The goal of the present work is to investigate the effect of supply pipe position on the heat transfer features of various active clearance control (ACC) geometries, characterized by different jet-to-jet distances. All geometries present 0.8 mm circular impingement holes arranged in a single row. The jets generated by such holes cool a flat target surface, which is replicated by a metal plate in the experimental setup. Measurements are performed using the steady-state technique, obtained by heating up the target plate thanks to an electrically heated Inconel foil applied on the side of the target opposite to the jets. Temperature is also measured on this side by means of an IR camera. Heat transfer is then evaluated thanks to a custom designed finite difference procedure, capable of solving the inverse conduction problem on the target plate. The effect of pipe positioning is studied in terms of pipe-to-target distance (from 3 to 11 jet diameters) and pipe orientation (i.e. rotation around its axis, from 0° to 40° with respect to target normal direction), while the investigated jet Reynolds numbers range from 6000 to 10000. The obtained results reveal that heat transfer is maximized for a given pipe-to-target distance, dependent on both jet-to-jet distance and target surface extension. Pipe rotation also affects the cooling features in a non-monotonic way, suggesting the existence of different flow regimes related to jet inclination.

For simulations of liquid jets in crossflows, the primary atomization can be treated with the quadratic formula, which has been derived from integral form of conservation equations of mass and energy in our previous work. This formula relates the drop size with the local kinetic energy state, so that local velocity data from the volume-of-fluid (VOF) simulation prior to the atomization can be used to determine the initial drop size. This initial drop size, along with appropriately sampled local gas velocities, is used as the initial conditions in the dispersed-phase simulation. This procedure has been performed on a coarse-grid platform, with good validation and comparison with available experimental data at realistic Reynolds and Weber numbers, representative of gas-turbine combustor flows. The computational procedure produces all the relevant spray characteristics: spatial distributions of drop size, velocities, and volume fluxes, along with global drop size distributions. The primary atomization module is based on the conservation principles and is generalizable and implementable to any combustor geometries for accurate and efficient computations of spray flows.

The effects of partial premixing on a reacting jet-in-crossflow is investigated in a five atmosphere axially staged combustor at stationary gas turbine relevant conditions. The facility consists of a dump style headend burner that provides a crossflow with a quasi-uniform velocity and temperature profile to the axial stage to isolate the effects of the jet-in-crossflow. The headend burner is run with methane and air at a lean equivalence ratio to match industry emission standards. For this work, the total air to the headend and axial stage is kept constant, and fuel is split between the headend and axial stage to represent different gas turbine loading conditions. For the cases analyzed, the fuel split to the axial stage went up to 25%. The axial stage consists of an optically accessible test section with a coaxial injector that provides variability to how long the methane and air can mix before entering the facility. Three different premixed levels are studied: fully premixed, nonpremixed, and partially premixed. The flow-field characteristics of the reacting jet-in-crossflow are analyzed using particle image velocimetry (PIV), and flame behavior is quantified by employing CH* chemiluminescence. NO measurements are made at the exit of the facility using a Horiba emissions analyzer. Two different flames are observed: flames that burn in the leeward recirculation region and flames that burn at the core of the jet.

The ever increasing performance requirements of modern aeroengines necessitate the development of effective ways to improve efficiency and reduce losses. Casing temperature control is particularly critical from this point of view, since thermal expansion directly affects the blade tip clearance and thus the associated leakages. To limit the turbine tip flows, active clearance control (ACC) systems have been implemented over the last decades. These systems are usually based upon impingement cooling, generated by a series of perforated manifolds enclosing the turbine casing. When dealing with aeroengine low pressure turbines, the current trend in increasing the engine bypass ratio, so as to enhance the system propulsive efficiency, pushes the limits of ACC traditional design performance. The reduction of the pressure head at the ACC system inlet requires lower nozzle-to-target distances as well as denser impingement arrays to compensate the reduction of the jets' Reynolds number. Literature correlations for the impingement heat transfer coefficient estimation are then out of their confidence range and also RANS numerical approaches appear not suitable for future ACC designs. In this work, methodologies for the development of accurate and reliable tools to determine the heat transfer characteristics of low pressure ACC systems are presented. More precisely, this paper describes a custom designed finite difference procedure capable of solving the inverse conduction problem on the target plate of a test sample. The methodology was successfully applied to an experimental setup for the measurement of the thermal loads on a target plate of a representative low pressure ACC impinging system. The experimental outcomes are then used to validate a suitable numerical approach. Results show that RANS model is not able to mimic the experimental trends, while scale-resolving turbulence models provide a good reconstruction of the experimental evidences, thus allowing to obtain a correct interpretation of flow and thermal phenomena for ACC systems.

