J. Eng. Power. 1981;103(2):257-263. doi:10.1115/1.3230713.

This paper describes the results of an experimental study of the effect of Mach number and end wall cooling on the secondary flow in a straight nozzle cascade with an aspect ratio of h/c = 0.83. The tests were performed for the outlet Mach numbers M2 = 0.1, 0.6 and 0.8. The cooling tests are done for one specific cooling configuration consisting of three double rows of injection holes positioned at the leading edge, inside the blade passage and in the throat. The flow was injected into the direction of the free stream direction. The tests are carried out for two cooling mass flow ratios: ṁ/ṁtot = 0.02 and 0.045. The corresponding injection to main flow pressure ratios are: 1.0, 1.16 for M2 = 0.6 and 1.0, 1.26 for M2 = 0.8. The flow surveys are made with three-directional pressure probes.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):265-270. doi:10.1115/1.3230716.

With increasing interest in the burning of coal in industrial gas turbines, there is also concern for the precise determination of the erosive effects on the turbine components. Series of experiments were conducted to determine the effects of fly ash constituents, particle size, particle velocity, angle of attack and target temperature on the erosion of iron and nickel base alloys. Based on the experimental results, a semi-empirical equation has been obtained for the prediction of the erosion losses. This equation provides a new technique for predicting the metal erosion due to the fly ash produced by the conventional burning of coal.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):271-278. doi:10.1115/1.3230717.

The pressure distribution along the shroud is measured for three types of centrifugal impeller at seven different values of tip clearance each. The change of input power due to a change of tip clearance is related to the effective blockage at the impeller tip. Since the change of input power is little for the test cases, the variation of local pressure gradient along the shroud is mostly attributed to the change of local pressure loss. The local pressure loss is related to the local tip clearance ratio and to the local pressure gradient based on the deceleration of relative velocity in the impeller. Since the local pressure gradient is largest near the throat of the impeller, for many impellers the clearance ratio at the throat is used as the representative value. The tip clearance loss is related to the clearance ratio and the pressure rise based on the deceleration of relative velocity in the impeller. A good correlation is observed in all cases at various flow rate.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):279-287. doi:10.1115/1.3230718.

A general analysis is proposed for studying the fluid-mechanical behavior of blade wakes from an annular blade-row in highly swirling flow. The coupling between the centrifugal force and the vorticity, which is inherent to highly swirling flows, can significantly modify the wake behavior from that in a two-dimensional situation. In steady flow, theoretical considerations show that a blade wake consists primarily of two distinct types of vorticity: (1) trailing vorticity shed from the blade due to a spanwise variation in blade circulation; and (2) vorticity associated with defects in stagnation pressure (or rotary stagnation in relative coordinate system). Three types of disturbances can be identified in the resulting three-dimensional disturbance field: (1) the exponentially decaying type (potential, irrotational), (2) the purely convected type (rotational), and (3) the nonconvected type (both rotational and irrotational parts). Type (3) arises because of the interaction of centrifugal and Coriolis forces with (1) and (2). It is found that near the blade row the nonconvected disturbances grow linearly in magnitude with the axial distance. However, although those nonconvected disturbances associated with the trailing vorticity (also called Beltrami vorticity) persist for moderate distances downstream, they eventually decay inversely with the axial distance, irrespective of the types of swirl distribution. In contrast, those parts of nonconvected disturbances which are induced by the vorticity caused by (rotary) stagnation pressure defects persist indefinitely downstream for any type of swirl other than free-vortex. In the limit of free-vortex swirl, all disturbances decay at least inversely with the axial distance downstream.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):288-293. doi:10.1115/1.3230719.

A published history of rotordynamic instability problems with centrifugal compressors is documented from the literature. Established theory for computerized stability analysis is reviewed, and the use of cross-coupled stiffness and damping coefficients to represent destabilizing forces is explained. A derivation of cross-coupled stiffness coefficients for torquewhirl is presented. Experimental measurements made on a simple test rig with a radial-vaned impeller are described, which show that a working fluid can exert destabilizing forces on such an impeller. These forces must be considered, in addition to journal bearing and internal hysteresis excitations, if accurate predictions of rotordynamic instability thresholds are to be made. Several classifications of these forces are hypothesized.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):294-307. doi:10.1115/1.3230720.

