Evolution. From “wired wings” to cyclorotor


  After I formulated the “flying elevator” concept, I tried to implement it in a straightforward manner: as a some system of wings connected to a common fuselage by their wires with the possibility of winding these wires. I reference such a system as "wired wings" configuration. I considered the following four variants of this kind. They are represented in the chart below.

"Wired wings" diagram

  The simplest variant A. has only one wing connected by a wire to a fuselage. The fuselage has a powered winding system, which is placed inside the fuselage next to its CG. This winding system also has locking capabilities when it is motionless. Also, the fuselage has a stabilizer that permits adjusting its attitude during the flight. The wing is designed with elements providing longitudinal and transverse stability, like a hang glider wing, and has a central, where the wire is connected, with the possibility of moving the connection point in the longitudinal and transverse directions under remote control, which processes the handling commands from the pilot.


  The variant B. has two wings equal to the wing described for the variant A., which are connected by wires to a common fuselage, which is differed from the fuselage of variant A., having a set from two winding systems instead of one. Here, the wings are referenced as Wing 1 and Wing 2 with winding speeds WS1 and WS2, respectively. In this configuration, there is the problem of transition of one wing in the vicinity of the wire of another. An avoidance of this is too tricky to handle. Thus, this practically disables the use of this aircraft. Nevertheless, this configuration is useful for a flight simulation analysis.


  The variant C. resolves the neighbor wire avoidance problem of the variant B. by permanently placing one wing over the other, and the Wire 2 from the upper Wing 2 passes through a pulley of an enhanced central node of the lower Wing 1.


  The variant D. is like as the variant A., but the fuselage has its own Wing 1 pivotally connected to it, and the upper wing is referenced as Wing 2. So, the fuselage here becomes a glider. Also, the stabilizer here is moved up to a tail from the proximity of the main wing. The main wing here is pictured as a kind of a symmetrical airfoil to decrease its steering moment, although for the simulation I used an asymmetric airfoil. Also, a standard glider scheme can be used here, having a fixed main wing and a stabilizer with an elevator surface.


  A variant of implementation of the enhanced central node from the variant C. is shown in the chart below.

Enhanced central node


  he main element of the node is a caret, which can be moved in the X-direction on the rods mounted on the Wing 1, using the X-screw, which is rotated by the corresponding servo. For the Y-direction, a gimbal can be moved on the respective rods of the caret, using the Y-screw, which is rotated by the corresponding servo. By this way, the connection point for the wires could be displaced, providing the desired steering of the Wing 1. The Wire 1 is connected to a pulley assembly, which is used to conduct movement of the Wire 2. The pulley assembly is mounted in the gimbal. This implementation provides zero-moment footprint on the entire Wing 1 from the wires.


  All four variants of the "wired wings" configuration were tested in a flight dynamics simulation program. I used angles of attack (AoA) of the wings and the winding speeds as input handling parameters. In addition, I brought the simulation as close as possible to reality by including the strain dynamics of the wires themselves, as well as the aerodynamic drag of the wires and the fuselage. The self-explaining diagram below represents constraints of the "wired wings" simulations, grouped by their modalities. I tried to find optimal handling parameters for each variant of the aircraft.

Constraints for "wired wings" simulation

  I prepared the result of these simulations in a form of composite charts, where the upper side presents a flight profile of each component of the respective aircraft, including the wires that keep connectivity of the data. Also, there is a labeling of the numbers of the resulted samples, one per five. The lower part is a plot composed from the handling AoA of the respective wings and components of the acceleration of the fuselage, which are normalized on the gravitational acceleration. The horizontal axis of this plot is simply the number of a sample corresponding to the number labeled on the flight profiles. I placed labels of the sampling in appropriate places instead of the axis itself. Also, keep in mind that the zero lifting AoA for the used airfoil is about -4°. The result of the entire simulation is represented below with respective analysis for each variant.

Variant A.

