![Design and Development of a Hybrid UAV
x | P a g e
ME5308 – Major Group Project
List of Figure
Figure 1 The IAI Mini Panther in level cruise flight [1] . .............................................. 4
Figure 2 Sitter type UAV the V-Bat from MLB [4]. ...................................................... 5
Figure 3 The Orbis from Santos Labs in hover [6]...................................................... 5
Figure 4 The Latitude Engineering HQ hybrid Prototype [8]....................................... 6
Figure 5 The Wingcopter V13CH in VTOL mode. ...................................................... 7
Figure 6 Advanced VTOL Technologies' Hammerhead [10] ...................................... 7
Figure 7 Bell Eagle Eye Tiltrotor UAV [11]. ................................................................ 8
Figure 8 QTW-UAV developed by Chiba University, Japan [14]. ............................... 8
Figure 9 Graph showing basic relation between small UAVs..................................... 9
Figure 10 Maximum Endurance vs. Maximum Take-off weight for a range of UAVs
[19] ........................................................................................................................... 11
Figure 11: The STOL mission profile........................................................................ 14
Figure 12: The VTOL mission profile........................................................................ 15
Figure 13 approximate sketch of the concept idea................................................... 18
Figure 14 Individual Proposal Concept by Bennie Mwiinga...................................... 21
Figure 15 The Doak VZ-4 by Doak Aircraft Company [24]. ...................................... 21
Figure 16 Blown Flight Control Concept................................................................... 23
Figure 17Tube and Tray Fuselage Concept............................................................. 25
Figure 18 Tube and Tray Fuselage as Used by Hobby Flyers ................................. 26
Figure 19 Concept Sketch of an initial idea.............................................................. 29
Figure 20 Concept Design Sketch............................................................................ 33
Figure 21 A quality function deployment (QFD) Matrix............................................. 35
Figure 22 Flow Diagram showing Design Stages..................................................... 42
Figure 23 Force Diagrams: a) Forces on a climbing aircraft, b) Forces on aircraft at
constant bank angle [27]. ......................................................................................... 44
Figure 24 Constraint Analysis for the UAV............................................................... 46
Figure 25 Wing Geometry, Note: dimensions in millimetres..................................... 50
Figure 26 Induced Drag Factor Vs. Taper and Aspect Ratio [28]............................. 51
Figure 27 Typical Cambered Aerofoil [31]................................................................ 52
Figure 28 Lift Coefficient Vs. Moment Coefficient Analysis of different profiles........ 53
Figure 29 Initial and Final profile Comparison. ......................................................... 54](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-11-320.jpg)
![Design and Development of a Hybrid UAV
xi | P a g e
ME5308 – Major Group Project
Figure 30 XFLR results for the final wing configuration. (a) Moment Force and chord
wise lift distribution. (b) Spanwise lift distribution. (c) ISO view of lift and lift
distribution................................................................................................................ 55
Figure 31 Polars: (a) Variation of Lift coefficient with AoA. (b) Variation of the drag
coefficient with lift coefficient. (c) Variation of lift to drag ratio with AoA. .................. 56
Figure 32: Tail design procedure as illustrated by Mohammad Sadraey. [40].......... 59
Figure 33 Total Drag Decomposition........................................................................ 63
Figure 34 Drag velocity curve................................................................................... 66
Figure 35 CD Vs. CL Polar for the wing and the aircraft........................................... 68
Figure 36 Comparison of the Lift Curve Slopes using different predicting methods:
Online database, XFLR5 and ESDU sheets............................................................. 74
Figure 37 How to obtain Trailing Edge Angle ...................................................... 76
Figure 38 Wing Curve slopes with control surface deflections. ................................ 77
Figure 39 Force balance kit to acquire aircraft CG location...................................... 79
Figure 40 Front load with dual dead weight batteries and back-up 4000 mah main
battery ...................................................................................................................... 80
Figure 41: Tail incidence angle vs. Moments generated. ......................................... 82
Figure 42: Graphs indicating the derivatives and for stable and instable
aircraft conditions. .................................................................................................... 84
