Research Papers

J. Turbomach. 2017;139(10):101001-101001-13. doi:10.1115/1.4036297.

This study comprehensively illustrates the effect of Reynolds number, hole spacing, nozzle-to-target distance, and target plate thickness on the conjugate heat transfer (CHT) performance of an impinging jet array. Test models are composed of a specific thermal-conductivity material which exerts a matched model Biot number to that of engine condition. High-resolution temperature measurements are conducted on the impinging-target plate utilizing steady liquid crystal (SLC) with Reynolds numbers ranging from 5000 to 27,500. Different streamwise and spanwise jet-to-jet spacing (i.e., X/D and Y/D: 4–8), nozzle-to-target plate distance (Z/D: 0.75–3), and target plate thickness (t/D: 0.75–2.75) are employed to compose a total of 108 different geometries. Experimental measured temperature is utilized as boundary conditions to conduct finite element simulation. Local and averaged nondimensional temperature and averaged temperature uniformity of target plate “hot side” are obtained. Optimum hole spacing arrangements, impingement distance, and target plate thickness are pointed out to minimize hot side temperature, amount of cooling air and to maximize temperature uniformity. Also included are 2D predictions with different convective boundary conditions, i.e., local 2D distribution and row-averaged heat transfer coefficients (HTCs), to estimate the accuracy of temperature prediction in comparison with the conjugate results.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101002-101002-10. doi:10.1115/1.4036436.

The results of this investigation come from two linear cascades at high diffusion factors (DFs). The measurements presented for each low-pressure turbine (LPT) profile were conducted at midspan under a range of Reynolds- and exit Mach numbers. The exit Mach number was varied in a range covering low subsonic up to values where a transonic flow regime on the suction side of the blade could be expected. This work focuses on two profiles with a diffusion factor in a range of 0.18DF0.22, where values in this range are considered as a comparable for the two cascades. Profile A is a front-loaded design and has shown no obvious flow separation on the suction side of the blade. Compared to the design A, design B is a more aft-loaded profile which exhibits flow separation on the suction side for all Reynolds numbers investigated. The integral total pressure losses were evaluated by wake traverses downstream of the airfoil. To determine the isentropic Mach numbers and the character of the boundary layer along the suction side of the profile, the static pressure measurements and traverses with a flattened Pitot probe were carried out. A correlation between the position of maximum Mach number on the suction side and the integral total pressure losses has been successfully established. The results show that the optimum location of peak Mach number to minimize integral total pressure losses is significantly dependent on the Reynolds number. However, the correlation presented in this paper, which is based on the data of the integral total pressure losses of an attached boundary layer, is not able to predict the integral total pressure loss or the location of the maximum Mach number on the suction side of the blade when an open separation bubble occurs.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101003-101003-15. doi:10.1115/1.4036341.

The effect of stator clocking has been experimentally and computationally investigated using a low-speed, two-stage, low-pressure turbine (LPT) which was specifically designed to maximize the clocking potential by aligning the stator 1 wake segments with the stator 2 leading edge along the span. It was verified that the wake segments are aligned to within 10% of stator pitch across the span. The measured clocking effect on the work extraction is 0.12% and on efficiency is 0.08%. Although the effect of clocking is small, it is repeatable, periodic across four stator pitches and consistent between independent measurements. Furthermore, factors to consider for a reliable clocking investigation are discussed. The measurements revealed that the majority of the clocking effect on the work extraction occurs in stage 2 and it originates at stator 2 exit. This indicates that the flow is being processed differently within stator 2. There is also an effect on the stage 1 work. In each blade row, the measured clocking effect on the lost work is similar across the span. The computations with meshed cavities do not capture any clocking effects in stage 1. This indicates that an unsteady viscid phenomenon within rotor 1 is not captured by the fully turbulent calculation, e.g., unsteady transition. However, the computations do capture the measured clocking effect on the stage 2 work extraction. It is hypothesized that the clocking effect on stator 2 flow turning is dominated by a steady, inviscid process.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101004-101004-9. doi:10.1115/1.4036437.

