0

Aircraft Engine

1985;():V001T01A001. doi:10.1115/85-IGT-11.
FREE TO VIEW

The development and progress of aero-derivative gas turbines for ground usage in China is reviewed. Several models derived and models in development are mentioned. as a result of some feasibility studies the prospect for future gas turbine applications is also pointed out.

Topics: Gas turbines , China
Commentary by Dr. Valentin Fuster
1985;():V001T01A002. doi:10.1115/85-IGT-12.
FREE TO VIEW

A study of the response of a turbojet engine to the steady-state and the turbulence-type dynamic inlet distortion is presented in this paper. The steady-state distortion is generated by a 180° extent, 36 mesh screen, and the turbulence-type dynamic distortion by a 180° extent plate with 50% blockage ratio at the engine face. This plate can produce a very strong pressure fluctuation at the engine face. The statistical analysis shows that the APD of pressure fluctuation follows approximately the Normal Distribution except those cases near rotating stall or surge.

Results from testing show: 1) inlet distortion generated by screen will produce a classical-surge or deep-surge (defined in ref. 1); 2) the degree of distortion by screen can change the mode of surge, e.g. from the classical-surge to the deep-surge and vice versa; 3) both the inlet distortion and the decrease in first-stage-turbine-nozzle area will change the compressor performance maps; 4) the turbulence-type dynamic distortion causes a “drift-surge” (defined in ref. 2).

Topics: Engines , Turbojets
Commentary by Dr. Valentin Fuster
1985;():V001T01A003. doi:10.1115/85-IGT-57.
FREE TO VIEW

The predicted potential performance of the Prop-Fan offers a major improvement in the energy efficiency of future, short-range to medium-range transports. This paper describes the approach taken in designing an optimum Prop-Fan propulsion system. Trade-offs in the configuration(s) and performance are discussed, as are the important aspects of integrating the propeller, gearbox, engine, inlet, exhaust, and nacelle. Realizing the impressive potential fuel savings of the Prop-Fan will require very careful engine/airframe integration. Design options that will be compared are: a single-rotation versus counter-rotation arrangement, a tractor versus pusher installation, and wing versus fuselage mounting. In summary, the performance of turbofan powered and Prop-Fan powered, short-haul transports will be compared in detail by using fuel burn, operating costs, and noise as criteria.

Commentary by Dr. Valentin Fuster
1985;():V001T01A004. doi:10.1115/85-IGT-82.
FREE TO VIEW

A method of order reduction of multivariate non-linear system to calculate twin spool turbojet and turbofan performance is recommended. With the proper incorporation of other approaches, it can achieve satisfactory results in any flight condition and throttle setting. Compare with the experimentation data, it can models the engine very well.

Topics: Turbofans , Turbojets
Commentary by Dr. Valentin Fuster
1985;():V001T01A005. doi:10.1115/85-IGT-83.
FREE TO VIEW

A rapid calculation procedure for design and off-design performance of turbojet and turbofan engine is developed. It peculiarity is that the general characteristics of components are established based on statistical data and the engine working conditions are searched according to matching of these general characteristics. This method can be used to select cycle parameters in engine design, and has been employed in engine performance calculation program used in the preliminary phase of engine design or airframe/engine integration design.

Topics: Engines
Commentary by Dr. Valentin Fuster
1985;():V001T01A006. doi:10.1115/85-IGT-139.
FREE TO VIEW

The JT15D Turbofan Engine was created to power fast economical low noise executive jets for the general aviation market and currently powers the Cessna Citation I & II, Mitsubishi Diamond series and the Aerospatiale Corvette. This paper describes a higher thrust more fuel efficient version, the JT15D-5, with electronic fuel control systems and power up to a thrust level of 3200 lb. The Engine and its development program are described, including flight test and customer installations. Development problems and their solutions are covered.

Commentary by Dr. Valentin Fuster
1985;():V001T01A007. doi:10.1115/85-IGT-150.
FREE TO VIEW

This paper presents a method to calculate the design point performance of both single rotation and dual contra-rotation propellers. Major propeller variables such as, power loading (shaft horsepower/prop diameter2), propeller efficiency (Prop Thrust × Acft Veloc/shaft horsepower), propeller tip speed and propeller adiabatic compression efficiency are accounted for and correlated. The resulting propeller performance is then combined with given shaft power producer (engine) performance to yield propfan engine performance. The performance trades between power loading, propeller efficiency, propeller tip speed and propeller adiabatic compression efficiency are presented. Resulting performance, with a given engine, are presented in the form of Figs 7 and 8 for both single and dual contra-rotation propellers. These engine performances can then be compared to advanced turbofan engines that utilize identical technology shaft power producer engines.

Commentary by Dr. Valentin Fuster
1985;():V001T01A008. doi:10.1115/85-IGT-151.
FREE TO VIEW

Allison currently has over 20,000 Model 250 small gas turbine engines operating throughout the world in a variety of turboprop and turboshaft installations. The experience base has contributed to the understanding and development of small engine installation requirements. Some of the critical interface systems will be reviewed to highlight some of the small engine considerations of engine installation design.