  The chart above represents the result for one wired wing connected to the fuselage. This configuration has some similarities with the bird's flight. There is only one possibility to recover the altitude of the wing, because the wing is only one. It is a partial weightlessness for a short time. But the bird has an advantage in this operation, because its wings aren't "wired". So, to keep the aircraft at least in horizontal flight, I need to use a high magnitude of the winding speed. By that, the vertical acceleration changes from 3g when the fuselage is going up with increased AoA to -1g when the fuselage is going down with decreased AoA. The horizontal acceleration changes from 0.3g to -1.4g. Let's look at sample 18, where the phase of the positive powering begins, when the fuselage has a significant sink after a partial fall. The wing is placed significantly more forward than the fuselage, so the accelerating force is inclined and acts as an inertial force. The horizontal component of the inertial force inclines the normal gravitational vertical, so it becomes an inertial vertical. The fuselage begins to accelerate in both directions. The slowly flying wing promptly reaches the speed of the fuselage under the gravitic acceleration, and they continue to move together, keeping a constant inclination of the wire up to sample 35. Now the fuselage has a significant positive vertical speed and increased horizontal speed. On sample 39 the phase is finished, the wing and fuselage are almost upright, but I don't switch to the negative powering phase. I locked the wire and wait for the vertical speed of the fuselage to be maximal. During this intermediate phase the wing accelerates and undergoes pendulum oscillations with a short period, which are reflected in the oscillations of the acceleration. On sample 53 the recovery phase begins. By this time the horizontal speed of the fuselage decreased. Previous recovery phase begins on sample 7. AoA was decreased to -1°, and to -3° on the next sample. Prior to this, the wing was in strong acceleration due to the high inclined flight path and has reached high speed. The high speed has induced a high aerodynamic force reflected in the mentioned vertical acceleration of 3g, which was possible since the wire was locked. Remember that there is a slipping constraint of 1.4g without this locking. In the recovery phase the wing continues to move forward and up, winding out the wire. Its flight path angle switches to the positive direction and the gravitational force begins to decelerate it. The speed of the wing significantly drops, and also the aerodynamic force. The fuselage enters in almost weightlessness and begins to increase its sink until the end of this phase. So, it isn't a comfortable flight. Also, it is too dangerous.

Variant B.

  The chart above represents the result for two equal wings connected to the fuselage. This aircraft permits less level of the oscillations of the acceleration, below 2g for the vertical component and 0.5g for the horizontal with both signs. The positive phase of one wing overlaps with the recovery phase of the other. But this overlapping induces mutual dependence in phases. This dependence leads to a higher resonance of long-periodic pendulum oscillations of the fuselage and wings. So, the amplitude of the speed-oscillations for the wings is very high, because the mass of the wings is small. This leads to the periodical occurrences of very low speeds of the wings, when a wing almost cannot support its own weight. This is very dangerous, since the wire begins to become forceless at the end of the recovery phase.

Variant C.

  The chart above represents the result of simulation for the aircraft with two wings at separate levels. This aircraft performs a bit better than it needs for a cruise flight. The Wing 1 through the pulley assembly of the central node applies additional constraints on the horizontal position of the upper wing and vice versa. So, the amplitude of the horizontal oscillations of the wings is reduced significantly. Here I succeeded in the simple handling of the aircraft. Each wing has AoA of 5° while the fuselage goes up toward it. And in this time the opposite wing has AoA of -1.5°, when it is flaring up, winding its wire out. However, this regular pattern of the handling isn't symmetrical. The phase with the upper sustaining wing is longer than the phase with its recovering. The vertical acceleration is reduced there to a range from -0.3g to 0.65g, and the horizontal acceleration - to a range from -0.3g to 0.15g. Also, these accelerations have the pattern of decrementing oscillations replenished after each transition between the handling phases. I use a low winding speed here, so the phases are long, permitting to see details of these oscillations. However, the system has a drawback: the winding speed that I use is maximal. An additional increasing in the winding speed leads to a transition to a mode of highly increased and irregular fluctuations with a significant loss of altitude and to increasing of the rotational energy of the entire system, i.e., to a high-entropy behavior. So, gaining the cruise altitude for this variant is still problematic.

Variant D.

  The chart above represents the result of simulation for the glider with an additional wired wing. This aircraft performs a just enough for a cruise flight. The pattern of the handling is also regular like for FIG. 5C, but there is a prolonged intermediate state for both main and recovery AoAs of the wired wing only. The system has a prolonged recovery phase, when the glider is mainly sustained by its own wing with AoA of 6°, and the wired wing is flaring up with AoA from -2.5° to -2°. After this, a shortened lifting phase occurs, when the wing of the glider is idle with AoA of -2.5°, and the wired wing sustains the glider with AoA from 3° to 6°. The vertical acceleration here lies in the range from -0.28g to 0.18g, and the horizontal acceleration - in the range from -0.17g to 0.15g. Although the winding speed that used here is higher than for the previous variant, the short powering phase doesn't permit to gain cruise altitude at all.


