This study provides, for the first time, an ex vivo comparative evaluation of formulations of budesonide with crystallised β-form mannitol, commercial DPI grade mannitol and lactose. The lactose-budesonide was the marketed Easyhaler® 200 g formulation. Ex vivo assessment of deposition using the Easyhaler® multi-dose high resistance inhaler with reservoir was compared with the RS01® single dose capsule low resistance inhaler at two different inhalation rates. Aerodynamic characteristics, flow and surface energies were investigated together with in vitro and ex vivo assessment of drug deposition. Dose emission was greater for all formulations with higher inhalation flow, indicating greater detachment of drug from carrier, and greater with the Easyhaler®, highlighting the importance of correct device for formulation. Emission was lowest at both inhalation rates for crystallised mannitol due to poor flowability associated with elongated particle shape which resulted in interception deposition. Surface energies were also implicated; closely matched polar surface energy of carrier and drug may be an important inhibiting factor. The promising aerodynamic characteristics of crystallised mannitol with the RS01® inhaler and lactose-budesonide from in vitro assessment were not supported by ex vivo results, highlighting the need for careful selection of device.

No description available.

No description available.

No description available.

No description available.

In the thesis, a lifting body has been designed aiming to generate lift force for the pentacopter, called TILT LR (Long Range), at higher velocities during flights to improve the aerodynamic performances. The configuration, which is used as the skeleton of the long range drone for up to 75 kilometers flights, is based upon a tilting system allowing the rotors to rotate around their own axis in both pitch and roll angles. This offers the possibility to the TILT LR flying withoutany vertical excess thrust at a proper angle of attack and velocity. This new drone can be directly applied to missions require long flight time or cover long distance, such as Search & Rescue(SAR), power lines and off-shore structures inspection, fire monitoring or surveillance. Several main CAD models have been created during the process of design and presented in the report together with the final design. For each model in the process, CFD simulations have been applied to observe the behaviors of the flows around the surfaces of the body during steady flights, followed by a brief analysis for further modification. A series of simulations with varying velocities and angle of attack have been performed for the final design, analyzing its performances under different air conditions. Flight envelope of the design has been presented also, together with some ideas of possible further studies on the pentacopter.

A new gage which can measure skin friction and heat flux simultaneously was designed, constructed, and tested. This gage is the combination of a non-nulling type skin friction balance and a heat flux microsensor. By mounting the heat flux microsensor directly on the surface of the floating element of the skin friction balance, it was possible to perform simultaneous measurements of the skin friction and the heat flux. The total thickness of the heat flux microsensor is less than 2 μm, so the presence of this microsensor creates negligible disruption on the thermal and the mechanical characteristics of the air flow. Tests were conducted in the Virginia Tech supersonic wind tunnel. The nominal Mach number was 2.4, and Reynolds number per meter was 4.87 x 10⁷ with total pressure of 5.2 atm and total temperature of 300 °K. Results of the tests showed that this new gage was quite reliable and could be used repeatably in the supersonic flow. This gage also has an active heating system inside of the cantilever beam of the skin friction balance so that the surface temperature of the floating element can be controlled as desired. With these features, the effects of a temperature mismatch between the gage surface and the surrounding wall on the measurements of the skin friction and the heat flux were investigated. An infrared radiometer was used to measure the surface temperature distributions. Without the active heating, the amount of temperature mismatch generated by the gage itself was from 2.5 °K to 4.5 °K. The active heating produced the temperature mismatch of 18.7 °K. The largest temperature mismatch corresponds to the levels typically found in high heat flux cases when it is expressed in dimensionless terms. This temperature mismatch made sizable effects — a 24 % increase in the skin friction measurement and a 580 % increase in the heat flux measurements. These experimental results were compared with the computational results using the Computational Fluid Dynamics code GASP. The input flow conditions were obtained from the boundary layer measurements. The temperature mismatch was input by specifying the density and the pressure at each grid point on the wall. The Baldwin-Lomax algebraic turbulence model was used with the thin layer approximations. The comparison showed that the difference in the skin friction and heat flux was less than 10 % of the measured data when the temperature mismatch was less than 8.5 °K, but the difference was increased as the amount of the temperature mismatch increased. It is presumed that the disagreement between the measurements and the calculations was caused mainly by deficiencies in the turbulence model for this complex, developing viscous flow, because the Baldwin-Lomax model cannot account for the multiple length scale in this flow. / Ph. D.