Topics: Gas turbines
Commentary by Dr. Valentin Fuster

Turbomachinery

1985;():V001T02A001. doi:10.1115/85-IGT-7.
FREE TO VIEW

Time-averaged and time-dependent velocity distributions have been measured in the the vaneless diffuser of an experimental compressor. Changes in these distributions have been identified with the onset of rotating stall and fully developed stall. At very small flow rates, an approach to equilibrium has been observed. Frequency spectra analyses have revealed the existence of low frequency disturbance waves which, together with the large adverse pressure gradient prevailing in the diffuser, could lead to selective wave amplification and surging.

Commentary by Dr. Valentin Fuster
1985;():V001T02A002. doi:10.1115/85-IGT-8.
FREE TO VIEW

An investigation was made to compare the performance of a highly-loaded transonic turbine stage with and without compound leaned vanes. In both cases, velocity distribution along the vane surfaces was calculated from a full 3 D time-marching finite volume method. Nozzles were tested in a wind tunnel. Through rig test, velocity profile at the stage exit was measured and the stage overall performance obtained. Performance in both tip and hub regions was improved by using the compound leaned vanes so that the stage efficiency increased by 1% approximately. The improvement is payticylarly remarkable at off-design points.

Topics: Turbines
Commentary by Dr. Valentin Fuster
1985;():V001T02A003. doi:10.1115/85-IGT-9.
FREE TO VIEW

The evolution of Controlled Diffusion Airfoils is traced from inception of the theoretical design model to demonstration of significant performance gains at engine operating conditions in a multistage compressor rig. The proven aerodynamic benefits and versatility of first-generation Controlled Diffusion Airfoil blade elements are extended to the endwall flow region using an Integrated Core/Endwall Vortex design model to produce a new full span optimized second-generation Controlled Diffusion design. Highlighted are the essential roles of extensive cascade, low speed, large scale and high Mach number compressor rig testing in developing and substantiating the second generation Controlled Diffusion technology resulting in a 1.5% increase in efficiency and 30% increase in surge-free operation relative to first-generation Controlled Diffusion Airfoils.

Commentary by Dr. Valentin Fuster
1985;():V001T02A004. doi:10.1115/85-IGT-15.
FREE TO VIEW

This paper provides a description of experimental studies of various dynamic distortion properties in a two-dimensional transonic diffuser. Based upon the measured dynamic pressure across the diffuser discharge section, the Δ PRMS, amplitude power density and probability density function have been analyzed. The results indicate that the dynamic distortion is closely related to the terminal shock wave stability of transonic diffuser, and the distribution of Δ PRMS and PSD are important Properties for dynamic distortion.

Commentary by Dr. Valentin Fuster
1985;():V001T02A005. doi:10.1115/85-IGT-16.
FREE TO VIEW

The effect of non-uniform inlet velocity and temperature profile on the aerodynamic performance of straight wall annular diffuser for turbofan augmentor has been investigated.

The distribution of static pressure, stagnation pressure and temperature has been measured, thus pressure recovery coefficient, velocity profile and temperature profile at different axial station along the diffuser center line can be determined.

The experimental results showed that the momentum ratio Display Formulaρ¯eV¯e2/ρ¯iV¯i2 of two streams across the diffuser inlet flow splitter is the non-dimensional flow parameter controlling diffuser aerodynamic performance. Thus, it is possible to simulate turbofan augmentor annular diffuser perfomance by using low temperature air flow aerodynamic test under the condition that the diffusers are of similar geometry, have the same inlet velocity profile and maintain the momentum ratio constant.

A correlation for the velocity distribution in the diffuser was also obtained.

Topics: Diffusers , Turbofans
Commentary by Dr. Valentin Fuster
1985;():V001T02A006. doi:10.1115/85-IGT-22.
FREE TO VIEW

A study is presented in the paper on the applicability of the surge/rotating stall criterion, or parameter B (B=U/2ωLc), to practical engineering problems. This criterion, while being of great theoretical value, appears to have some uncertainty in application to engineering compressor systems with the difficulties arising from changeable compressor characteristics and the diversity in the geometry and the time lag in different systems. Therefore, a series of calculations were conducted to show the influence of the above-mentioned factors. It is concluded from the numerical results that there exists a single critical value of parameter B for a given compressor working in possibly different systems while no identical critical value could be employed to different compressors in different systems.

Topics: Surges
Commentary by Dr. Valentin Fuster
1985;():V001T02A007. doi:10.1115/85-IGT-23.
FREE TO VIEW

A new method for predicting the off-design performance and the surge line of axial-flow compressors has been developed. In this method the stall points of stages are determined by generalized curves, which are derived from the criterion of rotating stall based on the neutral condition of stability, and the off-design characteristic of stages is calculated by a reasonable iterative method. The stage-stacking method is adopted to obtain the off-design performance of a multistage compressor. The surge line of the compressor is predicted with the “actuator-delay-volume” model and the method of numerical integration. Two examples are given. The results are satisfactory.