Considering the complicated geometry in an active clearance control (ACC) system, the design of an improved cooling feed pipe with a covering plate for a high pressure ribbed turbine case was investigated. Numerical calculations were analyzed to obtain the interactions between the impinging jet arrays fed by the pipe. Experimental tests were performed to explore the effect of the Reynolds number (2000–20,000) and the jet-to-surface spacing ratio (6–10) on the streamwise-averaged Nusselt numbers. Additionally, the effect of the crossflow produced by the configuration was investigated. Results showed a confined curved channel was formed by the pipe and ribbed case, which resulted in crossflow. The crossflow evolved into vortices and the streamwise-averaged Nusselt number on the high ribs was subsequently increased. Furthermore, the distribution of the heat transfer on the entire surface became more uniform compared with that of traditional impinging jet arrays. A higher Nusselt number was achieved by decreasing the jet-to-surface spacing and increasing the Reynolds number. This investigation has revealed a cooling configuration for controlling the wall flow and evening the heat transfer on the case surface, especially for the ribs.

Imaging of aeroacoustic noise sources is routinely accomplished with geometrically fixed phased arrays of microphones. Several decades of research have gone into improvement and optimization of sensor layouts, selection of basis models, and deconvolution algorithms to produce sharper and more localized images of sound-producing regions in space. This paper explores an extension to conventional phased array measurements that uses slowly, continuously moving microphone arrays with and without coupling to rigid fixed arrays to improve image quality and better describe noise mechanisms on turbofan engine sources such as jet exhausts and turbomachinery components. Three approaches are compared in the paper: fixed receiver beamforming (FRBF), continuous-scan beamforming (CSBF), and multireference CSBF (MRCSBF). The third takes advantage of transfer function matrices formed between fixed and moving sensors to achieve effective virtual arrays with spatial density one to two orders of magnitude higher, with practical sensor budgets and scan speeds. The MRCSBF technique produces array sidelobe rejection that approaches the theoretical array pattern of a continuous two-dimensional (2D) aperture. The implications of this finding are that better source localization may be achieved with conventional delay and sum (DAS) beamforming (BF) with practical sensor budgets, and that an improved starting image of the sound source can be provided to deconvolution algorithms. These findings are demonstrated on analytical and experimental examples from a low-cost rotating phased array using point sound sources, as well as linear scanning array experiments of an impinging jets point source and a near-sonic jet nozzle exhaust.

In order to further study the effects of the target channel shape on the cooling performance of the double swirl cooling (DSC), five double swirl channels formed by two overlapping elliptic cylinders with different length ratio between the vertical semi-axis and the horizontal semi-axis are applied. Numerical studies are carried out under three Reynolds numbers. The flow characteristics and heat transfer performance of five DSC cases are compared with the benchmark impingement cooling case. The flow losses, cross-flow development, generated vortices, and velocity distributions inside target channels are illustrated, analyzed, and compared. The spanwise averaged Nusselt number, Nusselt number distributions, and thermal performance are discussed and compared. Results indicate that the largest length ratio between the vertical semi-axis and the horizontal semi-axis of the target channel yields the lowest flow loss, largest overall averaged Nusselt number, and best thermal performance. With the decrease in the length ratio, the heat transfer distribution on the target surface becomes more uniform. The maximum enhancement of overall averaged Nusselt number and thermal performance in DSC is about 30% and 33%, respectively.

