Characteristics of Stable Rotating Stall Cells in an Axial Compressor

Flow field measurements of a high-speed axial compressor are presented during pre-stall and post-stall conditions. The paper provides an analysis of measurements from a circumferential array of unsteady shroud static pressure sensors during stall cell development. At low-speed, the stall cell approached a stable size in approximately two rotor revolutions. At higher speeds, the stall cell developed within a short amount of time after stall inception, but then fluctuated in circumferential extent as the compressor transiently approached a stable post-stall operating point. The size of the stall cell was found to be related to the annulus average flow coefficient. A discussion of Phase-Locked Average (PLA) statistics on flow field measurements during stable operation is also included. In conditions where rotating stall is present, flow field measurements can be Double Phase-Locked Averaged (DPLA) using a once-per-revolution (1/Rev) pulse and the period of the stall cell. The DPLA method provides greater detail and understanding into the structure of the stall cell. DPLA data indicated that a stalled compressor annulus can be considered to contained three main regions: over-pressurized passages, stalled passages, and recovering passages. Within the over-pressured region, rotor passages exhibited increased blade loading and pressure ratio compared to pre-stall values.

An Experiment-Based Sticking Model for Heated Sand

A hot sand model has been developed to predict the rebounding and sticking behavior of environmental particulates in the hot section of a gas turbine. This paper will focus on the sticking part of the model with rebounding effects to be discussed in a future paper. The key element of the model is determining the probability of the particle sticking to the surface when it comes into contact. Recent studies have suggested this sticking probability is a function of temperature, particle size, normal and tangential velocities of the impacting particle. Previous studies have predicted the sticking probability using theories for mechanical properties of the particles. These methods rely on idealized particle shapes and compositions which does not match the variable nature of sand in the environment. The current model attempts to take this randomness into account and ensure prediction accuracy by matching the model to results of a series of controlled coupon tests. The framework for the modeling approach and validation results of the developed predictive model are both presented.

Investigation of Aerodynamics and Heat Transfer of a Highly Loaded Turbine Blade Using the Universal Intermittency Function

Efficiency and performance of gas turbine engines are affected by the flow field around the blades. The flow field inside a gas turbine engine is very complex. One of the characteristics of the flow inside an engine is existence of periodic unsteady wakes, originating from the upstream stator blades. The unsteady wakes, with their highly vortical core, impinge on the downstream blade surfaces and cause an intermittent transition of the flow regime from laminar to turbulent. This study aims at investigating and modeling the behavior and development of the boundary layer along the suction surface of a highly loaded low-pressure turbine blade under steady and unsteady inlet flow condition. The current paper includes results of a computational work substantiated by the experimental verifications. For the experimental investigations, the linear cascade facility in Turbomachinery Performance and Flow research Lab (TPFL) at Texas A&M University was used to simulate the periodic unsteady flow condition inside gas turbine engine. Moving wakes, originating from upstream blades, were simulated in this facility by moving rods attached to two parallel timing belts. Measurements and calculations were conducted at Reynolds number of 110,000. This Reynolds number pertains to cruise condition of a low-pressure turbine. At this Reynolds number, the flow around the blades is transitional and highly susceptible to flow separation. Aerodynamics experiments include measuring the boundary layer, locating its transition, separation and finally re-attachment using miniature hot wire probes. Heat transfer measurements along the suction and pressure surfaces were conducted utilizing a specially designed heat transfer blade that was instrumented with liquid crystal coating. To numerically simulate the transitional behavior of the boundary layer under periodic unsteady flow condition, a new intermittency function is developed which is based on the universal intermittency function developed by Chakka and Schobeiri [1]. Accurate prediction of the boundary layer behavior under the above conditions requires minimum and the maximum intermittency functions. These functions were developed inductively using the experimental results that were obtained in the absence of flow separation. In the current investigation the impact of the separation on the minimum and maximum intermittency are accounted for. The enhanced minimum and maximum intermittency functions along with the universal intermittency are implemented in a RANS based solver for computational simulation. The computational results are compared with (a) experimental ones and (b) with the computational results from RANS that involves Langtry-Menter [2, 3] method.

