As a civil gas turbine cools down, asymmetric natural convective heat transfer causes the bottom sector of the rotor to cool faster than the top; this circumferential thermal gradient can potentially cause the shaft to deflect — a phenomenon called thermal or rotor bow. Rotor bow is tremendously difficult to predict due to its dependence on a number of engine design parameters, in addition to the complex nature of natural convective flows. A novel experimental facility has been developed to gain further understanding into shutdown cooling of a gas turbine. The scope of this paper is to quantify the effect of basic design features on natural convective cooling in an engine annulus during shut-down; specifically the influence of the thermal boundary wall conditions and the annular diameter ratio. In addition to this, a low-cost, robust thermocouple probe has been developed and validated, which allows for accurate temperature measurements in a natural convective boundary layer. An extensive experimental campaign has been completed. The key finding is that the local radial wall temperature difference was found to be the most influential parameter on the local heat transfer. Non-isothermal walls did not alter the overall distribution of the inner wall equivalent conductivity. This was true for both cylindrical and conical sections. An appropriate characteristic length for use in the Rayleigh number definition for both the concentric cylinder and conical sections have been validated. The conical inner section (5° hade angle) did not affect the overall heat transfer in the range of conditions tested. Therefore, the mean surface heat transfer for non-isothermal inner and outer profiles, within the range −0.4 < ΔRa/Ra L c < 0.4, where the thermal gradient is negative in the clockwise from top-dead-centre, can be predicted using isothermal correlations for Ra L c < 5.0 × 10 5 and Dr <= 1.5.

High pressure turbine blade tips are critical for gas turbine performance and are sensitive to small geometric variations. For this reason, it is increasingly important for experiments and simulations to consider real geometry features. One commonly absent detail is the presence of welding beads on the cavity of the blade tip, which are an inherent by-product of the blade manufacturing process. This paper therefore investigates how such welds affect the Nusselt number, film cooling effectiveness and aerodynamic performance. Measurements are performed on a linear cascade of high pressure turbine blades at engine realistic Mach and Reynolds numbers. Two cooled blade tip geometries were tested: a baseline squealer geometry without welding beads, and a case with representative welding beads added to the tip cavity. Combinations of two tip gaps and several coolant mass flow rates were analysed. Pressure sensitive paint was used to measure the adiabatic film cooling effectiveness on the tip, which is supplemented by heat transfer coefficient measurements obtained via infrared thermography. Drawing from all of this data, it is shown that the weld beads have a generally detrimental impact on thermal performance, but with local variations. Aerodynamic loss measured downstream of the cascade is shown to be largely insensitive to the weld beads.

The Transient Heat Transfer Facility (THTF) was developed at the University of Oxford to test full-scale high pressure compressor and turbine casing air systems using gas turbine engine representative secondary air system mass flow rates, total temperatures and total pressures. Transient casing response together with blade and disc responses governs achievable tip clearances in both compressors and turbines. Traditionally, casing thermal response is faster than disc response and so requires cooling. However, future engine designs may have thermal responses of the discs that are faster than the casings. In this paper we investigate the use of air impingement as a means to speed up the casing response to match such future designs. Two different impingement configurations were selected from a total of eight designs based on steady Reynolds-averaged Navier-Stokes (RANS) predictions of maximum heat transfer coefficient (HTC). The 3D thermal growth of the casing was characterised by the surface temperature rise over a given period of time to assess achievable dynamic response. The configurations were tested in the THTF as a rainbow set. The resulting casing metal temperatures were used in conjunction with an explicit solution of the conduction in the casing wall to calculate HTC due to impingement. The experimental set-up resembles a typical aircraft engine and therefore includes measurement uncertainties arising from features such as fixtures, seals, geometries and large surface areas that are then subjected to varying thermal inertias. These can lead to circumferential temperature non-uniformities, as evident from the experimental results. An uncertainty analysis was performed to quantify these effects on the overall casing thermal response. The experimental data was then compared against numerical predictions from an axisymmetric, 90° sector, conjugate heat transfer model of the facility using the two impingement plate designs. Overall, the values of HTC agreed well across the experimental and numerical results. Both approaches predicted differences in which of the two designs was more effective; however, advantages and limitations were identified in both approaches. The combined experimental and numerical study shows the significance of analysing the full annulus, at engine representative conditions and the benefit of an impingement array to potentially speed up casing response for future engines.

