During the static testing of a single spool turbojet engine (400 N Titan Engine) running on Jet A-1 as fuel, the starting sequence was unsuccessful, however, upon substituting the fuel with Diesel the engine started successfully. Till that point, the engine operation/performance was normal, and a series of tests were conducted for a total time of 5 hours. These 5 hours of testing spanned over five months during which ambient temperature varied from 5 °C to 37 °C, and relative humidity varied from 40% to 90%. The starting sequence was unsuccessful at the maximum ambient temperature (37 °C) and moderate relative humidity condition (50%). During the test, various parameters including pressure, temperature, rotor speed and mass flow rate were measured and recorded using a computer based data acquisition. For the failed start-up case, during the starting sequence, a sudden decrease in rotor speed was observed as the rotor reached ∼80% of the calibration value followed by flameout in combustion chamber. This sudden drop in rotor speed occurred due to mild-surge in compressor which continued up to flameout. The cause of mild-surge is attributed to the momentary blockage at the exit of the compressor caused by the combustion process. These pressure fluctuations propagated downstream leading to flame out. The combination of high ambient temperature (35 °C) and moderate humidity (∼50%) resulted in higher combustion chamber temperature which resulted in the momentary blockage at compressor exit during the transient operation. The substitution of Diesel as fuel resulted in lower combustion chamber temperature, and resolved the issue. It is important to note that, the mild-surge observed in this case is caused by downstream condition, and would occur only at some special situations (inlet conditions). Hence, it may be concluded, that the compressor surge in a jet engine may be caused by downstream components (under some special conditions), and design modifications to these downstream components may resolve the issue in those cases.

An engine test facility, capable of being operated with different nozzles, has been developed. The shape of this component has evolved substantially, over the years, from seemingly simple circular geometries to very complex geometries designed to address different requirements. There are numerous test facilities that facilitate an extensive standalone component level testing of these components. However, such testing does not offer an insight into the effect these complex geometries will have on the performance of a gas turbine engine. This facility will serve as a demonstration for graduate and doctoral students, enhancing their understanding of engine performance. The core of the test facility is a single spool turbojet engine from AMT, Netherlands. The engine is instrumented with pressure and temperature measurements at every inter-component location. A conventional intake duct is designed, for measurement of air flow rate. The entire engine is mounted on a 6-axis force measurement device, for measurement of thrust during engine operations. Towards the rear of the engine, a straight duct is attached after the jet pipe, on which any new nozzle can be retrofitted to the engine. To address changes in operating conditions when using different nozzles, the engine is equipped with a suitably designed bleed duct, which is attached onto the jet pipe. The engine, successfully integrated with all the components, has been subjected to multiple tests at different power settings. The first test was done using the baseline engine, operated at various rotor speeds. The next test was done with a custom nozzle having an area ratio of 2, attached to the jet pipe. Owing to the successful design of the components, the engine performance was measured when operated with a smaller area nozzle.

