Instantaneous wall-pressure data were recorded for a 25% eccentric whirling annular seal for rotor speeds of 1800RPM and 3600RPM, axial Reynolds numbers of 24000 and 12000, and whirl ratios of 0.1-1.0 following the procedure set forth by Winslow (1994), Robic (1999) and Suryanarayanan (2003). Overall, the phase averaged wall-pressure distributions were consistent with previous results. The “switch” in the pressure distribution measured by Suryanarayanan (2003) and Robic (1999) from pressure to suction between the seal entrance and exit occurs at and above a whirl ratio of 0.7 for 1800RPM and 0.4 for 3600RPM. For both rotor speeds, decreasing the flow rate by one-half also decreases the wall pressure fluctuation distributions by one-half. For whirl ratios less than 0.5, the phase averaged pressure field was relatively constant which leads to minimal forces being imparted on the rotor by the fluid in the annulus. Talyor-Gortler vortices are measured for 1800RPM and Re=24000 at whirl ratios 0.1 and 0.3-0.7. As the whirl ratio increases past 0.5, longitudinal vortices begin to emerge in the pressure contours and skew axially with 180° shifts occurring when the Taylor-Reynolds ratio is large enough. Longitudinal vortices were measured for both rotor speeds when the whirling motion is greater than 1400RPM for Re=12000 and greater than 1600RPM for Re=24000. Attempts were made to apply Childs (1983) procedure for finding the rotordynamic coefficients of annular seals; however, the seal moves in a non-circular orbit about the seal’s center so Childs analysis cannot be applied.

Instantaneous wall-pressure data were recorded for a 25% eccentric whirling annular seal for rotor speeds of 1800RPM and 3600RPM, axial Reynolds numbers of 24000 and 12000, and whirl ratios of 0.1-1.0 following the procedure set forth by Winslow (1994), Robic (1999) and Suryanarayanan (2003). Overall, the phase averaged wall-pressure distributions were consistent with previous results. The “switch” in the pressure distribution measured by Suryanarayanan (2003) and Robic (1999) from pressure to suction between the seal entrance and exit occurs at and above a whirl ratio of 0.7 for 1800RPM and 0.4 for 3600RPM. For both rotor speeds, decreasing the flow rate by one-half also decreases the wall pressure fluctuation distributions by one-half. For whirl ratios less than 0.5, the phase averaged pressure field was relatively constant which leads to minimal forces being imparted on the rotor by the fluid in the annulus. Talyor-Gortler vortices are measured for 1800RPM and Re=24000 at whirl ratios 0.1 and 0.3-0.7. As the whirl ratio increases past 0.5, longitudinal vortices begin to emerge in the pressure contours and skew axially with 180° shifts occurring when the Taylor-Reynolds ratio is large enough. Longitudinal vortices were measured for both rotor speeds when the whirling motion is greater than 1400RPM for Re=12000 and greater than 1600RPM for Re=24000. Attempts were made to apply Childs (1983) procedure for finding the rotordynamic coefficients of annular seals; however, the seal moves in a non-circular orbit about the seal’s center so Childs analysis cannot be applied.

The present work investigates the nature of dry-friction whip and whirl through experimental and numerical methods. Efforts of the author, Dyck, Pavalek, and coworkers enabled the design and construction of a test rig that demonstrated and recorded accurately the character of multi-mode dry-friction whip and whirl. These tests examined steady state whip and whirl characteristics for a variety of rub materials and clearances. Results provided by the test rig are unparalleled in quality and nature to those seen in literature and possess several unique characteristics that are presented and discussed. A simulation model is constructed using the Texas A and M University (TAMU) Turbomachinery Laboratory rotordynamic software suite XLTRC2 comprised of tapered Timoshenko beam finite elements to form multiple degree of freedom rotor and stator models. These models are reduced by component mode synthesis to discard highfrequency modes while retaining physical coordinates at locations for nonlinear interactions. The interaction at the rub surface is modeled using a nonlinear Hunt and Crossley contact model with coulomb friction. Dry-friction simulations are performed for specific test cases and compared against experimental data to determine the validity of the model. These comparisons are favorable, capturing accurately the nature of dryfriction whirl. Experimental and numerical analysis reveals the existence of multiple whirl and whip regions spanning the entire range of frequencies excited during whirl, despite claims of previous investigations that these regions are predicted by Black's whirl solution, but are not excited in simulations or experiments. In addition, spectral analysis illustrates the presence of harmonic sidebands that accompany the fundamental whirl solution. These sidebands are more evident in whip, and can excite higher-frequency whirl solutions. Experimental evidence also shows a strong nonlinearity present in the whirl frequency ratio, which is greater than that predicted by the measured radius-toclearance ratio at the rub location. Results include whirl frequencies 250% of that predicted by the measured radius-to-clearance ratio.