DDES Analysis of Wake Vortex Related Unsteadiness and Losses in the Environment of High-Pressure Turbine Stage

In this work, the flows inside the high pressure turbine (HPT) vane and stage are studied with the help of a high-fidelity delayed detached eddy simulation (DDES) code. This work intends to study the fundamental nozzle/blade interaction with special attention paid to the development and transportation of the vane wake vortex. There are two motivations for this work. On the one hand, the high pressure turbine operates at both transonic Mach numbers and high Reynolds numbers, which imposes a great challenge to modern computational fluid dynamics (CFD), especially for scale-resolved simulation methods. An accurate and efficient high-fidelity CFD solver is very important for a thorough understanding of the flow physics and the design of more efficient HPT. On the other hand, the periodic wake vortex shedding is an important origin of turbine losses and unsteadiness. The wake and vortex not only cause losses themselves, but also interact with the shock wave (under transonic working condition), pressure waves, and have a strong impact on the downstream blade surface (affecting boundary layer transition and heat transfer). Built on one of our previous DDES simulations of a HPT vane VKI LS89, this work further investigates the development and length characteristics of the wake vortex, provides explanations of the length characteristics and reveals the transportation of the wake vortex into the downstream rotor passage along with its impact on the downstream aero-thermal performance.

Size and Temperature Dependent Deposition Model of Micro-Sized Sand Particles

Sand ingestion and deposition in gas turbine engine components can lead to several operational hazards. This paper discusses a physics based model for modeling the impact and deposition of sand particles. The collision model divides the impact process into three stages, the elastic stage, the elastic-plastic stage, and full plastic stage. The recovery stage is assumed to be fully elastic. The contact force, contact radius and work of contact force are conformed to the Hertzian theory, using “Young’s modulus similarity” rule to predict the recovery displacement. The adhesion loss in the recovery stage is considered using Dunn’s model, which describes the adhesion force as an idealized line force with the contact radius. The validation case of steel spherical particle impact on a glass surface with the maximum Stokes number of 10000, shows that the adhesion model with elastoplastic impact model describes the experimental result well. When the Stokes number is less than 12, the particle deposits on the surface. Sand properties are characterized by size and temperature dependencies. Model predictions for particle sizes ranging from 0.5 to 50 micron, impact velocities up to 80 m/s, and temperatures above 1300 K are given and discussed. It is shown that both size and temperature have an effect on the deposition characteristics.

Heat Transfer in the Core Compressor Under Ice Crystal Icing Conditions

It has been recognised in recent years that high altitude atmospheric ice crystals pose a threat to aircraft engines. Instances of damage, surge and shutdown have been recorded at altitudes significantly greater than those associated with supercooled water icing. It is believed that solid ice particles can accrete inside the core compressor, although the exact mechanism by which this occurs remains poorly understood. Development of analytical and empirical models of the ice crystal icing phenomenon is necessary for both future engine design and this-generation engine certification. A comprehensive model will require the integration of a number of aerodynamic, thermodynamic and mechanical components. This paper studies one such component, specifically the thermodynamic and mechanical processes experienced by ice particles impinging on a warm surface. Results are presented from an experimental campaign using a heated and instrumented flat plate. The plate was installed in the Altitude Icing Wind Tunnel (AIWT) at the National Research Council of Canada (NRC). This facility is capable of replicating ice crystal conditions at altitudes up to 9 km and Mach numbers up to 0.55 [1]. The heated plate is designed to measure the heat flux from a surface at temperatures representative of the early core compressor, under varying convective and icing heat loads. Heat transfer enhancement was observed to rise approximately linearly with both total water content and particle diameter over the ranges tested. A Stokes number greater than unity proved to be a useful parameter in determining whether heat transfer enhancement would occur. A particle energy parameter was used to estimate the likelihood of fragmentation. Results showed that when particles were both ballistic and likely to fragment, heat transfer enhancement was independent of both Mach and Reynolds numbers over the ranges tested.

Numerical Investigations on Partial Surge Initiated Inception and Stall Evolution in a Transonic Compressor

The instability has been the largest barrier of the high performance axial compressor in the past decades. Stall inception, which determines the route and the characteristics of instability evolution, has been extensively focused on. A new stall inception, “partial surge”, is discovered in the recent experiments. In this paper full-annulus transient simulations are performed to study the origin of partial surge initiated inception and explain the aerodynamic mechanism. The simulations show that the stall inception firstly occurs at the stator hub region, and then transfers to the rotor tip region. The compressor finally stalled by the tip region rotating stall. The stall evolution is in accord with the experiments. The stall evolution can be divided into three phases. In the first phase, the stator corner separation gradually merged with the adjacent passages, producing an annulus stall cell at the stator hub region. In the second phase, the total pressure rise of hub region emerges rapid decline due to the fast expansion of the annulus stall cell, but the tip region maintains its pressure rise. In the third phase, a new rotating stall cell appears at the rotor tip region, leading to the onset of fast drop of the tip region pressure rise. The stall cells transfer from hub region to the tip region is caused by two factors, the blockage of the hub region which transfers more load to the tip region, and the separation fluid fluctuations in stator domain which increase the circumferential non-uniformity in the rotor domain. High load and non-uniformity at the rotor tip region induce the final rotating stall.

