dispatch capability in the event of a critical FADEC system failure. In addition, the FADEC includes advanced diagnostics for fault identification to the line replaceable unit (LRU) level without specialized test equipment.

This paper describes the architecture, primary features, and development process of the engine control system. Emphasis is placed on the design characteristics and technical challenges unique to the development of an inexpensive control system for the low thrust turbofan market.

Topics: Turbofans
Commentary by Dr. Valentin Fuster
1994;():V005T15A012. doi:10.1115/94-GT-184.
FREE TO VIEW

This paper details the design, construction and calibration of a heat-fluxmeter to measure at one end of an idealised model of an aero-engine drive shaft. The constraints on instrument design of manufacture and calibration, together with the matching of instrument and measurement characteristics are discussed.

Sample heat-transfer results based upon fluxmeter measurements are presented. A model for the heat-transfer to the nose-cone of an aero-engine for anti-icing applications, derived from these results, is shown to compare favourably with analysis of engine icing performance test data.

Commentary by Dr. Valentin Fuster
1994;():V005T15A013. doi:10.1115/94-GT-200.
FREE TO VIEW

Using a two dimensional compressible flow representation of axial compressor dynamics, a control-theoretic input-output model is derived which is of general utility in rotating stall/surge active control studies. The derivation presented here begins with a review of the fluid dynamic model, which is a 2D stage stacking technique that accounts for blade row pressure rise, loss and deviation as well as blade row and inter-blade row compressible flow. This model is extended to include the effects of the upstream and downstream geometry and boundary conditions, and then manipulated into a transfer function form that dynamically relates actuator motion to sensor measurements. Key relationships in this input-output form are then approximated using rational polynomials. Further manipulation yields an approximate model which is in standard form for studying active control of rotating stall and surge. As an example of high current relevance, the transfer function from an array of jet actuators to an array of static pressure sensors is derived. Numerical examples are also presented, including a demonstration of the importance of proper choice of sensor and actuator locations, as well as a comparison between sensor types. Under a variety of conditions, it was found that sensor locations near the front of the compressor or in the downstream gap are consistently the best choices, based on a quadratic optimization criterion and a specific 3-stage compressor model. The modeling and evaluation procedures presented here are a first step toward a rigorous approach to the design of active control systems for high speed axial compressors.

Topics: Compressors , Modeling
Commentary by Dr. Valentin Fuster
1994;():V005T15A014. doi:10.1115/94-GT-268.
FREE TO VIEW

A numerical method has been preposed to determine optimum laws to control gas turbine engine (CTE) variable components, including an independent control of blade rows in a multistage axial flow compressor under strong non-stationary flow disturbances at the inlet, optimum laws to control a turbofan under non-stationary thermal effects at the inlet have been obtained using mathematical models with various degree of filling in detail the flow in an engine flow path. There is shown a possibility to considerably increase a range of the CTE stable operation through the use of dynamic control of stator blades in a multistage axial flow compressor, also possibilities of practical use of optimum laws to control engine variable components in the system of preventing an unstable operation are being discussed.

Commentary by Dr. Valentin Fuster
1994;():V005T15A015. doi:10.1115/94-GT-269.
FREE TO VIEW

This paper investigates the concept of thrust rating as a means towards reducing the life cycle costs of turbo-jet engine ownership. Towards this end, the concepts of life usage and thrust rating will be discussed to provide a clear understanding of the goals and methods involved in this investigation. In general, all “serviceable” engines will experience varying levels of performance above baseline values. Thrust rating ideally seeks to limit engine performance to baseline values, thereby converting performance into life.

In particular, this paper will present a general computer simulation technique that may be applied to any turbo-jet engine to quantify the life cycle savings resulting from the adoption of thrust rating techniques. The inter-relationships that exist between thrust rating and many practical in-service aspects of turbo-jet usage including performance degradation, operational roles and maintenance practices and policies will be investigated. Some strengths and limitations of the simulation technique will also be identified. This paper will discuss general guidelines for investigating the feasibility of performing thrust rating on various turbo-jet engines. While this technique is likely to be of greatest interest to users of high performance military turbo-jet engines, its principles can be adapted for other gas turbine uses.

