0


RESEARCH PAPERS

J. Turbomach. 1987;109(2):151-154. doi:10.1115/1.3262077.

This paper presents the test performance of a lightly loaded, combination radial/axial turbine for a 420-hp, two-shaft gas turbine. This two-stage turbine configuration, which included an interstage duct and an exhaust duct discharging vertically to ambient pressure conditions, was shown to be capable of attaining an overall isentropic efficiency of 89.7 percent. The influence of exhaust diffuser struts on the turbine performance under stalled power turbine conditions was shown to significantly affect compressor and turbine matching.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):155-161. doi:10.1115/1.3262078.

This paper describes the performance of a highly loaded single-stage transonic turbine with a pressure ratio of 3.76 and a stage loading factor of 2.47. Tests were carried out with three rotors, covering a range of blade Zweifel coefficient of 0.77 to 1.18. Detailed traversing at rotor inlet and exit allowed an assessment of rotor and stage performance as a function of blade loading under realistic operating conditions. The effect of stator endwall contouring on overall stage performance was also investigated using two different contours with the same vane design.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):163-169. doi:10.1115/1.3262081.

The results of an experimental study of the three-dimensional flow field in a radial inflow turbine scroll are presented. A two-color LDV system was used in the measurement of three orthogonal velocity components at 758 points located throughout the scroll and the unvaned portion of the nozzle. The cold flow experimental results are presented for through-flow velocity contours and the cross velocity vectors.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):170-176. doi:10.1115/1.3262082.

The interaction between consecutive blade rows can be expected to have important effects on the heat transfer in cooled gas turbine cascades. In determining the local heat transfer under the influence of wake flow, nonintrusive optical measuring techniques were used to obtain the flow velocities and turbulence structures in the cascade inlet flow as well as along the test blade’s surface. The main purpose of the measurements is to provide accurate experimental data for the development of predictive codes. The applicability primarily of the laser-Doppler technique is discussed and problems arising from the use of laser-dual-focus anemometry are reported. In simulating the effects of wake flow, a plane airfoil was traversed in front of the cascade. Both the axial distance between the airfoil and the cascade and the position in circumferential direction were changed in discrete steps. Turbulence intensities between 1.4 and 15 percent were recorded in cold gas flow. The effects on the blade heat transfer are illustrated.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):177-185. doi:10.1115/1.3262083.

This paper describes an experimental study of the three-dimensional flow within a high-speed linear cascade of low-pressure turbine blades. Data were obtained using pneumatic probes and a surface flow visualization technique. It is found that in general, the flow may be described using concepts derived from previous studies of high-pressure turbines. In detail, however, there are differences. These include the existence of a significant trailing shed vortex and the interaction of the endwall fluid with the suction surface flow. At an aspect ratio of 1.8, the primary and secondary losses are of equal magnitude.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):186-193. doi:10.1115/1.3262084.

The present study intends to give some experimental information on secondary flows and on the associated total pressure losses occurring within turbine cascades. Part 1 of the paper describes the mechanism of production and development of the loss caused by secondary flows in a straight stator cascade with a turning angle of about 65 deg. A full representation of superimposed secondary flow vectors and loss contours is given at fourteen serial traverse planes located throughout the cascade. The presentation shows the mechanism clearly. Distributions of static pressures and of the loss on various planes close to blade surfaces and close to an endwall surface are given to show the loss accumulation process over the surfaces of the cascade passage. Variation of mass-averaged flow angle, velocity and loss through the cascade, and evolution of overall loss from upstream to downstream of the cascade are also given. Part 2 of the paper describes the mechanism in a straight rotor cascade with a turning angle of about 102 deg.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):194-200. doi:10.1115/1.3262085.

Part 1 of this paper [1] presents the detailed mechanism of secondary flows and the associated losses occurring within a straight stator cascade with a relatively low turning angle of about 65 deg. The significant contribution of secondary flows on the loss production process was shown only near the blade suction surface downstream from the cascade throat (Z/Cax = 0.74) in which regional flows decelerated due to adverse pressure gradient. In the second part, the same experimental analysis is applied to a straight rotor cascade with a much larger turning angle of 102 deg. Flow surveys were made at 12 traverse planes located throughout the rotor cascade. The larger turning results in a similar but much stronger contribution of the secondary flows to the loss developing mechanism. Evolution of overall loss starts quite early within the cascade, and the rate of the loss growth is much larger in the rotor case than in the stator case.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):201-209. doi:10.1115/1.3262086.

