FUNDAMENTAL UNDERSTANDING OF THE CYCLOIDAL-ROTOR CONCEPT FOR MICRO AIR VEHICLE APPLICATIONS

Abstract

The cycloidal-rotor (cyclorotor) is a revolutionary flying concept which has not

been systematically studied in the past. Therefore, in the current research, the

viability of the cyclorotor concept for powering a hover-capable micro-air-vehicle

(MAV) was examined through both experiments and analysis. Experimental study

included both performance and flow field measurements on a cyclorotor of span

and diameter equal to 6 inches. The analysis developed was an unsteady large

deformation aeroelastic analysis to predict the blade loads and average aerodynamic

performance of the cyclorotor. The flightworthiness of the cyclorotor concept was

also demonstrated through two cyclocopters capable of tethered hover.

Systematic performance measurements have been conducted to understand

the effect of the rotational speed, blade airfoil profile, blade flexibility, blade pitching

amplitude (symmetric and asymmetric blade pitching), pitching axis location,

number of blades with constant chord (varying solidity), and number of blades at

same rotor solidity (varying blade chord) on the aerodynamic performance of the

cyclorotor. Force measurements showed the presence of a significant sideward force

on the cyclorotor (along with the vertical force), analogous to that found on a spinning

circular cylinder. Particle image velocimetry (PIV) measurements made in the

wake of the cyclorotor provided evidence of a significant wake skewness, which was

produced by the sideward force. PIV measurements also captured the blade tip

vortices and a large region of rotational flow inside the rotor.

The thrust produced by the cyclorotor was found to increase until a blade

pitch amplitude of 45 was reached without showing any signs of blade stall. This

behavior was also explained using the PIV measurements, which indicated evidence

of a stall delay as well as possible increase in lift on the blades from the presence

of a leading edge vortex. Higher blade pitch amplitudes also improved the power

loading (thrust/power) of the cyclorotor. When compared to the flat-plate blades,

the NACA 0010 blades produced the highest values of thrust at all blade pitching

amplitudes. The NACA blades also produced higher power loading than the flat

plate blades. However, the reverse NACA 0010 blades produced better power loadings

at lower pitching amplitudes, even though at high pitch amplitudes, regular

NACA blades performed better. Among the three NACA sections (NACA 0006,

NACA 0010 and NACA 0015) tested on the cyclorotor, NACA 0015 had the highest

power loading followed by NACA 0010 and then NACA 0006.

The power loading also increased when using more blades with constant chord

(increasing solidity); this observation was found over a wide range of blade pitching

amplitudes. Asymmetric pitching with higher pitch angle at the top of the blade

trajectory than at the bottom produced better power loading. The chordwise optimum

pitching axis location was approximately 25-35% of the blade chord. For

a constant solidity, the rotor with fewer number of blades produced higher thrust

and the 2-bladed rotor had the best power loading. Any significant bending and

torsional flexibility of the blades had a deleterious effect on performance. The optimized

cyclorotor had slightly higher power loading when compared to a conventional

micro-rotor when operated at the same disk loading. The optimum configuration

based on all the tests was a 4-bladed rotor using 1.3 inch chord NACA 0015 blade

section with an asymmetric pitching of 45 at top and 25 at bottom with the

pitching axis at 25% chord.

The aeroelastic analysis was performed using two approaches, one using a

second-order non-linear beam FEM analysis for moderately flexible blades and second

using a multibody based large-deformation analysis (especially applicable for

extremely flexible blades) incorporating a geometrically exact beam model. An

unsteady aerodynamic model is included in the analysis with two different inflow

models, single streamtube and a double-multiple streamtube inflow model. For the

cycloidal rotors using moderately flexible blades, the aeroelastic analysis was able

to predict the average thrust with sufficient accuracy over a wide range of rotational

speeds, pitching amplitudes and number of blades. However, for the extremely flexible

blades, the thrust was underpredicted at higher rotational speeds and this may

be because of the overprediction of blade deformations. The inclusion of the actual

blade pitch kinematics and unsteady aerodynamics was found crucial in the accurate

sideward force prediction.