Commentary by Dr. Valentin Fuster
1985;():V001T02A008. doi:10.1115/85-IGT-24.
FREE TO VIEW

The “actuator-delay-volume” is used as a physical model for stages of compressors instead of “actuator-lumped volume” model to predict the stability of multistage compressors. The stall line of a multistage compressor is predicted with these two models respectively. The results of simulation are compared with the experimental data of a compressor rig. In order to identify the investigation, the tests, in which the engine is forced into stall and supplied with distorted inflow, have been conducted. The investigation shows that the model is well improved by incorporating an “inertia link” into the model between the “actuator” and the “lumped volume”.

Commentary by Dr. Valentin Fuster
1985;():V001T02A009. doi:10.1115/85-IGT-25.
FREE TO VIEW

This paper concentrates solely on the effect of end bend airfoil on performance of an axial flow compressor. The purpose is to reduce flow loss in the region of endwall boundary layer and to enlarge surge margin of an axial flow compressor by employing end bend airfoil on the stator blades in the last stages of the compressor. The experimental results show that surge margin is increased by 4.9–5.2 percent, efficiency by 2.6–3.2 percent at operational speed n=1.0–1.1.

Topics: Axial flow , Surges , Airfoils
Commentary by Dr. Valentin Fuster
1985;():V001T02A010. doi:10.1115/85-IGT-27.
FREE TO VIEW

Experimental datas were used to observe how the overall picture of the diffuser performance is affected by the presence of two different kind of swirl. The theoretical model considered were Cockrell & Markland (4), Sovran & Klomp (5), and Tyler & Williamson (6). In the case of Sovran & Klomp, the theoretical analysis used was one extended by Thakker (1) to take account of 2-D flow. It was found that the performance of the conical diffusers is best improved for a diffuser of total cone angle of 10 and 20 degrees by optimum Rankine vortex addition, and for cone angle of 30 degree by “solid-body” swirl 3.

Topics: Diffusers
Commentary by Dr. Valentin Fuster
1985;():V001T02A011. doi:10.1115/85-IGT-32.
FREE TO VIEW

This paper presents the calculation of optimal restagger and test method for ten-stage subsonic axial flow compressor with four rows of variable stator. The test results show that both are in close agreement.

The test results show that the overall compressor performance with optimal restaggers was significantly improved in comparison with the original. The efficiency at mid-speed increased by 12.7%. Surge margin reaches 31.2%. The total range of operation for compressor was expanded by 34%.

Topics: Axial flow , Stators
Commentary by Dr. Valentin Fuster
1985;():V001T02A012. doi:10.1115/85-IGT-34.
FREE TO VIEW

Means of predicting aerothermodynamic characteristics of turbines and expanders with gas-wet steam working medium and having condensation and release of latent heat of vaporization during expansion process are studied. After discussing the basic relations and fundamental equations of such process, a simple method is suggested on the basis of introducing the concept of replacing the condensing and heat release process thermodynamically by an equivalent polytropic process with an equivalent index.n. By such replacement the usual gas medium calculation program could be used and performance prediction could be easily made. Details of calculation procedures are recommended. The calculated performance of a blast furnace top gas turbo expander by this method shows good agreement with tested results.

Topics: Turbochargers , Steam
Commentary by Dr. Valentin Fuster
1985;():V001T02A013. doi:10.1115/85-IGT-35.
FREE TO VIEW

End wall total pressure losses downstream of a low-speed turbine cascade have been measured at several planes in order to determine the changes in secondary flow loss coefficients and the growth of the mixing loss with distance downstream. The results obtained are compared with various published secondary flow loss correlations in an attempt to explain some of the anomalies which presently exist. The paper includes some new correlations including one for the important gross secondary loss coefficient YSG with loading and aspect ratio parameters as well as the upstream boundary layer parameters.

Commentary by Dr. Valentin Fuster
1985;():V001T02A014. doi:10.1115/85-IGT-36.
FREE TO VIEW

This paper describes a correlation for the evaluation of the spanwise distribution of the secondary deviation angle. This correlation is made through the use of a theoretical method for the calculation of secondary flow which includes the effects of viscosity.

The correlation considers the influence of the spacing and height of the blade, inlet and outlet flow angles, inlet boundary layers and velocity ratio.

The results are compared with the experimental data of various turbine cascade. The agreement is good for most of the cases.

Commentary by Dr. Valentin Fuster
1985;():V001T02A015. doi:10.1115/85-IGT-41.
FREE TO VIEW

The largest member of the PT6 turboprop engine family, the PT6A-65, was developed in the early 1980’s and went into production in September 1982. The compressor for this engine consisted of four new axial stages combined with an existing centrifugal stage on a single shaft. This paper gives a brief description of the studies leading up to the choice of the compressor configuration and a more detailed examination of the development of the chosen compressor to the required performance level.

