The VKI Compression Tube Annular Cascade Facility CT3

Author(s):  
C. H. Sieverding ◽  
T. Arts

The purpose of this paper is to present the new transonic, annular facility developed at the von Karman Institute to investigate the aerodynamic heat transfer performances of real size advanced aero-engine and gas turbine components at correctly simulated operating conditions. The facility operates under the principle of an Isentropic Light Piston Compression Tube. Its definite advantage over classical blowdown wind tunnels is to independently model the freestream Mach and Reynolds numbers as well as the gas/wall/coolant temperature ratios. Its running time ranges between 0.1 and 1 s. The first part of the paper describes the design, the manufacturing and the installation of the different components of the wind tunnel and the test section. The second part deals with the different measurement techniques applied for aerodynamic and heat transfer measurements; it also describes some examples of the flow quality obtained in this new facility.

Author(s):  
M. D. Polanka ◽  
R. J. Anthony ◽  
David G. Bogard ◽  
Mark F. Reeder

Film cooling technology has been around for many decades and many significant advances in cooling effectiveness have been made at many different facilities using several different methods. A large proportion of film cooling research is successfully carried out using simplified scaled-up models in wind tunnels coupled with novel measurement techniques. These tests have been very effective in assessing basic film cooling parameters for many cooling hole geometries, patterns, and blowing ratios. In real engines, however, film cooling designs are ultimately subjected to highly unsteady 3-D secondary flows and rotational effects. Few film cooling experiments have quantified these effects on real, true scale turbine hardware in a rotating test environment. The Turbine Research Facility (TRF) at the Air Force Research Laboratory has been acquiring uncooled heat transfer measurements on full scale metallic airfoils both with and without rotation for several years. The addition of cooling flow to this type of facility has provided new capability, and new challenges. The primary two issues being that the film temperature is unknown and that the airfoil is no longer semi-infinite. This makes it more difficult to extract the adiabatic effectiveness and the heat transfer coefficient from the measurements of surface temperature and surface heat transfer since conventional methods used in most other experiments are not valid in this case. In contrast another cooling parameter, the overall effectiveness, is readily obtained from measurements of surface temperature, internal coolant temperature, and mainstream temperature. The overall effectiveness is a normalized measure of metal surface temperatures expected for actual operating conditions. It is the goal of this paper to evaluate how measurements, obtained from a transient blowdown facility like the TRF, can be used to quantify the expected performance of a film cooled turbine airfoil. Additionally, it is imperative to properly correlate these experimental results to the true engine conditions. The data required for this analysis has been collected using an array of surface mounted thermocouples and thin film gauges in a series of experiments where freestream temperatures and coolant temperatures and mass flow rates were varied. The airfoil used in this investigation was a thin walled metallic airfoil with a showerhead cooling scheme and several rows of normal holes on both the pressure and suction sides of the airfoil. The flow is typical of that seen in a modern high pressure turbine — that is an inlet Mach number of about 0.1 accelerating toward sonic at the throat with a high inlet freestream turbulence level of about 20%.


Author(s):  
D. J. Stankiewicz ◽  
T. R. Kirkham

A technique of heat transfer enhancement is investigated whereby the internal span-wise cooling passages of a typical first stage gas turbine blade are modified by the introduction of circumferential ribs. The technique is verified by the use of a test rig incorporating a heated internally ribbed tube operating at the same range of Mach and Reynolds numbers as the turbine blade as well as by a test rig incorporating actual production blades immersed in a heated oil bath.


Author(s):  
G. A. Sorokin ◽  
G. P. Bogoslovskaya ◽  
E. F. Ivanov ◽  
A. P. Sorokin

Boiling experiments on eutectic sodium-potassium alloy in the model of fast reactor subassembly under conditions of low-velocity circulation carried out at the IPPE call for further investigations into numerical modeling of the process. The paper presents analysis of pin bundle liquid metal boiling, stages of the process, its characteristics (wall temperature, coolant temperature, flow rate. pressure void fraction and others), that allowed the pattern map to be drawn. The problem of conversion of the data gained in Na-K mock-up experiments to in-pile sodium reactor operating conditions is analyzed here, as well as thermodynamic similarity of liquid metal coolants and eutectic Na-K alloy. Data on bundle boiling in Na-K are presented in comparison with those in different liquid metals. Analysis of data on liquid metal heat transfer in cases of pool boiling, boiling in tubes, in slots, and in pin bundles, as well as data on critical heat flux in tubes was performed and discussed in the paper. The relationship for calculation of critical heat flux in liquid metal derived by the authors is presented. Results of numerical modeling of liquid metal boiling heat transfer during accident cooling of reactor core applied to experimental conditions of going from forced to natural circulation are presented, too.