Jet in a cross-flow (JICF) is a flow arrangement found in many engineering applications, especially in gas turbine air–fuel mixing. Understanding of scalar mixing in JICF is important for low NO x burner design and operation, and numerical simulation techniques can be used to understand both spatial and temporal variation of air–fuel mixing quality in such applications. In this paper, mixing of the jet stream with the cross-flow is simulated by approximating the jet flow as a passive scalar and using the large eddy simulation (LES) technique to simulate the turbulent velocity field. A posteriori test is conducted to assess three dynamic subgrid scale models in modeling jet and cross-flow interaction with the boundary layer flow field. Simulated mean and Reynolds stress component values for velocity field and concentration fields are compared against experimental data to assess the capability of the LES technique, which showed good agreement between numerical and experimental results. Similarly, time mean and standard deviation values of passive scalar concentration also showed good agreement with experimental data. In addition, LES results are further used to discuss the scalar mixing field in the downstream mixing region.

Experimental research was conducted into a scooped rotor system that captures oil from a stationary jet and directs it through passages within the shaft to another axial location. Such a system has benefits for delivering oil via under-race feed to aeroengine bearings where direct access is limited. Oil capture efficiency was calculated for three jet configurations, a range of geometric variations relative to a baseline and a range of operating conditions. Flow visualization techniques yielded high-speed imaging in the vicinity of the scoop leading edge. Overall capture efficiency depends on the amount of oil initially captured by the scoop that is retained. Observation shows that when the jet hits the tip of a scoop element, it is sliced and deflected upward in a “plume.” Ligaments and drops formed from this plume are not captured. In addition, some oil initially captured is flung outward as a consequence of centrifugal force. Although in principle capture of the entire supply is possible over most of the shaft speed range, as demonstrated by a simplified geometric model, in practice 60–70% is typical. Significant improvement in capture efficiency was obtained with a lower jet angle (more radial) compared to baseline. Higher capture efficiencies were found where the ratio of jet to scoop tip speed was lower. This research confirms the capability of a scoop system to capture and retain delivered oil. Additional numerical and experimental work is recommended to further optimize the geometry and increase the investigated temperature and pressure ranges.

This paper presents volumetric velocimetry (VV) measurements for a jet in crossflow that is representative of film cooling. VV employs particle tracking to nonintrusively extract all three components of velocity in a three-dimensional volume. This is its first use in a film-cooling context. The primary research objective was to develop this novel measurement technique for turbomachinery applications, while collecting a high-quality data set that can improve the understanding of the flow structure of the cooling jet. A new facility was designed and manufactured for this study with emphasis on optical access and controlled boundary conditions. For a range of momentum flux ratios from 0.65 to 6.5, the measurements clearly show the penetration of the cooling jet into the freestream, the formation of kidney-shaped vortices, and entrainment of main flow into the jet. The results are compared to published studies using different experimental techniques, with good agreement. Further quantitative analysis of the location of the kidney vortices demonstrates their lift off from the wall and increasing lateral separation with increasing momentum flux ratio. The lateral divergence correlates very well with the self-induced velocity created by the wall–vortex interaction. Circulation measurements quantify the initial roll up and decay of the kidney vortices and show that the point of maximum circulation moves downstream with increasing momentum flux ratio. The potential for nonintrusive VV measurements in turbomachinery flow has been clearly demonstrated.