The effects of potential impurities, such as Na, K, Mg, Ca, and Cl, in coal-derived liquid fuels on accelerated corrosion of IN-100, U-700, IN-792, and Mar M-509 were investigated using a Mach 0.3 burner rig for times to 200 hr in 1 hr cycles. These impurities were injected in combination as aqueous solutions into the combustor. Other variables were time, temperature, and fuel-to-air ratio. The experimental matrix was a central composite fractional factorial design divided into blocks to allow modification of the design as data was gathered. The extent of corrosion was determined by metal consumption, τ.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):308-312. doi:10.1115/1.3230721.

Fatigue cracking of an aircraft engine labyrinth seal occurred during pre-flight factory testing. Testing in a static rig revealed that the seal could be aeroelastically excited by the labyrinth leakage air flow. An earlier analytical model used for stability analysis was extended to account for the effect of acoustic natural frequency on the aeroelastic stability. The new model predicted that the ratio of acoustic and mechanical natural frequencies was of vital importance in determining if the nature of the pressure fluctuations within the labyrinth seal teeth provided either positive or negative aerodynamic damping to the seal. The analytical results were verified by further rig testing and also by correlation with test results for several other seals tested as part of a labyrinth seal technology program. A mechanical friction damper sleeve was designed to suppress the aeroelastic instability. The damper sleeve was tested in a rotating rig to evaluate its damping characteristics. The aircraft engine was qualified with the newly designed damper which has demonstrated its effectiveness for eight years of service and half a million hours of operation without incident.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):313-318. doi:10.1115/1.3230722.

An earlier paper analyzed aeroengine fan flutter in terms of twin orthogonal vibration modes in the fan. This paper extends the analysis by incorporating variable aerodynamic phase lags between a blade motion and the resulting aerodynamic forces. The extended analysis predicts two independent kinds of flutter. If the total damping within each mode is negative, each mode develops a self-sustaining vibration at its own natural frequency. This was not predicted by the earlier twin mode analysis. Even when the damping in each mode is positive an interaction between the modes can cause flutter under some conditions. This behavior is basically of the form described in the earlier paper, but there are significant detailed changes when variable aerodynamic phase lags are introduced.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):319-325. doi:10.1115/1.3230723.

A previously developed technique for redesigning the vibrational properties of structures, by inverting the first-order perturbation analysis of the equations of motion, has been applied to a NASTRAN finite element analysis for plates and shells. The program finds the minimal changes to the thicknesses of the plate elements necessary to effect a given set of changes in the modal frequencies and shapes. Results have been obtained for a flat cantilever plate, a cantilever segment of a cylinder, and for a compressor blade for a jet engine.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):326-330. doi:10.1115/1.3230724.

An extension of the author’s earlier method of analyzing multi-shaft jet engine dynamics accounting for flexibility of bladed disks is outlined to calculate the first whirl harmonic for given nonlinear characteristics of squeeze-film dampers. A second whirl harmonic, of which experimental verification is found in Campbell’s paper of 1924, is shown induced by orbit ellipticity. The possibility that this harmonic, especially due to backward whirling, may be a source of blade excitation at higher frequencies than currently recognized from linear analysis is discussed by relating some engine experience.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):331-337. doi:10.1115/1.3230725.

Five turbine materials, IN100, 713LC, MAR-M-509, MA-754 and TZM were selected as candidate materials for use in a Compact Closed Cycle Brayton System (CCCBS) study in which helium served as the working fluid. The suitability of the alloys to serve in the CCCBS environment at 927 C (1700 F) was evaluated on the basis of creep-rupture tests conducted in air, ultra-high purity helium (>99.9999 percent), and a controlled impurity helium environment. Baseline reference creep rupture properties for times up to 10,000 hr were established in a static ultra-high purity helium environment.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):338-344. doi:10.1115/1.3230726.

