SpinChem AB's rotating bed reactor (RBR) is a small cylindrical centrifuge that allowsfor eective convective reactions between chemical substances. The aim of my master'sdissertation was to develop a measurement system that could estimate the uid velocity insidethe RBR. This report covers a quick pre-study of dierent uid measurement methods,details how we designed and constructed the system using integrated circuits and other electroniccomponents, and illustrates calibration results and nal velocity measurements. Thesystem design was based on the architecture of hot-wire constant temperature anemometry(CTA), a technique that measures the uid velocity over a heated metal wire. The resultsof this project is the measurement system itself, and the calibration and velocity measurementsgathered from experimentation. Two issues we encountered was that the calibrationprocess resulted in an overestimation of the velocity in the order of 4000 cm/s, and thatthe response to changes in velocity rates was quite slow. However, a slight tuning of thetwo calibration constants A0 & B0 smoothed the velocity estimation out to 40 cm/s, whichis more what we would expect. This illustrated how sensitive the results were to errors inthe calibration. Further investigation of the slow response, and a more stringently executedcalibration setup, could make the CTA a potential candidate for estimating the uid velocityinside the RBR.

Este trabalho apresentada um estudo experimental do fenômeno de biestabilidade que ocorre na geometria simplificada de dois tubos dispostos lado a lado submetidos a um escoamento cruzado turbulento. Na análise consideram-se a condição de tubos fixos e com um grau de liberdade (rotacional), submetidos a diferentes números de Reynolds. A técnica experimental consiste na medição de flutuações de velocidade do escoamento através da anemometria de fio quente em canal aerodinâmico, na avaliação dos esforços do escoamento sobre os tubos com o uso de uma célula de carga e na visualização dos efeitos do fenômeno biestável sobre os tubos livres para rotacionar. Os dados obtidos da medição no canal são tratados com o uso de ferramentas estatísticas, espectrais e de ondaletas. Os resultados das visualizações do escoamento são apresentados através de imagens obtidas das filmagens realizadas com uma câmera digital. A investigação experimental consiste de ensaios de cilindros rígidos elasticamente montados e pivotados na base e posteriormente liberados para oscilar e/ou rotacionar em um eixo central aos dois tubos, submetidos ao escoamento perpendicular permanente de ar no canal Nos resultados experimentais observam-se a presença da biestabilidade e nota-se o comportamento não simultâneo deste fenômeno ao longo dos cilindros. A formação assimétrica das esteiras também é estudada e atribuída à interação entre os vórtices das esteiras, já que, inicialmente, o desprendimento dos vórtices atrás dos cilindros é simétrico. Na avaliação dos esforços por extensiometria, as etapas de condicionamento e aquisição do sinal foram elaboradas e os resultados de deformação mecânica apresentaram similaridade com o fenômeno observado na anemometria experimental. Para um número de Reynolds de 27635 encontrou-se uma força de 26,1 mN com um torque desenvolvido na célula de carga de 84,95Nmm. A técnica de visualização do fenômeno empregada corrobora a interpretação dos resultados obtidos através da técnica de anemometria de fio quente e extensiometria em canal aerodinâmico. / This paper presents an experimental study of the bistability phenomenon that occurs in the simplified geometry of two tubes arranged side by side submitted to a turbulent cross flow. In the analysis we consider the condition of fixed tubes and with a degree of freedom (rotational), submitted to different Reynolds numbers. The experimental technique consists in the measurement of velocity fluctuations through the aerodynamic channel with the hot wire anemometry technique, the evaluation of the stresses of the flow on the tubes with the use of a load cell and the visualization of the effects of the bistable phenomenon on the tubes free to rotate. The data obtained from the measurement in the channel are treated with the use of statistical, spectral and wavelet tools. The results of the flow visualizations are presented through images obtained with a high speed digital camera. The experimental investigation consists of tests of rigid cylinders mounted elastically and pivoted at the base and later released to oscillate and / or to rotate in a central axis to the two tubes, submitted to the flow of air in the channel In the experimental results we observe the presence of bistability and note the non-simultaneous behavior of the phenomenon along the cylinders. The asymmetric formation of the wakes is also studied and attributed to the interaction between the vortices from wakes behind the cylinders, since, initially, the wake detachments is symmetrical. In the evaluation of the forces by means of the extensiometry technique, the steps of conditioning and acquisition of the signal were elaborated and the results of mechanical deformation presented similarity with the phenomenon observed in the experimental anemometry. For a Reynolds number of 27635 a force of 26.1 mN was found with a torque developed in the load cell of 84.95 Nmm. The technique used for the visualization of the phenomenon corroborates the interpretation of the results obtained through hot wire anemometry and extensiometry in aerodynamic channel.

