Serpentine diffusers have become a common feature in modern aircraft as they allow for certain benefits that are impossible with a traditional linear configuration. With the benefits, however, come certain disadvantages, namely flow distortions that reduce engine efficiency and decrease engine surge stability margins. These distortions are now being researched comprehensively to determine solutions for mitigating the adverse effects associated with them. This study investigates how varying the freestream turbulence intensity of the flow entering a serpentine diffuser affects the distortion patterns that are produced by the diffuser. Experiments were performed with a model serpentine diffuser on the Annular Cascade Facility of the Air Force Research Laboratory at Wright-Patterson Air Force Base. Hot wire anemometry was used to measure inlet turbulence, while static pressure probes located axially along the upper and lower surface of the model diffuser and total pressure probes located across the aerodynamic interface plane (AIP) were used to measure the distortion patterns of the flow passing through the diffuser. Varying levels of inlet freestream turbulence, ranging from 0 to 4%, were generated using square and round bar turbulence screens in three distinct test configurations. Axial static pressure measurements indicate that increasing turbulence slightly affects flow separation development downstream of the second turn. This effect is also seen at the AIP where the total pressure recovery increases with increasing level of inlet turbulence in the region of flow separation at the upper surface. The total pressure recovery along the lower surface is also seen to be increased with higher inlet turbulence. However, total pressure recovery increase across the entire AIP is almost negligible. Overall, the inlet freestream turbulence has a minor effect on the distortion patterns caused by the serpentine diffuser when compared with proven active inlet flow control methods.

<p>The objective of this research was to develop stability enhancing design features for aeroengine centrifugal compressors. The motivation for this research is based on climate change and fuel-efficiency concerns, which call for improvements in achievable pressure ratios and surge margins. Specifically, this research aimed to develop diffuser recirculation channels and provide more insight into their design space. These channels are passive casing treatments in the diffuser and have been successfully demonstrated to improve stage surge margin. Diffuser recirculation channels are secondary flow paths that connect an opening near the diffuser inlet to one further down in the passage. Flow is recirculated by relieving the static pressure differential between the two openings. The basic design concept of these features is to add blockage upstream of the diffuser inlet, reducing the amount of diffusion in the vaneless space. In addition, channel geometries can be optimized to specifically target adverse flow properties, such as high incidence on the diffuser vane leading edge.</p> <p><br></p> <p>This design development was purely computational and served as the first approach to implementation of these features in a future generation of the Centrifugal Stage for Aerodynamic Research (CSTAR) at the Purdue Compressor Research Lab. Design development consisted of a computational design study, which quantified the effects of changing diffuser recirculation channel geometries on stage stability and performance metrics. Moreover, the CFD model for this future configuration of CSTAR was created and served as the baseline comparison for design iterations. The design study was comprised of controlled variation of channel geometry parameters and iterative solving of those cases in unsteady full stage single passage CFD models. Further design optimization studies were completed on specific down-selected recirculation channel geometry configurations. In total, 16 unsteady CFD cases with varied geometry configurations and 43 steady models were solved. Once a final optimized design was confirmed, a pressure characteristic at 100 % corrected design speed was generated. Compared to the baseline speed line, the implementation of diffuser recirculation channels resulted in a more gradual numerical surge and apparent numerical surge margin enhancement. Furthermore, the variation in incidence at the diffuser vane leading edge near the shroud was significantly reduced with diffuser recirculation. For the baseline compressor, incidence grew by about 70 degrees from the design aerodynamic loading to numerical surge at that location. However, flow stabilization due to diffuser 16 recirculation resulted in a change of approximately 2 degrees through that range. In conclusion, a first approach design recommendation for diffuser recirculation channels is CSTAR was generated through computational studies. Using this recommendation, diffusers with this recirculation channel design can be manufactured and tested for experimental concept validation.  </p>