The complex flowfield inside a gas turbine combustor creates a difficult challenge in cooling the combustor walls. Many modern combustors are designed with a double-wall that contain both impingement cooling on the backside of the wall and effusion cooling on the external side of the wall. Complicating matters is the fact that these double-walls also contain large dilution holes whereby the cooling film from the effusion holes is interrupted by the high-momentum dilution jets. Given the importance of cooling the entire panel, including the metal surrounding the dilution holes, the focus of this paper is understanding the flow in the region near the dilution holes. Near-wall flowfield measurements are presented for three different effusion cooling hole patterns near the dilution hole. The effusion cooling hole patterns were varied in the region near the dilution hole and include: no effusion holes; effusion holes pointed radially outward from the dilution hole; and effusion holes pointed radially inward toward the dilution hole. Particle image velocimetry (PIV) was used to capture the time-averaged flowfield at approaching freestream turbulence intensities of 0.5% and 13%. Results showed evidence of downward motion at the leading edge of the dilution hole for all three effusion hole patterns. In comparing the three geometries, the outward effusion holes showed significantly higher velocities toward the leading edge of the dilution jet relative to the other two geometries. Although the flowfield generated by the dilution jet dominated the flow downstream, each cooling hole pattern interacted with the flowfield uniquely. Approaching freestream turbulence did not have a significant effect on the flowfield.

The complex flow field in a gas turbine combustor makes cooling the liner walls a challenge. In particular, this paper is primarily focused on the region surrounding the dilution holes, which is especially challenging to cool due to the interaction between the effusion cooling jets and high-momentum dilution jets. This study presents overall effectiveness measurements for three different cooling hole patterns of a double-walled combustor liner. Only effusion hole patterns near the dilution holes were varied, which included: no effusion cooling; effusion holes pointed radially outward from the dilution hole; and effusion holes pointed radially inward toward the dilution hole. The double-walled liner contained both impingement and effusion plates as well as a row of dilution jets. Infrared thermography was used to measure the surface temperature of the combustor liners at multiple dilution jet momentum flux ratios and approaching freestream turbulence intensities of 0.5% and 13%. Results showed that the outward and inward geometries were able to more effectively cool the region surrounding the dilution hole compared to the closed case. A significant amount of the cooling enhancement in the outward and inward cases came from in-hole convection. Downstream of the dilution hole, the interactions between the inward effusion holes and the dilution jet led to lower levels of effectiveness compared to the other two geometries. High freestream turbulence caused a small decrease in overall effectiveness over the entire liner and was most impactful in the first three rows of effusion holes.

In this paper, pilot-ignited high pressure dual-fuel combustion of a natural gas jet is investigated on a fundamental basis by applying two separate single-hole injectors to a rapid compression expansion machine (RCEM). A S hadowgraphy system is used for optical observations, and the combustion progress is assessed in terms of heat release rates (HRRs). The experiments focus on the combined influence of injection timing and geometrical jet arrangement on the jet interaction and the impact on the combustion process. In a first step, the operational range for successful pilot self-ignition and transition to natural gas jet combustion is determined, and the restricting phenomena are identified by analyzing the shadowgraph images. Within this range, the combustion process is assessed by evaluation of ignition delays and HRRs. Strong interaction is found to delay or even prohibit pilot ignition, while it facilitates a fast and stable onset of the gas jet combustion. Furthermore, it is shown that the HRR is governed by the time of ignition with respect to the start of natural gas injection—as this parameter defines the level of premixing. Evaluation of the time of gas jet ignition within the operability map can therefore directly link a certain spatial and temporal interaction to the resulting heat release characteristics. It is finally shown that controlling the HRR through injection timing variation is limited for a certain angle between the two jets.

The catalytic generation of ammonia from a liquid urea solution is a critical process determining the performance of selective catalytic reduction (SCR) systems. Solid deposits on the catalyst surface from the decomposition of urea have to be avoided, as this leads to reduced system performance or even failure. At present, reactor design is often empirical, which poses a risk for costly iterations due to insufficient system performance. The presented research project proposed a performance prediction and modeling approach for SCR hydrolysis reactors generating ammonia from urea. Different configurations of hydrolysis reactors were investigated experimentally. Ammonia concentration measurements provided information about parameters influencing the decomposition of urea and the system performance. The evaporation of urea between injection and interaction with the catalyst was identified as the critical process driving the susceptibility to deposit formation. The spray of urea solution was characterized in terms of velocity distribution by means of particle-image velocimetry. Results were compared with theoretical predictions and calculation options for processes in the reactor were determined. Numerical simulation was used as an additional design and optimization tool of the proposed model. The modeling approach is presented by a step-by-step method, which takes into account design constraints and operating conditions for hydrolysis reactors.