Commentary by Dr. Valentin Fuster
1994;():V005T15A016. doi:10.1115/94-GT-289.
FREE TO VIEW

A method for establishing signatures of faults in the rotating blades of a gas turbine compressor is presented. The method employs a panel technique for the calculation of the flow field around blade cascades, with disrupted periodicity, a situation which is encountered when a blade fault has occurred. From this calculation, time signals of the pressure at a location on the casing wall, facing the rotating blades, are constituted. Processing these signals, in combination with “healthy” pressure signals allows the constitution of fault signatures. The proposed method employs geometrical data, as well as data about the operating point of the engine. It gives the possibility of establishing the fault signatures without the need of performing experiments with implanted faults. The successful application of the method is demonstrated by comparison of signatures obtained by simulation to signatures derived from experiments with implanted blade faults, in an industrial gas turbine.

Commentary by Dr. Valentin Fuster
1994;():V005T15A017. doi:10.1115/94-GT-317.
FREE TO VIEW

This review surveys knowledge needed to develop an improved method of modelling the dynamics of gas turbine performance for fault diagnosis applications. Aerothermodynamic and control models of gas turbine processes are examined as complementary to models derived directly from test data. Extensive, often proprietary data are required for physical models of components, while system identification (SI) methods need data from specially-designed tests. Current methods are limited in: tuning models to test data, non-linear effects, component descriptions in SI models, robustness to noise, and inclusion of control systems and actuators. Conclusions are drawn that SI models could be formulated, with parameters which describe explicitly the functions of key engine components, to offer improved diagnostic capabilities.

Commentary by Dr. Valentin Fuster
1994;():V005T15A018. doi:10.1115/94-GT-319.
FREE TO VIEW

The design of high-performance gas turbines requires the reliable prediction of blade tip clearances. Excess clearances allow a portion of the hot gas to flow over the blade tips without performing useful work. The tip leakage flow disturbs the flow field which results in additional losses. Moreover, insufficient blade tip clearance may cause interference which can reduce turbine life. In conventional turbomachines, the blade tip clearances vary markedly with the operating condition of the turbine, essentially as a result of variations in gas temperatures and rotor speed.

Siemens tests prototype gas turbines in its own test facility. An extensive experimental program is devised to verify design calculations regarding strength, aerodynamics and thermodynamics. Among other measurements, the minimum operating tip clearance is measured by abrasion pins. Electro-mechanical sensors measure transient tip clearance during a selected duty cycle consisting of turning-gear operation, cold start, idle operation, as well as part-load, full-load, and most importantly, hot-start. In the present paper, the compressor and turbine tip clearances measured during such a load cycle are compared with calculated predictions. The experimental instrumentation for the prototype gas turbine, as well as design calculations, are presented.

The results show that the new Model V84.3 gas turbine does not exhibit critically small clearances during cold start nor during hot-start due to the careful matching of magnitude and the time constants of the thermal expansion of the blades, discs, blade-ring carriers and casing.

Commentary by Dr. Valentin Fuster
1994;():V005T15A019. doi:10.1115/94-GT-372.
FREE TO VIEW

In the present paper a method for deriving stage characteristics, which can provide accurate prediction of a multistage compressor map is presented. The method combines optimization techniques with the principles of stage stacking. The stage characteristics are produced by modifying some initial generic ones, until the desired accuracy in the prediction at selected points on the overall map is achieved. There are several reasons why prediction based on the initial stage characteristics can be inaccurate, the following three being the main ones. The first is due to inaccuracy in the representation of the stage characteristics themselves. The second is due to lack of exact knowledge of the geometric data of the various stages. Finally the third reason consists of the “weak” modelling representation of complex physical phenomena with one dimensional approaches. Therefore, even when the exact stage characteristics are known, this does not guarantee the accurate prediction of the compressor performance. On account of the above, it is preferable to acquire realistic “effective” stage characteristics which can be used for synthesizing overall compressor characteristics and assessing the effects of stage faults. In this paper, both of these aspects are successfully tackled as demonstrated by applying the method to different test cases.

Topics: Compressors
Commentary by Dr. Valentin Fuster
1994;():V005T15A020. doi:10.1115/94-GT-414.
FREE TO VIEW

The paper presents some examples of simulation research carried out in view of large turbomachinery diagnostics. On the basis of computer simulations of some typical damages, such as “unbalanced” and hydrodynamic instability, it was possible to obtain the trajectories of the journal and the disc centre, as well as the amplitudes and the phase angles.

