Performance Characterization of a Twin Scroll Volute for Turbocharging Applications

The use of twin scroll volutes in radial turbine for turbocharging applications has several advantages over single passage volute related to the engine matching and to the overall compactness. Twin scroll volutes are of increasing interest in power unit development but the open scientific literature on their performance and modelling is still quite limited. In the present work the performance of a twin scroll volute for a turbocharger radial turbine are investigated in some detail in a wide range of operating conditions at both full and partial admission. A CFD model for the volute have been developed and preliminary validated against experimental data available for the radial turbine. Then the numerical model has been used to generate the database of solutions that have been investigated and used to extract the performance. Different parameters and indices are introduced to describe the volute aerodynamic performance in the wide range of operating conditions chosen. The above parameters can be used for volute development or matching with a given rotor or efficiently implemented in automatic design optimization strategies.

Surface Roughness Impact on Secondary Flow and Losses in a Turbine Exhaust Casing

This paper experimentally addresses the impact of surface roughness on losses and secondary flow in a Turbine Rear Structure (TRS). Experiments were performed in the Chalmers LPT-OGV facility, at an engine representative Reynolds number with a realistic shrouded rotating low-pressure turbine (LPT). Outlet Guide Vanes (OGV) were manufactured to achieve three different surface roughnesses tested at two Reynolds numbers, Re = 235000 and Re = 465000. The experiments were performed at on-design inlet swirl conditions. The inlet and outlet flow of the TRS were measured in 2D planes with a 5-hole probe and 7-hole probe accordingly. The static pressure distributions on the OGVs were measured and boundary layer studies were performed at the OGV midspan on the suction side with a time-resolved total pressure probe. Turbulence decay was measured within the TRS with a single hot-wire. The results showed a surprisingly significant increase in the losses for the high level of surface roughness (25–30 Ra) of the OGVs and Re = 465000. The increased losses were primary revealed as a result of the flow separation on the OGV suction side near the hub. The loss increase was seen but was less substantial for the intermediate roughness case (4–8 Ra). Experimental results presented in this work provide support for the further development of more advanced TRS and data for the validation of new CFD prediction methods for TRS.

Critical Suction Peformance Test of Turbopump Assembly for a Liquid-Propellant Rocket Engine

This study investigates the behavior of a turbopump assembly during critical cavitation of the propellant pumps in the upper rocket engine of the Korea Space Launch Vehicle-II. Turbopumps operate under conditions involving low pressure at the pump inlet and high rotational speeds to allow for a lightweight design. This severe environment can easily cause cavitation to occur in the pump. This cavitation can then cause the pump operation to fail. As the cavitation number in the pump decreases below the critical point, the pump fails to operate. There is concern regarding the behavior of the turbopump assembly arising from pump failure due to cavitation. It is necessary to verify the problems that may occur if the turbopump assembly operates under extreme conditions, such like the critical cavitation. This study performed tests to investigate the breakdown of pumps in the turbopump assembly. Tests were conducted with liquid nitrogen, water, and high-pressure air instead of the mediums used during actual operation of liquid oxygen, kerosene, and hot gas. The turbopump was tested at the design point of 27,000 rpm, while the inlet pressure of each pump was controlled to approach the critical cavitation number. The turbine power output was maintained during the tests. The results show that the breakdown point of the oxidizer pump using liquid nitrogen, which is a cryogenic medium, occurred at a lower cavitation number than during an individual component suction performance test using water. The fuel pump using water, meanwhile, experiences breakdown at similar cavitation numbers in both tests. As the breakdown of the pump occurs, the power required by that pump decreases, and the rotational speed of the turbopump increases. Compared with individual pump suction performance tests, this breakdown test can be used to determine the limit of the propellant inlet pressure of the turbopump and to characterize the behavior of the turbopump assembly when a breakdown occurs. Vibrations were also analyzed for tests at a high cavitation number and at the critical cavitation number. The vibration increased with breakdown and notable frequencies were analyzed.

