Patent Grant
10421538
The invention relates to aircraft having both Vertical Take-Off and Landing (VTOL) capability and non-rotary-wing horizontal airplane flight.
The present invention combines aspects of helicopter and fixed-wing propeller-driven aircraft design and aerodynamics. As will be seen, advances in control as developed particularly for quadcopters enable greater flexibility in flight modes and simplifications of mechanical and aerodynamic designs. Given the very large patent literature relevant to this invention, most of it being old and now basic to the education of VTOL engineers, this Background section will largely ignore the specifics of the patent literature and instead will recall known, applicable general concepts and terminology.
A motivating principle behind the present invention involves induced drag and its relationship to forward speed through the air. In low-subsonic aerodynamics, the significance of the formula for the induced drag of an ideal elliptic lift distribution is readily understood in terms of a picture. Viewing an airplane from directly behind, consider a circle whose diameter extends from wingtip to wingtip. As the aircraft moves forward, this circle sweeps out a cylinder of increasing volume and a corresponding rate of increase of contained air mass: d(air_mass)/d(time). As the pair of wings pushes down on the air it is passing through, a flat wing with an elliptic lift distribution across its span will encounter a downward velocity component, “induced_velocity.” Two applicable formulas are given:
Lift=−d(momentum)/d(time) for downward momentum imparted to the air
Lift=−d(air_mass)/d(time)·induced_velocity for velocity of sinking air in the wing wake
For a wing with an elliptic spanwise lift distribution, the effective mass to which downward momentum is imparted is equivalent to the mass of air passing through this imaginary circle whose diameter matches the wingspan. As the aircraft moves faster, that circle sweeps through an increasing d(air_mass)/d(time), so for constant lift, the induced velocity decreases inversely with the increase in speed and swept air mass. Since the air mass is sinking down under the wings, the aircraft must effectively climb uphill through the sinking air in order to maintain level flight. Viewing the drag problem in terms of power, the power loss varies as the time derivative of mass multiplied by the square of the induced velocity:
Power=d(energy)/d(time)=½·d(air_mass)/d(time)·(induced_velocity)2
These formulas may vary by factors-of-2, depending on whether the induced velocity is specified at the wing or in the far wake, but the core principle is clear. To maintain constant lift, one must maintain a constant product of d(air_mass)/d(time) times (induced_velocity) while seeking to minimize the square-law power term. Induced drag is reduced as the craft travels faster through the air, so that more air mass is pushed down with a lesser induced velocity. This improvement in energy performance ceases at high speeds where the increase in form drag, varying roughly as the square of velocity, overtakes the decrease in induced drag. In a helicopter slowing to hovering flight, the fixed-wing induced drag formula ceases to be applicable. The rotary wings (e.g. two rotary wings for a two-bladed helicopter) cease to engage air mass with forward motion through the air. The spanwise horizontal-axis circle described above becomes a vertical-axis circle which is the swept diameter of the rotor. A relatively small d(air_mass)/d(time) is propelled directly downward with a correspondingly high velocity. Recalling the previous two equations, one sees that to maintain a constant lift according to the first equation while engaging a relatively low mass flow through the rotor at a correspondingly high velocity, the ratio of Power/Lift becomes large. A skilled helicopter pilot will minimize hovering and favor forward flight, where the helicopter's induced drag approaches the applicable fixed-wing formula. For a helicopter, the Aspect Ratio, defined as span2/area, becomes Diameter2/((π/4)·Diameter2)=4/π. This is obviously a low aspect ratio when compared to fixed-wing aspect ratios typically varying from 6 to 50. The rotor blade form drag associated with rotary forward flight is also particularly high, since the effective average tangential velocity of the rotor blade through the air must be significantly higher than the forward speed of the helicopter. In a single-rotor helicopter, the rotor blades traveling “backwards” opposite the direction of forward flight must have sufficient motion through the air to develop lift that balances that of the forward-moving rotor blade or blades. For a given rotational tangential tipspeed of the helicopter blades, a practical maximum forward flight speed is approximately 25% of that tipspeed. The maximum airspeed of the advancing rotor tip would therefore be roughly 4+1=5 times the forward speed, while for the receding (with the wind) rotor tip the airspeed multiple would be 4−1=3 times. The power dissipation associated with form drag of an airfoil varies roughly as the square of speed through the air, in the present example implying a power-dissipation multiple of 52=25 for the advancing wingtip and 32=9 for the receding wingtip. The energy dissipation multiples become more extreme at smaller radii, while the lack of adequate net airspeed to balance lift on the receding rotary wing becomes more extreme, thus setting the approximate 25% upper practical limit for forward speed as a fraction of tipspeed.
A comparison of rotary- versus fixed-wing aircraft energetics goes as follows. Consider aircraft of equal weight and where the rotor diameter matches the wingspan. In forward flight with equal engine power, the fixed-wing craft will fly more than twice as fast and achieve more than double the mileage. Clearly there is ample incentive to develop an aircraft combining the advantages of rotary-wing vertical takeoff and landing with efficient fixed-wing horizontal flight. Historic examples of such aircraft are seen in the experimental Bell Helicopter XV-3 and XV-15 aircraft, steps in the design lineage leading to the Bell-Boeing V-22 Osprey and the AgustaWestland AW609. In these related aircraft, vertical takeoff and low-speed horizontal flight are achieved with twin side-by-side helicopter rotors. An aircraft transitions to horizontal fixed-wing flight as the rotor planes tilt to provide increasing forward thrust, completing the transition with the rotors serving as propellers and fixed wings providing lift. Comparing similar aircraft, various models of the Bell UH-1 “Huey” series cruise between 120 and 140 knots, with the related, very high powered AH-1 “SuperCobra” having a cruise speed of about 150 knots and a maximum forward speed of 190 knots. A V-22 Osprey in airplane flight mode cruises at around 300 knots. The Osprey has separate systems for developing lift in its helicopter and airplane modes: two rotors for helicopter mode and wings for airplane flight. In helicopter mode, the wings interfere with the rotor downwash, with a substantial fraction of the wing chord hinging down and out of the way of the downwash as an oversize flap. In airplane flight, the rotors are extremely over-sized as propellers, while their twist is a compromise between relatively low twist for helicopter mode and much higher optimum twist for the high prop advance ratios of forward flight. The complexity, cost and high empty weight (in relation to wingspan and payload) of the Osprey are indications of the disadvantages of this aircraft design approach.
Other aircraft examples represent different compromises for achieving VTOL capability and forward flight in airplane mode. Harrier Jump Jets lift off by focusing downward thrust through a very small cross-section of airflow at very high downward velocity and very high power. To conserve fuel, this aircraft is forced to make quick transitions from takeoff to horizontal flight and back from horizontal flight to quick vertical landing.
There is a great need for an aircraft design that combines VTOL capability with the advantages of fixed-wing airplane-mode forward flight. The following Specification will teach such a physical design with its essential and optional degrees of freedom, along with a method for controlling its flight in VTOL and airplane modes and in transitions between the two modes.
The invention is an aircraft with two wings that function both as rotary wings in a helicopter mode and as fixed wings in an airplane mode, those wings being propelled in both modes by thrusters located along the wing spans or at the wing tips. The wings and thrusters feather together, controllably and independently about the wing-pitch-change axes, including large feathering rotations that align the wing leading edges and thrust vectors approximately oppositely for rotary-wing flight and in the same forward direction for airplane flight. The aircraft may include a normally non-rotating fuselage joined to the sometimes rotary wing pair through a mast bearing with optional means to control non-rotating yaw headings of the fuselage in relation to the rotations of the wings in helicopter mode. The wing and pitch change structure may be mounted to allow helicopter-mode rotary wing flapping and associated rotor-plane tilt with respect to the fuselage.