Thrust vectoring is a requirement for fifth generation fighters, giving them super-maneuverability capabilities, allowing them to execute tactical maneuvers that are not possible using conventional aerodynamic mechanisms. The most widely used and successful method for achieving this is by using gimbaled engines or nozzles. The complexities involved in this method, have encouraged future engine designers to explore different avenues for achieving thrust vectoring, one of which is fluidic thrust vectoring. In fluidic thrust vectoring, jet deflection is achieved by fluid injection at various locations on the nozzle. During thrust vectoring operations, the engine performance is affected. This is primarily due to the change in effective nozzle area. When a nozzle is gimbaled, as is the method used in currently operational thrust vectored engines, or during fluidic thrust vectoring operations, there is a change in effective nozzle area. This impacts the engine mass flow rate, thus affecting the engine operation. The change in performance is similar to that of an engine fitted with a variable area nozzle. In this study, we attempted to retrofit a thrust vectoring nozzle to an existing engine designed for a fourth-generation fighter aircraft, in order to give it fifth-generation fighter aircraft capabilities. A Twin spool mixed flow turbofan engine with a convergent nozzle is selected and its performance is simulated using Gasturb 13. The baseline engine consists of a low pressure spool, high pressure spool, combustion chamber and convergent-divergent nozzle. For the sake of simplicity, the convergent-divergent nozzle is replaced with a convergent nozzle, with no loss in thrust at design point. The design point is arrived at based on engine data available in open literature. Following this, offdesign performance is simulated, for studying the effect of thrust vectoring operations, which are modeled as a nozzle area change. Suitably scaled generic maps provided in Gasturb are used for off-design simulations. The effect of nozzle area change on engine performance is studied at sea level static conditions. The nozzle area is decreased by a maximum of 15%, in steps of 1%. During thrust vectoring operations, there is a significant change in bypass ratio and fan surge margin, with the other performance parameters being relatively constant. Following this, simulations are conducted at different flight conditions to understand the effect of nozzle area change for different flight regimes. A total of seven different flight conditions are selected to understand the operational envelope of thrust vectoring operation. It is found that at all flight conditions, thrust vectoring has a significant influence on bypass ratio and fan surge margin. While for most conditions, there is an improvement in fan surge margin, there are two conditions where fan surge margin decreases substantially.

This study deals with the investigations on the sources and the control of combustion noise, in an atmospheric gas turbine combustor. Combustion noise encountered here is also termed as hooting , as it occurs within a limited bandwidth of frequencies ranging from 300–450 Hz. Combustion noise is usually classified as direct and in-direct combustion noise. The present study emphases on the direct combustion noise which occurs when a volume of gas expands at constant pressure, as soon as it is heated by combustion; this results in a sound wave which propagates outside the boundary of the flame. At certain conditions, if the unsteady heat release rate drives the acoustic oscillations, satisfying Rayleigh criterion, pressure oscillations grow leading to discrete tonal sound and this phenomena is termed as combustion instability. Experiments are conducted in a liquid fuelled swirl stabilized atmospheric gas turbine combustor, whose aspect ratio is 2.5, combustion intensity varies from 25MW/m 3 atm to 50MW/m 3 atm. Air is passed through various stages: primary, secondary, quenching and atomizing air. Aviation turbine fuel is injected through an air-blast atomizer. An unsteady pressure transducer is located at the primary zone to measure the acoustic oscillations. The frequency of sound generated during the combustion process is compared with a microphone located at 1.25 m away from the combustor at an angle of 45° from the axis of the combustor. The main objective of this paper is to investigate the cause of hooting and the passive control techniques in order to reduce it. This is achieved by two ways, systematically by two ways, i) reducing the quenching air methodologically reducing the quenching air and ii) varying the air to liquid (fuel) ratio (ALR). By imparting these methods the sound pressure level inside the combustor is reduced from 143dB to 128dB. Since, combustion noise occurs in a broad-band of frequencies, the unsteady data obtained with the pressure transducer are analyzed using octave bands, where it shows linear decrement of energy present in-between the two frequencies. In order to perform parametric study, two swirlers of swirl numbers.0.77 and 0.86 are used. Established by the physics of Helmholtz oscillator, the combustor hooting is dictated, by which the sound produced by the combustor is explained.