Effects of Cavity Purge Flow on Intermediate Turbine Duct Flowfield

To increase the power output without adding additional stages, ultra-high bypass ratio engine, which has larger diameter low pressure turbine, attracts more and more attention because of its huge advantage. This tendency will lead to aggressive (high diffusion) intermediate turbine duct design. Much work has been done to investigate flow mechanisms in this kind of duct as well as its design criterion with numerical and experimental methods. Usually intermediate turbine duct simplified from real engine structure was adopted with upstream and downstream blades. However, cavity purge mass flow exists to disturb the duct flow field in real engine to change its performance. Naturally, the wall vortex pairs would develop in different ways. In addition to that, purge flow rate changes at different engine representative operating conditions. This paper deals with the influence of turbine purge flow on the aerodynamic performance of an aggressive intermediate turbine duct. The objective is to reveal the physical mechanism of purge flow ejected from the wheel-space and its effects on the duct flow field. Ten cases with and without cavity are simulated simultaneously. On one hand, the influence of cavity structure without purge flow on the flow field inside duct could be discussed. On the other hand, the effect of purge flow rate on flow field could be analyzed to investigate the mechanisms at different engine operating conditions. According to this paper, cavity structure is beneficial for pressure loss. And the influence concentrates near hub and duct inlet.

Implicit-LES Simulation of Variable-Speed Power Turbine Cascade for Low Free-Stream Turbulence Conditions

It is a challenge to simulate the flow in a Variable Speed Power Turbine (VSPT), or, for that matter, rear stages of low pressure turbines at low Reynolds numbers due to laminar flow separation or laminar/turbulent flow transition on the blades. At low Reynolds numbers, separation induced-transition is more prevalent which can result in efficiency lapse. LES has been used in recent years to simulate these types of flows with a good degree of success. In the present work, very low free stream turbulence flows at exit Reynolds number of 220k were simulated. The geometry was a cascade which was constructed with the midspan section of a VSPT design. Most LES simulations to date, have focused on the midspan region. As the endwall effect was significant in these simulations due to thick incoming boundary layer, full blade span computation was necessitated. Inlet flow angles representative of take-off and cruise conditions, dictated by the rotor speed in an actual design, were analyzed. This was done using a second order finite volume code and a high resolution grid. As is the case with Implicit-LES methods, no sub-grid scale model was used. Blade static pressure data, at various span locations, and downstream probe survey measurements of total pressure loss coefficient were used to verify the results. The comparisons showed good agreement between the simulations and the experimental data.

Using Turbulence Intensity and Reynolds Number to Predict Flow Separation on a Highly Loaded, Low-Pressure Gas Turbine Blade at Low Reynolds Numbers

The low-pressure turbine has become more important in the last few decades because of the increased emphasis on higher overall pressure and bypass ratios. The desire is to increase blade loading to reduce blade counts and stages in the low-pressure turbine of a gas turbine engine. Increased turbine inlet temperatures for newer cycles results in higher temperatures in the low-pressure turbine, especially the latter stages, where cooling technologies are not used. These higher temperatures lead to higher work from the turbine and this, combined with the high loadings, can lead to flow separation. Separation is more likely in engines operating at high altitudes and reduced throttle setting. At the high Reynolds numbers found at takeoff, the flow over a low-pressure turbine blade tends to stay attached. At lower blade Reynolds numbers (25,000 to 200,000), found during cruise at high altitudes, the flow on the suction surface of the low-pressure turbine blades is inclined to separate. This paper is a study on the flow characteristics of the L1A turbine blade at three low Reynolds numbers (60,000, 108,000, and 165,000) and 15 turbulence intensities (1.89% to 19.87%) in a steady flow cascade wind tunnel. With this data, it is possible to examine the impact of Reynolds number and turbulence intensity on the location of the initiation of flow separation, the flow separation zone, and the reattachment location. Quantifying the change in separated flow as a result of varying Reynolds numbers and turbulence intensities will help to characterize the low momentum flow environments in which the low-pressure turbine must operate and how this might impact the operation of the engine. Based on the data presented, it is possible to predict the location and size of the separation as a function of both the Reynolds number and upstream freestream turbulence intensity (FSTI). Being able to predict this flow behavior can lead to more effective blade designs using either passive or active flow control to reduce or eliminate flow separation.