The Unsteady Flow Field of a Purged High Pressure Turbine Based on Mode Detection

This paper presents an experimental study of the unsteady flow field downstream of a high pressure turbine with ejected purge flows, with a special focus on a flow field discussion using the mode detection approach according to the theory of Tyler and Sofrin. Measurements were carried out in a product-representative one and a half stage turbine test setup, which consists of a high-pressure turbine stage followed by an intermediate turbine center frame and a low-pressure turbine vane row. Four independent purge mass flows were injected through the forward and aft cavities of the unshrouded high-pressure turbine rotor. A fast-response pressure probe was used to acquire time-resolved data at the turbine center frame duct inlet and exit. The interactions between the stator, rotor, and turbine center frame duct are identified as spinning modes, propagating in azimuthal direction. Time-space diagrams illustrate the amplitude variation of the detected modes along the span. The composition of the unsteadiness and its major contributors are of interest to determine the role of unsteadiness in the turbine center frame duct loss generation mechanisms and to avoid high levels of blade vibrations in the low-pressure turbine which can in turn result in increased acoustic emissions. This work offers new insight into the unsteady flow behavior downstream of a purged high-pressure turbine and its propagation through an engine-representative turbine center frame duct configuration.

Assessment of the Severity of Unsteady Mach Number Effects in a 3-Stage Transonic Compressor

In this paper results from steady and unsteady CFD simulations of an industrial transonic compressor are compared, in order to gain a better understanding of the cause of the differences in the predicted efficiencies between the steady and unsteady simulations. Initially the first stage is simulated as an isolated compressor stage with inlet guide vanes in order to analyse the effect of individual blade rows on the stage performance. It is found that the rotor efficiency is lower for steady simulations than for unsteady simulations due to stronger shock waves. The stator efficiency is greater in the steady simulations due to not being able to model the interaction of the rotor wakes with the stator blade leading edge and boundary layers. Greater variation between steady and unsteady predictions is found at higher operating speeds. In the 3-stage unsteady simulations, the front stage efficiency characteristic is the same as the efficiency calculated from the isolated unsteady simulations. This shows that the unsteady pressure potential propagating from the downstream stages has no significant effect on the front stage efficiency meaning that the designer does not need to give great consideration to the downstream blade rows when predicting the characteristics of the front stage.

Numerical Investigation Into the Mechanism of Tip Flow Unsteadiness in a Transonic Compressor

It is well known that tip flow unsteadiness has profound effects on both performance and stability of axial compressors. A number of numerical simulations have been performed in transonic compressors to uncover the nature of tip flow unsteadiness. From this research, tip flow unsteadiness can be attributed to many factors, such as the movement of the primary and secondary leakage flow, the interaction between shock and vortex, and the tip leakage vortex breakdown. However, no final conclusion has yet been reached on this matter. The current investigation is carried out to explore the origin of tip flow unsteadiness from the perspective of the evolution and development of tip leakage vortex breakdown. In this paper, unsteady RANS simulations have been performed to investigate the fluid dynamic processes in a tip-critical transonic compressor, NASA Rotor 35. A vortex core visualization method based on an eigenvector method is introduced as an important tool to identify the vortex arising from tip leakage flow. As the flow rate varies, three critical operating points with distinctive features of flow unsteadiness are observed. At the first critical operating point, bubble-type breakdown occurs, and gives rise to a weak unsteadiness with high frequency in the rotor passage due to the oscillation of the recirculation region induced by the tip leakage vortex breakdown. At the second critical operating point, the vortex breakdown has transformed from bubble-type to spiral-type, which leads to the frequency of the pressure oscillation reduced almost by half and the amplitude increased significantly. At the third critical operating point, a new vortex that is perpendicular to the pressure surface comes into being in the tip region, which leads to a prominent pressure oscillation of the tip flow and another jump in amplitude. As a result, the evolution and development of tip leakage vortex breakdown are closely related to the tip flow unsteadiness of the investigated rotor.