In the present paper, the influence of the presence of an inlet flow nonuniformity on the aerodynamic and thermal performance of a film cooled linear nozzle vane cascade is fully assessed. Tests have been carried out with platform cooling, with coolant ejected through a slot located upstream of the leading edge. Cooling air is also ejected through a row of cylindrical holes located upstream of the slot, simulating a combustor cooling system. An obstruction was installed upstream of the cascade at variable tangential and axial position to generate a flow nonuniformity. The cascade was tested at a high inlet turbulence intensity level (Tu1 = 9%) and at a constant inlet Mach number of 0.12 and nominal cooling condition. Aerothermal characterization of vane platform was obtained through five-hole probe and end wall adiabatic film cooling effectiveness measurements. Results show a relevant negative impact of inlet flow nonuniformity on the cooled cascade aerodynamic and thermal performance. Higher film cooling effectiveness and lower aerodynamic losses are obtained when the inlet flow nonuniformity is located at midpitch, while lower effectiveness and higher losses are obtained when it is aligned to the vane leading edge. Moving the nonuniformity axially or changing its blockage only marginally influences the platform thermal protection.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101005-101005-12. doi:10.1115/1.4036342.

It is common to assume that the performance of low-speed turbines depends only on the flow coefficient and Reynolds number. As such, the required operating point is achieved by controlling the values of these two nondimensional quantities by, for example, appropriate choices for the mass flow rate and applied brake torque. However, when the turbine has an atmospheric inlet and uses unconditioned air, variations in ambient pressure, temperature, and humidity are introduced. While it is still possible to maintain the required values for the flow coefficient and Reynolds number, the ambient variations affect additional nondimensional quantities which are related to the blade speed and gas properties. Generally, the values of these additional nondimensional quantities cannot be controlled and, consequently, they affect the turbine performance. In addition, thermal effects, which are exacerbated by the use of plastic blades, can cause changes in the blade row seal clearance and these also affect the performance. Therefore, to obtain measurements with greater accuracy and repeatability, the changes in the uncontrolled nondimensional quantities must be accounted. This paper contains four parts. First, it is described how suitable data acquisition parameters can be determined to eliminate short time scale facility unsteadiness within the measurements. Second, by the analysis of models, the most appropriate forms for the additional nondimensional quantities that influence turbine performance are obtained. Since the variations in the uncontrolled nondimensional quantities affect repeatability, the size of the effect on the turbine performance is quantified. Third, a best-fit accounting methodology is described, which reduces the effects of the uncontrolled nondimensional quantities on turbine performance, provided sufficient directly related measurements are available. Finally, the observations are generalized to high-speed turbomachines.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101006-101006-12. doi:10.1115/1.4036344.

Noise originating from the core of an aero-engine is challenging to quantify since the understanding of its generation and propagation is less advanced than that for noise sources of other engine components. To overcome the difficulties associated with dynamic measurements in the crowded core region, dedicated experiments have been set up in order to investigate mainly two processes: the propagation of direct combustion noise through the subsequent turbine stage, and the generation of indirect combustion noise by the passage of inhomogeneities of entropy and vorticity through the turbine stage. In the current work, a transonic turbine stage was exposed to isolated and well-characterized acoustic, entropic, and vortical disturbances. The incoming and outgoing sound fields were analyzed in detail by two large arrays of microphones. The mean flow field and the disturbances were carefully mapped by several aerodynamic and thermal probes. The results include transmission and reflection characteristics of the turbine stage, the latter was found to be much lower than commonly assumed. The modal decomposition of the acoustic field in the upstream and downstream section shows additional modes besides the expected rotor–stator interaction modes. At the frequency of entropy or vorticity excitation, respectively, a significant increase of the overall sound power level was observed.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101007-101007-9. doi:10.1115/1.4036357.