This study investigates the effect of inlet velocity profiles and rib orientations on the Nusselt number distribution within ribbed radial turbine cooling passages representative of systems used in current engines. A triple-pass serpentine passage is investigated, which includes rib turbulators angled at 45° and 180° bends. The first two passes are radially outward and inward respectively and both have an aspect ratio of 1:4, with the third pass radially outward with an aspect ratio of 1:2. Multiple inlet velocity profiles are studied in RANS CFD simulations under both stationary and rotating conditions. The rotating simulations have Reynolds, Rotation and Buoyancy numbers representative of a passage within a HP turbine blade of a gas turbine engine. The flow structure and Nusselt number distributions are discussed in detail with the inlet velocity profile found to have a very large influence in the first pass under both stationary and rotating conditions, with smaller differences observed in the later passes. The rib orientation in the second pass was also investigated, with simulations of reversed and non-reversed rib orientation compared. Improved heat transfer characteristics were found in simulations where the ribs were orientated in the same direction for all three passages. These simulations are compared to experimental results in order to explain previous discrepancies found between experimental and CFD data from an experimental setup with complex inlet geometry.

This study investigates the effect of rotation on the Nusselt number distribution within ribbed radial turbine cooling passages representative of systems used in current jet engines. The results are unusual in that the cooling passage length to diameter ratio is engine representative and full distributions of local Nusselt number have been measured using the transient liquid crystal method. The results are compared to RANS CFD simulations and the level of agreement discussed in detail. A triple-pass serpentine passage is investigated, which includes 45° filleted rib-turbulators and 180° curved bends. The first two passes have an aspect ratio of 1:4 which are radially inward and outward respectively, with the final pass being radially outward with an aspect ratio of 1:2. The Reynolds, Rotation and Buoyancy numbers are all representative of a passage within a HP turbine blade of a gas turbine engine at 97000/108000, 0.081/0.088 and 0.052/0.035 respectively for the 1:4/1:2 aspect ratio passages. CFD simulations are found to give good predictions under stationary conditions however significant differences are observed when rotation is introduced. The Nusselt number distributions depend strongly on both rotation and upstream flow conditions created by the specific geometry.

Numerical modelling of internal cooling passages in gas turbine blades is a challenging task due to their physical characteristics, such as rounded duct corners, the presence of rib turbulators and their staggered locations between surfaces. This results in complex fluid dynamic phenomenon such as counter-rotating vortices and other secondary flow structures that can drive the heat transfer. Heat transfer mechanisms in such passages are inherently coupled with momentum transport and diffusion. Current industry practices for numerical modelling of such passages use unstructured mesh generation tools, steady Reynolds-averaged Navier-Stokes (RANS) equations and two-equation turbulence models such as k-ε and k-ω SST. This paper investigates two generic, engine-representative rib geometries using current numerical practices to determine their limitations. Three mesh generation tools and two turbulence models are compared across two rib geometries. The results are qualitatively and quantitatively compared to detailed experimental Nusselt numbers on the passage walls. It was found that as long as the rib geometry results in a secondary flow that directly impinges onto the wall, the meshing tools and turbulence models agree reasonably well with experiments. When the passage includes wall-wrapped ribs resulting in more complex secondary flows, this decreases the validity of the numerical tools, suggesting that more sophisticated modelling techniques are required as rib geometries continue to evolve.

This paper describes the design and development of a major new facility capable of reproducing flow conditions for large civil aero engine turbine and compressor segment cavities. The facility reproduces in 3-D, typical secondary air system cruise temperatures, pressures and mass flows. The facility allows better understanding of the circumferential heat transfer effects of air system flows on turbine and compressor casings, as these have a significant influence on both steady-state and transient blade tip clearance behavior. Different air system architectural solutions are considered and 1-D transient air system modelling is used to design architecture to ensure that they meet tight performance requirements in terms of allowable pressure and temperature ripples when subject to fast switching of flows. An explanation of the design of the double-skin pressure vessel using 2-D axisymmetric thermo-mechanical analysis to understand the life-limiting features of the design is discussed. Finally we conclude the paper with some experimental results from rig commissioning and the use of these results in the determination of engine casing transients is discussed.

Complete distributions of local heat transfer coefficient have been measured over the full surface of a pedestal bank at a number of Reynolds numbers. The transient heat transfer method with thermochromic liquid crystals used as the surface thermometer was used to obtain the data. The pedestal geometry included fillet radii representative of those used in engine blade cooling passages. The heat transfer coefficient distributions are compared to previously reported local measurements made on a bank of plain cylinders. Averaged values based on the local mixed bulk gas temperature are compared to established correlations. The distributions of local heat transfer coefficient show a remarkable periodicity when based on a local temperature. These patterns are discussed in terms of the interpreted flow field. Measurements of pressure drop are also reported and, for a range of engine representative Reynolds numbers, agree with an existing correlation for prismatic pedestals.

