| In previous work, fully-pulsed jets have been researched as an alternative to steady propulsion. Research has shown that propulsive efficiency, etaP, of a pulsed jet can be improved by decreasing the jet slug length-to-diameter ratio, L/D, or increasing the duty cycle, StL (Moslemi and Krueger, 2009 and Moslemi and Krueger, 2010). While pulsed jets lose efficiency as vehicle Reynolds Number, Renu, is decreased, the losses are smaller than that of steady jets (i.e. hPhP,sj , the ratio of propulsive efficiency of a fully-pulsed jet to an equivalent steady jet, increases as Renu decreases) making it a more attractive means of locomotion for vehicles that operate at intermediate Reynolds numbers (1<Renu<1000.0) (Moslemi and Krueger, 2011).;Using a robotic squid, dubbed Robosquid, developed for pulsed-jet experimentation at the Experimental Fluid Dynamics Lab at Southern Methodist University (SMU), additional inlets and jet outlets were retrofitted to test the effects these configuration changes have on etaP and hPhP,sj , in the intermediate Renu range mentioned above. The baseline configuration used radial inlets and a smoothly tapering 0.75-inch (1.91 cm) diameter nozzle design. Additional propulsive devices tested were directionalized inlets that could be faced forward or backward, a blunted nozzle design of identical diameter to the baseline, and two orifice designs. One of the orifice designs maintained the same diameter as the nozzles, and the other used a 50% larger diameter at 1.13 inches (2.86 cm).;The drag of each configuration was determined experimentally for 130< Renu<7100. A jet velocity program with an isosceles profile (IP) in which equivalent acceleration and deceleration to a peak jet velocity, UM was used, L/D and StL were varied as digital particle image velocimetry (DPIV) measurements of the jet flow were taken along with time-averaged vehicle velocity, Unu, measurements. This data was used to calculate etaP and hPhP,sj for each unique configuration and pulsing condition. For the larger orifice design, the pulsing was varied slightly since the change in diameter, D, prevented exact matching of the pulsing parameters. Two variations were considered: matching the mass flow of the pulse, m˙p, of the baseline pulsing and matching the L/D and UM of the baseline pulsing. All efficiency data was collected at 25<Renu<110 in which the jet Reynolds number, Rej was 275 for all the standard diameter outlets, 413 for the large orifice with matched L/D and UM, and 184 the large orifice with matched m˙p.;Efficiency gains on average of 2.03% were seen in the blunted nozzle over the smoothly tapering baseline nozzle. Drag results further confirmed that viscous effects begin to dominate at the intermediate Re nu range, and that blunted aft-sections can be more efficient at sufficiently low Renu. Data also showed that inlet directionality plays a part in etaP. Forward-facing inlets seem to have the best etaP, followed by radial inlets and aft-facing inlets in descending order. This indicates vehicle motion and pressure effects at the inlet still play a significant enough role at the intermediate Renu range to reverse momentum effects. Results also showed that the orifice designs are more efficient than nozzle designs. The compact energy storage from the vena contracta of an orifice-generated vortex ring leads to an average etap of 1.02% higher than a nozzle design of identical diameter, assuming identical drag characteristics for the vehicle (i.e. a blunted nozzle). The large orifice design with matched m˙ p in which Rej was 184 consistently showed the highest etap for all L/D and StL conditions tested, including the other large orifice condition at which Rej was higher. This suggests that lowering Rej for a fixed diameter also maximizes efficiency. Though this resulted in the lowest Unu, it made the disparity between the velocities across the shear layers of the jet and the Eex shed into the jet lower relative to the other configurations tested. |