The highly flexible HALE (High Altitude Long Endurance) aircraft analysis methodology is of interest because early studies indicated that HALE aircraft might have different vibration and aeroelastic characteristics from those of conventional aircraft. Recently the computer code Nonlinear Aeroelastic Trim And Stability of HALE Aircraft (NATASHA) was developed and used to the flight dynamics and aeroelastic analysis of flying wing HALE aircraft. Further analysis improvements were required to extend its capability to the ground vibration test (GVT) environment and to both GVT and aeroelastic behavior of HALE aircraft with other configurations. First, the geometrically exact fully intrinsic beam theory was extended to treat other aircraft configurations modeled as an assembly of beam elements. It includes auxiliary elevator input in the horizontal tail and fuselage aerodynamics. Second, the methodology was extended to treat the GVT environment to provide modal characteristics for model validation. A newly developed bungee formulation is coupled to the intrinsic beam formulation for the GVT modeling. After the coupling procedures, the whole formulation cannot be fully intrinsic because the geometric constraint by bungee cords makes the system statically indeterminant. Third, because many HALE aircraft are propeller driven, the methodology was extended to include an engine/nacelle/propeller system using a two-degree-of-freedom model. This step was undertaken to predict a dynamic instability called ``whirl flutter," which can be exhibited in such HALE aircrafts. For simplicity, two fundamental assumptions are made: constant approximation on the propeller aerodynamics and the use of equivalent three-bladed counterpart for two-bladed propeller system to obviate the need for Floquet theory. The validity of these assumptions is verified by investigating the periodic effect of side forces and hub moments and the periodic inertia effect. Finally, parametric studies show how the current methodology can be utilized as a unified preliminary analysis tool for the vibration and aeroelastic analysis of highly flexible HALE aircraft.

The present work investigates the phenomena of whip and whirl for a rigid rotor contacting at two bearing locations. The idea originated from an anemometer consisting of a rotor with an elastically supported stator undergoing the phenomena of dry friction whip and whirl at the two bushing contacts. To analyze the behavior, a mathematical model similar to the anemometer is developed and analyzed assuming two possible solutions, Mode1 (normal reaction forces in phase at two contacts) and Mode 2 (normal reaction forces out of phase at two contacts). Analytical solutions are only possible for the models with same RCl (Radius to Clearance ratio) at the two rub locations. A simulation model is constructed using the Texas A&M University (TAMU) Turbomachinery Laboratory rotordynamics software suite XLTRC² comprised of Timoshenko beam finite elements to form multiple degrees of freedom rotor and stator models. The nonlinear connections at the rub surface are modeled using Hunt and Crossley‘s contact model with coulomb friction. Dry friction simulations are performed for three separate models depending on the rotor‘s mass disk location with respect to the contact locations. The three models used have (1) Disk at center location (2) Disk at 3/4 location (3) Disk at overhang location. The adequacy of the analytical solution is investigated using the above simulations. Also, cases are explored where the general assumed solution would not solve the mathematical model, e.g. different RCl ratios at the two contacts. Simulations are performed for increasing as well as decreasing running speeds. There is partial agreement between simulation predictions and the analytical solutions for the cases with the mass center at centered and at 3/4 location. First, whirl-to-whip transitions occur at near the combine rotor-stator bounce frequency for both disk at center and disk at 3/4 location. The case with overhang mass disk predicts the two contacts to whip and at different frequencies simultaneously. Neither of the analytical solutions predicts a case where precession occurs at two different frequencies at the two contact points. Predictions for models with different RCl on the Backward Precessional (BP) graph imitate whirling. The BP graph predicts increasing BP frequency with increasing rotor speeds which is a characteristic of whirling, whereas investigation of individual contact velocities suggest that they are slipping at all conditions, one of them slipping more than the other netting a whirling like motion. For the overhang model with different RCl, apart from whipping at different frequency the two contacts also whirl at different frequencies corresponding to the RCl at the respective contacts. Simulations for decreasing rotor speed predict jump down from whirl- to-whip different BP frequency as compared to the jump up from whip-to-whirl for the speed up.