Este trabalho apresenta um estudo acerca dos principais parâmetros que caracterizam o fenômeno da biestabilidade em dois tubos dispostos lado a lado submetidos a escoamento cruzado turbulento. A técnica experimental da anemometria de fio quente em canal aerodinâmico é aplicada na medição das flutuações de velocidade do escoamento após os tubos. As séries temporais obtidas são utilizadas como dados de entrada para determinação das funções densidade de probabilidade (PDF) usando um modelo de mistura finita, de acordo com uma função t de Student assimétrica e com o auxílio do método de Monte Carlo. Transformadas de ondaletas discretas e contínuas são aplicadas na filtragem das séries temporais para determinadas bandas de frequências e na análise do conteúdo de energia destes sinais. Através de conceitos de sistemas caóticos, é realizada a reconstrução do atrator do problema pelo método dos atrasos temporais, a partir das séries experimentais de velocidade, permitindo a determinação da dimensão de imersão e o cálculo do maior expoente de Lyapunov. Os resultados mostram a existência de dois patamares distintos de velocidade média nas séries temporais, correspondentes aos dois modos do escoamento, cada qual com números de Strouhal e funções densidade de probabilidade distintas. Uma análise conjunta das componentes axial e transversal do escoamento e suas PDF indicam as regiões no plano de medições onde o fenômeno se manifesta, sendo que reconstruções da trajetória filtrada das séries temporais para determinadas bandas de frequências apresentam características caóticodeterminísticas. O maior expoente de Lyapunov das séries experimentais é positivo, o que é um indício de comportamento caótico. / This work presents a study of the main parameters that characterize the phenomenon of bistability in two tubes placed side by side submitted to turbulent crossflow. The experimental technique of hot wire anemometry in aerodynamic channel is applied in the measurement of velocity fluctuations of the flow after the tubes. The time series obtained are used as input data for determining the probability density functions (PDF) using a finite mixture model, according to an asymmetric Student t function and with the aid of a Monte Carlo method. Wavelet transforms are applied in discrete and continuous filtering of time series for certain frequency bands and in the analysis of the energy content of these signals. By means of chaotic systems concepts, the attractor reconstruction of the problem is performed using the method of time delays from the experimental series of velocity, allowing the determination of the embedding dimension and calculating the largest Lyapunov exponent. The results show the existence of two different levels of mean velocity in time series, corresponding to two flow modes, each one with different Strouhal numbers and probability density functions. A joint analysis of axial and transverse components of flow and its PDF indicate the regions in the measurement plan where the phenomenon is manifested, and reconstructions of the trajectory of the filtered time series for certain frequency bands have chaotic-deterministic characteristics. The largest Lyapunov exponent of experimental series is positive, which is an indication of chaotic behavior.

An instrumented unmanned aerial vehicle (UAV) was developed and employed to observe the full range of turbulent motions that exist within the inertial subrange of atmospheric surface layer turbulence. The UAV was host to a suite of pressure, temperature, humidity, and wind sensors which provide the necessary data to calculate the variety of turbulent statistics that characterize the flow. Flight experiments were performed with this aircraft, consisting of a large square pattern at an altitude of 100 m above ground level. In order to capture the largest turbulent scales it was necessary to maximize the size of the square pattern. The smallest turbulent scales, on the other hand, were measured through the use of a fast response constant temperature hot wire anemometer. The results demonstrates that the UAV system is capable of directly measuring the full inertial subrange of the atmospheric surface layer with high resolution and allowing for the turbulence dissipation rate to be calculated directly.

Hot-wire anemometry has been routinely employed for laboratory measurements of turbulence for decades. This thesis presents a hot-wire anemometer suitable for use with the Boundary Layer Data System (BLDS). BLDS provides a unique platform for in- flight measurements because of its small, self-contained, robust design and flexible architecture. Addition of a hot-wire anemometer would provide BLDS with a sensor that could directly measure flow velocity fluctuations caused by turbulence. Hot-wires are commonly operated in constant-temperature mode for high frequency response, but require a carefully tuned bridge. The constant-voltage anemometer (CVA) uses a simple op-amp circuit to improve frequency response over constant-current operation. Due to its balance between ease of operation and performance, a CVA system built for this project was tested with a 3.8 micron diameter, platinum-coated tungsten probe. The CVA was calibrated in a steady jet and a power-law curve fit accurately represented the calibration data. The CVA successfully measured velocity fluctuations in a turbulent jet, as well as in laminar and tripped turbulent boundary layers over a flat plate in a 110 MPH wind tunnel. CVA frequency response was investigated using a thermal/electrical model, controlled oscillation in a steady flow, and with a square wave test; these three methods showed agreement. The CVA is selected for integration with BLDS.