  So finally, the "wired wings" configurations permit to have only an aircraft with the ability to perform a cruise flight, low ability to gain cruise altitude and zero ability to perform operations on the runway for takeoff. These limitations follow from the constraint of the self-sustaining abilities of the wired wings themselves, and from the lack of control of their angular kinetic energy relative to the center of gravity of the entire aircraft. So, for the correct implementation of the "flying elevator" concept, there needs an aircraft with wings that are being fully controlled in their movement and steering. Ideally, the wings of such an aircraft should be in some conveying movement with a certain winding speed along their pivots, where their path has a segment where the lift powering is performed, and another segment to perform a simple return to the upper position, maintaining a low level of aerodynamic force. So, I designed a variant of such a "conveyer" configuration, which is represented in the chart below.

"Conveyer" configuration

  The aircraft is pictured in cruising flight and has a standard fuselage with an upper tail stabilator, which is used to compensate for variations of the moment of the "conveyer" actuators on both sides in a wide range of flight operations. The pathway of the wings on the actuator is a rounded rectangle, which is inclined backward at an angle of the Skew from its vertical position. This inclining is used to distribute a load of the lifting wings along the fuselage direction, and to reduce the overall height of the aircraft and the driving force of the entire actuator along its pathway. The pathway of the actuator has a segment "I", where the lift powering performs, a segment "II", where the recovering of the altitude of the wings performs, a segment "III" to perform a transition from the recovering to the powering, and a segment "IV" to perform a transition from the powering to the recovering. Also, due to the duality representation of the power, the lifting segment "I" can be considered as performing a propulsion powering. And also, the same possibility exists for segment "II", when its wings have a negative load. Such a negative load was not possible in the "wired wings" configurations, but now it is possible for this aircraft. I suppose that the number of the wings per the actuator pictured there is close to optimal, since having a smaller number can lead to a high level of vibrations, and having a larger number leads to too weak wings. I also assume that the separation of the wings pictured there is also close to optimal, in order to have a compact enough actuator with a fairly low level of wings' interference.


  Although from the operational point of view, this aircraft looks perfect, it has a significant drawback. It is almost impossible to implement. The main challenge for this is to solve the problem of the presence the illustrated movement of the wings with their simultaneous rigidness and their long length for the desired high aspect ratio. Indeed, the aircraft should have a high aspect ratio of the wings (AR) in order to be effective enough. But its wings must be rigid enough to withstand the high load on the segment "I" and the high level of centrifugal forces on the segments "III" and "IV". The best method to get sufficient rigidness is to bring the wings to a common support at their free ends. Doing this for a circular path can simply be resolved by a ring. But that doesn't work here. Thus, one way to resolve this problem is to place the wings between two fuselages, which has many disadvantages, such as the presence of additional transverse elements for the frame rigidness. I cannot exclude that one day this problem will be resolved, but currently I don't have a multi-tiered correct solution for this.


  So, the remained way to correctly implement the "flying elevator" concept in an aircraft is to use a circular actuator. An exemplary variant of this kind aircraft is represented in the chart below.

Cyclorotor configuration

  The aircraft has a fuselage that is similar to the fuselage of the previous aircraft, and now the stabilator is used to compensate for variations of the moment implied by the circular actuator of the aircraft, which I reference as a rotor. The rotor has the same number of wings as the previous aircraft, and the same wings' separation. Now those specific segments of the wings pathway have an overlapped placement, since the lifting ability of a wing has some variativity over the forward side of this pathway, so do those equivalent propulsion abilities of the negative loaded wings on the rear side of this pathway. Also, the mentioned ring, which provides rigidness to the rotor, is not shown here so as not to obscure the wings in the view.


  Aircraft with circular actuators are known as cyclorotor aircraft. I discuss the matter of them in a separated topic "Cyclorotor aircraft".


Copyright © 2018-2021 Yuri Feldman - All Rights Reserved