Labyrinth seal force estimates are important to the prediction of the stability of turbomachinery. The force prediction methods fall into several categories: experiments, bulk flow analysis, and finite volume analysis. Finite volume analysis can be split into two subcategories: self-developed and commercial. In this research, a commercial computational fluid dynamics (CFD) program called ANSYS-CFX was used to predict the forces generated in a labyrinth seal whirling at specified speeds. The results were compared to data from VT-FAST, a bulk flow code, and TASCflow, another commercial CFD program. It was shown that there were discrepancies among the results, and several hypotheses were made as to the reason for these discrepancies. Additionally, ANSYS-CFX was used to study the effect of labyrinth seal eccentricity ratio on the resultant force generated. It was shown that the radial force component within the seal behaved linearly with respect to eccentricity ratio. However, the tangential force component had no distinguishable relationship with the eccentricity ratio. It was hypothesized that the lack of a relationship was caused by the small fluctuations in the inlet swirl. Although the inlet swirl varied very little at each eccentricity ratio, it was shown there was a relationship between the tangential force and inlet swirl. / Master of Science

This work deals with experimental analysis of fluid film instability in rotor bearing systems stored in hydrodynamic plain bearings with focus on bearing surface modification. The experiments were performed on a journal bearing dynamics simulator. The aim was to determine the influence of individual operating parameters on the occurrence and course of turbulence instability. The emphasis here was on the effect of surface micro textures. The work also included the preparation of textured samples. Experiments were divided into experiments monitoring the influence of temperature, pressure, bearing geometry and surface modification. The output of the individual experiments were the values of the threshold speed of formation and termination of instability or the magnitude of the amplitude depending on the above described parameters. These values were plotted in graphs. This data was then confronted with previously published articles. The results showed that instabilityin rotor systems is very sensitive to temperature, pressure and bearing geometry changes. The fundamental influence on the unstable behavior of the rotor system has been demonstrated by the use of textured surfaces, where the threshold speed was significantly shifted to a higher speed range compared to the non-textured bearing.

No description available.

No description available.

[pt] Neste trabalho um modelo dinâmico para um rotor vertical em balanço, considerando o fenômeno de contato com a sua guarda, é analisado. A conjunto é modelado como um sistema eixo-rotor-estator com contato. A análise do contato é particularmente complexa pela não linearidade nas equações de movimento. O impacto com o estator é levado em conta através do modelo de contato tipo Kelvin-Vôigt, e, as equações de movimento, do rotor, são deduzidas através da formulação Lagrangeana; estas equações podem capturar os fenômenos devido à vibração lateral, como: precessão direta, precessão retrograda, rolamento e escorregamento. Pela existência de diferentes parâmetros combinados e devido à não linearidade da equação de movimento, a resposta dinâmica não é simples de ser obtida apriori. Portanto, métodos numéricos são empregados para a solução, especificamente emprega-se o método de Runge-Kutta Fehlberg de passo variável. Os resultados da simulação mostram que para certas condições, o rotor pode mudar de orbita devido aos impactos com o estator, podendo chegar a realizar precessão retrograda. Este tipo de fenômeno é considerado como o mais violento e perigoso nas maquinas rotativas. Com o fim de estudar a dinâmica lateral do sistema, um rotor vertical em balanço com guarda anular é investigado. A passagem dela através de sua velocidade critica, quando conduzida por um motor elétrico, é analisada (e também quando o sistema opera em velocidades constantes). Além disso, neste trabalho, os resultados experimentais obtidos da bancada de experimentação são usados para estudar o fenômeno da precessão. / [en] In this work a dynamic model for the overhung rotor, considering the contact phenomenon between the rotor and the stator is analyzed. It is modeled as a shaft-rotor- stator system with contact. The analysis of contact is particularly complex, due to the high nonlinearity of motion equations. Impact with the stator is accounted by a consistent contact model, as Kelvin-Vôigt model, and, rotor`s motion equation is encountered employing Lagrangean`s method; this equations are capable of capturing the phenomenon due to lateral vibration, as: forward whirl, backward whirl, rolling or sliding along the stator. Due to the combined parameters and the effect of nonlinearity in motion equations, the dynamical response is not simple or easily predictable. Numerical simulation is the preferred method of analysis, exactly is used the Runge-Kutta Fehlberg method with variable step. Simulation results show that under certain conditions, a rotor changes its orbit due the impacts with the stator and after that, it executes backward whirl motion. It is a kind of phenomenon, which is considered as the most violent and dangerous in rotating machines. To this end, the analysis of a vertical overhung shaft-disc system with annular guard is investigated. The passing through its critical speed is analyzed when driven by an electric motor (also when the system operates under a constant rotational velocity). In addition, in this work the results obtained with an experimental test rig are used to investigate the whirl phenomenon.