Lamyaa A. El-Gabry

Experimental Investigation of Local Heat Transfer Distribution on Smooth and Roughened Surfaces Under an Array of Angled Impinging Jets

Measurements of the local heat transfer distribution on smooth and roughened surfaces under an array of angled impinging jets are presented. The test rig is designed to simulate impingement with crossflow in one direction. Jet angle is varied between 30, 60, and 90 deg as measured from the target surface, which is either smooth or randomly roughened. Liquid crystal video thermography is used to capture surface temperature data at five different jet Reynolds numbers ranging between 15,000 and 35,000. The effect of jet angle, Reynolds number, gap, and surface roughness on heat transfer and pressure loss is determined along with the various interactions among these parameters.

Validation and Analysis of Numerical Results for a Two-Pass Trapezoidal Channel With Different Cooling Configurations of Trailing Edge

The fact that thermal efficiency of the gas turbine is related directly to the gas entry temperature has led to the operation of gas turbines at elevated temperatures. It is therefore necessary for the first stage vanes and blades to withstand temperatures higher than the melting point of the components. Air from the compressor is fed to the internal channels of the vanes/blades to keep the metal temperatures below the melting point. The channel is usually divided into three parts as leading edge, midspan, and trailing edge sections. The trailing edge region is difficult to cool as it is narrow and has very little space for coolant to flow. The treatment is difficult not only due to heat transfer problems but also due to the aerodynamic losses. The aerodynamic losses associated with the trailing edge enforce the requirement of a narrow and smooth trailing edge. Therefore a compromise is forced between blade cooling and aerodynamic losses. The conventional method of cooling the trailing edge is to provide the trailing edge slots where from the coolant leaves the blade. This ejected coolant mixes with the gas path and not only reduces the temperature of the main flow but also adds to the aerodynamic losses. On the other hand, the internal cooling for the trailing edge has limitations like small flow area. A better design of the internal cooling of the trailing edge is therefore required. In the case of internal cooling, the coolant is fed to the two-pass channels which are cast in the blade. The phenomenon like impingement, flow separation, and recirculation are characterized by the flow in such channels. The heat transfer enhancement is obtained on expense of the pressure drop. An acceptable two-pass channel requires optimized heat transfer enhancement and pressure drop. The literature survey has shown that these are modeled as rectangular or trapezoidal channels depending upon their location in the blade. Metzger and Sahm [1] varied the divider location and the gap at the 180 deg turn in a two-pass smooth rectangular channel and studied forced convection. They observed a nonuniform enhancement of heat transfer at the bend region due to flow characteristics at the bend. Park and Lau [2] studied the effect of sharp turning flows in a two-pass square channel and found very large spanwise variation in heat transfer at the turn and upstream of the outlet pass. Han et al. [3] and Liou et al. [4] showed that heat transfer enhances in a two-pass channel after the turn and this is due to the secondary flow generated by the centrifugal force at the turn. Many researchers [5–9] have studied the effect of channel aspect ratio on heat transfer characteristics. They concluded that heat transfer and pressure drop is influenced by the aspect ratio of channel. It was found that the pressure drop in a wide channel is more than the narrow channel. The two-pass channels are turbulated with ribs to enhance heat transfer. This leads to an increase in pressure drop as well. There have been many fundamental studies to understand the heat transfer enhancement phenomena by the flow separation caused by ribs. Han [10,11] studied the effect of rib pitch to rib height ratios (P/e) in stationary channels of different aspect ratios and found that a rib pitch to rib height ratio of 10 is optimum for heat transfer in these channels. Chandra et al. [12] showed that increase in the number of ribbed walls reduces the heat transfer performance. Wright et al. [13] studied the thermal performance of three different types of ribs (45 deg angled, V- and W-shaped) in a high aspect ratio (W/H = 4:1) channel with Reynolds number varied from 10,000 to 40,000. They found that W-shaped ribs performed better than the other two types of ribs. The shape of the trailing edge is such that it can be modeled as the trapezoidal channel. Taslim et al. [14] modeled the trailing edge as trapezoidal channel and studied the effect of bleed holes and tapered ribs on the heat transfer and pressure drop. Taslim et al. [15] found