In the present thesis a study on the thermal state of steelmaking ladles was undertaken. The transient hot wire method was verified for thermal conductivity measurements on metallurgical slags and applied to ladle slag measurements. Temperature measurements on ladles in an industrial environment were carried out. The emissivities of the outer and inner shells of steelmaking ladles were investigated. Two dynamic models were developed to predict the heat transfer and fluid flow in a preheating and teeming ladle. The gathered thermal conductivity values for ladle slag were used to study the effect of the slag layer on the top surface of the melt on heat transfer and fluid flow in a teeming ladle. In the first stage, the transient hot-wire method was verified to measure the thermal conductivity of metallurgical slags at steelmaking temperatures. A numerical model was developed, cold model experiments were conducted and test measurements using a high temperature experimental setup were carried out. To minimize natural convection and to obtain more reliable measurements, the crucible diameter, the hot-wire diameter, the applied current, the position of the wire in the crucible and the cooling on the upper surface of the crucible were studied. Investigations into the choice of sheathing material of the circuit exposed to the slag were also made. It was found that only certain materials were suitable for slag measurements depending on slag composition and temperature. The electrical resistivity of the hot wire was measured to make the thermal conductivity calculation more reliable. The wire diameter also played a major role due to the heat generation per surface area. The thermal conductivity should be derived from the values measured during the first seconds. In this initial stage, the effect of the natural convection as a function of the wire position in the crucible, the cooling on the top surface and the diameter of the crucible are negligible. A compromise has to be made in choosing the electrical current, since higher current results in higher sensitivity but at the same time in more natural convection. In the second stage, the thermal conductivities of four different ladle slags were measured at 1773 K, 1823 K, 1873 K and 1923 K using the transient hot wire method. Very good reproducibility was obtained. The thermal conductivity did not vary substantially with the variation of slag composition at 1873 K and 1923 K, at which the slag samples were all entirely liquid. The thermal conductivities were low. It was found that the precipitation of solid phase resulted in a considerable increase of thermal conductivity. In the third stage, a two dimensional model was developed in order to predict the temperature distribution in the ladle wall during the preheating process. The model calculated the heat transfer and the velocity field in the gas phase inside the ladle as well as the heat transfer in the solid walls during the preheating process. Measurements of the temperature profiles in an industrial ladle were carried out using an infrared thermography. The measurements were made both inside and outside the ladle. The model predictions were found to be in reasonably good agreement with the measured temperatures. It was found that the preheating time could be minimized when the working lining became thinner. The effect ofthe distance between the lid and the ladle was also studied by the model. The results indicated that there was no significant temperature change on the upper side wall of the ladle. On the lower side wall and bottom the temperature changed slightly. The temperature difference in the lower part of the ladle could be explained by the larger flame distance from the bottom layer. In the fourth stage, a two dimensional axisymmetric model was developed to predict the heat flux in a steelmaking ladle during the teeming process. The model predicts dynamically the flow fields in both the liquid phase and the gas phase along with the movement of the liquid upper surface. The model also predicts the temperature distributions in the liquid metal, gas phase and all layers in the ladle wall. Again, industrial measurements were performed using an infrared thermography, both inside the ladle after teeming and at the wall outside the ladle during the whole process sequence. The model predictions were found to be in agreement with the measured data. It was found that the heat transfer to the surrounding atmosphere and the conductivity of the highly insulating layer were the most important factors for the heat loss. The decrease of the thickness of the working lining was found to have limited effect on the total heat flux. In the fifth and final stage, the effect of the slag layer on the top surface of the melt, on fluid flow and on heat transfer in a teeming ladle was investigated theoretically. The two dimensional axisymmetric model developed in the fourth stage was used. To predict the effect of the slag layer a stationary heat conduction boundary condition including thermal conductivity and slag layer thickness was employed. Different calculations with differing thermal conductivity values for the slag layer were carried out. The calculations showed that the effect of the slag layer was insignificant. This could be explained by the similarity of the thermal conductivity of slag and gas phase. / <p>QC 20121010</p>