Film cooling is a method used to protect surfaces exposed to high-temperature flows such as those that exist in gas turbines. It involves the injection of secondary fluid (at a lower temperature than that of the main flow) that covers the surface to be protected. This injection is through holes that can have various shapes; simple shapes such as those with a straight circular (by drilling) or straight square (by EDM) cross-section are relatively easy and inexpensive to create. Immediately downstream of the exit of a film cooling hole, a so-called horseshoe vortex structure consisting of a pair of counter-rotating vortices is formed. This vortex formation has an effect on the distribution of film coolant over the surface being protected. The fluid dynamics of these vortices is dependent upon the shape of the film cooling holes, and therefore so is the film coolant coverage which determines the film cooling effectiveness distribution and also has an effect on the heat transfer coefficient distribution. Differences in horseshoe vortex structures and in resultant effectiveness distributions are shown for circular and square hole cases for blowing ratios of 0.33, 0.50, 0.67, 1.00, and 1.33. The film cooling effectiveness values obtained are compared with experimental and computational data of Yuen and Martinez-Botas (2003a) and Walters and Leylek (1997). It was found that in the main flow portion of the domain immediately downstream of the cooling hole exit, there is greater lateral separation between the vortices in the horseshoe vortex pair for the case of the square hole. This was found to result in the square hole providing greater centerline film cooling effectiveness immediately downstream of the hole and better lateral film coolant coverage far downstream of the hole.

This study attempts to understand one of the most fundamental and challenging problems in fluid flow and heat transfer for rotating machines. The study focuses on gas turbines and electric generators for high temperature and high energy density applications, respectively, both which employ rotating cooling channels so that materials do not fail under high temperature and high stress environment. Prediction of fluid flow and heat transfer inside internal cooling channels that rotate at high rotation number and high density ratio similar to those that are existing in turbine blades and generator rotors is the main focus of this study. Both smooth-wall and rib-roughened channels are considered here. Rotation, buoyancy, bends, ribs and boundary conditions affect the flow inside theses channels. Ribs are introduced inside internal cooling channel in order to enhance the heat transfer rate. The use of ribs causes rapid increase in the supply pressure, which is already limited in a turbine or a generator and requires high cost for manufacturing. Hence careful optimization is needed to justify the use of ribs. Increasing rotation number (Ro) is another approach to increase heat transfer rate to values that are comparable to those achieved by introduction of ribs. One objective of this research is to study and compare theses two approaches in order to decide the optimum range of application and a possible replacement of the high-cost and complex ribs by increasing Ro. A fully computational approach is employed in this study. On the basis of comparison between two-equation (k-[epsilon] and k-[omega]) and RSM turbulence models, against limited available experimental data, it is concluded that the two-equation turbulence models cannot predict the anisotropic turbulent flow field and heat transfer correctly, while RSM showed improved prediction. For the near wall region, two approaches with standard wall functions and enhanced near wall treatment were investigated. The enhanced near wall approach showed superior results to the standard wall functions approach. Thus RSM with enhanced near wall treatment is validated against available experimental data (which are primarily at low rotation and buoyancy numbers). The model was then used for cases with high Ro (as much as 1.29) and high-density ratios (DR) (up to 0.4). Particular attention is given to how turbulence intensity, Reynolds stresses and transport are affected by Coriolis and buoyancy/centrifugal forces caused by high levels of Ro and DR. Variations of flow total pressure along the rotating channel are also predicted. The results obtained are explained in view of physical interpretation of Coriolis and centrifugal forces. Investigation of channels with smooth and with rib-roughened walls that are rotating about an orthogonal axis showed that increasing Ro always enhances turbulence and the heat transfer rate, while at high Ro, increasing DR although causes higher turbulence activity but does not necessarily increase Nu and in some locations even decreases Nu. The increasing thermal boundary layer thickness near walls is the possible reason for this behavior of Nu. The heat transfer enhancement for smooth-wall cases correlates linearly with Ro (with other parameters are kept constant) and hence it is possible to derive linear correlation for the increase in Nu as a function of Ro. Investigation of channels with rib-roughened walls that rotate about orthogonal axis showed that 4-side-average Nur correlates with Ro linearly, where a linear correlation for Nur/Nus as a function of Ro is derived. It is also observed that the heat transfer rate on smooth-wall channel can be enhanced rapidly by increasing Ro to values that are comparable to the enhancement due to the introduction of ribs inside internal cooling channels. This observation suggests that ribs may be unnecessary in high-speed machines, and has tremendous implications for possible cost savings in these machines. In square channels that rotate about parallel axis, the heat transfer rate enhances with Ro on three surfaces of the square channel and decreases on the inner surface (that is the one closest to the axis of rotation). However, the four-sides average Nu increases with Ro. Increasing wall heat flux at high Ro does not necessarily increase Nu on walls although higher turbulence activity is observed. This study examines the rich interplay of physics under the simultaneous actions of Coriolis and centrifugal/buoyancy forces in one of the most challenging internal flow configurations. Several important conclusions are reached from this computational study that may have far-reaching implications on how turbine blades and generator rotors are currently designed. Since the computation study in not validated for high Ro cases, these important results call for a experimental investigation.