The effects of effusion and film cooling momenta on combustor flow fields are investigated. Steady, compressible three-dimensional (3D) simulations are performed on a single-swirler combustor using Reynolds-averaged Navier–Stokes (RANS) with flamelet generated manifold and Lagrangian–Eulerian multiphase spray, while accounting for dome and liner cooling. Two simulations are performed on the same mesh. One simulation is conducted using a parallelized, automated, predictive, imprint cooling (PAPRICO) model with dynamic flux boundary conditions and downstream pressure probing (DFBC-DPP). PAPRICO involves removing the cooling jet geometry from the dome and liner while retaining the cooling hole imprints. The PAPRICO model does not require a priori knowledge of the cooling flow rates through various combustor liner regions nor specific mesh partitioning. The other simulation is conducted using the homogenously patched cooling (HPC) model, which involves removing all the cooling jets. The HPC model applies volumetric sources adjacent to the combustor wall regions where cooling jets are present. The momentum source, however, becomes negligible. The HPC model is not predictive and requires tedious ex situ mass flow measurements from an auxiliary flowbench experiment, afflicted with systematic errors. Hence, the actual in situ air flow splits through the several combustor regions is not known with absolute certainty. The numerical results are compared with measurements of mass flow rates, static pressure drops, and path-integrated temperatures. The results demonstrate that it is critical to account for the discrete dome and liner cooling momentum to better emulate the reacting flow in a combustor.

Experiments to investigate the effect of varying jet hole diameter and jet spacing on heat transfer and pressure loss from jet array impingement on a curved target surface are reported. The jet plate configurations studied have varying hole diameters and geometric spacing for spatial tuning of the heat transfer behavior. The configuration also includes a straight section downstream of the curved section, where the effect on heat transfer and pressure loss is also investigated. The jet plate holes are sharp-edged. A steady-state measurement technique utilizing temperature-sensitive paint (TSP) was used on the target surface to obtain local heat transfer coefficients. Pressure taps placed on the sidewall and jet plate of the channel were used to evaluate the flow distribution in the impingement channel. For all configurations, spent air is drawn out in a single direction which is tangential to the target plate curvature. First row jet Reynolds numbers ranging from 50,000 to 160,000 are reported. Further tests were performed to evaluate several modifications to the impingement array. These involve blocking several downstream rows of jets, measuring the subsequent shifts in the pressure and heat transfer data, and then applying different turbulator designs in an attempt to recover the loss in the heat transfer while retaining favorable pressure loss. It was found that by using W-shaped turbulators, the downstream surface average Nusselt number increases up to ∼13% as compared with a smooth case using the same amount of coolant. The results suggest that by properly combining impingement and turbulators (in the post impingement section), higher heat transfer, lower flow rate, and lower pressure drop are simultaneously obtained, thus providing an optimal scenario.

New experimental data are provided for full-coverage effusion cooling and impingement array cooling, as applied simultaneously onto the respective external and internal surfaces of a single instrumented test plate. For the effusion cooled surface, presented are spatially resolved distributions of surface adiabatic film cooling effectiveness, and surface heat transfer coefficients. For the impingement cooled surface, presented are spatially resolved distributions of surface Nusselt numbers. Impingement jet arrays at different jet Reynolds numbers, from 7930 to 18,000, are employed. Experimental data are given for spanwise and streamwise impingement hole spacing such that coolant jet hole centerlines are located midway between individual effusion hole entrances. For the effusion cooling, streamwise hole spacing and spanwise hole spacing (normalized by effusion hole diameter) are 15 and 4, respectively. Effusion hole angle is 25 deg, and effusion plate thickness is 3.0 effusion hole diameters. In regard to the impingement cooled cold-side surface of the effusion plate, associated surface Nusselt number variations provide evidence that impingement jets are turned and redirected as they cross the impingement passage, just prior to the entrance of coolant into individual effusion holes. In regard to the effusion cooled hot-side surface of the effusion plate, when compared at particular values of injectant and mainstream Reynolds numbers, streamwise location x / d e and blowing ratio BR, significantly increased thermal protection is provided when the effusion coolant is provided by an array of impingement cooling jets (compared to a cross flow channel supply arrangement).