The paper refers to an example of two-point-supported rotor modelled using the Finite Element Method. The calculations were made on the basis of the complex elastodia-thermic model of the hydrodynamic slide bearing and also on the non-linear analysis of the system motion equations. The non-linear analysis made it possible to generate a characteristic vibration spectrum for every diagnostic relation. It refers especially to the computer simulations of the oil whirls and the oil whip, carried out for the various conditions of the rotor-machinery operation.

Commentary by Dr. Valentin Fuster
1994;():V005T15A021. doi:10.1115/94-GT-422.
FREE TO VIEW

Current and future aircraft engines are increasingly relying upon the use of multivariable control approach for meeting advanced performance requirements. A multivariable adaptive control (MRAC) scheme is proposed in this paper. The adaptation law is derived using only input and output (I/O) measurements. Simulation studies are performed for a two–spool turbojet engine. The satisfactory transient responses are obtained at different operating points from idle to maximum dry power within the flight envelope. These show insensitivity of the design to engine power level and flight condition. Simulation results also show high effectiveness of reducing interaction in multivariable systems with significant coupling. Using the multivariable MRAC controller, the engine acceleration time is reduced by about 19 percent in comparison with the conventional engine controller.

Commentary by Dr. Valentin Fuster
1994;():V005T15A022. doi:10.1115/94-GT-470.
FREE TO VIEW

Software commonly used to predict Aeroengine performance has been adapted for the design point simulation of two large industrial compressor sets. A NOx gas compressor and an air compressor on a common shaft driven by a steam turbine and a tail gas turbine are modelled accurately after the inclusion of both exothermic and endothermic reactions in the NOx gas compressor. Experiment and theory agree to within 1.5%.

Results for the simulation of a three spool Ammonia Synthesis compressor with a deliver pressure of 207 bar agree to within 1% of experimental data. Since complex gas mixtures were used, Mollier diagrams or Compressibility factor techniques were not used and, instead, Departure Function theory was adopted to accommodate the real gas behaviour found at the high pressure.

Commentary by Dr. Valentin Fuster

Education

1994;():V005T16A001. doi:10.1115/94-GT-083.
FREE TO VIEW

An integral part of the U. S. Department of Energy’s Advanced Turbine Systems Program is the Industry/University consortium, formed to strengthen the gas turbine technology base among American universities. The Advanced Gas Turbine Systems Research program is sponsored by DOE, directed by industry leaders, and the universities conduct the research. This close alliance assures that the research is accomplished in a timely fashion, in areas critical to industry needs.

In the past year and a half the consortium has been organized, members solicited, Requests For Proposals issued, and contracts awarded. Some 60 universities in the U. S. are participating. This paper will discuss how the program was initiated, how it functions, and will review some of the interesting research that is under way in support of the overall ATS goals.

Commentary by Dr. Valentin Fuster
1994;():V005T16A002. doi:10.1115/94-GT-225.
FREE TO VIEW

Osborne Reynolds’ developments of the concepts of Reynolds averaging, turbulence stresses, and equations for mean kinetic energy and turbulence energy are viewed in the light of one-hundred years of subsequent flow research. Attempts to use Reynolds energy-balance method to calculate the lower critical Reynolds number for pipe and channel flows is reviewed. The modern use of turbulence-energy methods for boundary layer transition modelling is discussed, and a current European Working Group effort to evaluate and develop such methods is described. The possibility of applying these methods to calculate transition in pipe, channel and sink flows is demonstrated using a one-equation, q-L, turbulence model. Recent work using the equation for the kinetic energy of mean motion to gain understanding of loss production mechanisms in three-dimensional turbulent flows is also discussed.

Commentary by Dr. Valentin Fuster
1994;():V005T16A003. doi:10.1115/94-GT-226.
FREE TO VIEW