In particular, The present invention is an aircraft capable of (a) sustained powered rotary-wing-mode VTOL flight and (b) sustained powered non-rotary forward airplane-mode flight, that includes a hub, a pair of wings coupled to the hub and serving as both rotary wings in rotary-wing-mode flight and as a pair of lifting wings in airplane-mode flight and a propulsion component configured to provide propulsion for the aircraft sufficient for sustained powered flight, wherein the propulsion component includes a plurality of thrusters including at least one thruster on each wing of said pair of wings, wherein said plurality of thrusters provides sustained rotation of the pair of wings in rotary-wing-mode flight and sustained forward motion of the pair of wings and of the entire aircraft in airplane-mode flight, wherein said at least one thruster on a first one of said pair of wings is able to thrust in substantially the opposite direction from propulsion from said at least one thruster on a second one of said pair of wings in rotary-wing-mode flight and, wherein said propulsion from said at least one thruster on said first one of said pair of wings is able to thrust in substantially the same direction as propulsion from said at least one thruster on said second one of said pair of wings in airplane-mode flight. The thrusters may be unshrouded propellers or shrouded turbines, for example. The propulsion from each of said plurality of thrusters may be varied differentially in said airplane-mode flight to control yaw rotations of said aircraft. In a transition between said rotary-wing-mode flight and said airplane-mode flight, said at least one thruster on said first one of said pair of wings is affixed to said first one of said pair of wings and rotates in a feathering rotation with said first wing about a pitch change axis of said wing, said pitch change axis being substantially parallel to a span of said wing, and where said thruster and wing rotate in said feathering rotation through an angle change exceeding about 120 degrees relative to the hub. The hub may be located between said pair of wings and attached to each wing of said pair of wings, wherein attachment between said pair of wings permits feathering rotations in both wings about said pitch change axes of each of said two wings, wherein said feathering rotations control aircraft roll in said airplane-mode and, said feathering rotations control variations in the plane of rotation of said pair of wings in rotary-wing-mode. The attachment may control and power said feathering rotations. The thrusters on said wings may cause gyroscopic torsions acting on said wings about the respective pitch change axes of said wings when operating in said rotary-wing-mode, and said gyroscopic torsions may be controllably varied cyclically in a one-per-rev cycle to augment or entirely effect said feathering rotations to control said variations in the plane of rotation in rotary-wing-mode. One example of accomplishing that functionality involves including for at least one of the thrusters at least two rotary propeller or turbine components rotating in substantially opposite vector rotation senses, wherein variable angular momenta arising from said opposite rotation senses at least partially cancel one another and, said gyroscopic torsions about said pitch change axes, arising from said variable angular momenta in rotary-wing-mode flight, are controllably varied through alteration of said rotating in substantially opposite vector rotation senses, thereby acting to augment or entirely effect said feathering rotations. Further in that instance, said at least two rotary components of each of said thrusters provide independently variable thrust vectors acting through differing moment arms with respect to the associated one of said respective pitch change axes, whereby a variable wing-pitch-control moment arising from said thrust vectors and moment arms controls wing pitch in airplane-mode flight. Optionally, the aircraft includes a normally non-rotary fuselage, operating below said hub in normal sustained flight in both rotary-wing-mode and airplane-mode flight, with rotatable attachment to said hub, wherein said fuselage maintains a controllable non-rotary yaw angle in said rotary-wing-mode flight. In that instance, said rotatable attachment to said hub includes torsion actuation operating through said attachment to control said non-rotary yaw angle. Alternatively, in the embodiment with the rotatable attachment of the fuselage to the hub, the aircraft may include a controllable aerodynamic thrust in said fuselage, said thrust acting through a radius from the axis of said attachment for said maintaining a controllable yaw angle. In that case, said thrust acting through a radius is provided by a tail component undergoing angular changes, said angular changes interacting with downwash from said rotary wing to provide said thrust acting through a radius. The aircraft may include decoupling and re-attachment of said rotatable attachment, whereby said hub can fly independently of said fuselage or with said attachment to carry said fuselage. It may also include tilt-angle decoupling between said fuselage and said rotary wing, whereby the plane of rotation of said rotary wing can tilt in pitch and roll directions independent of the pitch and roll angles of said fuselage. In that version, optionally said tilt decoupling includes a flapping hinge in said hub, allowing said pair of wings to flap through variable angles with respect to said hub or it includes a universal hinge permitting rotation and suspension of said fuselage at arbitrary suspension angles in pitch and roll with respect to said plane of rotation.
The invention is also a method for controlling an aircraft in a rotary-wing-mode of flight and an airplane-mode of flight and in transitions between the two modes, the method comprising the steps of independently controlling the pitch angles of each of two wing-plus-thruster systems of the aircraft in feathering about their respective wing-pitch-change axes, including feathering to cause angle changes exceeding about 120 degrees relative to a hub that is common to both of said two wing-plus-thruster systems, controlling said pitch angles to direct the thrusts of said wing-thruster systems over a range of angles in pitch as related to changes in upward lift, in approximately opposite directions for rotary-wing-mode flight, controlling said pitch angles to direct the thrusts of said wing-thruster systems over a range of angles in approximately the same directions for airplane-mode flight, controlling said pitch angles cyclically and differentially to control rotary-wing-mode plane of rotation, controlling said pitch angles differentially to control aircraft roll in airplane-mode flight, controlling said pitch angles continuously in flight in transition from said approximately opposite thrust directions to said approximately the same thrust directions for transition from said rotary-wing-mode to said airplane mode and, controlling said pitch angles continuously in flight in transition from said approximately the same thrust directions to said approximately opposite thrust directions for transition from said airplane-mode to said rotary-wing-mode. The method further optionally includes controlling angular speeds of the thrusters of said wing-thruster systems differentially and cyclically in said rotary-wing-mode in a one-per-rev cycle, thereby cyclically varying gyroscopic wing-pitching moments in said thrusters to control wing cyclic pitch. It further includes the option of independently controlling the thrusts of said wing-thruster systems differentially in said airplane-mode for controlling aircraft yaw.
Another method of the invention provides for controlling the pitch angles of a pair of wings with corresponding attached thrusters, those wings being joined to a control hub and constituting an aircraft, each wing and thruster combination being capable of independent pitch angle rotation through large angles under servo control, the method comprising the steps of determining a target common-mode lift from the pair of wings, determining a target differential-mode lift from the pair of wings, determining indicated airspeeds of the two wings, determining, from said indicated airspeeds, from said common-mode lift and from said differential-mode lift, target theoretical lift coefficients for the two wings, determining actual lift coefficients of the two wings and servo controlling the pitch angles of the two wings to cause said actual lift coefficients to approach said target lift coefficients. In that method, said target differential-mode lift may be varied cyclically at a one-per-rev rate, synchronized to rotations of said aircraft in a rotary wing mode, thereby to control the plane of rotation of said aircraft. If one of said indicated airspeeds of the two wings falls below a predetermined threshold, then said servo control of pitch angles may revert to an alternate control method. The alternate control method may be used to control the vector forces of said thrusters. The method for controlling pitch angles may also include the step of limiting said target lift coefficients to magnitudes achievable within aerodynamic capabilities of the two wings.
These and other features will become clear from the following Specification.
A sequence of figures represents a sequence of views, through time, of conversion maneuvers from helicopter mode to airplane mode and back to helicopter mode. Starting from the reference view of
The present invention is an aircraft design and related control methods for Vertical Take Off and Landing (VTOL) and in-flight conversion to and from a fixed-wing horizontal flight mode. The VTOL mode will usually be described as helicopter mode or helicopter flight and the fixed-wing mode as airplane mode or airplane flight. In helicopter mode, the rotary wings may sometimes be called rotor blades or simply blades.
Referring to
Degrees of Freedom in Single Images and Images Paired for Comparison
Not shown in the figures, the yaw-inducing differential prop thrust variations of
The following figure sequence may be regarded as selected frames from a movie showing flight conversion from helicopter mode to airplane mode and back to helicopter mode.
Selected “Movie Frames” of Flight Conversion:
FIGS: 2, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 2.
Note that several complete rotor revolutions may transpire in the sequence between
Referring again to
Degrees of Freedom, Descriptions:
1. Controlled independent feathering of the two wings about their pitch change axes, including:
a. Helicopter collective pitch change
b. Helicopter cyclic pitch change
c. Airplane common-mode pitch change to control altitude, speed (elevator equivalent)
d. Airplane differential-mode pitch change to control roll (aileron equivalent)
2. Controlled independent power-speed variation of the motors/props on each wing, including:
a. Helicopter common-mode power-speed variation to control rotor rotation speed
b. Helicopter differential cyclic speed and gyroscopic torque variation to drive cyclic pitch (optional)
c. Airplane common-mode power-speed variation to control speed, altitude
d. Airplane differential-mode power-speed variation to control yaw
3. Passive hinge to allow flapping degree of freedom (optional)
a. Helicopter mode, allows rotor-plane tilt independent of mast-fuselage tilt
b. Airplane mode, allows wing roll independent of mast-fuselage roll
4. Rotation decoupling of rotor mast and hub from fuselage
a. Omission option (2600): no decoupling, everything spins with the rotor
b. Passive/aerodynamic option (2700): wind aligns fuselage during forward flight
c. Active/aerodynamic option (2900, 3000): rotor downwash pushes variable control surface
d. Active/mechanical option 2220: motor drives counter-rotation of fuselage via bearing suspension 2210
The items of the above outline are now described.