Author(s):  
Muzafar Hussain ◽  
Shahbaz Tahir

Abstract Nanofluids are widely adopted nowadays to enhance the heat transfer characteristics in the solar applications because of their excellent thermophysical properties. In this paper, a modified Eulerian-Eulerian model recently developed based on experiments was validated numerically to account for the deviations from the experimental data. The modified Eulerian-Eulerian model is compared with the single-phase model, Eulerian-Eulerian models for TiO2-water at different operating conditions and deviation from the experimental data for each of the model was documented. However, the modified Eulerian-Eulerian model gave much closer results when compared to the experimental data. For the further extension of work, the modified Eulerian-Eulerian model was applied to different nanofluids in order to investigate their heat transfer characteristics. Three different nanoparticles were investigated namely Cu, MgO, and Ag and their heat transfer characteristics is calculated based on the modified Eulerian-Eulerian model as well as the single-phase model for the comparison. For lower values of Reynolds numbers, the average heat transfer coefficient was almost identical for both models with small percentage of error but for higher Reynolds numbers, the deviation got larger. Therefore, single-phase model is not appropriate for higher Reynolds numbers and modified Eulerian-Eulerian model should be used to accurately predict the heat transfer characteristics of the nanofluids at higher Reynolds numbers. From the analysis it is found that the Ag-water nanofluid have the highest heat transfer characteristics among others and can be employed in the solar heat exchangers to enhance the heat transfer characteristics and to further improve the efficiency.


Author(s):  
S. M. Guo ◽  
M. L. G. Oldfield ◽  
A. J. Rawlinson

The influence of localized pin-shaped surface roughness (P-Pins) on heat transfer and aerodynamics of a fully film cooled engine aerofoil has been studied in a transonic annular cascade. The “P-Pins”, present on some casting film cooled turbine blades and vanes, are the residues left in the manufacturing process. This paper investigates the effect of the P-Pins on the aerodynamic performance and measures the heat transfer consequences both for the aerofoils and the P-Pins. The effect on performance was determined independently on the pressure and suction surface of the aerofoil. For comparison, the aerofoil without P-Pins was also tested to provide baseline results. The measurements have been made at engine representative Mach and Reynolds numbers. Wide band liquid crystal and direct heat flux gauge technique were employed in the heat transfer tests. A four-hole pyramid probe was used to obtain the aerodynamic data. The aerodynamic and thermodynamic characteristics of the coolant flow have been modelled to represent engine conditions by using a heavy “foreign gas” (30.2% SF6 and 69.8% Ar by weight). The paper concludes that P-Pins as usually placed on the blade do not have detrimental effects to the heat transfer performance of film-cooled aerofoil. P-Pins, located in thick boundary layer regions of the aerofoil, such as the later portion of the suction surface, do not cause any reduction of aerofoil aerodynamic efficiency. For contrast, the P-Pins located in the thin boundary layer regions on the pressure side of the aerofoil cause noticeably more losses.


1992 ◽  
Vol 114 (4) ◽  
pp. 734-740 ◽  
Author(s):  
S. P. Harasgama ◽  
C. D. Burton

Heat transfer and aerodynamic measurements have been made on the endwalls of an annular cascade of turbine nozzle guide vanes in the presence of film cooling. The results indicate that high levels of cooling effectiveness can be achieved on the endwalls of turbine nozzle guide vanes (NGV). The NGV were operated at the correct engine nondimensional conditions of Reynolds number, Mach number, gas-to-wall temperature ratio, and gas-to-coolant density ratio. The results show that the secondary flow and horseshoe vortex act on the coolant, which is convected toward the suction side of the NG V endwall passage. Consequently the coolant does not quite reach the pressure side/casing trailing edge, leading to diminished cooling in this region. Increasing the blowing rate from 0.52 to 1.1 results in significant reductions in heat transfer to the endwall. Similar trends are evident when the coolant temperature is reduced. Measured heat transfer rates indicate that over most of the endwall region the film cooling reduces the Nusselt number by 50 to 75 percent.