This paper reports results from the use of laser Doppler anemometry (LDA) to measure the mean and the fluctuating flow field in a U-bend of strong curvature, Rc/D = 0.65, that is either stationary or rotating in orthogonal mode (the axis of rotation being parallel to the axis of curvature). The data acquisition system enables a stationary optical fibre probe to collect flow data from a rotating U-bend sweeping past it. Three cases have been examined all concerning a flow Reynolds number of 100,000; a stationary case, a case of positive rotation (the pressure side of the duct coincides with the outer side of the U-bend) at a Rotational number (ΩD/Um) of 0.2 and a case of negative rotation at a Rotational number of −0.2. Measurements have been obtained along the symmetry plane of the duct and also along a plane near top wall. The most important influence on the development of the mean and the turbulence flow fields is exerted by the streamwise pressure gradients that occur over the entry and exit regions of the U-bend. In the stationary case a 3-dimensional separation bubble is formed along the inner wall at the 90° location and it extends to about 2 diameters downstream of the bend causing the generation of high turbulence levels. Along the outer side, opposite the separation bubble, turbulence levels are suppressed due to streamwise flow acceleration. For the Rotation numbers examined, the Coriolis force also has a significant effect on the flow development. Positive rotation doubles the length of the separation bubble and generally suppresses turbulence levels. Negative rotation causes an extra separation bubble at the bend entry, raises turbulence levels within and downstream of the bend, increases velocity fluctuations in the cross-duct direction within the bend and generates strong secondary motion after the bend exit. It is hoped that the detailed information produced in this study will assist in the development of turbulence models suitable for the numerical computation of flow and heat transfer inside blade-cooling passages.

Commentary by Dr. Valentin Fuster
1994;():V005T16A005. doi:10.1115/94-GT-267.
FREE TO VIEW

Reynolds stresses and mean flow velocities were measured throughout a tip leakage vortex in a large scale linear turbine cascade using hot-wire anemometry. An X-wire endflow probe was rotated about the probe axis using a computer controlled stepper motor. An improved solution procedure reduced the approximations involved in the Reynolds stress analysis. Measurements were made on a plane just upstream of the blade trailing edge. The results show the turbulence intensity, the turbulence kinetic energy, and all six Reynolds stresses. The non-isotropy of the Reynolds normal stresses is also presented. Turbulence production rates are evaluated and the contributions to the dissipation of mean kinetic energy from Osborne Reynolds’ nine terms are found. The turbulence production due to the strain rate of the secondary flow is also considered.

Commentary by Dr. Valentin Fuster
1994;():V005T16A006. doi:10.1115/94-GT-323.
FREE TO VIEW

This paper describes the Dutch educational system for gas turbine engineering. The history of the gas turbine industry and education is given briefly. Recently the industry took the initiative of creating a foundation ‘to stimulate gas turbine education’. Courses are implemented in the general educational structure. The cooperation of academia and industry has proved to be very fruitful.

Commentary by Dr. Valentin Fuster

General

1994;():V005T17A001. doi:10.1115/94-GT-029.
FREE TO VIEW

By considering time dependent terms as external excitation forces, the approximate dynamic response of a cracked horizontal rotor is analysed theoretically and numerically. The solution is good for small cracks and small vibrations in the stable operating range. For each steady state harmonic component the forward and backward whirl amplitudes, the shape and orientation of the elliptic orbit and the amplitude and phase of the response signals arc analysed, taking into account the effect of crack size, crack location, rotor speed and unbalance. It is found that the crack causes backward whirl, the amplitude of which increases with the crack. For a cracked rotor, the response orbit for each harmonic component is an ellipse, the shape and orientation of which depends on the crack size. The influence of the crack on the synchronous response of the system can be regarded as an additional unbalance whereupon, depending on the speed and the crack location, the response amplitude differs from that of the uncracked rotor. The nonsynchronous response provides evidence of crack in the sub-critical range, but is too small to be detected in the supercritical range. Possibilities for crack detection over the full speed range include the additional average (the constant) response component, the backward whirl of the response, the ellipticity of the orbit, the angle between the major axis and the vertical axis and the phase angle difference between vertical and horizontal vibration signals.

Commentary by Dr. Valentin Fuster
1994;():V005T17A002. doi:10.1115/94-GT-189.
FREE TO VIEW

Deterioration in the burners or combustion chamber of a gas turbine can result in uneven and unstable air flow and excessive temperature profiles, any of which will produce unnecessary and potentially damaging stress cycles.

Detection relies on comparisons of thermocouples located circumferentially at a convenient point in the hot gas path. The techniques usually recommended by engine manufacturers compare absolute values taking no account of any initial asymmetry due to manufacturing tolerances, thermocouple positioning or turbulence of the gas flow at the measurement point.

The initial profile of measured temperatures can be established empirically over the operating range and it is deviation from this profile which provides a sensitive measure of deterioration in the combustion system. This more sensitive technique provides earlier detection of impending failure.