The development of this compressor presented a two-fold technical challenge. Firstly, the limited space in the small compressor gas path did not permit the effective use of conventional total pressure and temperature probes for performance evaluation. Secondly, the short time available for development excluded some attractive corrective measures such as the redesign of some of the axial blade rows because the time required would have jeopardized the meeting of the tight development deadline. The first problem was overcome by a combination of limited wall static pressure measurements and an extensive use of numberical flow analysis codes. This approach proved to be quite cost-effective. The second was solved by the adaptation of an existing fully analytically-designed research axial stage to the first stage position in the axial compressor.

Topics: Compressors
Commentary by Dr. Valentin Fuster
1985;():V001T02A016. doi:10.1115/85-IGT-42.
FREE TO VIEW

Through-flow analysis, which is at the heart of the aerodynamic design of turbomachinery, requires as aerodynamic input a row-by-row description of the airfoil loss, deviation, and blockage. Loss and deviation have been investigated extensively in both cascades and rotating rigs as well as in numerous two- and three-dimensional analytical studies. Blockage, however, has received far less attention. As defined herein, blockage is a measure of the departure of the flow field from the condition of axisymmetry which is assumed in the through-flow analysis. The fullspan blockage distributions calculated from measured single-stage rotor wake data were used to provide the input to the through-flow analysis, along with the measured fullspan distributions of loss and deviation. Measured and computed results are compared for the single-stage rotor operating with both thick and thin inlet hub and tip boundary layers. It is demonstrated that both the level and the spanwise and streamwise distributions of blockage have a strong impact on the computed rotor exit flow field.

Commentary by Dr. Valentin Fuster
1985;():V001T02A017. doi:10.1115/85-IGT-43.
FREE TO VIEW

An analytic model has been developed to predict the performance of axial compressors with an exit static pressure perturbation. The model uses a two dimensional compressible semi-actuator disc model. This method can be applied to the compressor with known circumferential variation in exit static pressure which is measured or predicted by an analytical method. The analytical results are found to be in good agreement with experiments carried out on two transonic fans.

Topics: Pressure , Compressors
Commentary by Dr. Valentin Fuster
1985;():V001T02A018. doi:10.1115/85-IGT-44.
FREE TO VIEW

Reliable cascade data are essential to the development of highspeed turbomachinery, but it has long been suspected that the tunnel environment influences the test results. This has now been investigated by testing one plane gas turbine rotor blade section in four European wind tunnels of different test sections and instrumentation. The Reynolds number of the transonic flow tests was Re2 = 8 · 105 based on exit flow conditions. The turbulence was not increased artificially. A comparison of results from blade pressure distributions and wake traverse measurements reveals the order of magnitude of tunnel effects.

Commentary by Dr. Valentin Fuster
1985;():V001T02A019. doi:10.1115/85-IGT-45.
FREE TO VIEW

The hydraulic analogy was employed in a rotating water table for simulating the compressible two dimensional flow in a low pressure turbine stage. Both steady and unsteady forces were measured directly on a rotating blade in a blade row rotating concentrically with a row of stator vanes. With proper modeling of the simulation, it is shown that the rotating water table can yield results that agree favorably with the analytical predictions and turbine test results. Using this test facility, the effects of axial spacing between rotor and stator rows on the nozzle wake excitation have been investigated for two different stator vane profiles. The water table test results correlate qualitatively with the turbine test data. The cancellation of nozzle passing frequency excitation by off-setting nozzle pitch was demonstrated in the water table and the results compared with both the analytical predictions and the laboratory turbine test results.

Topics: Pressure , Wakes , Nozzles , Turbines , Water
Commentary by Dr. Valentin Fuster
1985;():V001T02A020. doi:10.1115/85-IGT-63.
FREE TO VIEW

The purpose of this paper is to show, for both rotating and non-rotating blade rows, the importance of including circumferential non-uniform flow effects in a quasi-three-dimensional blade design system. The paper follows on from previous publications on the system in which the mathematical analysis and computerised system are detailed. Results are presented for a different stack of the nozzle guide vane presented previously and for a turbine rotor. In the former case it is again found that the blade force represents a major contribution to the radial pressure gradient, while for the rotor the radial pressure gradient it is dominated by centrifugal effects. In both examples the effects of circumferential non-uniformities are detailed and discussed.

Commentary by Dr. Valentin Fuster
1985;():V001T02A021. doi:10.1115/85-IGT-64.
FREE TO VIEW

In a test facility for rotating annular cascades with three conical test sections of different taper angles (0°, 30°, 45°), experiments are conducted for two geometrically different turbine cascade configurations, a hub section cascade with high deflection and a tip section cascade with low deflection. The evaluation of time averaged data derived from conventional probe measurements upstream and downstream of the test wheel in the machine-fixed absolute system is based on the assumption of axisymmetric stream surfaces. The cascade characteristics, i.e. mass flow, deflection and losses, for a wide range of inlet flow angles and outlet Mach numbers are provided in the blade-fixed relative system with respect to the influence of annulus taper. Some of the results are compared with simple theoretical calculations. To obtain some informations about the spatial structure of the flow within the cascade passages, surface pressure distributions on the profiles of the rotating test wheels are measured at three different radial blade sections. For some examples those distributions are compared with numerical results on plane cascades of the same sweep and dihedral angles and the same aspect ratios. The computer code used is based on a three-dimensional time-marching finite-volume method solving the Euler equations. Both experimental and numerical results show a fairly good qualitative agreement in the three-dimensional blade surface pressure distributions. This work will be continued with detailed investigations on the spatial flow structure.