Analysis and design of a catalytic ceramic element and its support structure in a turbine combustor for low emission application have been performed. Preliminary analysis including a survey of literature has helped identify certain design considerations and conceptual designs of the catalytic ceramic element. A thermo-mechanical analysis of the major components in these conceptual designs has been performed for both steady-state and transient (shut-down) situations. Consequently, an arrangement to build a viable catalytic ceramic combustor element has been identified which is expected to perform its mechanical functions.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):345-347. doi:10.1115/1.3230727.

When testing large fans in the field, it is often necessary to make measurements at locations where the distributions of gas velocity, pressure, and temperature are highly irregular. When this occurs the fan pressure rise or fan specific energy is usually expressed in terms of average values of pressure, density, and temperature. This paper proposes criteria for the definition of such averages and develops defining equations for them. The averages proposed allow the continuity, energy, and ideal gas state equations to be written in simple one-dimensional form without correction factors. Performance variables such as total pressure are defined in the customary way from the averages of the fundamental flow properties.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):348-357. doi:10.1115/1.3230728.

There is no question that, with the increasing costs of Fossil-Fuels, overall production of electricity, and replacement energy costs, the need to develop an operator’s complete familiarity with the specific plant he will be required to operate is a necessity. The major benefit is that of reduced operator errors which maximizes unit availability as dictated to by the operating staff. The approach presented utilizes the concept of developing an operator’s understanding of each system within the plant through the use of a visual tool called an operational schematic. Coupled with the use of condensed word instructions to explain both normal and abnormal operator actions in addition to a system’s control logic, the operator experiences an environment very similar to his working atmosphere. Finally, the effectiveness of this approach, considering that classroom training is coupled with continual control room and in-plant inspections, can be seen to be competitive with that obtainable from the use of a simulator but for a fraction of the cost.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):358-366. doi:10.1115/1.3230729.

In order to optimize a steam generator tube bundle support system, it is necessary to understand the differences in vibration behavior of tube arrays subjected to a two-phase flow regime as compared to a single phase flow regime. The relationships discussed in this paper are based on findings derived from a comprehensive vibration testing program which included both water and simulated two-phase (air-water) flow regimes. Tube bundles of various configurations and spacings were tested in crossflow using 3/4 in. (1.9 cm) and 7/8 in. (2.2 cm) o.d. tubes with 36 in. (91 cm) span lengths and as many as 40 tubes per array. Sufficient test loop capacity was provided to drive most test array configurations up to and beyond fluid-elastic instability. Parameters obtained included displacement, effective force coefficients, damping coefficients, Strouhal numbers, and instability constants for various tube array configurations in both liquid and simulated two-phase flow mediums.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):367-372. doi:10.1115/1.3230730.

A partially-parabolic calculation procedure is used to calculate flow in a centrifugal impeller. This general-geometry, cascade-flow method is an extension of a duct-flow calculation procedure. The three-dimensional pressure field within the impeller is obtained by first performing a three-dimensional inviscid flow calculation and then adding a viscosity model and a viscous-wall boundary condition to allow calculation of the three-dimensional viscous flow. Wake flow, resulting from boundary layer accumulation in an adverse reduced-pressure gradient, causes blockage of the impeller passage and results in significant modifications of the pressure field. Calculated wake development and pressure distributions are compared with measurements.

Commentary by Dr. Valentin Fuster

RESEARCH PAPERS: Symposium Papers: Measurement Methods in Rotating Components of Turbomachinery

J. Eng. Power. 1981;103(2):374-392. doi:10.1115/1.3230732.

The objective of this paper is to review the techniques employed for the measurement of flow in turbomachinery rotors. Only nonoptical techniques are included. Measurement of the following properties are covered: three-dimensional mean velocity, turbulence intensity and Reynolds stresses, static and stagnation pressures, and blade surface measurement. Both the conventional as well as the hot-wire/film techniques, including both the rotating- and the stationary-probe measurements, are reviewed. A brief description of various rotating-probe traverse mechanisms and rotor flow data transmission systems in use is also included.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):393-399. doi:10.1115/1.3230733.