1. Controlled Independent Feathering of the Two Wings
Each wing rotates about a pitch change axis, this axis typically being aligned with or nearly with the airfoil lifting line at or near the 25% chord. Called wing feathering, this pitch change degree of freedom is illustrated in
In related configurations 300 and 400 of
In the present invention, the wing airfoils are typically symmetric, since one of the two wings must reverse its direction of lift when transitioning in either direction between helicopter and airplane modes. An aerodynamic symmetry of the two wings is desirable, considering that one of the two wings must operate “right side up” in helicopter mode and then “upside down” airplane mode. For similar reasons, the wings are typically not twisted. A potentially desirable twist with decreasing pitch angle going from root to tip would become reversed and would go the wrong way following transition from one flight mode to the other. Variable wing geometry opens up more options, for example when a wing can be cambered one way for helicopter flight and the camber removed or reversed as needed during conversion to airplane flight. Independent pitch control of each wing is an aspect of this invention, including through pitch change angle ranges relative to the hub of about 120 degrees or more for at least one wing.
Pitch control requirements fall roughly into two categories: rapid controlled cyclic pitch change through relatively small angles, as is familiar for controlling the plane of rotation of a helicopter rotor; and relatively slow pitch changes to vary collective pitch in helicopter mode, net lift angle in airplane mode, and left-right differential lift angle for airplane roll control—the equivalent of aileron control but using whole-wing rotations in the context of this invention. As mentioned, the relative pitch angle changes used for cyclic pitch control are illustrated by a comparison of the wings in
While yaw rotation might optionally be controlled as indicated in
2. Controlled Independent Power-Speed Variation of the Motors/Props
Clearly, varying motor power and prop thrust on both wings in helicopter flight will alter rotor speed and power, as needed in conjunction with collective pitch changes to control lift and vertical velocity.
Less obviously, as the props spin about their own axes and simultaneously rotate about the rotor hub, the resulting gyroscopic torques tend to alter the rotor blade pitches. If the rotor blades and props both function as right-hand airscrews, with the rotary wings lifting upward when rotating counterclockwise as viewed from above (and as shown in these figures) and with the props also rotating counterclockwise as viewed from in front (thrusting toward the observer), then the gyroscopic moments on the props will tend to pitch the rotary wings up. The same pitch-up torsion occurs if blades and props are both left-hand airscrews. Since the inertias of flat untwisted rotor blades tend to lie flat in the plane of rotation and resist upward pitching, the pitch-up propeller gyroscopic interaction is potentially beneficial. On the roughly one-meter-span scale of the illustrated aircraft, gyroscopic pitching moments from the props can represent a significant fraction of the total pitch-up moment needed for hovering flight or, depending on design details, the gyroscopic moments can be excessive compared to the so-called “tennis racket” wing-flattening effect of helicopter parlance. Cyclic variation in prop rotation speeds can augment or entirely drive cyclic pitch. Previewing view 2500 of
Aircraft with twin engines, four engines (two per wing) or six engines (three per wing) commonly derive equal power from the left-wing and right-wing engines. In the present invention, it is an extension of common practice to use variable symmetric or common-mode prop motor power variations for speed and altitude control. Use of variable antisymmetric or differential-mode prop power variations for yaw control is less obvious, as conventional aircraft rely on rudder control to produce yaw. In designs of the current invention intended optionally to eliminate the necessity of a trailing rudder or leading yaw control surface, a left-right differential in engine power and/or blade pitch is used to produce yaw. This operating mode will generally call for active electronic yaw sensing and feedback, which is readily accomplished in modern aircraft with potential reductions in mechanical complexity, weight and cost.
One optional mode for sensing indicated airspeed, aerodynamic pitch and yaw in airplane mode is illustrated in
3. Passive Hinge to Allow Flapping
For flight control in helicopters, a flapping hinge degree of freedom is commonly provided, though some rotary wing aircraft fly without it. The flapping degree of freedom is illustrated in views 800 and 900 going from
4. Rotation Decoupling of Rotor Mast and Hub from Fuselage
While
In common embodiments of this invention, the fuselage is coupled to the rotor mast via bearings, for example inside bearing sleeve 2210 of
Extending fuselage yaw control to the hovering situation, a variable pitch rudder 2910 of
In the contexts of
By contrast with
As contrasted with the aerodynamic yaw control solutions variously shown for the fixed vertical stabilizer 2720 with hinged rudder 2721 and for fully rotatable rudder 2910, a more direct mechanical way to control fuselage yaw is illustrated in
These detail illustrations of embodiment 100 show all the powered system components and their powering batteries (
Control Dynamics, General Discussion
In the embodiment 100 of this invention, as shown variously in whole or in part in
This new aircraft, however, has additional dynamic pitch control in the form of variable propeller thrust, which is exerted along a line passing above the aircraft c.g. to cause a variable pitch moment. In the absence of prop thrust and with an un-stalled low-drag wing, the lift vector is directed nearly at right angles to the relative wind angle. The pilot (whether human, remote or robotic) can vary pitch to vary the magnitude of the lift vector, while the line of that vector tends usually to pass nearly through the craft c.g. The wing pitch variations therefore have very little short-term effect on aircraft pitch change. Fuselage pitch variation comes with delay as the wing pitch variations cause the flight path to slope upward or downward. As sensed by someone riding in the aircraft, the apparent gravitational “down” vector is always nearly perpendicular to the flight path, not “down” towards the center of the earth. In smoothly curving gliding flight without prop thrust, the line from the wing center to the pendulous aircraft c.g. usually points nearly perpendicular to the flight path with centrifugally-driven tilt away from the curvature center of a circling flight path.
The poor short-term pitch control situation just described is improved by the addition of variable prop thrust in a high wing. Propeller thrust variation acts nearly perpendicular to lift variation, where the line of the thrust vector in embodiment 100 passes well above the aircraft c.g., giving a significant moment arm for rapid dynamic pitch control. This pitch control is used in concert with the slower pitch change associated with wing lift and the flight path slope. In a conversion maneuver that includes a steep climb with airspeed approaching zero at the peak height, the wings lose control authority as the airspeed decreases. The spinning props, however, retain their control influence and enable the craft to controllably re-orient both pitch and yaw angles for completion of the conversion. The following text and figures describe how this works in the context of flight conversions from helicopter to airplane flight and back again.
Control Dynamics Applied to Flight Mode Conversion
An aircraft with prop-rotors plus wings, such as the V-22 Osprey and its research predecessors and the more recent AW609 aircraft, has redundant aerodynamic surfaces: wings that are a hindrance to helicopter performance since they are pushed down by the prop-rotor downwash, and prop-rotors that are sufficiently large to carry the craft in helicopter mode and therefore much too large to make efficient propellers. An advantage of this redundancy is that the aircraft can make smooth flight transitions between helicopter and airplane flight modes, transferring aerodynamic lift continuously from the prop-rotors to the wings and back again. The aircraft of the present invention lacks this redundancy in two ways:
1) Having propellers that are appropriately sized to spin the rotary wing and also to power forward flight with high efficiency, but generally too small to lift the aircraft by themselves; and,
2) Having two wings that fully serve both rotary wing and fixed wing functions.
The consequence is that flight conversions entail temporary partial-to-total loss of lift as the available aerodynamic surfaces reorient to different functions. Since the vertical lift component must average one-g over time (where a lower average implies cumulative downward momentum and hitting the ground), the conversion maneuver must include periods of significantly elevated g-forces to compensate for the momentary loss of lift. A particular consequence is that in a sufficiently large version of this aircraft, a human passenger must be prepared for a brief roller-coaster ride where the beginning and ending flight conversions each include one near-zero-g moment, preceded and/or followed by elevated-g time period(s). While this precludes use of the human-scale aircraft in general aviation service including passengers who will not tolerate the rough moments, this leaves a wide variety of practical uses, including with passengers.
Flight conversion maneuvers are now described in greater detail, recognizing that departures from the following description are possible within the scope of the invention. The description begins with reference to pictorial figures, seeking a general understanding. Further description follows with reference to the charts of method steps in
Transition from vertical takeoff to horizontal airplane flight optionally starts with a hovering rotation as in image 200 of
An option for both take-off and landing is to provide a landing gear with rolling components (tires, etc.) appropriate for take-off and/or landing in airplane mode, offering some fuel or battery-energy saving compared to VTOL mode. The VTOL option remains as needed.