that in a trapezoidal channel, half-length ribs on two opposite walls enhance the heat transfer on the two walls with full-length ribs. Moon et al. [16] studied the local distributions of the heat transfer coefficient on all of the walls at the turn of a smooth two-pass channel with a trapezoidal cross section for various rates of airflow through the channel. The heat transfer was found to be higher at the turn and the outlet pass. The lowest flow rate resulted in highest heat transfer due to turn. Lee et al. [17] used naphthalene sublimation technique to study the heat (mass) transfer distribution in a two-pass trapezoidal channel with a 180 deg turn. The results were obtained over a range of Reynolds numbers for the channel with smooth walls and with ribs on one wall and on two opposite walls. They found that for all cases, the average heat transfer was higher on the downstream of the turn compared to that on the upstream of the turn. In addition, the shape of the local heat transfer distribution was found to be unaffected by the variation in flow rates. The trapezoidal shape of the cross-section of two-pass channels results in differences in the 3D fluid flow and heat transfer patters in them as compared to the square or rectangular two-pass channels. Cravero et al. [18] analyzed the flow field and heat transfer in a three-pass trapezoidal channel and showed that the geometry of the channel has a strong influence on flow field, especially at the regions of flow separation and recirculation. Taslim et al. [19] investigated trapezoidal cooling channels and showed that the trapezoidal channel has higher thermal performance compared to the square channel. It was concluded that the stronger interaction of the adjacent walls results in the increased heat transfer of the trapezoidal duct. Ekkad et al. [20] and Murata and Mochizuki [21] investigated heat transfer in straight and tapered (from hub-to-tip) two-pass ribbed channels. They found that at low Reynolds number, the heat transfer augmentation in the inlet pass is comparable in both cases, but at high Reynolds numbers the acceleration effect in the tapered channel leads to higher heat transfer as compared to the straight channel. At the outlet pass, the heat transfer was found to be comparable. Kiml et al. [22] studied the rib-induced secondary flow structure inside a trapezoidal channel with rib height proportional to the channel cross-section (proportional ribs) and constant height ribs (nonproportional ribs) at four rib inclinations, i.e., 90 deg, 75 deg, 60 deg, and 45 deg. They concluded that the proportional ribs offer less pressure losses, but they deteriorate the strength of the secondary flow rotational momentum as a result of wider space for the air flow between the rib and the opposite wall. Furthermore, the strength of the secondary flow rotational momentum increases with change of rib inclination from 90 deg to 45 deg. There have been many numerical studies performed to analyze the flow and heat transfer in the two-pass channel with a 180 deg bend. Lucci et al. [23] studied the performance of k-ɛ, k-ω, and RSM (Reynolds Stress Model) in the computation of the turbulent flow in a two-pass smooth channel. For the Reynolds number 100,000, all three models showed similar results. Su et al. [24] applied the RANS (Reynolds-averaged Navier–Stokes) approach in combination with a near wall, second moment turbulence closure to validate experiments performed at Re = 100,000. Pape et al. [25] successfully modeled a two-pass channel with a 180 deg bend and 45 deg ribs at Re = 100,000 using the realizable version of k-ɛ model with enhanced wall treatment. Shevchuk et al. [26] used the same model for a varying aspect ratio two-pass channel. The inlet channel aspect ratio (W/H) was 1:2, while for the outlet channel it was 1:1. The channels were roughened with 45 deg ribs and connected together with a 180 deg bend. The simulations were performed for Re = 100,000. They found the model to be effective and time efficient for high Reynolds number flows. To get more accurate results, advanced numerical methods like LES (Large Eddy Simulation) and DES (Detached Eddy Simulation) were used by Sewall and Tafti [27] and Viswanathan and Tafti [28]. They concluded that due to computational economy, it is more practical to use RANS methods based on turbulence models to study heat transfer in the convective cooling channel. In the present study, three different configurations of two-pass trapezoidal channels have been studied, keeping the trailing edge as the focus. Table ​Table11 shows the details of the three configurations. Table 1 The three configurations for augmented heat transfer at trailing edge The purpose of this study is to analyze the effect of placing ribs at the trailing edge on heat transfer and pressure drop in the channel for the three configurations. The results of this study should help gas turbine designers to improve the thermal performance of the gas turbine.