Most research involving arrays of impinging jets was conducted using steady state techniques which allow the impingement plate (through which the gas flows) to achieve an equilibrium (adiabatic) temperature during the test. Invariably, the impingement plate temperature was not reported for these tests as the floating temperature condition was taken to be representative of conditions in the application being modeled. Thermal analysis of gas turbine conditions showed the present authors that conditions in the engine could often be significantly different from this floating plate temperature state. Such conditions include engine operating point transients and situations in which the plate is fixed to the aerofoil in such a way to achieve good thermal contact. Furthermore, the capacity of the impingement plate to contribute to enhanced heat transfer by paying attention to the thermal boundary conditions at its support has not been realized. The influence of the impingement plate temperature on local target surface heat transfer was fully quantified by Van Treuren et al. (1993, 1994 and 1996), using a transient liquid crystal heat transfer technique. Superposition was used to show that the target surface heat flux can be written as the summation af two separate heat transfer coefficients. These temperature difference products quantify the contributions of the impingement plate and the target surface thermal boundary conditions. In other words: (1) q = h j T w - T j + h p T w - T p Van Treuren et al.’s experiments showed the heat transfer coefficient for target surface heat flux and impingement plate to target surface temperature difference, h p , can be up to 40% of the heat transfer coefficient for plenum to target surface temperature difference, h j , in crossflow areas away from the jet stagnation zone. The present report covers steady state experiments conducted at three average jet Reynolds numbers (10,000, 14,000, and 18,000) and two impingement to target plate spacings (1 and 4) for an inline array of jets. The purpose of the experiments was to determine the adiabatic impingement plate temperature expressed as a non-dimensional temperature difference, θ. The data allow the difference in thermal boundary conditions between the steady state experiments and the transient heat transfer experiments to be accounted for.

Recent work, Van Treuren et al. (1993, 1994), has shown the transient method of measuring heat transfer under an array of impinging jets allows the determination of local values of adiabatic wall temperature and heat transfer coefficient over the complete surface of the target plate. Using this technique, an inline and staggered array of impinging jets was tested over a range of average jet Reynolds numbers (10,000–40,000) for three impingement plate to target plate separations (1, 2 and 4). The array was confined on three sides and spent flow was allowed to exit in one direction. Local and average values are presented. These values for the two array configurations are then compared with each other as well as with previously published data in related geometries. A new correlation technique is presented, based on the local data, which breaks the target surface into jet and crossflow areas of interest, with excellent results. The correlation uses the local jet Reynolds number and local jet-to-crossflow mass velocity ratio. This new technique compared favourably with published correlations. Also presented is the influence of the impingement plate on the target plate heat transfer in the form of an effectiveness parameter. This influence is accounted for in the correlation.

To optimize the efficiency of modern aero-gas turbine engines the turbine tip clearances must be tightly controlled so as to minimize leakage losses. In addition, the clearance control system must be able to respond with sufficient rapidity to engine thermal transients. One method of achieving turbine tip-clearance control is to manipulate the turbine casing temperature, and thereby radial growth, by convective cooling. The consequent clearance control system represents a particularly complex thermo-mechanical design problem. The current experimental study aims to simulate the heat loads to which the internal surfaces of the casing are typically exposed and to characterize the radial and axial displacement of the free-body casing under varying external cooling conditions. Importantly, the newly commissioned test facility allows a realistic assessment of the casing cooling impact on dimensional control, and also the rapid characterization and comparison of different concepts. The test facility comprises a model of a high-pressure/intermediate-pressure turbine casing with generic impingement cooling manifolds. A radiant heater is mounted within the casing model such that a near-uniform heat flux condition can be established on the casing wall inner surface. Extensive surface mounted thermocouples are welded to the casing wall to monitor variations in metal temperature. Radial and axial displacement of the casing is monitored using laser triangulation and linear variable differential transformer sensors. Experiments have been conducted over a range of heat load conditions and with engine representative levels of casing cooling applied. Importantly, the new test facility allows for the characterization of the casing cooling system as a whole.

A study of a large-scale model of an engine representative impingement cooling system has been performed. A series of tests have been carried out to fully characterise the behaviour of the system. These include cold flow diagnostic tests to determine the pressure loss and the static pressure distribution, and flow visualisation to assess surface shear. The surface shear stress pattern provided by multiple stripes of coloured paint applied to the target surface yielded important information on the near wall flow features far from the jet axis. The row solved flow and pressure distributions are compared to industry standard predictions. Heat transfer tests using the transient liquid crystal technique were also conducted using coatings comprised of a mixture of three thermochromic liquid crystals. Analysis of the thermochromic liquid crystal data was enhanced by recent developments in image processing. In addition, an energy balance approach to analysing signals from fast response thermocouples for air temperature measurement was applied to verify the levels of heat transfer coefficients on surfaces not coated with the temperature sensitive liquid crystal.

Detailed heat transfer coefficient distributions have been measured on both surfaces of the impingement plate of an engine-representative impingement cooling system using the thermochromic liquid crystal (TLC) transient technique. The color images of the TLC on the impingement downstream surface provide evidence of a re-impingement flow. The re-impingement flow is found to contribute to local increases in the heat transfer on the impingement plate downstream surface. It was found that the average heat transfer coefficient on the impingement downstream surface is about 50% of the average target surface heat transfer coefficient. The results are compared with a previously reported correlation.