To transition from horizontal aircraft flight to helicopter flight, a typical conversion maneuver begins by increasing prop power and putting on extra speed in the flight mode of
The pictorially illustrated dynamic sequences of flight conversion just described are now described in greater detail and in more optional variations with reference to the control method steps of chart 3400 in
3401: On Ground: Set Max Power, Servo Zero CL Until Max-Rpms
In this step, as would be viewed in configuration 200 of
Care must be taken for safe operation in this phase. Cyclic gyroscopic bending moments in propeller blades with be highest under these conditions of high rotor RPMs multiplied by high prop RPMs. Centrifugal force on the rotary wings will also be at a maximum, while wings optionally pre-coned to minimize bending stresses in rotary wing flight will experience unusual downward-bending stresses during this phase of kinetic energy accumulation.
3402: Servo High CL for Sustainable-Rpms, Ascend, Max Power
Continuing the energy-conserving strategy, a rapid initial lift-off maximizes the vertical velocity gained while still in ground effect. Although power demand is at a maximum during rapid vertical ascent, at least some of the rotor power is usefully lifting the aircraft vertically. If the aircraft were hovering above ground effect, the rotor would be pulling air down at high speed just to sustain the weight of the aircraft, a most inefficient situation.
3403, Optional: Tilt Rotor Disk for Angled Ascent or Level Flight, Rotary Mode
One-per-rev cyclic pitch, as illustrated by comparison of
3404: Servo to Max CL, Max Power, Ascend, Decreasing RPMs
Substantial rotary-wing pitch-up, normally seen to this extent only in rapid ascent, is illustrated in
3405: Geometric Pitch Differential to Parallel Wings, Max Power
As the aircraft rises rapidly while rotor RPMs decrease and while the rotary wings maintain a high angle of attack at a maximum efficient CL, the rotary wings will pitch differentially, i.e. in opposite rotation senses (recall the oppositely directed rotation symbols 1010 and 1020 of
3406: Optional: Geometric Pitch Differential Past Parallel to Stop Rotor, Return to Parallel
With wing rotary motion still significant, the wings pitch in differential geometric fashion past the parallel orientation of 1100, with the props or turbine thrusters continuing to operate at full power so that their thrusts are angled to oppose and stop the rotation. As rotation approaches zero, the wings and thrusters return to parallel alignment. At this point the aircraft may be rising or, if heavily loaded, likely falling, possibly with a significant component of horizontal motion.
3407: Geometric Pitch Common Mode Down, Max Power
As shown in configuration 1200 of
3408: Servo Common Mode Max-CL, Descend, Gain Airspeed, Max Power
The aircraft gains speed rapidly (1300 of
3409: Servo Common Mode CL for Level Flight
As shown in configuration 1500 of
3410: Servo Power to Bring Servoed-CL to Cruise-CL
If indicated airspeed exceeds the desired cruise speed, then the target CL for wing-pitch servo control is reduced under servo control. This situation involves two layers of servo control: of CL, in relation to the overspeed, to maintain the amount of lift required to sustain horizontal flight; and of wing pitch angle, to track the servo-target for CL. Thruster power is reduced, allowing the aircraft to slow to its intended cruise speed while the wings automatically increase their aerodynamic attack angle to develop the same lift force as the airspeed decreases. Alternatively, the aircraft might use excess speed to regain some of the altitude that was lost during the flight conversion. The chosen combination of lift, thruster power and altitude gain (or loss) will depend on optimization factors of ordinary airplane-mode flight, also depending on the intended flight plan including the desired final cruise altitude.
3501: Increase to Max Power, Servo Common Mode CL for Level Flight, Gain Speed
As the aircraft previously stored extra kinetic energy in rotary motion for vertical takeoff and conversion to airplane-mode flight, similarly a strategy for conversion back to rotary-wing flight may begin by adding kinetic energy, here in the form of forward speed. Height above the ground must also be sufficient for safe conversion with any expected altitude loss during conversion. As the aircraft packs on kinetic energy as extra speed, and to the extent that altitude gain is not needed, the wings pitch somewhat downward together under aerodynamic servo control, lowering their attack angles and CL-coefficients while maintaining constant lift. The aircraft configuration will continue to resemble 1500 of
3502: At Max Speed, Servo Common Mode Max-CL, Ascend, Lose Speed, Max Power
The conversion maneuver will initially resemble an aerobatic hammerhead stall, wherein the aircraft puts on extra speed, angles its flight path upward to the vertical while losing speed, and comes to a stop at maximum altitude. In the present instance, the aircraft may intentionally retain some horizontal velocity as its vertical velocity goes to zero. Highly powered model airplanes will be seen vertically oriented and hovering entirely on the support of prop thrust. An expected goal for most design variations within the scope of the present invention will be to design for less than sufficient thrust to hover on prop or turbine thrust alone (without help from rotary wing lift)—thus avoiding over-sized thrusters that would be inefficient for utilitarian high-mileage cruising flight. Thus, the expected conversion maneuver would combine steep altitude gain with loss of vertical speed through zero and into descending flight. Configuration 1600 of
3503: Servo Common-Mode CL for Steep or Vertical Ascent, Lose Speed, Max Power
As shown in configuration 1700 of
3504: At Zero Speed, Geometric Pitch Common Mode Down, Descend
As vertical descent begins, the wings pitch down together in common mode, i.e. rotating with the same direction sense as viewed externally, pitching down rapidly to minimize stall or avoid it entirely. Configuration 1800 of
3505: Gain Speed Down, Servo Differential Max-CL, Gain RPMs
Shifting from common mode to differential pitch change (
3506: Optional Tilt Rotor for Angled Descent
The descent path of the aircraft may target a slowing vertical descent to landing or optionally an angled descent, taking advantage of the aerodynamic power savings of combining horizontal with vertical flight. If vertical landing is desired, horizontal velocity may be arrested shortly before landing. In a descent with braking of vertical velocity, care must be taken to avoid vortex-mode loss of lift, where the aircraft descends and drops to the ground in a self-generated bubble of falling air. Maintaining a horizontal flight velocity component and avoiding excessive velocity braking will avoid this catastrophic vortex mode.
3507: Arrest Descent, Servo CL for Hover or Level Flight
At this stage, the aircraft enters a flight plan for normal helicopter-mode flight. Flight configuration 200 of
3508: Servo Power to Bring Servoed-CL to Hover or Cruise-CL
The goal here is to use rotary-wing flight for slow forward travel, hover or vertical descent through a speed-restricted airspace. Wing geometric pitch is determined by servo-feedback to achieve a lift coefficient, CL, appropriate for the flight condition.
This ends the detailed description of flight mode conversions for aircraft 100. Conversions will differ somewhat for configurations 3000 and 3100 or
The aircraft invention being described here, in its various configurations and embodiments, is scalable over a wide range from “model airplane” size to large enough to carry human passengers willing to handle the transition maneuvers between flight modes. Besides the necessity of those maneuvers, there is another inherent design limitation worth mentioning. In helicopters and other rotary wing aircraft such as quadcopters, problems arise when propeller or rotary wing tips approach sonic velocity with steep increases in drag, power dissipation, noise generation and other problems associated with transonic flow, supersonic flow and shock waves. As a practical matter, tip speeds exceeding about Mach 0.8 are problematic. In helicopter rotors, required tip speeds tend to increase with disk loading. As geometries scale from small to large, disk loading tends to increase. Thus, tip speed Mach limitations tend to become more prominent with increasing scale. In the present aircraft invention, the propellers that drive the wings develop prop tip speeds considerably larger than the rotary wing tips in helicopter mode, potentially creating high Mach number problems.