Piezoelectric-based energy harvesting devices provide an attractive approach to powering remote devices as ambient mechanical energy from vibrations is converted to electrical energy. These devices have numerous potential applications, including actuation, sensing, structural health monitoring, and vibration control -- the latter of which is of particular interest here. This work seeks to develop an understanding of energy harvesting behavior within the framework of a semi-active technique for reducing turbomachinery blade vibrations, namely resonance frequency detuning. In contrast with the bulk of energy harvesting research, this effort is not focused on maximizing the power output of the system, but rather providing the low power levels required by resonance frequency detuning. The demands of this technique dictate that harvesting conditions will be far from optimal, requiring that many common assumptions in conventional energy harvesting research be relaxed. Resonance frequency detuning has been proposed as a result of recent advances in turbomachinery blade design that have, while improving their overall efficiency, led to significantly reduced damping and thus large vibratory stresses. This technique uses piezoelectric materials to control the stiffness, and thus resonance frequency, of a blade as the excitation frequency sweeps through resonance. By detuning a structure*s resonance frequency from that of the excitation, the overall peak response can be reduced, delaying high cycle fatigue and extending the lifetime of a blade. Additional benefits include reduced weight, drag, and noise levels as reduced vibratory stresses allow for increasingly light blade construction. As resonance frequency detuning is most effective when the stiffness states are well separated, it is necessary to harvested at nominally open- and short-circuit states, corresponding to the largest separation in stiffness states. This presents a problem from a harvesting standpoint however, as open- and short-circuit correspond to zero charge displacement and zero voltage, respectively, and thus there is no energy flow. It is, then, desirable to operate as near these conditions as possible while still harvesting sufficient energy to provide the power for state-switching. In this research a metric is developed to study the relationship between harvested power and structural stiffness, and a key result is that appreciable energy can be harvested far from the usual optimal conditions in a typical energy harvesting approach. Indeed, sufficient energy is available to power the on-blade control while essentially maintaining the desired stiffness states for detuning. Furthermore, it is shown that the optimal switch in the control law for resonance frequency detuning may be triggered by a threshold harvested power, requiring minimal on-blade processing. This is an attractive idea for implementing a vibration control system on-blade, as size limitations encourage removing the need for additional sensing and signal processing hardware.