There are two coupled primary objectives for this study of buoyancy-driven turbulence. The first objective is to create a new diagnostic for collection of measurements to capture the physics of Rayleigh-Taylor (RT) mixing. The second objective is to use the new diagnostic to specifically elucidate the physics of large Atwood number, ( )( )2 1 2 1 / ρ ρ ρ ρ + − = t A , RT mixing. Both of these objectives have been satisfied through the development of a new hot-wire diagnostic to study buoyancy-driven turbulence in a statistically steady gas channel of helium and air ( 6 . 0 03 . 0 ≤ ≤ t A ). The capability of the diagnostic to simultaneously and instantaneously measure turbulent velocity and density fluctuations allows for a unique investigation into the dynamics of Rayleigh-Taylor mixing layers at large At, through measurements of turbulence and mixing statistics. The new hot-wire diagnostic uses temperature as a fluid marker for helium and air, which is possible due to the Lewis number ~ 1 (Le = ratio of thermal diffusivity to mass diffusivity) for helium and air, and the new diagnostic has been validated in an At = 0.03 mixing layer. The energy density spectrum of v′ ′ ρ , measured experimentally for the first time in RT mixing, is found to closely follow the energy distribution of v′ , up to the Reynolds numbers investigated ( ( ) mix t h gA h υ 6 2 Re 2 / 3 = ~ 1450). Large At experiments, with At = 0.6, have also been achieved for the first time in a miscible RT mixing layer. An asymmetric penetration of the bubbles (rising fluid) and spikes (falling fluid) has been observed, resulting in measured self similar growth parameters αb = 0.060 and αs = 0.088 for the bubbles and spikes, respectively. The first experimental measurements of turbulent velocity and density fluctuations for the large At case, show a strong similarity to lower At behaviors when normalized. However conditional statistics, which separate the bubble (light fluid) and spike (heavy fluid) dynamics, has highlighted differences in v′ ′ ρ and rms v′ in the bubbles and spikes. Larger values of v′ ′ ρ and rms v′ were found in the downward falling spikes, which is consistent with the larger growth rates and momentum of the spikes compared to the bubbles. These conditional statistics are a first in RT driven turbulence.

<p>Despite its importance in various industrial applications there is still a lack of experimental studies on the dynamic and thermal field of swirling jets in the near-field region. The present study is an attempt to close this lack and provide new insights on the effect of rotation on the turbulent mixing of a <i>passive scalar</i>, on turbulence (joint) statistics as well as the turbulence structure.</p><p>Swirl is known to increase the spreading of free turbulent jets and hence to entrain more ambient fluid. Contrary to previous experiments, which leave traces of the swirl generating method especially in the near-field, the swirl was imparted by discharging a slightly heated air flow from an axially rotating and thermally insulated pipe (6 m long, diameter 60 mm). This gives well-defined axisymmetric streamwise and azimuthal velocity distributions as well as a well-defined temperature profile at the jet outlet. The experiments were performed at a <i>Reynolds</i> number of 24000 and a swirl number (ratio between the angular velocity of the pipe wall and the bulk velocity in the pipe) of 0.5.</p><p>By means of a specially designed combined X-wire and cold-wire probe it was possible to simultaneously acquire the instantaneous axial and azimuthal velocity components as well as the temperature and compensate the former against temperature variations. The comparison of the swirling and non-swirling cases clearly indicates a modification of the turbulence structure to that effect that the swirling jet spreads and mixes faster than its non-swirling counterpart. It is also shown that the streamwise velocity and temperature fluctuations are highly correlated and that the addition of swirl drastically increases the streamwise<i> passive scalar</i> flux in the near field.</p>