1977 ◽  
Vol 11 (5) ◽  
pp. 736-739 ◽  
Author(s):  
M. M. Ardasheva ◽  
V. Ya. Borovoi ◽  
P. I. Gorenbukh ◽  
M. V. Ryzhkova

2014 ◽  
Vol 137 (3) ◽  
Author(s):  
Ignacio Mayo ◽  
Tony Arts ◽  
Ahmed El-Habib ◽  
Benjamin Parres

The convective heat transfer distribution in a rib-roughened rotating internal cooling channel was measured for different rotation and Reynolds numbers, representative of engine operating conditions. The test section consisted of a channel of aspect ratio equal to 0.9 with one wall equipped with eight ribs perpendicular to the main flow direction. The pitch to rib height ratio was 10 and the rib blockage was 10%. The test rig was designed to provide a uniform heat flux boundary condition over the ribbed wall, minimizing the heat transfer losses and allowing temperature measurements at significant rotation rates. Steady-state liquid crystal thermography (LCT) was employed to quantify a detailed 2D distribution of the wall temperature, allowing the determination of the convective heat transfer coefficient along the area between the sixth and eighth rib. The channel and all the required instrumentation were mounted on a large rotating disk, providing the same spatial resolution and measurement accuracy as in a stationary rig. The assembly was able to rotate both in clockwise and counterclockwise directions, so that the investigated wall was acting either as leading or trailing side, respectively. The tested Reynolds number values (based on the hydraulic diameter of the channel) were 15,000, 20,000, 30,000, and 40,000. The maximum rotation number values were ranging between 0.12 (Re = 40,000) and 0.30 (Re = 15,000). Turbulence profiles and secondary flows modified by rotation have shown their impact not only on the average value of the heat transfer coefficient but also on its distribution. On the trailing side, the heat transfer distribution flattens as the rotation number increases, while its averaged value increases due to the turbulence enhancement and secondary flows induced by the rotation. On the leading side, the secondary flows counteract the turbulence reduction and the overall heat transfer coefficient exhibits a limited decrease. In the latter case, the secondary flows are responsible for high heat transfer gradients on the investigated area.


Author(s):  
Joel H. Wagner ◽  
Jay C. Kim ◽  
Bruce V. Johnson

Internal convective cooling is used to maintain acceptable gas turbine rotor blade temperatures. The heat transfer from the blade coolant passage walls is governed by forced convection, Coriolis forces and buoyance due to wall and coolant temperature differences. Currently little data is available to designers regarding the combined effects of these three parameters. To obtain required data, a rotating heat transfer facility was developed for experiments with large scale models and run at rotation and flow parameters typical of current gas turbine operating conditions. Analysis of the equations of motion showed that the perinent nondimensional parameters were Reynolds number, Rossby number, the difference in wall fluid and bulk fluid density and geometric ratios. The models were instrumented to measure average heat transfer rates on the coolant passage wall elements, and with pressure taps for friction data. An initial set of experiments have been conducted with rough wall geometries, typical of those used in blades. Results from the rotating experiments showed large heat transfer coefficient increases and decreases on the coolant passage leading and trailing surfaces compared to nonrotating heat transfer coefficients. The heat transfer was shown to be a function of inward or outward flow direction and Rossby number for the experiments conducted.


Author(s):  
M. C. Spencer ◽  
G. D. Lock ◽  
T. V. Jones ◽  
N. W. Harvey

Aerodynamic and heat transfer measurements have been made on the hub and casing endwalls of an annular cascade of high pressure nozzle guide vanes. The measurements have been made over a range of engine representative Mach and Reynolds numbers and with large levels of freestream turbulence intensity. The transient liquid crystal technique has been employed, which has the advantage of yielding full surface maps of heat transfer coefficient. Computational predictions and aerodynamic measurements of Mach number distributions on the endwall surfaces are also presented, along with surface-shear flow visualisation using oil and dye techniques. The heat transfer results are discussed and interpreted in terms of the secondary flow and Mach number patterns.


Sign in / Sign up

Export Citation Format

Share Document