This research developed two innovative test methods that were used to experimentally evaluate the performance of a novel blade tip timing (BTT) system from Prime Photonics, LC. The research focused on creating known blade tip offsets and tip vibrations so that the results from a BTT system can be validated. The topic of validation is important to the BTT field as the results between many commercial systems still are not consistent. While the system that was tested is still in development and final validation is not complete, the blade tip offset and vibration frequency validation results show that this BTT system will be a valuable addition to turbomachinery research and development programs once completed. For the first test method custom rotors were created with specified blade tip offsets. For the blade tip offset alternate measurement, the rotors were optically scanned and analyzed in CAD software with a tip location uncertainty of 0.1 mm. The BTT system agreed with the scanned results to within 0.13 mm. Tests were also conducted to ensure that the BTT system identified and indexed the blades properly. The second developed test method used an instrumented piezoelectric blade to create known dynamic deflections. The active vibration rotor was able to create measureable deflection over a range of frequencies centered on the first bending mode of the blade. The results for the 110 Hz, 150 Hz, 180 Hz first bending resonance, 200 Hz, and 1036 Hz second bending resonance cases are presented. A strain gage and piezoelectric sensor were attached to the active blade during the dynamic deflection tests to provide an alternate method for determining blade vibration frequency. The BTT system correctly identified the active blade excitation frequencies as well as a 120 Hz frequency from the drive motor. This thesis also explored applying BTT methods and testing to more realistic blade geometry and vibration. Blade vibrations are usually classified by their frequency relative to the rotation speed. Synchronous vibrations are integer multiples of the rotational speed and are often excited by struts or vanes fixed to the engine case. For this reason, special probe placement algorithms were explored that use sine curve fitting to optimize the probe placement. Knowing how the blade will vibrate at operation before testing is critical as well. In preparation for future research, ANSYS Mechanical was used to predict the first three modes of a PT6A-28 first stage rotor blade at 1,966, 5,539, and 7,144 Hz. These frequencies were validated to within 4% using scanning laser vibrometry. The simulation was repeated at speed to produce a Campbell Diagram to highlight synchronous excitation crossings. / Master of Science

This Thesis work focused on the acquisition and evaluation of data on mean velocity, turbulence intensity and length scales obtained via hot wire measurements in a high pressure single-stage land-based turbine available at the Department of Heat and Power Technology in KTH. Turbulence was generated by the use of grids and twoprobes, upstream and downstream of the turbine stage, were used. Different cases with different features were taken into consideration in this work: data were acquiredfor three different grids (I, II and III), two different blisks (5 and 6) and two operatingpoints. After post-processing all the information obtained from the anemometer, interesting comparisons between sub-cases could be made, and conclusions could bedrawn. In particular, it was found that Grid III generates a higher turbulence intensityupstream (around 13 %) with respect to Grid II (6 %) and I (3 %). Also, turbulenceintensity downstream did not seem to be affected by the conditions upstream butonly from the blisk type: turbulence levels were very similar when the same bliskwas mounted regardless of the type of grid upstream. Furthermore, the presence ofturbulence resulted in a slight decrease of the stage total-to-static efficiency ηT −S ofless than 0.5 % for Grid III while almost no change in efficiency was noticed when GridII or I were mounted. Furthermore, three different length scales were computed andanalyzed: Integral Length Scale, Taylor Microscale and Kolmogorov Microscale. Whilethe first two were found to be both in the order of 10−3 m, the last one -as expected- ismuch smaller, around 10−6 m. / Detta examensarbete fokuseradepå insamling och utvärdering av data om medelhastighet, turbulensintensitet och längdskalor erhållna via varmtrådsmätningar i ett högtryck turbinsteg tillgänglig påavdelningen för Kraft- och värmeteknologi på KTH. Turbulens genererades medelstanvändningen av olika perforerade plåtar i inloppet och två sonder, uppströms och nedströms om turbinsteget, användes för mätningar. Olika driftpunkter ochkonfigurationer med olika egenskaper togs i beaktande i detta arbete: data inhämtadesför tre olika inloppturbulensfall, Grid (I, II och III), två olika rotorbliskar (5 och 6)och två driftpunkter. Efter att ha efterbehandlat all information som erhållits frånmätningarna, kan intressanta jämförelser mellan delfall göras och slutsatser dras. I synnerhet fann man att Grid III genererar en högre turbulensintensitet uppströms (cirka 13 %) med avseende på Grid II (6 %) och I (3 %). Dessutom verkar turbulensintensitet nedströms inte påverkas av förhållandena uppströms men endast av typ avrotorblisk: turbulensnivåerna nedströms var mycket lika när samma blisk monteradesoavsett typ av galler uppströms. Dessutom resulterade förekomsten av turbulens i enliten minskning av stegets total-till-statiska verkningsgrad ηT−S med mindre än 0,5 %-enheter för Grid III medan nästan ingen förändring i verkningsgrad märktes när GridII eller I var monterade. Vidare beräknades tre olika längdskalor och analyserades: Integral Length Scale, Taylor Microscale och Kolmogorov Microscale. Medan de två första visade sig vara båda i storleksordningen 10−3 m, är den sista, som förväntat, är mycket mindre, cirka 10−6m.