The flow from a small forward curved fan rotor was studied to provide guidance for fan design. The single width fan had a 0.23 m rotor containing 48 blades. It is typical of the centrifugal fans used in small air-conditioning units. Visualization techniques and hot wire measurements showed the rotor flow to be highly turbulent and strongly three-dimensional. The inlet flow was found to fill only 3/4 of the blade span. The shroud end of the rotor was an inactive or separated region. A jet-wake pattern occurred at the blade exit in the active flow region. The inlet flow separated at the leading edge of the blade suction surface. Design changes are offered to improve the performance of FC fans.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):400-405. doi:10.1115/1.3230734.

Measurement methods for obtaining various types of experimental data on a turbine rotor blade are discussed in this paper. A variety of different types of measurements have been taken in the rotating frame of reference, including: airfoil surface static pressure distributions, the radial distribution of total pressure in the incident flow, flow visualization of surface streamlines, and radial-circumferential traversing of a pneumatic probe aft of the rotor. Typical results are presented/showing interesting flow phenomena present on the rotor. In particular, results are shown which demonstrate the various viscous and inviscid mechanisms that give rise to strong radial flows.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):406-414. doi:10.1115/1.3230735.

This paper reports the measurement of the relative flow in the rotor blade passages of an axial-flow compressor and an axial-flow inducer using conventional probes such as five-hole, disk, and spherical pitot-static probes. The five-hole probe provides an inexpensive, yet accurate, method of deriving the three-dimensional flow field in a rotor; the disc probe is suitable for the blade boundary-layer measurement; and the pitot-static probe is useful for obtaining the static and stagnation pressures and the total velocity across the rotor passage. Typical data obtained from each of these probes demonstrate the complex nature of the turbomachinery rotor flow. A detailed discussion of various sources of probe errors and methods of estimating these errors is also included.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):415-423. doi:10.1115/1.3230736.

This paper describes the measuring methods developed at the ONERA Aerodynamics Department for the characterization of unsteady flows in turbomachines. They mainly concern the flow in the rotor, its wake, and boundary layers on stator or rotor blades. The means used consist of: • measurements using pressure probes or short response time pressure transducers, • measurements with hot wire probes or hot film gauges, and • direct, measurements using laser velocimeter.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):424-429. doi:10.1115/1.3230737.

Innovative features of the anemometer include: (1) a rapid and efficient data acquisition process, (2) a detailed real-time graphic display of the data being accumulated, and (3) input laser beam positioning that maximizes the size of the intra-rotor region being mapped. Results are presented that demonstrate the anemometer’s capability in flow mapping within a transonic axial-flow compressor rotor. Typically, a velocity profile, derived from 30,000 measurements along 1000 sequential circumferential positions covering 20 blade passages, can be obtained in 30 s. The use of fluorescent seed particles allows flow measurements near the rotor hub and the casing window.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):430-437. doi:10.1115/1.3230738.

A laser anemometer system employing an efficient data acquisition technique has been used to make measurements upstream, within, and downstream of the compressor rotor. A fluorescent dye technique allowed measurements within endwall boundary layers. Adjustable laser beam orientation minimized shadowed regions and enabled radial velocity measurements outside of the blade row. The flow phenomena investigated include flow variations from passage to passage, the rotor shock system, three-dimensional flows in the blade wake, and the development of the outer endwall boundary layer. Laser anemometer measurements are compared to a numerical solution of the streamfunction equations and to measurements made with conventional instrumentation.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):438-444. doi:10.1115/1.3230739.

A Laser Transit Anemometer (LTA) system for measurement of velocity, flow angle and turbulence in the rotating blade rows of turbomachinery and for other applications has been developed by the authors (SDL). Advanced optical, electronic, and computer components make the system sensitive to submicron scatterers, provide automated data collection and permit probing close to blade and wall surfaces, in some cases without seeding the flow. This paper describes new and practical developments of the optical system and presents test results from axial compressor measurements. Based upon these test results, improvements were incorporated into the system and these improvements are discussed briefly.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):445-450. doi:10.1115/1.3230740.