Within a given rotor design and a given disk loading requirement, prop tip speeds can be reduced while maintaining sufficient propeller power by moving the props inboard from the wing tips (as shown at 2730 and 2732 of
Finally, there is an issue with vibration in two-blade propellers. The gyroscopic propeller torsion described above manifests as cyclic bending stress in individual prop blades. This bending stress goes through a full reversing cycle with each prop revolution, with the greatest bending exerted when the blade passes nearly perpendicular to the disk of rotary wing rotation. Individual blades must be strong enough to endure the resulting cyclic stresses. Two-blade props transmit torque vibration to their hubs, shafts, motors and wings at two vibration cycles per prop revolution. These torque vibrations result in considerable noise generation as well as potential wing fatigue problems. Symmetric arrangements of three or more blades per prop avoid this transmitted vibration, though the individual blades still encounter the cyclic stresses and must be designed to withstand them. Hence, the embodiment 100 of
Calculations make it clear that cyclic gyroscopic blade bending stresses in the present invention can be severe and problematic. A first line of defense is to design blades that are very strong and very lightweight. Carbon fiber composite construction is a favorable design option. While mass reduction is emphasized toward the blade tips where the gyroscopic inertia contributions are high, the blade roots should be made very robust to withstand the cyclic bending moments, with transition from thin to thick airfoil sections moving from tip to root. Looking beyond individual blade geometry, one finds that blade gyroscopic bending moments vary as the product of rotary wing angular velocity multiplied by prop angular velocity and by the mass moment of inertia of each prop blade. For blades operating at the same tangential tipspeed, blades scaled up with geometric similarity to longer blade radius will rotate relatively more slowly (inversely as blade length for the same tangential tipspeed) and will have greater strengths in their cross-sections. In computing worst-case peak surface stresses, however, the greater mass moments of inertia of the larger blades generally overwhelm the advantages of lower rotation speed and greater cross-section strength. Thus, relatively small-diameter props are commonly desired. Prop blades of a given radius and designed to operate at relatively high advance ratios will be spinning relatively more slowly when advancing through the air at a given rotary-wing tangential speed (which becomes the advance speed of the props.) Thus, prop design for a high advance ratio confers a reduction in cyclic gyroscopic bending stress. In order for slower-spinning blades of similar radius and surface area to develop a given required net prop thrust, however, one must use a greater number of blades-per-prop. To manage the problem of excessive cyclic gyroscopic blade bending stresses, therefore, a favorable design direction is toward relatively small props using three, four or more blades per prop and operating at high advance ratios and correspondingly low rotation speeds. Potentially applicable ducted turbine designs carry this trend further, using many blades on each of several coaxial turbine elements and, at least in some cases, with all the turbine blades operating at high advance ratios.
This section has taught the structural topology, articulated degrees of freedom and associated actuation forces and moments of a new aircraft in embodiment 100, including variations within the context of that embodiment. It has described the essential aerodynamic features and indicated how those features can be controlled, via the actuation forces an moments, to achieve helicopter and airplane flight modes and transitions between modes. The discussion has put certain design limitations into perspective, particularly regarding Mach number limitations on speed and disk loading and also regarding propeller stresses, vibrations and associated noise. The following section provides more detailed descriptions of various embodiments of this invention, demonstrating the scope and variety of designs falling within the scope of this invention. The options chosen for these embodiments will not be construed as limiting, nor will all of these options necessarily be found to be optimal in light of future contexts of design and use. The teachings provided already and extended below, however, define a new aerodynamic and mechanical topology offering most of the best of two worlds: VTOL capability and propeller airplane capability in a relatively simple and economic package. The general discussion above has already covered the important functional design features of an embodiment of the invention as pictured in all the figures up thru 25. As mentioned briefly above in conjunction with
The details of the aircraft configuration depend on its intended use, with the motor and prop locations 2730 and 2732 of
The sequences of mode-transition events previously described entailed major wing configuration changes when the aircraft was stopped or nearly stopped. Wings 2800 are ineffective at very low airspeeds. Therefore any mode transition that significantly involves those wings needs to take place when the aircraft is moving through the air fast enough for wings 2800 to lift a significant fraction of the aircraft weight. For design simplicity and economy, one might hope to use fixed-pitch wings, but there are compelling reasons to include variable net pitch, with the wings optionally joined across the fuselage for matching pitch rotations or, in a more complicated configuration, with the wings having independent pitch angles. The feature at 2805 represents a pitch-change spar extending from the wing into the fuselage, with internal fuselage features controlling the pitch of that wing and similarly for a hidden spar emerging into the opposite wing. One motivating reason for variable secondary-wing pitch is to allow the wings to pitch steeply upward to streamline their areas to rotary-wing downwash. A more important reason for variable secondary-wing pitch is to achieve the control necessary for these wings to assist in flight mode transitions, as is now discussed. In an optional description for transition from rotary-wing to fixed-wing flight using wings 2800, the rotary wings tilt their plane of rotation forward to achieve a maximum practical forward speed in that mode. Aerodynamic drag will then tend to pull the fuselage back and into a nose-down pitch, especially if the wings 2800 are unable to pitch up into the downward wind angle caused by rotor downwash. With variable pitch, however, the wings 2800 can pitch up to come out of stall and develop lift, partially taking over from the lift of the rotary wing. Variable lift from tail assembly 2700, specifically from horizontal stabilizer 2710 with variable elevator control surfaces 2711, may be needed to pitch the fuselage so that the lower-wing lift vector balances the top-heavy (relative to wings 2800) mass of the aircraft. With further increases in forward speed in rotary-wing mode, the rotary wing on the side traveling backward relative to the aircraft's forward motion sees a reduced relative wind speed and, even with cyclic pitch becomes unable to match the lift of the opposite forward-traveling wing. The assistance of lower wing 2800 taking over part of the lift extends the range of forward rotary-wing speed. If the wings 2800 are further capable of differential pitch change, they can compensate for the developing roll moment as the rotary wings become unable to balance lift between the advancing and retreating wings. The increasingly rapid forward advance calls for a correspondingly steeper tilt of the rotation plane of the rotary wings, which in the illustrated configuration must be accomplished entirely by wing flapping and not by forward tilt of the aircraft body. Forward fuselage tilt would cause the lift vector from wings 2800 to pass behind the aircraft c.g., tipping the aircraft uncontrollably further nose-down. Pitch control from the elevator in 2700 plays a role here. To complete a transition to airplane mode, wings 2800 take over 100% of aircraft lift while the rotary wings pitch to zero lift angles and the aircraft glides forward, under control but temporarily un-powered. The propellers then act to slow rotary wing rotation, stopping the rotary wings with one wing facing forward and one aft. In this alignment, one of those wings can be flipped without being caught broadside to the wind, smoothly coming into parallel pitch alignment with the opposite wing. A 90-degree yaw rotation then swings the rotary wings into forward-facing alignment for airplane flight. The rotary wings then pitch up to share aircraft lift with the lower wings in a biplane configuration, while the props come up to speed and provide airplane-mode thrust.
A transition from airplane-mode to rotary-wing-mode flight optionally follows approximately the reverse of the sequence just described. The airplane is flying forward as a biplane. The aircraft speeds up sufficiently to fly entirely on the lift of lower wing pair 2800, possibly including some extra speed in anticipation of speed loss during conversion. With their aerodynamic pitch angles servoed to zero lift, the rotary wings are rotated by prop action into fore-and-aft alignment with respect to the wind. In this alignment, the wings rotate in pitch relative to each other until they are both pitch-flat and facing in opposite directions. The props then start to spin the rotary wing, enabling the forward-advancing wing to develop significant lift. The resulting roll-moment imbalance is compensated by differential pitch change in wings 2800. As the aircraft continues to slow and the rotary wing rotation speeds up, the retreating-side rotary wing becomes able to lift and begin to achieve roll-moment balance in the rotary wing. Lift from wings 2800 is progressively reduced as the aircraft continues to slow and transition entirely into rotary-wing forward flight.
The tail section 2700 shown in
As a practical note regarding clearances, the hub mechanisms of 140 and its streamline fairing 141 are relatively large in the figures up thru
Given the complexity of the flight controls of this invention, semi-robotic or fully robotic control is highly desirable. Some of the sensors listed above overlap in use. For example, time-integration of linear and angular accelerometer sensor outputs provides rapid dynamic indication of velocity and angle, but that indication is subject to drift. GPS indicates position and, with time delay, velocity or at least short-term-average velocity, which can be used to correct drift in the integration of accelerometer signals. Partial indications of angular orientation in space come from the 3-D magnetic compass and possibly from differing antenna signal strengths of GPS and radio beacons: signals that can be combined to give orientation independently or as corrections for drift in the integration of electronic gyro signals.
As shown and discussed with reference to
When the corresponding impact pressures from 3050 and 3051 are sensed differentially with respect to free-stream pressures or comparable barometric pressure references, these pressure differentials, divided into the corresponding top-to-bottom pressure differentials, provide quotient indications of wing aerodynamic attack angles, nearly independent of airspeed. The context of the present invention with its very wide variations in geometric wing pitch angles makes it particularly important to control pitch in relation to true aerodynamic flow angles, rather than in relation to hub geometry. One wants to control lift, not pitch, and one wants to assure that wing angles of attack do not exceed stall angles except possibly when the craft's overall airspeed is very low and aerodynamic control is given over to the thrusters, be they props or turbines. Servo control of wing pitch for aerodynamic attack angle is discussed below along with alternative approaches to measuring wing attack angles.