Aerodynamic Interaction Between Incoming Vortex and Tip Leakage Flow in a Turbine Cascade

In turbines, secondary vortices and tip leakage vortices develop and interact with each other. In order to understand the flow physics of vortices interaction, the effects of incoming vortex on the downstream tip leakage flow are investigated in terms of the aerodynamic performance in a turbine cascade. Experimental, numerical and analytical methods are used. In the experiment, a swirl generator was used upstream near the casing to generate the incoming vortex, which interacted with the tip leakage vortex in the turbine cascade. The swirl generator was located at ten different pitchwise locations to simulate the quasi-steady effects. In the numerical study, a Rankine-like vortex was defined at the inlet of the computational domain to simulate the incoming swirling vortex. Incoming vortices with opposite directions were investigated. The vorticity of the positive incoming swirling vortex has a large vector in the same direction as that of the tip leakage vortex. In the case of the positive incoming swirling vortex, the vortex mixes with the tip leakage vortex to form one vortex near the tip as it transports downstream. The vortices interaction reduces the vorticity of the flow near the tip, as well as the loss by making up for the streamwise momentum within the tip leakage vortex core. In contrast, the negative incoming swirling vortex has little effects on the tip leakage vortex and the loss. As the negative incoming swirling vortex transports downstream, it is separated from the tip leakage vortex and forms two vortices. A triple-vortices-interaction kinetic analytical model and one-dimensional mixing model are proposed to explain the mechanism of vortex interaction on the aerodynamic performance.

A High Fidelity Quality Assessment of High Pressure Turbine Blades Using Surface Curvature and Gradient-Based Adjoint

Gas turbine performance is highly dependent on the quality of the manufactured parts. Manufacturing variations in the parts can significantly alter the performance, especially efficiency and thus SFC. The legacy process is to accept variations within predefined profile tolerance limits and a few other qualitative parameters, mostly at a few, key two-dimensional aerofoil sections. With the widespread use of White light scans and other similar three-dimensional scans, this has improved to include the three-dimensional profile. The future however may lie with performance based quality assessment of manufactured parts, combined with quantitative surface quality assessment to implement an intelligent screening process for the parts. The adjoint method, typically used for shape optimization is adapted to provide a prediction of the impact on performance due to manufacturing variations. The work presented outlines a three stage quality assessment process for manufactured parts, involving three-dimensional profile tolerance based screening, followed by a surface curvature based screening and finally an Adjoint based performance prediction.

Influences of Tip Leakage Flows Discharged From Main and Splitter Blades on Flow Field in Transonic Centrifugal Compressor Stage

A transonic centrifugal compressor impeller is generally composed of the main and the splitter blades which are different in chord length. As a result, the tip leakage flows from the main and the splitter blades interact with each other and then complicate the flow field in the compressor. In this study, in order to clarify the individual influences of these leakage flows on the flow field in the transonic centrifugal compressor stage at near-choke to near-stall condition, the flows in the compressor at four conditions prescribed by the presence and the absence of the tip clearances were analyzed numerically. The computed results clarified the following noticeable phenomena. The tip clearance of the main blade induces the tip leakage vortex from the leading edge of the main blade. This vortex decreases the blade loading of the main blade to the negative value by the increase of the flow acceleration along the suction surface of the splitter blade, and consequently induces the tip leakage vortex caused by the negative blade loading of the main blade at any operating points. These phenomena decline the impeller efficiency. On the other hand, the tip clearance of the splitter blade decreases the afore mentioned acceleration by the formation of the tip leakage vortex from the leading edge of the splitter blade and the decrease of the incidence angle for the splitter blade caused by the suction of the flow into the tip clearance. These phenomena reduce the loss generated by the negative blade loading of the main blade and consequently reduce the decline of the impeller efficiency. Moreover, the tip clearances enlarge the flow separation around the diffuser inlet and then decline the diffuser performance independently of the operating points.

APU-Noise Reduction by Novel Muffler Concepts

For a GTCP36-28 auxiliary power unit (APU), a set of mufflers has been designed and tested for some representative operating conditions. The first muffler design uses cavities of different sizes in conjunction with a bias flow for efficient broadband sound absorption. The second design — also expected to perform well over a large frequency range — makes use of a variable perforation and some porous absorber material. The acoustic damping performance of the mufflers was assessed using a downstream section of dedicated microphone probes. Individual spectra and circumferential averages have been computed and are used for a comparison to a hard-walled duct section of the same length. Results show a reasonable broadband absorption for most configurations. For one operating point, significant differences were found while comparing the performance of the cavity muffler with and without bias flow. The results suggest, that a small amount of air — less than initially expected — is sufficient to obtain the desired noise reduction.