Commentary by Dr. Valentin Fuster
1985;():V001T02A022. doi:10.1115/85-IGT-65.
FREE TO VIEW

The tip clearance flow region of high pressure axial turbine blades for small gas turbine engines has been investigated in a water flow cascade. The blade model features variable clearance and variable endwall speeds. The cascade is scaled for Reynolds number and sized to give velocities suitable for visualization. Pressure profiles were measured on one blade, and correlated with the visualization. Unloading is found to be a major feature of the pressure field at both tip and midspan, and is intimately connected with scraping effects and the behavior of the clearance vortex. Some initial hot film velocity measurements are also presented.

Commentary by Dr. Valentin Fuster
1985;():V001T02A023. doi:10.1115/85-IGT-68.
FREE TO VIEW

It is found in the experiments that blowing at the lip separation of the inlet obviously reduces the turbulences at the inlet exit, and apparently reduces the intensity of pressure fluctuations caused by the shock-boundary layer interaction down-stream of throat. The coherence between pressure in the interaction region and total pressure at the exit is also reduced.

The coherence between the pressure in the lip separation region and total pressure at the exit is 0.32. If, in addition, there is a stronger shock down-stream of the throat the above mentioned coherence is reduced to 0.06.

Commentary by Dr. Valentin Fuster
1985;():V001T02A024. doi:10.1115/85-IGT-69.
FREE TO VIEW

The flow in an elastic duct has some special properties. The speed of sound in this kind of flow is discovered to be smaller than in the flow in a rigid duct. Since the speed of sound decreases, flows which include area change, friction, heat transfer and mass addition reach the critical condition at Mach number less than one (M=1). It is particularly interesting that in a flow with area change the critical section doesn’t coincide with the minimum section. The main property of the flow in an elastic duct, that is the speed of sound being smaller than the traditional one, has been proved by experiment.

Commentary by Dr. Valentin Fuster
1985;():V001T02A025. doi:10.1115/85-IGT-70.
FREE TO VIEW

In the measurement of aerodynamic parameters behind rotors, measurement errors may be caused by rotation of rotors. These errors are define as rotation state errors which contain two types of errors: weighted error and response error. In this paper, the equations of rotation state errors are derived by the sweep in tegral of the square wave for the first order system. It provides a clear concept for the measurement errors behind rotors and presents the method and equations for analysing and estimating the rotation state errors.

Topics: Rotation , Rotors , Errors
Commentary by Dr. Valentin Fuster
1985;():V001T02A026. doi:10.1115/85-IGT-74.
FREE TO VIEW

This paper introduces the features of using co-rotating vortex generators for controlling boundary layer and flow field in the inlet without flow separation. The principles of the arrangements of the blades and selection of constructional parameters of the generators that are applied to create the transverse flow between the high and low pressure regions and to reduce the secondary flow losses are analysed. The experimental results show that when the appropriate parameters of the co-rotating vortex generators are chosen for the inlet subsonic diffuser with apparent high and low pressure regions, not only the nonuniformity of the flow field is greatly improved but also the dynamic performance of the flow at exit is slightly improved.

Commentary by Dr. Valentin Fuster
1985;():V001T02A027. doi:10.1115/85-IGT-75.
FREE TO VIEW

The aerodynamic characteristics of a variable area nozzleless volute (VAV) for small radial inflow turbines are studied with the aid of cold air tests in which flow parameters inside the volute channel and at the outlet of the nozzleless ring are measured for cases of various cross-sectional areas. The experimental investigation shows that the desired aerodynamic parameters, especially the outlet flow angle α1 as well as the mass flow rate range can be obtained by varying the croes-sectional area of the VAV. A reasonable approximate method for calculating the average outlet flow angle α1 is presented in which the variation of the circulation in the flow is taken into account.

Commentary by Dr. Valentin Fuster
1985;():V001T02A028. doi:10.1115/85-IGT-76.
FREE TO VIEW

This paper presents the use of a single slanted hot wire to measure the three velocity components of a swirling flow field. In its usage, the hot wire probe has been calibrated in a subsonic air jet for both speed and direction. It was then used to measure the swirling flow field associated with the presence of an inlet model. Using Hybrid Method, the final velocity components are obtained via the solution of a set of simultaneous nonlinear equations given the voltage readings at three orientations of the hot wire probe at a point in the flow field. The distortions of total pressure and velooity were also measured in the same inlet model.