Harmonic oscillatory tests for a fighter aircraft using the Dynamic Plunge--Pitch--Roll model mount at Virginia Tech Stability Wind Tunnel are described. Corresponding data reduction methods are developed on the basis of multirate digital signal processing. Since the model is sting mounted, the frequencies associated with sting vibration are included in balance readings thus a linear filter must be used to extract out the aerodynamic responses. To achieve this, a Finite Impulse Response (FIR) is designed using the Remez exchange algorithm. Based on the reduced data, a state–space model is developed to describe the unsteady aerodynamic characteristics of the aircraft during roll oscillations. For this model, we chose to separate the aircraft into panels and model the local forces and moments. Included in this technique is the introduction of a new state variable, a separation state variable which characterizes the separation for each panel. This new variable is governed by a first order differential equation. Taylor series expansions in terms of the input variables were performed to obtain the aerodynamic coefficients of the model. These derivatives, a form of the stability derivative approach, are not constant but rather quadratic functions of the new state variable. Finally, the concept of the model was expanded to allow for the addition of longitudinal motions. Thus, pitching moments will be identified at the same time as rolling moments. The results show that the goal of modeling coupled longitudinal and lateral–directional characteristics at the same time using the same inputs is feasible. / Master of Science

The effects of grid generated free-stream turbulence on compressor cascade performance was measured experimentally in the Virginia Tech blow-down wind tunnel. The parameter of key interest was the behavior of the measured total pressure loss coefficient with and without generated free-stream turbulence. A staggered cascade of nine airfoils was tested at a range of Mach numbers between 0.59 and 0.88. The airfoils were tested at both the lowest loss level cascade angle and extreme positive and negative cascade angles about this condition. The cascade was tested in a Reynolds number range based on the chord length of approximately 1.2-2x106. A passive turbulent grid was used as the turbulence-generating device, it produced a turbulent intensity of approximately 1.6%. The total pressure loss coefficient was reduced by 11-56% at both the "lowest loss level" and more positive cascade angles for both high and low Mach numbers. Oil Visualization and blade static pressure measurements were performed in order to gain a qualitative understanding of the loss reduction mechanism. The results indicate that the effectiveness of an increasing turbulent free-stream on loss reduction, at transonic Mach numbers, depends on whether the shock wave on the suction surface is strong enough to completely separate the boundary layer. At negative cascade angles, increasing free-stream turbulence proved to have a negligible influence on the pressure loss coefficient. At cascade angles where transition exists within a laminar separation bubble, increasing free-stream turbulence suppressed the extent of the laminar separation bubble and led to an earlier turbulent reattachment. / Master of Science

Force and moment balances have proved to be essential in the measurement and calculation of aerodynamic properties during wind tunnel testing. With the recent advancements of technology, new fiber optic sensors have been designed to replace the conventional foil strain gage sensors commonly found on balances, thereby offering several distinct advantages. The use of fiber optic sensors on a balance brings with it some potential advantages over conventional strain gage balances including increased resolution and accuracy, insensitivity to electromagnetic interference, and the capability of use at high temperatures. By using the fiber optic sensors, some of the limitations of the conventional balance can be overcome, leading to a better overall balance design. This thesis considers an initial trial application of new fiber optic sensors on a conventional, six-component sting balance while retaining the original foil strain gage sensors for comparison. Tests were conducted with a blunt, 10º half-angle cone model in the Virginia Tech 9x9 inch Supersonic Wind Tunnel at Mach 2.4 with a total pressure of 48 psia and ambient total temperature of 25.3ºC. Results showed a close comparison between the foil strain gages and the fiber optic sensor measurements, which were set up to measure the normal force and pitching moment on the blunt cone model. A Finite Element Model (FEM) of the sting balance was produced in order to determine the best locations for the fiber optic sensors on the sting balance. Computational Fluid Dynamics (CFD) was also used in order to predict and compare the results acquired from all of the sensors. / Master of Science