A detailed aerothermal characterization of an advanced leading edge (LE) cooling system has been performed by means of experimental measurements. Heat transfer coefficient distribution has been evaluated exploiting a steady-state technique using thermochromic liquid crystals (TLCs), while flow field has been investigated by means of particle image velocimetry (PIV). The geometry key features are the multiple impinging jets and the four rows of coolant extraction holes, and their mass flow rate distribution is representative of real engine working conditions. Tests have been performed in both static and rotating conditions, replicating a typical range of jet Reynolds number (Re j ), from 10,000 to 40,000, and rotation number (Ro j ) up to 0.05. Different crossflow conditions (CR) have been used to simulate the three main blade regions (i.e., tip, mid, and hub). The aerothermal field turned out to be rather complex, but a good agreement between heat transfer coefficient and flow field measurement has been found. In particular, jet bending strongly depends on crossflow intensity, while rotation has a weak effect on both jet velocity core and area-averaged Nusselt number. Rotational effects increase for the lower crossflow tests. Heat transfer pattern shape has been found to be substantially Reynolds independent.

Driven by the need for higher cycle efficiencies, overall pressure ratios for gas turbine engines continue to be pushed higher thereby resulting in increasing gas temperatures. Secondary air, bled from the compressor, is used to cool turbine components and seal the cavities between stages from the hot main gas path. This paper compares a range of purge flows and two different purge hole configurations for introducing the purge flow into the rim cavities. In addition, the mate face gap leakage between vanes is investigated. For this particular study, stationary vanes at engine-relevant Mach and Reynolds numbers were used with a static rim seal and rim cavity to remove rotational effects and isolate gas path effects. Sealing effectiveness measurements, deduced from the use of CO 2 as a flow tracer, indicate that the effectiveness levels on the stator and rotor side of the cavity depend on the mass and momentum flux ratios of the purge jets relative to the swirl velocity. For a given purge flow rate, fewer purge holes resulted in better sealing than the case with a larger number of holes.

A novel parallelized, automated, and predictive imprint cooling model (PAPRICO) was developed for modeling of combustor liners using Reynolds-averaged Navier–Stokes (RANS). The methodology involves removing the film and effusion cooling jet geometry from the liner while retaining the cooling hole imprints on the liner. The PAPRICO can operate under two modalities, viz., two-sided and one-sided. For the two-sided PAPRICO model, the imprints are kept on the plenum and combustor sides of the liner. For the one-sided PAPRICO model, the imprints are retained only on the combustor side of the liner and there is no need for a plenum. The PAPRICO model neither needs a priori knowledge of the cooling flow rates through various combustor liner regions nor specific mesh partitioning. The imprint mass flow rate, momentum, enthalpy, turbulent kinetic energy, and eddy dissipation rate are included in the governing equations as volumetric source terms in cells adjacent to the liner on the combustor side. Additionally, the two-sided PAPRICO model includes corresponding volumetric sinks in cells adjacent to the liner on the plenum side. A referee combustor liner was simulated using PAPRICO under nonreacting flow conditions. The PAPRICO results were compared against predictions of nonreacting flow results of a resolved liner geometry, against a combustor liner with prescribed mass and enthalpy source terms (simplified liner) and against measurements. The results clearly conclude that PAPRICO can qualitatively and quantitatively emulate the local turbulent flow field with a reduced mesh size. The simplified liner fails to emulate the local turbulent flow field.