Topics: Wire , Swirling flow
Commentary by Dr. Valentin Fuster
1985;():V001T02A029. doi:10.1115/85-IGT-79.
FREE TO VIEW

The quasi-orthogonal surface method of calculation of three-dimensional flow of references (1, 2, 3) is extended to take account of the influences of upstream and downstream flow and to include various losses in the impeller. By this method, the three-dimensional flow distribution of two impellers were calculated and the results were compared with experimental data. It was found that the calculation results of the present method which takes account of the gradient of entropy and the upstream and downstream conditions are in closer agreement with the experimental results than the previous method.

Commentary by Dr. Valentin Fuster
1985;():V001T02A030. doi:10.1115/85-IGT-80.
FREE TO VIEW

An “oscillating mean-streamline method” is presented in this paper. It is used to calculate the unsteady aerodynamic force on the oscillating high-turning cascade. An oscillating streamline physical model is discussed, basic relations of the oscillating streamline are derived, the calculations of several examples are given and are compared with corresponding experimental data.

The validity and reliability of this method is also analysed.

Commentary by Dr. Valentin Fuster
1985;():V001T02A031. doi:10.1115/85-IGT-81.
FREE TO VIEW

In this paper, a time marching method is proposed to predict the propagation of steady inlet circumferential distortion through an axial flow compressor rotor (or stator). This method also adopt the “Semi-actuator disk” concept to replace the rotor. It is assumed that the flow fields outside of the disk are described by a set of ideal, compressible and unsteady quasi-linear partial differential flow equations. Furthermore, the unsteady response of rotor is also taken into account. This method can be easily developed to a stage or extended to investigate the effect of transient disturbance on the stability of a compressor. Through numerical experiments, it has been found that distortion amplitude, rotation of the rotor, the chord length and especially the characteristic curves of the blade row have crucial effect on the attenuation behavior of the rotor.

Topics: Compressors , Rotors
Commentary by Dr. Valentin Fuster
1985;():V001T02A032. doi:10.1115/85-IGT-86.
FREE TO VIEW

This paper presents a method for the computer aided design of the transonic turbine blade profile shaping and aero thermodynamic analysis.

A new technique of “Computational Geometry” was developed for blade flow path shaping. An improved method of “Streamline Curvature” for cascade analysis is used for blade-to-blade stream surface analysis.

Using above method, a high-loaded turbine cascade with anti-convex suction surface near the trailing edge has been successfully designed, and appreciable supersonic performance was achieved through wind tunnel test.

This method has been included in our “Turbine Blade Computer Aided Design System (GTCAD)” as one of the core modules.

Commentary by Dr. Valentin Fuster
1985;():V001T02A033. doi:10.1115/85-IGT-89.
FREE TO VIEW

It is understood that the flow field calculation for the high-loaded super- or transonic compressor in inviscid gas model is essential in order to minimize the complexity of viscid calculation and, hereof the characteristic method has been often adopted for its relative effectiveness. However, due to the irregular grids and complex boundary conditions in turbomachines some difficulties still remain. This paper introduces a simpler and more practical characteristic method adopting reverse directional differentials for solving supersonic S2 flow field in supersonic or transonic compressors.

Commentary by Dr. Valentin Fuster
1985;():V001T02A034. doi:10.1115/85-IGT-90.
FREE TO VIEW

A mixed finite difference method for calculating the external and internal transonic flow field around an s-shaped inlet is presented. Starting from the velocity potential equation and using Cartesian mesh and mixed finite difference schemes, the authors have obtained a system of finite difference equations and solved them with the aid of alternating line relaxations along two directions. Computations have been made for an s-shaped inlet with free stream Mach number M=0.8 at different angles of attack. Computed results are compared with those computed by perturbation method and with experimental results. Such a comparison shows that the present method is promising.

Topics: Transonic flow
Commentary by Dr. Valentin Fuster
1985;():V001T02A035. doi:10.1115/85-IGT-93.
FREE TO VIEW

For studying the secondary flow in a turbine cascade, the flow field is measured in detail. The measurements of pressure and velocity are taken at various axial planes upstream of, within, and downstream of the cascade by a 4-hole probe. The static pressures are taken on the endwall, suction and pressure surfaces. By treating the experiment data the mechanism of the secondary flow field and the loss model are proposed in this paper.

Commentary by Dr. Valentin Fuster
1985;():V001T02A036. doi:10.1115/85-IGT-94.
FREE TO VIEW

24 double circular are tandem blade cascades of three different chord-ratios were investigated under different displacements in peripheral and axial direction. The inlet Mach number was 0.3. The Reynolds number based on blade chord was 2.7×105. The characteristics of the tandem blade cascades, such as the dependence of turning angle and coefficient of total pressure loss on incidence angle were obtained. The ranges of main geometrical parameters under optimal conditions were recommended.

Topics: Blades
Commentary by Dr. Valentin Fuster
1985;():V001T02A037. doi:10.1115/85-IGT-96.
FREE TO VIEW

This article covers the analysis and study of the surge characteristic in a full-scale 10-stage subsonic compressor under the condition of performance test. The transient responses which arised after the compressor enters in non-stationary condition range are estimated through an improved nonlinear calculation model. There are two different kinds of response (or transient processe) model: stall pattern and surge pattern. The calculated results coincide with the test very well. When relative turning speed ñ ≤ 0.5, the system transient process is shown in stall pattern; when ñ ≥ 0.6, it is surge pattern. The obtainted calculation result has been drawn as a system transient process locus in the nondimensional flow-pressure rise phase plane. The nondimensional parameters affecting system transient process pattern were analysed and compared with the results made by E.M. Greitzer.