Turbulence and Heat Transfer Measurements in an Inclined Large Scale Film Cooling Array: Part II—Temperature and Heat Transfer Measurements

The second of a two-part paper, this study focuses on the temperature field and surface heat transfer measurements on a large-scale models of an inclined row of film cooling holes. Detailed surface and flow field measurements were taken and presented in Part I. The model consists of three holes of 1.9-cm diameter that are spaced 3 hole diameters apart and inclined 30° from the surface. Additionally, another model with an anti-vortex adaptation to the film cooling holes is also tested. The coolant stream is metered and cooled to 20°C below the mainstream temperature. A thermocouple is used to obtain the flow temperatures along the jet centerline and at various streamwise locations. Steady state liquid crystal thermography is used to obtain surface heat transfer coefficients. Results are obtained for blowing ratios of up to 2 in order to capture off-design conditions in which the jet is lifted. Film cooling effectiveness values of 0.4 and 0.15 were found along the centerline for blowing ratios of 1 and 2 respectively. In addition, an anti-vortex design was tested and found to have improved film effectiveness. This paper presents the detailed temperature contours showing the extent of mixing between the coolant and freestream and the local heat transfer results.Copyright

Turbulence and Heat Transfer Measurements in an Inclined Large Scale Film Cooling Array: Part I—Velocity and Turbulence Measurements

A large-scale model of an inclined row of film cooling holes is used to obtain detailed surface and flow field measurements that will enable future computational fluid dynamics code development and validation. The model consists of three holes of 1.9-cm diameter that are spaced 3 hole diameters apart and inclined 30° from the surface. The length to diameter ratio of the coolant holes is about 18. Measurements include film effectiveness using IR thermography and near wall thermocouples, heat transfer using liquid crystal thermography, flow field temperatures using a thermocouple, and velocity and turbulence quantities using hotwire anemometry. Results are obtained for blowing ratios of up to 2 in order to capture severe conditions in which the jet is lifted. This first part of the two-part paper presents the detailed velocity component and turbulence stresses along the centerline of the film-cooling hole and at various streamwise locations.Copyright

Measurements of Hub Flow Interaction on Film Cooled Nozzle Guide Vane in Transonic Annular Cascade

An experimental study has been performed in a transonic annular sector cascade of nozzle guide vanes (NGVs) to investigate the aerodynamic performance and the interaction between hub film cooling a ...

Numerical Modeling of Heat Transfer and Pressure Losses for an Uncooled Gas Turbine Blade Tip: Effect of Tip Clearance and Tip Geometry

A computational study has been performed to predict the heat transfer distribution on the blade tip surface for a representative gas turbine first stage blade. Computational fluid dynamics (CFD) predictions of blade tip heat transfer are compared with test measurements taken in a linear cascade, when available. The blade geometry has an inlet Mach number of 0.3 and an exit Mach number of 0.75, pressure ratio of 1.5, exit Reynolds number based on axial chord of 2.57×106, and total turning of 110 deg. Three blade tip configurations were considered; a flat tip, a full perimeter squealer, and an offset squealer where the rim is offset to the interior of the tip perimeter. These three tip geometries were modeled at three tip clearances of 1.25%, 2.0%, and 2.75% of the blade span. The tip heat transfer results of the numerical models agree well with data. For the case in which side-by-side comparison with test measurements in the open literature is possible, the magnitude of the heat transfer coefficient in the “sweet spot” matches data exactly and shows 20–50% better agreement with experiment than prior CFD predictions of this same case.

Numerical Study on the Sensitivity of Film Cooling CFD Results to Experimental and Numerical Uncertainties

Film cooling is used in a wide range of industrial and engineering applications; one of the most important is in gas turbine cooling. The intent of film cooling is to provide a layer of cool film between the surface and the hot gas. Predicting film-cooling characteristics, particularly at high blowing ratios where the film is likely to be detached from the surface, is a challenge due to the complex three-dimensional and possibly anisotropic nature of the flow. Despite the growth of more sophisticated techniques for modeling turbulence, such as large eddy simulation (LES), the most commonly used methods in design are Reynolds-Averaged Navier Stokes (RANS) methods that employ a two-equation turbulence model for specifying the eddy viscosity. Although these models have deficiencies, they continue to be used throughout industry because they offer reasonable turnaround time as compared to LES or other methods. This paper studies in detail two cases, one of high blowing ratio (off-design condition) of 2.0 and low blowing ratio of 0.5, and compares RANS-based computational fluid dynamics (CFD) results with experimental data for flow field temperatures and centerline, lateral, and span-averaged film effectiveness for a 35-degree circular jet. The effects of mainstream turbulence conditions, boundary layer thickness, and numerical dissipation are evaluated and found to have minimal impact in the wake region of separated films (i.e., they cannot account for the discrepancy between measured and predicted CFD results in the wake region). Analyses of low blowing ratio cases are in good agreement with data; however, there are some smaller discrepancies, particularly in lateral spreading of the jet.

