https://www.researchgate.net/publication/3270629_Harnessing_High-Altitude_Wind_Power
The page 2 mentions an angle of up to 50 degrees. This is likely due to the requirement of sweeping the larger possible frontal airspace. This system uses torque.
But the figure 18 of the following document about autorotation (like SkyMill Energy or what I try to do) shows that 30° is enough to optimize the force coefficient of the rotor, i.e. as if the swept surface was roughly a solid flat area.
https://dspace-erf.nlr.nl/xmlui/bitstream/handle/20.500.11881/1094/DY02.pdf?sequence=1
The force of a rotor of gyroplane increases when its rotor plane angle of attack (AoA) increases. But also its L/D ratio increases when its rotor plane angle of attack decreases when the airspeed increases. It is the reason why at low speed the AoA is high, while at high (translation) speed the AoA is very low.
I suggested on some previous post that my rotor of gyrokite has a low L/D ratio of 0.7 due to the low performances of this rotor. This is not quite accurate because an optimal thrust (lift + drag) is obtained at a high angle attack but not too high. And (roughly) 35 degrees elevation angle is an usual value for any AWES in operation. The problem is that the rotors are connected to the line and their respective angles of attack depend on said line: an elevation angle of 45 degrees imposes an angle of attack of 45 degrees, and the segments of less than 45 degrees correspond to an even higher angle of attack. This problem is not shared with the gyrokite. This explains why, at least from 2 rotors, the assembly can hardly go up because of a too high angle of attack, unless a solution such as that proposed by @Windy_Skies or myself can succeed. So to obtain a maybe desirable angle of attack of about 30 degrees, a lifter kite flying at an angle of elevation of at least 60 degrees would be required. But from a certain area of rotors the elevation angle would decrease significantly, leading to a too higher angle of attack…
According to Fig. 10, the L/D ratio of the MTOsport rotor is 2 at about 10 m/s speed then increases to reach 5 at 20 m/s. The angle of attack decreases and the thrust also, the L/D ratio of said rotor reaching approximately 11 at 33 m/s and a rotor plane AoA of 3.2 degrees (table 1), for approximately 0.1 of force coefficient according to Fig. 18. The maximum force coefficient of the rotor, then substantially perpendicular to the flow, is 1.2 or 1.4, even 2 for the parachutal drag coefficient of the MTOsport rotor, page 7.
The structure of the gyrokite allows to disconnect the angle of attack of the rotor plane from the elevation angle of the tether. If the bend rod system allows the same by using less material than for a gyrokite, that would be interesting. I think also that the curvature of the tether maybe could be used to settle the rotors at an appropriate place.