The benefits of minimizing the weight of an aircraft are substantial, due to which all aircraft components are designed so as to perform acceptably with minimal weight. As a result, modern compressor, fan and turbine blades are increasingly being designed with thinner airfoil profiles, while also being subjected to high loading, so as to improve the thrust-to-weight ratio of the engine. These conditions make the blades extremely likely to result in large-amplitude vibrations. A class of critical vibrations are caused due to aeroelastic instabilities within the turbomachinery blade rows. Forced response or the self-excited flutter can lead to high-cycle fatigue which can cause a catastrophic failure of the blades. The absence of a reliable prediction methodology for the occurrence of these instabilities signify the importance of unsteady aerodynamic studies in turbomachine cascades. The present experiments are conducted on a newly commissioned annular cascade tunnel. The test section consisting of 14 compressor blades is studied under subsonic conditions. The profiles of the blades have a constant span design, and the cascade parameters are chosen as EPFL’s Second Standard Test Configuration at the mid-span location. A set of guide vanes upstream of this blade row set the required incidence to the cascade by imparting a circumferential component to the velocity. For the present unsteady studies, selected blades are subjected to controlled vibrations while the unsteady response on a reference stationery blade is measured. Two blades are connected to individual servo motors through a mechanism so as to execute controlled, low-amplitude, torsional (pitching) oscillations about its mid-chord axis. A reference blade is mounted on a load cell to enable measurement of both the axial and transverse forces and moments. In order to simulate the effect of the inter-blade phase angle prevalent in a rotating turbomachine blade row, the phase angle between the vibrating blades is varied to all admissible values. The fundamental parameters for evaluating the stability are the phase difference between blade position and the forces responses. This is estimated from the Fourier transform of the displacement and load signals. The parameters are evaluated for a range of reduced frequencies and inlet velocities to evaluate the stability of the cascade at all specified flow conditions.

The presented study is on a laboratory scaled industrial gas turbine combustor of intensity 25MW/m 3 atm, where an open loop active control technique is investigated. Combustion noise is classified as direct and in-direct combustion noise. The present study is focused on the investigation of direct combustion noise. It occurs when the volume of the gas fluctuates due to the fluctuations in heat release rate, caused perhaps due to flow turbulence. This results in sound waves, which propagate outside the boundary of the flame. The radiated acoustic waves are reflected from the boundaries of the combustion chamber, perturbing the fuel flow rate and hence the spray characteristics. This eventually leads to perturbation in the heat release rate and thus a feedback loop is established. At certain conditions, if the unsteady heat release rate drives the acoustic oscillations, satisfying Rayleigh criterion, pressure oscillations grow leading to discrete tonal sound and this phenomena is termed as combustion instability. Experiments are performed in a scaled down swirl stabilized liquid fueled gas turbine combustor, where a new scheme for open-loop control of combustion noise using periodic fuel injection is employed without drastically altering the combustor design or forfeiting its performance. Fuel is modulated in the frequency range of 0.6 to 5 Hz with various duty cycles [25–75%] using square wave. Fuel modulation is achieved by passing fuel through a direct current (DC) powered solenoid valve, which is being controlled using a custom-made circuit. The modulated fuel enters the combustor through an air-blast atomizer and is metered through a turbine flow meter. The main objective of this paper is to investigate the potential of active control to reduce combustion noise in laboratory scaled gas turbine combustor. Pressure transducer is used to capture the sound pressure level inside the combustor. A reduction in overall sound pressure level of 14dB is achieved by modulating fuel with 50% duty cycle at 1.5Hz.

The leading edge serrations are a type of passive flow control techniques in a compressor cascade. They are particularly attractive as they have been observed to increase the stall angle. This stall postponing character of the serrations is helpful in preventing compressor surge and widens the operational window of the compressor. Due to the simpler geometry of the serration type used in this study, it can be easily implemented onto the existing compressor blades. An experimental study on the flow modifications and losses due to these serrations are conducted in a linear cascade tunnel. The experiments are conducted on blades of NACA 65209 airfoil with and without leading edge serrations at Re of 120,000. Four serration profiles of various width and amplitude are compared. End plane measurements taken with 5-hole probe are studied for the better serration profile and surface flow visualizations are conducted to study the variation in the surface flow pattern on the suction side. The surface flow visualization reveals the presence of local recirculation zones and stream wise vortices created from each wave of the serration leading to flow attachment. These serrated blades have higher losses at 0 deg incidence; the reason for the same is found to be the flat leading edge surfaces formed from serration.