Effect of Pulsed Film Cooling on Leading Edge Film Effectiveness

Detailed film effectiveness measurements have been made on a cylindrical leading edge surface for steady and pulsating flows. The film hole is off centered by 21.5 deg from the centerline and angled 20 deg to the surface and 90 deg from the streamwise direction. Two jet-to-cross-flow velocity ratios have been considered: VR=1 and 2, which correspond to blowing ratios of 1 and 2, respectively. The pulsating frequency is 10 Hz and the duty cycle is 50%. Comparisons between film effectiveness with a pulsating film and a continuous film show that for the same blowing ratio, the effectiveness of the film drops by a factor of 2 when the flow is pulsed. Hotwire measurements are made to characterize the pulsating velocity waveform at the exit of the film exit and verify the integrity of the pulse. The variation in the measured surface adiabatic wall temperature over the pulsing duration is very small, suggesting a large thermal inertia that keeps the wall surface largely unaffected by the time scale of the pulsations; this holds true for both blowing ratios tested.

Prediction of Heat Transfer Distribution Over the Surface of Nonfilm-Cooled Nozzle Guide Vane in a Transonic Annular Cascade

Having gas turbine components that can withstand high temperatures is a key factor in improving turbine efficiency; therefore, a deeper understanding of the heat transfer phenomena associated with the flow of hot gases over Nozzle Guide Vanes (NGVs) is crucial for proper vane design and implementation of adequate cooling schemes. In this study, the heat transfer distribution over the surface of a nonfilm-cooled NGV in a transonic annular cascade (Mexit=0.89, Reexit=2.6×106) is investigated numerically using a three-dimensional computational fluid dynamics (CFD) model and compared to results from a 2-D Boundary Layer (BL) code (TEXSTAN). The CFD model has been built and analyzed using a finite volume based commercial code (ANSYS CFX). Although the industrial turbine vane is film cooled, the analysis presented will be for the uncooled vane. In order to validate the CFD model against experimental data, a study is carried out on the NASA C3X vane; a CFD model of the C3X vane was built and several modeling parameters are varied in order to obtain good agreement with the experimental data. In addition, the numerical results are compared to those of other analytical and numerical simulations of the C3X vane. The methods found to yield the best agreement for the C3X are implemented in the modeling of the industrial NGV.Copyright

Numerical Modelling of Slot Film Cooling Using a Wall Function

CFD modeling of gas turbine film cooling remains a challenge for the computational arena due to the lack of robust accurate turbulence model or numerical technique to solve this highly complex problem. Modeling the exact behavior of the coolant jet is computationally expensive due to the complexity of the jet mainstream interaction, such as vortex generation, and separation. This paper, validation progress is presented using experimental data executed by the second author [GT2011-46491] and Thurman et-al [GT2011-46498] for blowing ratios of 1.0 and 2.0, and density ratio of 1.0. A wall function approach is chosen for a robust computation, and aiming for CPU time reduction. The in-house CFD code EOS is used to solve the RANS equations. A simple flow over flat plate validation problem was executed using experimental data of Klebanoff and El-Tahry as a code validation evidence. The computational results of the flow field were in agreement with the experimental measurements, with a slight over estimation of the thermal field due to over prediction of dissipation, resulting in less diffusion and mixing between the coolant jet and the mainstream. The wall function approach is shown to have great potentials for a robust accurate solution. The use of the continuous slot injection is known to be the best film cooling technique when compared to the other conventional circular coolant tubes, with the critical drawback of lowering the mechanical integrity of the blade. One way to overcome this problem is the use of discrete slots, which will optimize for better film cooling effectiveness, while maintaining the structural strength of the blade. Rectangular slots showed an increase in film cooling effectiveness for the same mass flow rate of coolant and blowing ratio when compared to the standard circular hole, which is due to the minor counter rotating vortex pair effect. An oval slot design showed higher film cooling effectiveness and more spreading, covering more surface area.Copyright