A Thesis Submitted to the School of Graduate Studies in Partial Fulfillment of the Requirements for the Degree of Master of Applied Science in Mechanical Engineering / In the wake of the rising global demand for more electric transportation, aerospace electrification is becoming a highly active research area as commercial fully electric aircrafts are becoming a reality. The transportation electrification industry is challenged to develop powerful, safe, and compact-sized machines that can replace fossil fuel powered engines in aircrafts. Axial Flux Permanent Magnets (AFPM) machines are currently being intensively developed as a great candidate for this purpose due to their inherently higher power density compared to other machine electric machines topologies. The efforts of further increasing AFPM machines power density add more thermal challenges as intensive cooling is required at a relatively small machine package to avoid machine failure. One of the most concerning failure modes in these machines is power output reduction due to overheating of the rotor-mounted permanent magnets or even complete failure due to irreversible demagnetization. This research discusses the design process of an integrated rotor air cooling system for a 100 kW AFPM machine designed for an electric aircraft propulsion system. The embedded cooling system allows the rotor to be self-cooled at a sufficient cooling rate while minimizing the impact on machine efficiency due to windage power losses. The presented design process includes several stages of cooling enhancement including the addition and fine-tuning of rotor fan blades and rotor vents design. These enhancements are done by studying the air flow over the rotor surfaces in conjunction with heat transfer through Conjugate Heat Transfer (CHT) Computational Fluid Dynamics (CFD) analyses. In an initial study, different rotors with different combinations of rotor cooling features are studied and their thermal performance is compared. The results show that using rotor embedded fan blades in throughflow ventilated rotor geometry offers the best performance balance, achieving sufficient rotor cooling rate within a reasonable increase of windage power loss. A parametric study is performed to improve the rotor blade geometry for a higher ratio of heat transfer to windage losses. Another study is performed where the rotor and the enclosure geometries are fine-tuned simultaneously to reduce the negative effect on rotor heat transfer imposed by the enclosure. The final geometry of the rotor enclosure assembly is generated based on the research results and the design is integrated into the final machine prototype to be tested. / Thesis / Master of Applied Science (MASc) / Axial-flux permanent magnets (AFPM) machines are gaining the transportation electrification industy attention as a greener alternative to combustion engines in aircraft propulsion systems due to their high power and torque density. The intense endeavors of the current research to further improve AFPM machines power densities brings thermal design challenges to ensure the safe operation of the machine. Rotor permanent magnets failure due to demagnetization as a result of overheating can impose a great risk to the machine operation and safety. Accordingly, special attention should be paid to rotor thermal management. This research discusses the design process of an integrated rotor air cooling system for an AFPM machine designed for an electric aircraft. The machine mechanical and thermal design parameters are used to set an initial rotor design with different rotor cooling features based on literature findings. Rotor fan blades and air vents are selected as the main rotor cooling features for the design. Several design iterations are then made to fine-tune the rotor geometry targeting low operating temperature of the permanent magnets at a low cost of windage losses. The thermal performance of the different designs is assessed and compared to each other using conjugate heat transfer (CHT) computational fluid dynamics (CFD) analyses. Safe operating temperature of the magnets is achieved at an acceptable windage losses value with the final design, and it is selected for prototyping.

No description available.

No description available.