An optimization has been performed on a well-proven radial compressor design known as the SRV4 impeller (the Krain impeller), which has been extensively tested in the past, using the autoopti tool developed at DLR's Institute of Propulsion Technology. This tool has shown its capability in several tasks, mainly for axial compressor and fan design as well as for turbine design. The optimization package autoopti was applied to the redesign and optimization of a radial compressor stage with a vaneless diffusor. This optimization was performed for the SRV4 compressor geometry without fillets using a relatively coarse structured mesh in combination with wall functions. The impeller geometry deduced by the optimization had to be slightly modified due to manufacturing constraints. In order to filter out the improvements of the new so-called SRV5 radial compressor design, two work packages were conducted: The first one was the manufacturing of the new impeller and its installation on a test rig to investigate the complex flow inside the machine. The aim was, first of all, the evaluation of a classical performance map and the efficiency chart achieved by the new compressor design. The efficiencies realized in the performance chart were enhanced by nearly 1.5%. A 5% higher maximum mass flow rate was measured in agreement with the Reynolds-averaged Navier–Stokes (RANS) simulations during the design process. The second work package comprises the computational fluid dynamics (CFD) analysis. The numerical investigations were conducted with the exact geometries of both the baseline SRV4 as well as the optimized SRV5 impeller including the exact fillet geometries. To enhance the prediction accuracy of pressure ratio and impeller efficiency, the geometries were discretized by high-resolution meshes of approximately 5 × 106 cells. For the blade walls as well as for the hub region, the mesh resolution allows a low-Reynolds approach in order to get high-quality results. The comparison of the numerical predictions and the experimental results shows a very good agreement and confirms the improvement of the compressor performance using the optimization tool autoopti.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101008-101008-12. doi:10.1115/1.4035520.

Adiabatic film cooling effectiveness measurements are obtained using pressure-sensitive paint (PSP) on a flat film cooled surface. The effects of blowing ratio and hole spacing are investigated for four multirow arrays comprised of eight rows containing 52 holes of 3.8 mm diameter with 20 deg inclination angles and hole length-to-diameter ratio of 11.2. The four arrays investigated have two different hole-to-hole spacings composed of cylindrical and diffuser holes. For the first case, lateral and streamwise pitches are 7.5 times the diameter. For the second case, pitch-to-diameter ratio is 14 in lateral direction and 10 in the streamwise direction. The holes are in a staggered arrangement. Adiabatic effectiveness measurements are taken for a blowing ratio range of 0.3–1.2 and a density ratio of 1.5, with CO2 injected as the coolant. A thorough boundary layer analysis is presented, and data were taken using hotwire anemometry with air injection, with boundary layer, and turbulence measurements taken at multiple locations in order to characterize the boundary layer. Local effectiveness, laterally averaged effectiveness, boundary layer thickness, momentum thickness, turbulence intensity, and turbulence length scale are presented. For the cylindrical holes, at the first row of injection, the film jets are still attached at a blowing ratio of 0.3. By a blowing ratio of 0.5, the jet is observed to lift off, and then impinge back onto the test surface. At a blowing ratio of 1.2, the jets lift off, but reattach much further downstream, spreading the coolant further along the test surface. A thorough uncertainty analysis has been conducted in order to fully understand the presented measurements and any shortcomings of the measurement technique. The maximum uncertainty of effectiveness and blowing ratio is 0.02 counts of effectiveness and 3%, respectively.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101009-101009-9. doi:10.1115/1.4036677.

The aerodynamic performance of a cavity-winglet tip is investigated in a high-pressure turbine cascade by experimental and numerical methods. The winglet tip has geometric features of a cavity and a suction side fore-part winglet. A cavity tip is studied as the baseline case. The aerodynamic performances of the two tips are investigated at three tip gaps of 0.8%, 1.7%, and 2.7% chord. At tip gaps of 1.7% and 2.7% chord, the loss near the blade tip is dominated by the tip leakage vortex (TLV) for both tips, and the winglet tip mainly reduces the loss generated by the tip leakage vortex. In the past, it was concerned that at a small tip gap, the winglet tip could introduce extra secondary loss and show little aerodynamic benefits. The winglet tip used in the current study is also found to be able to effectively reduce the loss at the smallest tip gap size of 0.8% chord. This is because at this small tip gap, the tip leakage vortex and the passage vortex (PV) appear simultaneously for the cavity tip. The winglet tip is able to reduce the pitchwise pressure gradient in the blade passage, which tends to suppress the formation of the passage vortex. The effects of the winglet tip on the flow physics and the loss mechanisms are explained in detail.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101010-101010-13. doi:10.1115/1.4036646.