As part of a study on the structure of a trailing vortex, laser doppler anemometer (LDA) measurements were made of the flow field near an open rotor having an inlet velocity gradient. The measurements were made in the 1.22 m dia water tunnel of the Applied Research Laboratory at The Pennsylvania State University. Velocity data were obtained for rotor inlet and outlet flow fields for several different inlet velocity gradients. Velocity data were also obtained downstream of the rotor plane that shows the vortex structure. Flow field measurements show the development of the downstream vortex motion. Small variations in the inlet velocity gradient near the rotor wall caused large differences in the structure of the trailing vortex. In addition, a measured downstream velocity profile is compared with a calculated velocity profile.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):451-456. doi:10.1115/1.3230741.

Using laser Doppler anemometry, two components of the gas velocity have been measured at the inlet of a centrifugal compressor impeller, operated at speeds typical of service conditions for a medium-sized turbocharger. The flow was found to be unstable, especially adjacent to the suction side of the blades, such that two predominant conditions existed in the flow. The unstable flow is illustrated in the paper by distributions of relative velocity and relative flow angle, and the effects of different operating conditions on these distributions are examined. The instability is believed to be caused by a pre-stall condition as the compressor operating point approaches a fully stalled condition which occurs during surge.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):457-460. doi:10.1115/1.3230742.

The need for blade tip clearance instrumentation has been intensified recently by advances in technology of gas turbine engines. A new laser-optical measurement system has been developed to measure single blade tip clearances and average blade tip clearances between a rotor and its gas path seal in rotating component rigs and complete engines. The system is applicable to fan, compressor and turbine blade tipe clearance measurements. The engine mounted probe is particularly suitable for operation in the extreme turbine environment. The measurement system consists of an optical subsystem, an electronic subsystem and a computing and graphic terminal. Bench tests and environmental tests were conducted to confirm operation at temperatures, pressures, and vibration levels typically encountered in an operating gas turbine engine.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):461-467. doi:10.1115/1.3230743.

Engine-order sampling was used to eliminate the integral harmonics from the flutter spectra corresponding to a case-mounted static pressure transducer. From the optical displacement data it was demonstrated that blade-order sampling of pressure data can yield erroneous results because of the interference caused by blade vibration. Two methods are presented that effectively eliminate this interference and yield the blade-pressure-difference spectra. The phase difference between the differential-pressure and displacement spectra was evaluated.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):468-472. doi:10.1115/1.3230744.

A high pressure (4 × 106 N/m2 ), high-temperature (2400 K) turbine test facility is being constructed at the NASA Lewis Research Center for use in turbine cooling research. Several recently developed temperature and pressure measuring techniques will be used in this facility. This paper will briefly describe these measurement techniques, their status, previous applications and some results. Noncontact surface temperature measurements will be made by optical methods. Radiation pyrometry principles combined with photoelectric scanning will be used for rotating components and infrared photography for stationary components. Contact (direct) temperature and pressure measurements on rotating components will be handled with an 80-channel rotary data package which mounts on and rotates with the turbine shaft at speeds up to 17,500 rpm. The data channels are time-division multiplexed and converted to digital words in the data package. A rotary transformer couples power and digital data to and from the shaft.

Commentary by Dr. Valentin Fuster
J. Eng. Power. 1981;103(2):473-479. doi:10.1115/1.3230745.

Measurements made on the rotating components of aero-gas turbines are among the most difficult of instrumentation problems. Success brings considerable reward, however, by providing a means of verifying design predictions and investigating unforeseen phenomena. This paper describes the Rolls-Royce approach to this problem and the principles used in the design of over 30 radio telemetry systems designed and manufactured within the company. A summary of operating experience reviews the problems encountered and measures taken to overcome them. The paper concludes with an impression of future requirements which are receiving attention.

Commentary by Dr. Valentin Fuster



J. Eng. Power. 1981;103(2):373. doi:10.1115/1.3230731.
Topics: Turbomachinery
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