There are possible redundant indications derived from expected flight dynamics. For example, a wing roll angle will result in a changing yaw angle so that (for instance) a steady compass heading will indicate flight trim with no roll, thus providing drift correction for roll angle indications from a gyro. Similarly, a wing pair's collective aerodynamic pitch is related to indicated airspeed and vertical acceleration, while the left-right wing-pair differential pitch is related to indicated airspeed and roll acceleration. These dynamic relationships provide redundant sensing information. There is value, however, in fast-responding direct sensing of aerodynamic pitch of the two wings from indicated airspeed and pressure differentials between different parts of the wing. A goal of dynamic integration of these various signals is to provide seamless indication of the parameters of flight, valid over short and long time frames. Such integration is being developed and applied to the sophisticated control of quadcopters and multicopters, providing methodologies that can be adapted to the flight control sensing and software needs of the present aircraft invention.
View 2200 of
If the centrifugal tension wire 2540 shown here extends for the full span of the wing and is thick enough to withstand the needed centrifugal force, but not excessively thick, then in a small lightweight aircraft, its torsional compliance can be high enough that wire torsion moments can be largely ignored. “Wire” 2540 may optionally be a stranded wire or a composite fiber-reinforced rod, for example a pultruded carbon fiber rod of glass fiber rod. Stranded and composite options like these provide high tensile strength with a greater torsional compliance than would be obtained using a solid wire. Indeed, in full scale helicopters, harnesses including many loops of wire withstand centrifugal forces while allowing pitch change, but their torsional stiffness is high and must be taken into account in the pitch control design. Such a conventional harness will typically be anchored to provide a pitch-up moment, thereby relieving some of the steady forces in the pitch control linkages. A similar torsional function can be realized in the present context, with similar harnesses or with the wire 2540 shown in
The above discussion regarding pitch-up moment in a twisted wire 2540 is applicable only where the thruster gyroscopic moments are smaller than the gyroscopic wing-flattening or “tennis racket effect” moments of the wings at their nominal rotary-wing pitch angles. Depending on design details, however, the gyroscopic torsion moment from a thruster (including the prop or turbine and its driving motor or engine) can exceed the inertial flattening moment of the driven rotary wing at a desired pitch angle. In such a context, the pitch control motors may be required to counter steady incompletely-canceled gyroscopic moments from the thrusters. Previewing embodiment 3100 of
Discussion now shifts from steady rotary-wing pitching moments to the cyclic pitching moments associated with dynamic cyclic pitch control. Cyclic pitch control in helicopter-mode flight can optionally be provided entirely by cyclic variation of prop speed and the associated variable gyroscopic moment. Consider a completely flat rotor blade that, at zero pitch, has no inertial extension along the vertical axis. It can then be shown that for small angular perturbations from zero pitch, there will be an inertial natural resonance between blade inertia and centrifugal pitching moment variations with pitch angle change, such that the blade will pitch up and down resonantly at a one-cycle-per-revolution or one-per-rev frequency. Blade inertia extending along the vertical or out-of-plane axis will lower this natural frequency below one-per-rev. Components that will add such out-of-plane rotational inertia include the props and their driving motors of this invention.
Each rotary wing blade and its associated pitch-inertia components can be dynamically re-tuned to one-per-rev by coupling it to a torsional spring restoration. View 2500 of
The steady-state rotational phase of the pitch change response to variable thruster motor speed will depend on the net un-balanced one-per-rev wing inertia (if the spring is too soft) or torsion spring rate (if the spring is too stiff) and on torsional damping, while in the short term, near-resonant pitch change response will build cumulatively over multiple wing revolutions. Further system response lag arises because the rotational inertia of the thruster will cause motor speed variations to lag behind changes in motor winding currents and resulting electromagnetic torques. Given these multiple interacting response lags and phase shifts, cyclic pitch control from motor speed variation must include dynamic corrective feedback and, for dynamic stability, also feed-forward. Consider for example a prop motor whose driving voltage phase is synchronized to the mechanical rotation phase of the motor. In such a motor, electromagnetic torque will vary predictably as a function of motor current alone. That magnetic torque will control angular acceleration, so thruster angular velocity response will exhibit a second-order dynamic lag behind motor current. In an algorithm of a feed-forward motor and cyclic pitch controller, that lag can be expressed as a function of the rotation phase of the rotary wing over a range of operating conditions. For example, one might find that in a given operating condition, thruster gyroscopic torque might lag behind applied motor current by some angle of rotation of the rotary wing. In order to control gyroscopic torque for cyclic pitch control, a feed-forward controller would therefore apply changes in motor current at substantially the same angle ahead of the desired gyroscopic moment. Further phase lead would be needed to account for the dynamic lag in response of left-right differential wing lift to gyroscopic moment, accounting for in-plane and out-of-plane components of wing rotational inertia, aerodynamic damping of mechanical response, and even the lag of aerodynamic lift behind pitch angle variation. Finally, changes in tilt of the rotor disk-plane lag geometrically by 90-degrees of rotation behind variations in differential wing lift. So, if one wants the rotor plane to pitch nose-down for forward acceleration, then rotary wing lift must be maximally reduced when the wing extends laterally to one side coming around to the front, and maximally increased when the wing extends to the opposite side headed to the back. The net effect of these multiple phase lags and variable amplitude responses might be evaluated by an adaptive controller, for example employing “fuzzy logic” rather than an analytic model to get the desired response from the system.
This kind of control method logic will need to be worked out in varying contexts for varying physical embodiments of the present invention. For example, in a motor and gear system such as 2340 and 2341, providing collective pitch bias while spring 2510 and its opposite-wing counterpart interact cyclically with varying dynamic current to control cyclic pitch and variation in the plane of rotation of rotary wings 110 and 112—this is the kind of situation that will call for an appropriate control method. The mechanical, inertial and aerodynamic systems disclosed here are amenable to this kind of control and may be resolved for specific application contexts.
It is unlikely that the resonant pitch control options described above would be worthwhile in a relatively small, simple version of the present aircraft invention. The added complexities involved could, however, become worthwhile in a larger, heavier aircraft intended for thousands of hours of service. Where a high premium is placed on reliability, helicopter-mode pitch control entirely through motor speed control could be achieved in a backup mode. If the aircraft is in helicopter mode and the pitch motors are fully disengaged, allowing the rotor blades to feather freely, then their collective pitch would be set by a combination of torsion bias in the centrifugal tension wire (possibly a pitch-down bias) plus gyroscopic pitch-up moment from the props. Increasing non-cyclic prop speed would immediately increase blade pitch, causing the craft to rise. As the rotor speed increases in response to the extra prop torque, the rotor blade inertia moments would bring the pitch angles back down with the increasing rotor speed, sustaining the extra lift while desirably bringing the blade pitch angles back down toward a design-optimum angle of attack. Conversely, slowing the props would immediately reduce blade pitch, causing the aircraft to drop, while the rotor speed would slow and the blade collective pitch would move partway back to the original design-optimum angle. Thus, prop speed could entirely control hovering vertical response. Cyclic prop speed variation could provide control of cyclic pitch, tilt of the rotor plane and horizontal motion. This mechanism could provide backup control in the event of certain failures, provided that the failed normal pitch control mechanism could be disengaged to allow this to work. Control of this sort is not expected to be sufficient to convert from airplane flight to helicopter flight. Note that in a system with relatively higher prop rotational inertia, the stabilizing vertical control response just described could give way to instability.
While the above discussions have engineering significance for particularly lightweight, low-angular-momentum thruster systems, their reasoning may be difficult or impractical in applications using high-angular-momentum thrusters whose gyroscopic pitching moments significantly exceed the strengths of the “tennis racket” wing-flattening inertial tendencies of the rotary wings themselves. Such significant variations on the previously-described embodiments are now discussed with reference to
Unlike the teardrop-shaped hubs of earlier embodiments, the central rotary hub 3010 of aircraft system 3000 is essentially a flying wing combining low drag properties with lift that is continuous with the spanwise lift distribution of the attached wings. Lateral cross sections through 3010 have airfoil shapes whose aerodynamically-balanced 25% chord points are at least approximately lined up with the balanced pitch change axes of the wings. The middle lateral sections, however, extend further aft behind the regular section trailing edges into a thin aerodynamic tab 3015. This tab may be fixed and serve as a horizontal stabilizer or may optionally be angled up and down as indicated at 3016, providing variable elevator control. The tab is drawn bending upward toward the trailing edge, thus normally providing a pitch-up moment (in the absence of a significant angling-down elevator control change.) When this pitch-up moment is balanced against gravity working through a wing c.g. placed ahead of the dynamic lifting line of the wing section (generally meaning ahead of the 25% chord), the result is a dynamically pitch-stable airfoil that inherently seeks a design angle-of-attack. That angle can be intentionally varied by optional elevator angle control 3016.