Η παρούσα εργασία πραγματεύεται τη μελέτη και κατασκευή μιας μονάδας ανεμομέτρου θερμού σύρματος (Hot wire anemometer). Η εργασία αυτή εκπονήθηκε στο Εργαστήριο Θερμοκινητήρων του Τμήματος Μηχανολόγων και Αεροναυπηγών Μηχανικών της Πολυτεχνικής Σχολής του Πανεπιστημίου Πατρών. Σκοπός είναι η μελέτη, σχεδιασμός, κατασκευή και δοκιμή μιας μονοκάναλης μονάδας θερμού σύρματος σταθερής θερμοκρασίας (constant temperature anemometer CTA). Η μέθοδος του Hot wire anemometer χρησιμοποιείται για την έμμεση ταχύτητα της ροής, με βάση την συνεχή παρακολούθηση των θερμικών απωλειών γύρω από ένα θερμαινόμενο αισθητήρα. Το αποτέλεσμα δεν είναι μόνο συναρτήσει της ταχύτητας της ροής, αλλά επίσης και ορισμένων άλλων χαρακτηριστικών της ροής. Αρχικά μελετούνται και καταγράφονται οι βασικές αρχές που διέπουν την ανεμομετρία θερμού σύρματος. Στη συνέχεια αναπτύσσεται ο ηλεκτρονικός σχεδιασμός του ηλεκτρονικού κυκλώματος CTA. Ενώ αργότερα γίνεται η υλοποίηση του κυκλώματος στο λογισμικό Altium Designer μέσω της οποίας προκύπτει ο σχεδιασμός των πλακετών PCB. Επιπλέον περιγράφεται αναλυτικά η κατασκευή της διάταξης, περιγράφονται τα κυριότερα υποσυστήματα της και εξηγείτε αναλυτικά η αρχή λειτουργίας της. Τέλος γίνεται βαθμονόμηση του κυκλώματος, βαθμονόμηση του αισθητήρα και πιστοποιείται η σωστή λειτουργία της μονάδας λαμβάνοντας μετρήσεις στην έξοδο ενός τρισδιάστατου ακροφυσίου τύπου Boërger. / The present study deals with the design and construction of a "Hot Wire Anemometer" module. The work has been conducted in Thermal Engines Laboratory of the Mechanical Engineering and Aeronautics Department of the University of Patras, under the supervision of Professor Demos P. Georgiou. The purpose of the current study was the design, construction and testing of a single channel, Constant Temperature Hot Wire Anemometer unit (CTA). The Hot-wire anemometry is a method for indirectly measuring fluid flow velocity, based on monitoring the thermal losses in a heated measuring element. The result is a function of not only the flow velocity, but also of certain other flow parameters. Initially, some basic principles regarding the Hot Wire Anemometer are discussed and some aspects concerning the governing equations are also given. Furthermore, the development of the electronic CTA circuit shown, while later, the final circuit is implemented by employing the Altium Designer Software through which the Printed Circuit Boards (PCB) for the construction are finally obtained. After the detailed presentation of the device construction, the main components and subsystems are discussed in depth. In addition, the principle of operation is also explained in detail. Finally, a fine calibration is made for both, the electronic circuit and the sensor probe. The proper operation of the system as a final product is then certified by obtaining aerodynamic measurements at the exit of a three dimensional Boërger type nozzle.

A key remaining challenge in the design of hypersonic vehicles is the incomplete understanding of the process of boundary-layer transition. Turbulent heating rates are substantially higher than those for a laminar boundary layer, and large uncertainties in transition prediction therefore demand conservative, inefficient designs for thermal protection systems. It is only through close collaboration between theory, experiment, and computation that the state of the art can be advanced, but experiments relevant to flight require ground-test facilities with very low disturbance levels. To enable this work, a unique Mach 6 low-disturbance wind tunnel, previously of NASA Langley Research Center, is established within a new pressure-vacuum blow-down infrastructure at Texas A&M. A 40-second run time at constant conditions enables detailed measurements for comparison with computation. The freestream environment is extensively characterized, with a large region of low-disturbance flow found to be reliably present for unit Reynolds numbers Re < 11×10^6 m-1. Experiments are performed on a 5º half-angle flared cone model at Re = 10×10^6 m-1 and zero angle of attack. For the study of the second-mode instability, well-resolved boundary-layer profiles of mean and fluctuating mass flux are acquired at several axial locations using hot-wire probes with a bandwidth of 330 kHz. The second mode instability is observed to undergo significant growth between 250 and 310 kHz. Mode shapes of the disturbance agree well with those predicted from linear parabolized stability equation (LPSE) computations. A 17% (40 kHz) disagreement is observed in the frequency for most-amplified growth between experiment and LPSE. Possible sources of the disagreement are discussed, and the effect of small misalignments of the model is quantified experimentally. A focused schlieren deflectometer with high bandwidth (1 MHz) and high signal-to-noise ratio is employed to complement the hot-wire work. The second-mode fundamental at 250 kHz is observed, as well as additional harmonic content not discernible in the hot-wire measurements at two and three times the fundamental. A bispectral analysis shows that after sufficient amplification of the second mode, several nonlinear mechanisms become significant, including ones involving the third harmonic, which have not hitherto been reported in the literature.