Topics: Compressors , Surges
Commentary by Dr. Valentin Fuster
1985;():V001T02A038. doi:10.1115/85-IGT-97.
FREE TO VIEW

This paper provides a method of modelling the axial-flow compressors in the low speed starting regime of an engine from windmilling to idling. A structural formula for the model is established by means of reference (1). A method of step-by-step regression is provided by the author for determining the coefficient matrices of the structural formulae. Excellent agreement was obtained between the computational and experimental results.

Commentary by Dr. Valentin Fuster
1985;():V001T02A039. doi:10.1115/85-IGT-102.
FREE TO VIEW

A single-rotor axial flow compressor and a two-dimensional cascade have been tested with and without casing treatment. A three-dimensional flow mechanism of the onset of rotating stall is suggested. It gives good reason to explain the mechanism of stall-margin improvement of casing treatment.

Commentary by Dr. Valentin Fuster
1985;():V001T02A040. doi:10.1115/85-IGT-108.
FREE TO VIEW

Tests have been conducted on an annular, Hybrid Diffuser, of short length while operating in isolation. The purpose was to ascertain whether such a diffuser could operate efficiently without the downstream influence of the combustor dome which had been a feature of previous tests. The diffuser was fitted with a pre-diffuser in order to raise the static pressure level of the bleed air. A series of exit cones was used as the final stage of diffusion, covering a range of included angle from 15 to 45 degrees.

Results obtained demonstrate that this configuration of Hybrid diffuser operates efficiently, that performance is improved if pressures inside the inner and outer vortex chambers are balanced, and that larger quantities of bleed must be extracted in order to maintain a high pressure recovery as the exit included angle is increased.

Topics: Diffusers
Commentary by Dr. Valentin Fuster
1985;():V001T02A041. doi:10.1115/85-IGT-109.
FREE TO VIEW

Tests have been conducted on a Conical Hybrid Diffuser of short length in order to determine the effects of a range of inlet flow distortions of an axisymmetric nature.

As with most other types of diffuser, the performance was impaired as the level of distortion was increased, although some restoration could be obtained by either increasing diffuser length or rate of bleed-off.

In general, diffuser performance, in terms of flow stability and pressure recovery, was satisfactory.

Topics: Diffusers , Inflow
Commentary by Dr. Valentin Fuster
1985;():V001T02A042. doi:10.1115/85-IGT-113.
FREE TO VIEW

A Clausius-Rankine-cycle has been proposed to recover waste heat from a piston engine. This waste heat is then used to supercharge the cylinders by means of a steam turbocharger. The advantage of using this steam turbocharger system is to avoid the losses due to the engine back pressure which accompany the use of the conventional exhaust gas turbocharger. The mass flow rate of turbines for steam turbochargers in the range from 1 to 10 kW is about 0.03 to 0.08 kg/s. This implies a special turbine design, characterised by partial admission and supersonic flow, which unfortunately leads to low turbine efficiencies. A small Pelton turbine for steam has been designed and produced. The turbine is connected to the radial compressor of a conventional exhaust gas turbocharger which works, in this case, as a brake to dissipate the generated turbine power. A special test rig has been built to carry out the experimental investigations on the proposed Pelton turbine. The test rig is supplied with superheated steam from the University’s power plant. Two different rotors for this Pelton turbine have been tested under the same operating conditions (rotor 2 see Fig. 1). Some experimental test results of a special Pelton turbine are presented and discussed in this report.

Commentary by Dr. Valentin Fuster
1985;():V001T02A043. doi:10.1115/85-IGT-119.
FREE TO VIEW

Making use of functional variations with variable domain and natural boundary conditions, this paper presents a unified theory of various hybrid problems (being a unification as well as a generalization of direct- & inverse problems) for 3-D incompressible potential flow in a rotor-blading. Three families of variational principles (VPs) have been established and provide a series of new rational ways for blade design and a sound theoretical basis for the finite element method (FEM). This theory can be extended to compressible and rotational flows and also constitutes an important part of the optimum design theory of bladings (8).

Commentary by Dr. Valentin Fuster
1985;():V001T02A044. doi:10.1115/85-IGT-120.
FREE TO VIEW

Turbulent vortices occur everywhere in flowing fluids and possess the properties of dissipation and dispersion. A set of new control equations is presented featuring the interaction between dissipation and dispersion of turbulence. By analysis of instability the rate of turbulent energy production is established. Two 3rd order derivative momentum equations are derived, one for weak and the other for strong vorticity. By this new theory various turbulent flow problems can be solved such as: energy inversion in the vortex tail behind a bluff body, the coherent horseshoe vortices in a turbulent boundary layer, the delay in cascading down of turbulent energy through the spectrum, anisotropy of turbulence intensities, etc. Two computational examples are presented.