Partial surge is a new type of instability inception in the form of axisymmetric low-frequency disturbance located in the hub region and has been observed in transonic axial flow compressors. Previous studies on the evolution of instability in a transonic axial flow compressor at different rotor speeds found that partial surge occurs and leads to full compressor flow instability at high rotor speeds but not at low rotor speeds, and the blade loading at the hub increases with the rotor speed. A hypothesis is first made that the level of blade loading in the hub region could be highly correlated to the occurrence of partial surge. Experiments and numerical simulations are then conducted to test this hypothesis when the radial distribution of blade loading near the stall point is varied by introducing inlet distortion (i.e., alternately mounting specially designed screens at the inlet of the compressor). Both the experimental results of instability evolution and the numerical results of radial distribution of blade loading show that high hub loading near the stall point is the necessary condition for the occurrence of partial surge. In addition, the general effects of radial loading distribution on the type of stall inception are presented and discussed.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101011-101011-9. doi:10.1115/1.4036647.

This paper presents a continued study on a previously investigated novel winglet-shroud (WS) (or partial shroud) geometry for a linear turbine cascade. Various widths of double-side winglets (DSW) and different locations of a partial shroud are considered. In addition, both a plain tip and a full shroud tip are applied as the datum cases which were examined experimentally and numerically. Total pressure loss and viscous loss coefficients are comparatively employed to execute a quantitative analysis of aerodynamic performance. The effectiveness of various widths (w) of DSW set at 3%, 5%, 7%, and 9% of the blade pitch (p) is numerically investigated. Skin-friction lines on the tip surface indicate that different DSW cases do not alter flow field features including the separation bubble and reattachment flow within the tip gap region, even for the case with the broadest width (w/p = 9%). However, the pressure side extension of the DSW exhibits the formation of separation bubble, while the suction side platform of the DSW turns the tip leakage vortex (TLV) away from the suction surface (SS). Meanwhile, the horse-shoe vortex (HV) near the casing is not generated even for the case with the smallest width (w/p = 3%). As a result, both the tip leakage and the upper passage vortices are weakened and further dissipated with wider w/p in the DSW cases. Larger width of the DSW geometry is indeed able to improve the aerodynamic performance, but only to a slight degree. With the w/p increasing from 3% to 9%, the mass-averaged total pressure loss coefficient over an exit plane is reduced by only 2.61%. Therefore, considering both the enlarged (or reduced) tip area and the enhanced (or deteriorated) performance compared to the datum cases, a favorable width of w/p = 5% is chosen to design the WS structure. Three locations for the partial shroud (linkage segment) are devised, locating them near the leading edge, in the middle and close to the trailing edge, respectively. Results demonstrate that all three cases of the WS design have advantages over the DSW arrangement in lessening the aerodynamic loss, with the middle linkage segment location producing the optimal effect. This conclusion verifies the feasibility of the previously studied WS configuration.

Commentary by Dr. Valentin Fuster
J. Turbomach. 2017;139(10):101012-101012-14. doi:10.1115/1.4036765.

A general issue in turbomachinery flow computations is how to capture and resolve two kinds of unsteadiness efficiently and accurately: (a) deterministic disturbances with temporal and spatial periodicities linked to blade count and rotational speed and (b) nondeterministic disturbances including turbulence and self-excited coherent patterns (e.g., vortex shedding, shear layer instabilities, etc.) with temporal and spatial wave lengths unrelated to blade count and rotational speed. In particular, the high cost of large eddy simulations (LES) is further compounded by the need to capture the deterministic unsteadiness of bladerow interactions in computational domains with large number of blade passages. This work addresses this challenge by developing a multiscale solution approach. The framework is based on an ensemble-averaging to split deterministic and nondeterministic disturbances. The two types of disturbances can be solved in suitably selected computational domains and solvers, respectively. The local fine mesh is used for nondeterministic turbulence eddies and vortex shedding, while the global coarse mesh is for deterministic unsteadiness. A key enabler is that the unsteady stress terms (UST) of the nondeterministic disturbances are obtained only in a small set of blade passages and propagated to the whole domain with many more passages by a block spectral mapping. This distinctive multiscale treatment makes it possible to achieve a high-resolution unsteady Reynolds-averaged Navier–Stokes (URANS)/LES solution in a multipassage/whole annulus domain very efficiently. The method description will be followed by test cases demonstrating the validity and potential of the proposed methodology.

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