Fluids & Structural Mechanics > CM Projects > Turbomachinery & Vehicle Analysis (TVA)

Figure 1:

This is an example (c. 2001) of a fully coupled analysis of a complex, three blade row pumpjet. In these types of simulations the nominally axisymmetric hull flow is coupled to the three-dimensional flow through each blade row in a technique that we refer to as a powering iteration. The only inputs to a powering iteration are geometry, vehicle speed, and shaft rpm. The result of a powering iteration is a detailed picture of the time-mean performance of the propulsor at the specified advance ratio from which multiple forcing functions can be extracted for structural integrity and acoustic performance predictions. Also shown in this figure are comparisons of the three-dimensional blade row simulations with oil paint flow visualization of the unit operating in the Garfield Thomas Water Tunnel. Powering iterations are performed routinely as part of ARL’s propulsor design methodology.

Figure 2:

This figure shows an example of a more sophisticated form of a powering iteration (c. 2005). In this case, the hull is not a body of revolution and the propulsion pumps are deeply embedded within internal ducting. For this type of configuration, the powering iteration couples the three-dimensional hull flow with two mixed flow pumps each consisting of a rotor and a stator. The result is a detailed picture of the pump behavior when installed within the hull and ingesting the flow from the hull boundary layer. From these calculations one also gets a detailed picture of the hydrodynamic loads on the hull under powered conditions. Examining off-design conditions during maneuvers is also readily performed to examine, among other things, the tendency for pump starvation under severe conditions. Once again these types of simulations can be done as part of the pump design process.

Figure 3:

This figure is an example of validation exercise for a submerged, unpowered vehicle under off-design conditions (c. 2005). The vehicle is ONR Body-1 and the experimental data was taken at NSWC-CD. This particular simulation is an example of the use of overset mesh techniques to resolve geometric components and/or volumetric flow features. The left image shows surface static pressure contours and particle paths for ONR Body-1 at 12º of yaw. The particle paths illustrate the roll-up of the sail vortex and two hull vortices which are clearly evident on the mid-body slice showing contours of axial velocity. Shown on the right is a sample comparison between flow simulation results and measured data for the net transverse force and yaw moment over an 18º range of yaw angle.

Figure 4:

This figure shows the flexibility of the overset approach for simulating the flow field around very complex hull forms (c. 2004). The vehicle is the Infante UUV (upper left). This UUV has multiple control surfaces, ducted thrusters, various masts/antennas, and three tunnel thrusters all of which are evident in the CAD definition shown at the lower left. The upper right image shows the surface overset mesh for the fully appended vehicle. At the lower right the resulting flow field corresponding to a subset of the appendages is shown. The figure shows surface static pressure and particle paths for some flow particles ingested by the tunnel thrusters. Overset meshing is ideal for these types of stripping analyses of complex hulls. This is because, with overset, one can readily build-up an increasingly complex geometry while ensuring very high grid quality on each individual geometric component.

Figure 5:

This is an example of a non-Navy application of ARL’s turbomachinery and vehicle analysis flow simulation capability (c. 2006). These images show the flow around a non-lethal ring airfoil projectile (right) and its sabot (left). In this analysis the aerodynamic loads on the components were the metric of interest. For the ring airfoil projectile, multiple conditions were examined consisting of different magnitudes of cross winds and different projectile rotation rates from barrel rifling.

Figure 6:

This is another example of a non-Navy application of ARL’s computational toolset. This effort involved predicting the acoustic performance of a miniature unmanned air vehicle (c. 2006). The primary metrics from the CFD analyses, therefore, were acoustic forcing functions. The flow simulations involved powering iterations under various hovering and non-hovering conditions. The powering iterations coupled the three-dimensional duct and strut flow (lower left) with the detailed three-dimensional rotor flow (lower right). The downstream stator was modeled throughout using equivalent volumetric body forces reverse-engineered from the blade geometry.

Figure 7:

This figure shows two biomedical applications of ARL’s in-house turbomachinery analysis capability. Both are ventricular assist devices (VAD), i.e., blood pumps. The left image is an axial flow VAD; the right image is a magnetically-levitated Tesla VAD. An important hydrodynamic metric for the simulation of both devices is efficiency, i.e., the energy rise across the pump as a function of the work input. Efficiency is a universal concern in all turbomachinery design. However, for these types of devices, a very unique forcing function related to hemolysis (or blood damage) is also extracted from the simulations. An estimate of hemolysis is as important as efficiency for any VAD design.