The ability of a given blade profile to operate over a wide range of conditions is often of the utmost importance. This paper reports the off-design performance of a low-pressure turbine rotor root section in a linear cascade. Data were obtained using pneumatic probes and surface flow visualization. The effects of incidence (+9, 0, −20 deg), Reynolds (1.5, 2.9, 6.0 × 105 ), pitch-chord ratio (0.46, 0.56, 0.69), and inlet boundary layer thickness (0.011, 0.022 δ*/C ) are discussed. Particular attention is paid to the three dimensionality of the flow field. Significant differences in the detail of the flow occur over the range of operating conditions investigated. It is found that the production of new secondary loss is greatest at lower Reynolds numbers, positive incidence, and the higher pitch-chord ratios.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):210-219. doi:10.1115/1.3262087.

The optimization of the blade surface velocity distribution is promising for a reduction of turbine cascade losses. Theoretical and experimental investigations on three turbine cascades with the same blade loading show the important influence of the blade pressure gradient and the free-stream turbulence on the loss behavior. The results presented demonstrate that it is the boundary layer transition behavior that determines the losses on turbine cascades. An enormous effort in measuring technique is required in order to define the location of transition from cascade experiments very accurately.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):220-228. doi:10.1115/1.3262088.

A complete viscous–inviscid interaction is performed that reliably computes steady two-dimensional, subsonic and transonic attached and separated flows for cascades of airfoils. A full-potential code was coupled with both a laminar/transition/ turbulent integral boundary-layer/turbulent wake code and the finite-difference boundary-layer code using the semi-inverse methods of Carter and Wigton. The transpiration coupling concept was applied with an option for a porous airfoil with passive and active physical transpiration. Examples are presented which demonstrate that such flows can be calculated with engineering accuracy by these methods. Carter’s update formula gives smoother solutions for a strong shock than Wigton’s update formulas, although Wigton’s formulas are preferred in the early coupling cycles. The computations show that passive physical transpiration can lead to a lower drag coefficient and higher lift coefficient, a weaker shock, and elimination of shock-induced separation. The extent of the porous region and permeability factor distribution of the porous region must be chosen carefully if these improvements are to be achieved.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):229-236. doi:10.1115/1.3262089.

This paper describes the development of a semi-empirical model for estimating end-wall losses. The model has been developed from improved understanding of complex endwall secondary flows, acquired through review of flow visualization and pressure loss data for axial flow turbomachine cascades. The flow visualization data together with detailed measurements of viscous flow development through cascades have permitted more realistic interpretation of the classical secondary flow theories for axial turbomachine cascades. The re-interpreted secondary flow theories together with integral boundary layer concepts are used to formulate a calculation procedure for predicting losses due to the endwall secondary flows. The proposed model is evaluated against data from published literature and improved agreement between the data and predictions is demonstrated.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):237-244. doi:10.1115/1.3262090.

The paper examines in detail the structure of the tip leakage flow and its effect on the blade loading in a large-scale planar cascade of turbine blades. The tip clearance was varied from 0.0 to 2.86 percent of the blade chord. One of the blades is instrumented with 14 rows of 73 static taps which allowed a very detailed picture of the loading near the tip to be obtained. In addition to the measurements, extensive flow visualization was conducted using both smoke and surface oil flow. A new feature found in the present experiment was the formation of multiple, discrete tip-leakage vortices as the clearance was increased. Their presence is clearly evident from the surface oil flow and they account for the multiple suction peaks found in the blade pressure distributions. Integration of the pressure distributions showed that for larger values of the clearance the blade loading increases as the tip is approached and only begins to decline very near the tip. The increase was found to occur primarily in the axial component of the force.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):246-250. doi:10.1115/1.3262092.

Optimization of transonic turbine bladings over a broad range of operating conditions calls for better understanding of the relationship between blade profile loss and cascade geometric parameters. In fact, many of the experimental correlations published to date have failed to take into due consideration transonic effects, while others have considered far too few of the numerous geometric parameters affecting profile loss in transonic flows. Through examination of the experimental data gathered by some 20 authors regarding the effects of the most significant blading geometric parameters on profile losses, a loss correlation procedure has been developed. The procedure is especially advantageous in that it allows continuous updating as new experimental data become available.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):251-256. doi:10.1115/1.3262093.

Nonintrusive measurements near and within the rotor of a cold-air turbine showed a sudden increase of turbulence energy when the wake portion of the incoming fluid entered the rotor. It has been suggested that this was due to the cutting of the passage vortices and trailing-edge shed vortices which emerge from the stator row. Since these secondary vortices are located very close to the stator wakes, it was very difficult to distinguish between the effects of shed vortex and passage vortex cutting on turbulence intensification. In the present paper, a method is shown which, with the help of time–distance diagrams, made it possible to attribute the turbulence increase to the breakdown of the secondary vortices. Further, the time–distance diagrams made it possible to locate the origin of turbulence production and follow the spreading of the highly turbulent flow regions through the rotor channel.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):258-267. doi:10.1115/1.3262096.

