That report is a really enjoyable and honest read
it does suggest that a rotary design has a swept radius sweet spot …
Their tests with a single kite could not sweep a tight enough radius to be efficient.
Their analysis showed that sweeping too tight a radius with 2 kites linked at the tips is inefficient.
We’ve demonstrated that Kite rotors still work when you remove the central parts and replace the wing to wing connection with tensile line like the slightly later picture posted on tied flexible wing
That enables a tight autorotation with high speed on all blade elements and lower blockage with less chance of hitting your own wake
This design concerned rather C-shaped semi-rigid wings. I forgot this for two reasons: no single C-shaped wing has ever flown to my knowledge; the second reason was not correct because I thought that the tied wing blade could have a huge span while it remained narrow, being not able to fill the swept area, forgetting that the narrowness could be compensated by the number of the tied wing blades. So gigantic rotors could be built by tying very small flexible wings.
@Rodread , concerning your method by removing “…the central parts and replace the wing to wing connection with tensile line…”: indeed the wing power is more fully used (about 2 times per wing area for a complete rotor compared to a rotor of Rotokite or mine with tied wings everywhere, and it is a proven method but at small scale. I do not know if entanglements could be avoided at high scale without a fine computerized control as for any dancing kite systems.
Indeed tensile lines are efficient but only in traction, while tying flexible wings on the full span provide a supplementary shape cohesion by wind force over the entire surface of the rotor, giving also the possibility to add some lift in the central part comprising the central parachute.
In some way tying wing of the full rotor would reproduce the classic configuration of a three-blade ground-based wind turbine where the central part also does not produce much power, although it produces lift for the AWES version.
Now concerning the Rotokite report on my previous comment, I think the table 4.2 page 50 is a bit optimist.
Peak Power Output Class 20 kW 1 MW
Nominal Wind Speed 9 m/s 9 m/s
Wingspan Of One Kite 6.5 m 46.2 m
Rotor Area 130 m2 6700 m2
Kite Weight 9 kg 654 kg
Elevation Angle 30◦ 30◦
Reel-Out Speed Coefficient 0.38 0.38
Partial Load Efficiency 0.33 0.33
Nominal Force 5800 N 290 kN
Induction Factor 0.9 0.9
Nominal Ouput Power 20 kW 1 MW
Nominal Tether Length 150 m 300 m
Tether Safety Factor 5 5
Tether Weight 3.7 kg 360 kg
Hover Power 80 W 4 kW
Drag During Reel-In 1.1 1.1
Area Reduction from Rotor Area Solidity x 0.05 Solidity x 0.05
Knowing that in reeling yo-yo mode the power coefficient (CP) cannot be higher than 0.15 (4/27, so 1/4 Betz limit as documented), that leads to a value of 439.857 kW, then 285.731 kW by taking account of cosine³ coefficient (0.65) with an elevation angle of 30 degrees as stated, then half due to recovery phase, so about 1/7 the value of 1 MW as stated. It is true that 9 m/s wind speed is rather a low nominal value.
This paper presents a multidisciplinary framework for the design and analysis of gyrocopter-type airborne wind turbines. In this concept, four rotary wings provide lift to a flying vehicle, and excess power is extracted using gearboxes and generators before being transferred to the ground through electrical conductors embedded in a structural tether. A physical breakdown of the system was performed, and five models were constructed: wind model, rotor aerodynamics, structural mass, electri- cal system, and tether (structures and aerodynamics). A stochastic optimizer in the framework enforces interdisciplinary compatibility and maximizes electrical power transmitted to the ground under various operating conditions. The framework is then used to explore the design space of this advanced concept in numerous flight conditions. The effect of implementing new technologies was also studied in order to evaluate their effect on the overall performance of the system. It is shown through a 1.3 MW design that a gyrocopter-type airborne generator could provide more power than a ground-based wind turbine for a given blade radius, although only a fraction of the available wind power can be harvested using off-the-shelf technologies and components. The work presented in this study demonstrates the challenges of designing a high altitude wind generator and shows that performance is affected by complex interactions between each subsystem.