This paper presents an experimental study of primary breakup of liquid jet in an annular passage in a cross flow of air at a fixed Mach number of 0.12, at atmospheric pressures. The experiments were conducted for various velocities of liquid jet from 1.417 m/s to 7.084 m/s (based on orifice diameter = 1 mm) and the corresponding liquid-air momentum flux ratios varied from 1 to 25. The droplet sizes and velocities were measured using a Phase Doppler Particle Analyzer (PDPA) downstream of the liquid inlet port along the axial direction at the centerline of the annular passage along the plane of injection. Observed droplet sizes and velocity variations at different momentum flux ratios, in the axial direction, show three distinct zones. The first zone is the ligament formation zone represented by large variation in droplet Reynolds number with momentum ratio. The second zone is the primary droplet formation zone in which a fairly monotonic decrease in droplet size and droplet acceleration due to the breakup is observed. However, the Reynolds number of the droplets is almost invariant with momentum ratio. The third zone is where the spray attains the critical state where the size and velocity does not vary in the axial direction and the variation in size in this zone with the momentum ratio is primarily due to the initial conditions established in the ligament formation zone.

In the present study, the effectiveness of passive structures called tubercles on an axial compressor blade row is studied experimentally. Tubercles are the modifications to the leading edge of an airfoil in the form of blunt wave-like serrations. Although several studies on the effect of tubercles on isolated blades are available in literature, detailed study of their effect on a cascade of blades, such as in the case of an axial flow turbo-machine is lacking. Such an application in an axial compressor will result in a significant increase in the stall margin. Presently, experiments have been performed on a linear compressor cascade with a blade height of 0.15 m and mean chord of 0.06 m, on a NACA 65209 airfoil profile. The plain and modified blades are fabricated using rapid prototyping to ensure conformity to the required geometry. The cascade is designed in such a way that the incidence (angle of attack) and the stagger can be changed easily. The measurements are taken at the exit plane using a five-hole Pitot probe to obtain three-components of velocity and static pressure data over fine measurement grids. The effect is determined in terms of lift and drag coefficients, lift-to-drag ratio and total pressure loss coefficient. Experiments have been carried out for different pitch and amplitude (serration depth) of tubercles to understand their effect. The stall incidence angle for the best performing blade is found to increase up to 8.6° from that of the unmodified blade of 6.0°. Application of such structures in axial compressor blades may well be adequate to prevent stalling in axial compressors over a wide operating range.

This paper reports an experimental investigation of a non-reacting flow in a low aspect ratio dump combustor with taper exit. The separated flow reattachment length for this combustor was higher than the length of the combustor. In order to understand the flow behavior inside the combustor, the velocity measurement inside the combustor was carried out by calibrated one-dimensional hot wire probes for the same flow conditions. The turbulent intensity variation inside the combustor without swirl flow was found to be axi-symmetric. The turbulence intensity along the centre line is lower than at other radial locations because of the presence of the potential core. The intensity in the core region increases as one moves along the centerline in the axial direction, which can be attributed to the weakening of the potential core and the increase in the extent of the shear layer. The turbulence behavior significantly changed with the use of a swirler at the inlet of the combustor. The intensity variation inside the combustor is more uniform than the low swirl number SN = 0.23. The higher turbulence kinetic energy and turbulent dissipation rate was observed at this swirl number. Therefore, the effect of swirl is expected to improve the combustion through better mixing and flame stabilization.

This paper presents the results of an experimental investigation of liquid jet breakup in a cross-flow of air under the influence of swirl (swirl numbers 0 and 0.2) at a fixed air flow Mach No. 0.12 (typical gas turbine conditions). The experiments have been conducted for various liquid to air momentum flux ratios (q) in the range of 1 to 25. High speed (@ 500 fps) images of the jet breakup process are captured and those images are processed using MATLAB to obtain the variation of breakup length and penetration height with momentum flux ratio. Using the high speed images, an attempt has been made to understand the physics of the jet breakup process by identification of breakup modes — bag breakup, column breakup, shear breakup and surface breakup. The results show unique breakup and penetration behavior which departs from the continuous correlations typically used. Furthermore, the images show a substantial spatial fluctuation of the emerging jet resulting in a wavy nature related to effects of instability waves. The results with 15° swirl show reduced breakup length and penetration related to the non-uniform distribution of velocity that offers enhanced fuel atomization in swirling fuel nozzles.