In earlier discussions of mode-conversion maneuvers of aircraft embodiment 100, it was found that the aircraft wings lost control authority as the flight path approached a peak altitude and began to descend. Pitch control then relied on variations in thruster force acting through an effective pitching moment arm with respect to a pendulous c.g., somewhere well below the level of the props. Optional center prop 3020 of embodiment 3000, interacting with lifting body fuselage 3010 and tab 3015 and optionally with its elevator variations 3016, provides continuity of fully aerodynamic fuselage pitch control at low forward airspeed. Prop wash from 3020 streams back across the slowing hub fairing 3010 and continuing across tab 3015, providing controllable pitching moments and limited controllable lift. This optional system does not rely on a pendulous aircraft c.g. placed well below the lines of the thruster force vectors. This leads to a different kind of fuselage attachment, as is now described.
The flying-wing approach to pitch stability as just described is not nearly as robust, in countering load imbalances, as is a more conventional airplane structure with a horizontal stabilizer placed far aft of the main wing. The hub itself, however, can be carefully balanced by design, while imbalances that may arise in the payload pod 3060 are torsionally decoupled from hub 3010. Specifically, linking rod 3005 joining 3060 to 3010 includes a universal swivel at an attachment point inside 3010 that lies on or very close to the effective aerodynamic balance center of the entire wing-plus-hub system. This swivel at the top of 3005 allows 3060 to swing freely fore-and-aft as indicated by double-arrow 3008, also to swing side to side as indicated by double-arrow 3009, and finally allows free yaw rotation of 3060 in either direction and through multiple revolutions as indicated by double-arrow 3007. Given this mechanical freedom, the orientation and alignment of pod 3060 is controlled by the sloping rudder 2910 and the elevator 2901, which will be recognized from earlier appearance in
In all embodiments of the invention, rotary wing control of carrier orientation in space is derived from the stabilizing “platform” of not-easily-changed rotational inertia about the yaw axis, with robust aerodynamic moments to tilt that gyroscopic “platform” arising from cyclic pitch of the wings and resulting strong variations in wing lift acting through large moment-arm radii to apply plenty of torque. In airplane-mode flight, carrier roll orientation is controlled by comparably robust aerodynamic moments of the wings operating with left-right pitch angle differences. Carrier yaw orientation is strongly controlled by differences in thrust between the left and right wing thrusters. This leaves carrier pitch in airplane-mode flight as the weakest control dimension. Carrier airplane-mode pitch control is handled in three alternative modes in three embodiments, two of which have been described. These first two are now reviewed briefly for comparison with the third pitch control mode and embodiment.
In variations of embodiment 100 of the present invention, pitch control arises primarily from the interactions of the associated flight path and wing lift vector with the pendulous mass below the wings. This control is augmented by variable vector thruster forces operating more or less at right angles to the flight path. In flight mode conversions, this “augmentation” of variable vector thruster force momentarily becomes the sole source of pitch control. Fortunately, the conversion maneuvers are of brief duration, so that pitch variation can be largely a predetermined trajectory controlled largely by the initial conditions of linear and angular momentum upon emergence from wing-controlled flight into the conversion transition. In this system, the carrier unit, consisting of the joined thrusters and rotary wings and the hub system with flapping-hinge decoupling, may or may not be able to fly independently with its fuselage or payload pod detached. The joined hub and fuselage units have been treated as a single pendulous mass in pitch control discussions above. The detachment option depends on details of the design and its mass distribution. Within the scope of this invention, extra elevator-like surfaces may optionally be added to the carrier of embodiment 100 for more robust control. Such surfaces are generally more effective if mounted to the hub, either aft or canard-style forward (with active stabilization) to take advantage of greater lever arms aft or forward of the lift line through the wings. Common mode wing pitch control finds the “anchor” against which it works in the orientation and relatively high pitch-change inertia of the hub. Embodiment 3000 replaces the streamlined but otherwise non-aerodynamic “tear drop” hub 140 with the aerodynamic “flying wing” hub 3010, with optional control augmentation by prop 3020, tab 3015 and further optional elevator control 3016. Unlike in embodiment 100, the pitch of carrier or cargo pod 3060 is entirely decoupled from pitch of 3010. This 3010 hub body, in turn, serves as the “anchor” for wing pitch control, as with embodiment 100.
In contrast to embodiments 100 and 3000, pitch control in embodiment 3100 of
The wing flattening moment Mθ is roughly:
Mθ≈Ω2 sin(θ)lyy (1)
The gyroscopic pitching moment Mgyr is roughly:
Mgyr≈Ωω cos(θ)lax (2)
Pitch control by gyroscopic pitching inertia alone requires roughly the following:
Mgyr>Mθ (3)
This Eq. 3 inequality is equivalent to the following:
(ω/Ω)cot(θ)(lax/lyy)>1 (4)
The equivalent inequalities 3 and 4 are satisfied for very small positive angles θ, where the cotangent function goes to infinity. The inequalities are satisfied in many practical situations for rotary wing operation in hovering and forward flight, but not for the extreme pitch angles needed for flight mode conversions. The moment couples generated by angled prop thrust, as described below, might be sufficient to control flight conversions. Some pitching moment assistance coming from the hub-to-wing pitch couplings will commonly be desired for reliable operation. The pitch control approach described for embodiment 3100 will be applicable where the above inequalities are satisfied with a sufficient margin.
Considering thruster system 3112, there are two independently rotating motor-prop thruster subsystems 3130 and 3131. These are counter-rotating, so that the prop of 3130 pulls forward and the pusher prop of 3131, having the opposite handedness of 3130 (such as a left-hand prop for pushing and right-hand prop for pulling), will push. A near-coaxial alignment allows the pusher prop to recover some of the twisting wake energy of the puller prop. This loss recovery becomes significant with high-pitch props. Importantly, the angular momentum vectors of 3130 and 3131 tend to cancel. If the two motor-prop systems are identical, the angular momenta and associated gyroscopic pitching moment effects will cancel when the props run at equal speeds in opposite rotation senses. Unequal motor-prop systems might be considered within the scope of this invention, including where pitch-up moments are needed more than pitch-down moments. Another way to obtain average pitch-up moments is available, however.
Note that thruster 3130 is centered below the bracket joining 3130 and 3131 to the wing, while 3131 is centered above the bracket. The bracket itself is aligned with the chord of the wing. The offset motors are parallel to each other and to the bracket and wing chord. The opposite vertical offsets of the motors cause a tilt in the larger principal axis of inertia of the motors and props in the plane of the wing chord. This skewed inertia causes the thruster system 3112 using thrusters 3130 and 3131 to develop a pitch-up inertial moment when operating in rotary-wing-mode. Other modifications to the mass distribution in planes perpendicular to the pitch change axis can similarly bias a wing to naturally assume a desired pitch-up angle in rotary-wing flight. One can, for example, add weights near the wing roots and lying above or below the plane of the wing chord and also well ahead of or behind the pitch change axis. Centrifugal forces on such weights near the root are low, but the pitching moment effects are the same near the wing root as near the tip. The products of inertia of these weights skew the angle of the original chord-plane moment of inertia of the wing.
Wing pitch can then be varied from this skewed rotary equilibrium angle by relative speeding-up of 3130 and slowing-down of 3131 and vice versa, using the rotation speed differentials to unbalance the opposing angular momenta and drive pitch change. The opposite-wing thruster system 3110 is shown facing forward in airplane-mode. Unlike leading or puller thruster 3130 on the right wing, which is offset below the bracket center, leading puller thruster 3120 on the left wing is offset above the bracket center. The same kind of asymmetry applies in comparing thrusters 3131 and 3121. When the left-hand wing is flipped 180 degrees about the pitch change axis, however, 3120 is then below-center and 3121 above-center. Indeed, the left wing and thruster are found to be identical to the right wing and thruster rather than mirror images through the center-plane of the hub. In airplane-mode configuration, one might expect mirror symmetry of the two wings, but that is not the case here.
In rotary-wing-mode, the dual-component thrusters are seen to provide controllable pitch-up and pitch-down gyroscopic moments when the speeds of the two paired components are varied differentially, one faster and one slower. Their offsets above and below the plane of the wing chord serve inertially to bias both wings pitch-up in this mode, with the degree of bias rotation determined by the vertical and horizontal spacings of the motor-prop components. Speeded up or slowed down together, the thruster units provide variable thrust. The inertia distribution in the relatively flat hub will cause it to spin in the plane of rotation of the wings and thrusters. Thus, the entire system promises to have both stability and controllability in its rotary-wing-mode. Control in airplane-mode, however, is quite different.