Commentary by Dr. Valentin Fuster
1985;():V001T02A045. doi:10.1115/85-IGT-123.
FREE TO VIEW

A new method which can be used to construct a H-type grid for cascade flow calculation is presented in this paper. To test the feasibility of the grid, the subsonic and transonic full potential equations are solved by using finite difference approximations on a transformed coordinate system. It is found that in order to improve computing accuracy and reduce cost, four conditions should be fulfilled in constructing cascade grids. Also, the artificial viscosity scheme should be improved for cascade calculations.

Commentary by Dr. Valentin Fuster
1985;():V001T02A046. doi:10.1115/85-IGT-131.
FREE TO VIEW

A finite element scheme for two dimensional incompressible viscous flows in primitive variables is proposed in this paper. An upwind factor finite element method is devised to solve the momentum equations, and the continuity equation is satisfied by the correction of the pressure field.

Numerical experiments are carried out for a driven cavity and a diffuser. The Renolds Number for the cavity flow is 100.0, and for the diffuser is 50000.0. The numerical result of the scheme for the cavity flow is compared with that by another numerical method and satistactory agreement is found.

Commentary by Dr. Valentin Fuster
1985;():V001T02A047. doi:10.1115/85-IGT-132.
FREE TO VIEW

This paper presents a method for predicting the blade root loss in an annular nozzle cascade in which the consideration is given to the influence of the radial pressure gradient (RPG) on it. The variation of blade root losses under different RPG is obtained experimentally and finite element method is used to calculate the pressure distribution in the blade passage.

Commentary by Dr. Valentin Fuster
1985;():V001T02A048. doi:10.1115/85-IGT-133.
FREE TO VIEW

Time marching solutions of transonic flow are widely used for the analysis of turbomachinery cascades. Attempts to increase efficiency, accuracy and computational speed are, at present, the main goal of research in the field of two dimensional flow. The aim of the paper is the presentation of a new pseudo-time dependent method to try to achieve these goals. The method, basic idea and procedure used to develop the new formulation are described and discussed in brief. Some applications of the method to transonic turbine cascades are presented against experimental results. The accuracy and speed of the method is discussed and possibilities for further developments are shown.

Commentary by Dr. Valentin Fuster
1985;():V001T02A049. doi:10.1115/85-IGT-135.
FREE TO VIEW

A large scale (2 inch wide test vanes at 50-inch inlet diameter), low speed (100–200 ft/sec) steady flow radial cascade wind tunnel for diffuser studies was designed, built and tested. The apparatus was shown to provide flow angles from radial in the range 58–72 degrees with suitable spanwise profiles. The novel flow angle control mechanism was shown to work but measured flow angles were somewhat smaller than expected. The mechanism for controlling case-wall boundary layer profile did not behave as predicted. Attempts to predict the generated flow both analytically and with computational codes are compared with initial measurements. The low speed apparatus will be used to obtain detailed data for diffuser design and analysis code verification, and to provide experience toward the design of a high speed device.

Topics: Diffusers
Commentary by Dr. Valentin Fuster
1985;():V001T02A050. doi:10.1115/85-IGT-146.
FREE TO VIEW

The effects of varying radius along a streamsurface are included in a two dimensional blade-to-blade potential flow analysis of turbomachinery rotors and stators in order to model better three dimensional effects. On a cylindrical streamsurface, flows in rotors can be treated as steady flows in the rotating frame; the rotation can be ignored. On a streamsurface with varying radius, however, the flow, even in the rotating frame, is not potential. Thus the physical flow is related to a non-physical, potential flow for calculation purposes. Computed results for compressor stators and rotors show that the potential solution agrees well with an Euler solution, that effects of radius variation are very important for both stators and rotors, and that effects of radius contouring and of rotation can also be important.

Commentary by Dr. Valentin Fuster
1985;():V001T02A051. doi:10.1115/85-IGT-148.
FREE TO VIEW

A variational finite element method for solving the blade-to-blade flow in centrifugal compressor’s cascades with splitter blades on an arbitrary streamsheet of revolution is suggested in this paper. At first, the variational principles Ref.(1) is modified, then the variational principle after modification is discretized by eight node isoparametric finite elements to carry out the system of nonlinear algebraic equations for solving the velocity potential function. Finally, the flow field which agrees with Kutta condition and has an region behind the cascade of enough length has been worked out.

In this paper, it has been discovered that when the region behind cascade L3 spreads too long the system of equations might become unsolvable as a suitable exit angle β2 can't be found. The linear relation between the velocity defference of the two side of the trailing edge and the exit angle β2 has been found, it shows the range of linear variation of β2 decreases with the increasing of the length of the region behind cascades, in addition, the linear variational relation between Display Formulatanβ2 and L3 has also been obtained.

The iterative computational method for the flow field with different length L3 is used to get the solution of flow field satisfying Kutta condition and with enough length L3.

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