An experimental investigation was performed to measure Reynolds stresses in the turbulent flow downstream of a large-scale linear turbine cascade. A rotatable X-wire hot-wire probe that allows redundant data to be taken with solution for mean velocities and turbulence quantities by least-squares fitting procedures was developed. The rotatable X-wire was used to obtain the Reynolds stresses on a measurement plane located 10 percent of an axial chord downstream of the trailing edge. Here the turbulence kinetic energy exhibits a distribution resembling the contours of total pressure loss obtained previously, but is highest in the blade wake where losses are relatively low. The turbulent shear stresses obtained are consistent in sign and magnitude with the gradients of mean velocity. The measured Reynolds stresses are combined with measured distributions of velocity to show how and where losses are being produced. The mechanisms for the dissipation of mean kinetic energy in this swirling three-dimensional flow are revealed.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):268-277. doi:10.1115/1.3262097.

The SIMPLE method of Patankar and Spalding and its variants such as SIMPLER, SIMPLEC, and SIMPLEX are segregated methods for solving the discrete algebraic equations representing the equations of motion for an incompressible fluid flow. The present paper presents the extension of these methods to the solution of compressible fluid flows within the context of generalized segregated approach. To provide a framework for better understanding the segregated approach to solving viscous compressible fluid flows an interpretation of the role of pressure in the numerical method is presented. With this interpretation it becomes evident that the linearization of the equation for mass conservation and the approach used to solve the linearized algebraic equations representing the equations of motion are important in determining the performance of the numerical method. The relative performances of the various segregated methods are compared for several subsonic and supersonic compressible fluid flows.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):278-285. doi:10.1115/1.3262098.

An experimental investigation was conducted to characterize a symmetric horseshoe vortex system in front of and around a single large-diameter right cylinder centered between the sidewalls of a wind tunnel. Surface flow visualization and surface static pressure measurements as well as extensive mean velocity and pressure measurements in and around the vortex system were acquired. The results lend new insight into the formation and development of the vortex system. Contrary to what has been assumed previously, a strong vortex was not identified in the streamwise plane of symmetry, but started a significant angular distance away from it. Rather than the multiple vortex systems reported by others, only a single primary vortex and saddle point were found. The scale of the separation process at the saddle point was much smaller than the scale of the approaching boundary layer thickness. Results of the present study not only shed light on such phenomena as the asymmetric endwall flow in axial turbomachinery but can also be used as a test case for three-dimensional computational fluid mechanics computer codes.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):286-295. doi:10.1115/1.3262101.

Cascade testing tries to simulate the actual flow conditions encountered in a turbine. However, it is possible to reproduce neither the free-stream turbulence structure of the turbomachinery, nor the periodic wake effects of upstream blade rows. The usual understanding is that the latter in particular results in a significantly different behavior of the boundary layer in the engine. Experimental results from cascades and turbine rigs are presented. Grid-generated free-stream turbulence structure is compared to that in the turbine. Measurements of the profile pressure distribution, flush-mounted hot films, and flow visualization were used for the interpretation of the test results. Some observations of the boundary layer development in the cascade, on the guide vanes, and on rotor blades with typically skewed boundary layers are shown indicating essentially similar behavior in all cases.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):296-302. doi:10.1115/1.3262102.

Laminar-to-turbulent transition in the presence of Görtler vortices has been investigated experimentally, in the outer wall boundary layer of a curved water channel. Ratios of boundary layer thickness at the start of curvature to wall radius were around 0.05 and core flow turbulence intensities were between 1 and 3 percent. Measurements of intermittency factor were made by hot film probe and of mean and rms velocity by laser anemometer. At Reynolds numbers low enough to allow considerable nonlinear vortex amplification in the laminar region, transition was found to begin sooner and progress faster at a vortex upwash position than at a spanwise-adjacent downwash position. Measured Görtler numbers at transition onset bore little relationship to those often used as transition criteria in two-dimensional boundary layer prediction codes. Little spanwise variation in intermittency occurred at higher Reynolds numbers, where mean velocity profiles at upwash were much less inflected. Toward the end of curvature, favorable pressure gradients estimated to exceed the Launder relaminarization value corresponded with cases of incomplete transition.

Commentary by Dr. Valentin Fuster
J. Turbomach. 1987;109(2):303-309. doi:10.1115/1.3262103.

The heated thin-film method was adapted to meet the requirements of investigations on boundary layer behavior in a turbine rig. Special multisensor probes of vaporized nickel on a polyimide foil were developed and applied to the vanes. Basic experiments with an airfoil in a free stream were carried out and a reliable interpretation of the thin-film results was found by comparison with pressure distribution, flow visualization, and laser measurements. It can be shown that this measuring device is a suitable method for the investigation of separation bubbles and boundary layer transition.

Commentary by Dr. Valentin Fuster

DISCUSSIONS

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