This paper reviews the normally recommended approach, presents the case for detecting departure from the footprint and looks at a case study.

Commentary by Dr. Valentin Fuster
1994;():V005T17A003. doi:10.1115/94-GT-369.
FREE TO VIEW

The laboratory scale ram combustor test program has been investigating to obtain fundamental combustion characteristics of a ram combustor which operates from Mach 2.5 to 5 for the super/hypersonic transport propulsion system. In our previous study, combustion efficiency had been found poor, less than 70%, due to a low inlet air temperature and a high velocity at Mach 3 condition. To improve the low combustion efficiency a fuel zoning combustion concept was investigated by using a sub-scale combustor model first. Combustion efficiency more than 90% was achieved and the concept was found very effective. Then a laboratory scale ram combustor was fabricated and combustion tests were carried out mainly at the simulated condition of Mach 5. A vitiation technique was used to simulate a high temperature of 1263 K. The test results indicates that ignition, flame stability and combustion efficiency were not so significant but the NOx emissions are a critical problem for the ram combustor at Mach 5 condition.

Commentary by Dr. Valentin Fuster
1994;():V005T17A004. doi:10.1115/94-GT-371.
FREE TO VIEW

Gas turbine operation in urban environments creates new challenges in inlet filtration. The particulate concentration could be light, but significant quantities of the particles are sub-micron in size. Additionally, the presence of hydrocarbons can cause problems with air filter performance, both through short life and turbine fouling.

A new filter system for these environments is described. Development objectives for the medias and filter systems are discussed, test and evaluation procedures are defined, and the performance of the resulting product is documented.

Commentary by Dr. Valentin Fuster
1994;():V005T17A005. doi:10.1115/94-GT-380.
FREE TO VIEW

Rotor-bearing system dynamics can only be clearly understood if the magnitude of the structure internal properties (damping, stiffness) are known. Seldom are these properties known to users and researchers. Particularly, industrial users of machines are concerned with the in-situ structural (structure of machine and foundation) properties which provide better insight for understanding the machinery characteristics behaviors. In this paper, a simple method is outlined for calculating the damping factor with the help of Polar and Bodé plots. Pertinent equations are also developed. Experimental results shows the rotors supported by journal bearings (air and oil lubricated) have the damping factor in the range of 0.0349 ∼ 0.0386.

Topics: Bearings , Damping , Rotors
Commentary by Dr. Valentin Fuster
1994;():V005T17A006. doi:10.1115/94-GT-400.
FREE TO VIEW

This paper presents the unique design and operational aspects of the triple fuel system developed and implemented for the Westinghouse/Mitsubishi Heavy Industries 701-DA combustion turbine. Eight of these units have recently been put into service at the Enron cogeneration project in Teesside, England. This facility is currently the world’s largest gas fired combined cycle cogeneration project. The triple fuel system was designed to meet the customer’s unique operational requirements. A variable composition natural gas in combination with gaseous propane, and liquid naphtha presented considerable fuel and control system design challenges.

Commentary by Dr. Valentin Fuster
1994;():V005T17A007. doi:10.1115/94-GT-428.
FREE TO VIEW

Forging is one of the most widely used manufacturing process for making high strength, structurally integrative, impact and creep resistant Ti-6A1-4V compressor blades for jet engines. In addition, in modern metal forming technology, finite element analysis method and computer modeling are being extensively employed for initial evaluation and optimization of various processes, including forging.

In this study, DEFORM, a rigid viscoplastic two-dimensional finite element code, was used to study the effects of initial die temperature and initial ram velocity on the forging process. For a given billet, die temperature and ram velocity influence the strain rate, temperature distribution, and thus the flow stress of the material. The die temperature and the ram velocity were varied over the range 300–700°F and 15–25 in/sec, respectively, to estimate the maximum forging load and the total energy required to forge compressor blades. The ram velocity was assumed to vary linearly as a function of stroke.

Based on the analysis, it was found that increasing the die temperature from 300–700°F decreases the forging loads by 19.9% and increases the average temperature of the workpiece by 43°F. Similarly, increasing the initial ram velocity from 15–25 in/sec decreases the forging loads by 25.2% and increases the average temperature of the workpiece by 36°F. The nodal temperature distribution is bimodal in each ease. The forging energy required to forge the blades is approximately 18 kips*in/in.

Commentary by Dr. Valentin Fuster

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