Embodiment 3100 is pictured in airplane-mode configuration in
Additional features 3150 and 3152 are seen extending from the trailing edges of the left and right wings in embodiment 3100. These features, as drawn, are too small to be very effective ailerons and certainly too small to be elevators. Indeed, they are aerodynamic pitch sensors. 3152 is shown in detail in view and subsystem 3200 of
A magnetic sensor package 3260, e.g. a Hall effect device, responds to this magnetic signal with a signal output 3261. Advantageously in a Hall sensor, the sensing bridge component inside 3260 may be oriented in a vertical plane, such that when 3152 is angled at the center of its range, the permanent-magnet field will be parallel to the plane of the sensing bridge and produce a null output. Rotation of 3152 and the magnet will cause the magnetic field to pass through the bridge in one direction for pitch-up and the opposite direction for pitch-down, yielding a continuous analog signal that provides a good indication of aerodynamic pitch. When the wing is pitched up, the natural streamlines past the airfoil will bend upward going back past the trailing edge. This bend in the flow will rotate 3152 pitch-down, i.e. trailing edge up. An opposite wing pitch will give an opposite angle response in 3152 and an opposite-polarity output along 3261. A comparable pitch-sensing signal from sensor 3150 on the opposite wing is indicated along 3262.
The use of servo control of geometric pitch to control the lift coefficient CL or the equivalent aerodynamic angle of attack has been discussed extensively above. Diagram 3270 of
The pitch control system of 3100 certainly requires such compensations. In the discussion of embodiment 3000 and its diagram, pressure sensor apertures 3050, 3054, 3058, 3051, 3055 and 3059 were discussed in relation to lift coefficient computations from pressure signals. Indeed, the kind of computation taking place in computation subsystem 3270 is very general and can use inputs 3261 and 3262 from a variety of types of pitch sensors, including the pressure sensor system just mentioned. As indicated in
Direct trailing-edge measurement of wind flow is particularly sensitive to stall. Thus, in the event of wing stall, sensor aerodynamic surface 3152 will give an extreme angle response with fluctuations responding to turbulence. Even with servo control to avoid stall, air turbulence and gusts can induce stall, which needs to be sensed with corrective response. An alternative trailing edge flow sensor system 3300 is now discussed with reference to
A limitation of the system with just the “essential” single photosensor relates to the stiffness of sheet 3352 and to a potential stability problem. If it is too flexible, 3352 might chatter in turbulence from various sources including from instability in its own boundary layer. Stiffness in the sheet, however, can cause the sheet to significantly resist bending, giving reduced pitch-angle sensitivity at low indicated airspeeds. For example, if the sheet-absent unperturbed air flow pattern is uniformly curved in the sheet region, then with a stiff sheet the cumulative bending moment from pressure differential across the sheet will be highest at the sheet attachment in front and will go to zero at the sheet trailing edge, causing the curvature to decrease going from attachment to trailing edge. As airspeed increases with the same wing lift coefficient and the same unperturbed air flow pattern, the flow will act more strongly to bend the entire sheet toward the unperturbed flow contour, giving a more uniform curvature. When computation module 3372 uses both inputs 3361 and 3371 from respective sensors 3360 and 3370, then the two optical signals can be interpreted in combination to infer both the lift coefficient “CL” and a reference Bernoulli pressure or corresponding indicated airspeed “U”. As has been discussed above, the controller for the aircraft of the present invention uses measures of airspeed to decide, for example, when to transition from wing CL control to pitch control for directing thruster force.
Returning attention to lift coefficient sensor system 3200, an option is provided there to measure a combination of speed and “stiffness” of the air flow over surface 3152, leading to a measurement of Bernoulli pressure or corresponding indicated airspeed “U”. A test current applied to electrical coil 3245 generates a magnetic field passing through both field sensor 3260 and magnet 3230, the field being directed roughly at right angles to the poling of the magnet. This field produces torsion in the magnet, which in turn perturbs the angle of 3152. It is possible to calibrate the direct effect of the coil field on sensor 3260 and subtract that from signal 3261, so that the resultant signal is sensitive only to the angular response of 3152 to the magnetic torsion. The torsional stiffness exhibited by 3152 in resisting the known perturbing magnetic torsion provides a good measure of Bernoulli pressure. Alternatively, the magnetic perturbation can be pulsed, with dynamic variation in signal 3261 indicating the speed of angle recovery, overshoot and possible damped oscillation responding to pulses. Even a mechanical response spectrum of amplitude and/or phases versus frequency can be measured. It will be seen that multiple measurement variations can be used to determine the important pair of parameters “CL” and “U” or their correlates such as wing geometric attack angle and Bernoulli pressure.
Application of the just-discussed measurements in larger flight-control contexts includes the procedural steps listed in
From the many design options and variations presented here, falling within the scope of the present invention, one must not lose sight of the fundamental novelty. That is, an aircraft with two wings, thrusters such as powered props or ducted turbines on each wing, and independent pitch control for each wing through large angles, can be configured to operate as a helicopter and as a fixed wing aircraft, having the VTOL advantage of its helicopter mode and the efficient high-speed flight advantage of its fixed wing mode, and being able to undergo in-flight conversions in either direction between the two modes, combining those two major advantages.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US2015/067630 | 12/28/2015 | WO | 00 |
| Publishing Document | Publishing Date | Country | Kind |
|---|---|---|---|
| WO2016/109408 | 7/7/2016 | WO | A |
| Number | Name | Date | Kind |
|---|---|---|---|
| 1592740 | MacNeil | Jul 1926 | A |
| 1754910 | Serna | Apr 1930 | A |
| 2479125 | Leonard | Aug 1949 | A |
| 3035789 | Young | May 1962 | A |
| 3116040 | Petrides | Dec 1963 | A |
| 3141633 | MacKay | Jul 1964 | A |
| 3246861 | Curci | Apr 1966 | A |
| 3900176 | Everett | Aug 1975 | A |
| 5297759 | Tibor | Mar 1994 | A |
| 5516060 | McDonnell | May 1996 | A |
| 6471158 | Davis | Oct 2002 | B1 |
| 7093788 | Small | Aug 2006 | B2 |
| 8991751 | Page | Mar 2015 | B2 |
| 9567075 | Tighe | Feb 2017 | B2 |
| 9764828 | Ulrich | Sep 2017 | B2 |
| 20070187547 | Kelly | Aug 2007 | A1 |
| 20090032638 | Zhao et al. | Feb 2009 | A1 |
| 20090045294 | Richardson et al. | Feb 2009 | A1 |
| 20090206208 | Kennedy et al. | Aug 2009 | A1 |
| 20100230547 | Tayman | Sep 2010 | A1 |
| 20110315809 | Oliver | Dec 2011 | A1 |
| 20120248259 | Page et al. | Oct 2012 | A1 |
| 20140008498 | Reiter | Jan 2014 | A1 |
| 20150158581 | Fengler | Jun 2015 | A1 |
| 20170283052 | Moshe | Oct 2017 | A1 |
| 20180334241 | Long | Nov 2018 | A1 |
| Number | Date | Country |
|---|---|---|
| 1335248 | Feb 2002 | CN |
| 2060487 | May 2009 | EP |
| WO-9910235 | Mar 1999 | WO |
| 2011096850 | Aug 2011 | WO |
| 2015089679 | Jun 2015 | WO |
| WO-2015089679 | Jun 2015 | WO |
| Entry |
|---|
| International Preliminary Report on Patentability in corresponding PCT application serial No. PCT/US2015/067630, dated Aug. 4, 2017, 39 pp. |
| Examination report in corresponding Australia patent application serial No. 2015374294, dated Mar. 26, 2019, 7 pp. |
| Extended search report in corresponding European patent application serial No. 15876091.8, dated Nov. 13, 2018, 9 pp. |
| International Preliminary Report on Patentability in corresponding PCT application No. PCT/US2015/067630, Aug. 4, 2017, 39 pages. |
| Transmittal regarding foreign literature, dated Aug. 17, 2017, 1 page. |
| International Search Report and Written Opinion in corresponding PCT application No. PCT/US2015/067630, May 31, 2016, 10 pages. |
| Horansky-Mckinney, Amanda, Stop-Rotor Rotary Wing Aircraft, Navel Research Laboratory, located on www.techbriefs.com, Jul. 2017, 2 pages. |
| Number | Date | Country | |
|---|---|---|---|
| 20180370624 A1 | Dec 2018 | US |
| Number | Date | Country | |
|---|---|---|---|
| 62099477 | Jan 2015 | US |