Figure 43 Wing and tail forces.................................................................................. 86
Figure 44 Statically Stable and Unstable pitching moment curves........................... 87
Figure 45 Final aircraft CG Lift configuration............................................................ 88
Figure 46: control surface effectiveness parameter vs. control surface to lifting
surface chord ratio. [40]............................................................................................ 92
Figure 47 Shows the rudder curve slope with deflection angles of ±20 degrees...... 97
Figure 48 Shows the elevator curve slope with deflection angles of ±20 degrees. .. 97
Figure 49 Longitudinal CG Envelope for Project vehicle .......................................... 98
Figure 50 Tri-copter configuration with reference axes. ......................................... 101
Figure 51 Pitch up by using Rotor 1. ...................................................................... 102
Figure 52 Roll in the Clockwise direction................................................................ 102
Figure 53 Roll in the Counter Clockwise direction.................................................. 102
Figure 54 Yaw authority of a tri-copter. .................................................................. 103
Figure 55 Mass Flow of air through rotor in hover.................................................. 104
Figure 56 Altitude Hold (Hover) with all 3 rotors..................................................... 106](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-12-320.jpg)
![Design and Development of a Hybrid UAV
xii | P a g e
ME5308 – Major Group Project
Figure 57 Mass Flow of air through rotor in vertical climb. ..................................... 107
Figure 58 A level vertical climb by the tri-copter..................................................... 108
Figure 59 Flow of air through the rotor in forward flight.......................................... 109
Figure 60 Rotor Disc showing Azimuth angle......................................................... 110
Figure 61 Full model of UAV at a hover. ................................................................ 111
Figure 62 Full model of UAV in transition............................................................... 111
Figure 63 UAV model in full horizontal flight........................................................... 112
Figure 64 Monocoque fuselage design [61] ........................................................... 114
Figure 65 Truss fuselage structure [32].................................................................. 114
Figure 66 Semi-monocoque Fuselage [32] ............................................................ 115
Figure 67 Global Hawk Cutaway [64]..................................................................... 115
Figure 68 Falco Cutaway diagram [64]................................................................... 116
Figure 69 Cutaway of the ScanEagle [64].............................................................. 116
Figure 70 Bonding in progress of the Demon UAV composite structure [65] ......... 117
Figure 71 Loading on a triangular structure [68]..................................................... 118
Figure 72Skeletal frame of the fuselage................................................................. 119
Figure 73 Landing Gear Positioning for Proper Weight Distribution [71] ................ 121
Figure 74 Moveable Landing Gear Concept........................................................... 122
Figure 75 ABS Landing Gear Mount - Broken During Aircraft Assembly................ 123
Figure 76 ANSYS principle stress analysis on bulkhead displaying key on the left 124
Figure 77 demonstration of typical wing structure [75] ........................................... 127
Figure 78 Single Spar Wing Connection ................................................................ 131
Figure 79 Double Spar Wing Connection............................................................... 132
Figure 80 Moveable Landing Gear Mount.............................................................. 133
Figure 81 Computational Stress Test Result for Basic Landing Gear Mount ......... 134
Figure 82 Computational Stress Test Result for Lightweight Landing Gear Mount 134
Figure 83 Nose Vertical Lift Fan Skeletal Structure................................................ 135
Figure 84 Detail View of the Tongue and Groove Assembly Method ..................... 135
Figure 85 Initial fuselage concept........................................................................... 136
Figure 86 First Full Group CAD Aircraft Design ..................................................... 136
Figure 87 Structure and connections of various components within the aircraft..... 137
Figure 88 Top, front and side views of the final CAD model................................... 138
Figure 89 T/W vs Maximum amp draw................................................................... 140
Figure 90 T/W vs EDF price ................................................................................... 141](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-13-320.jpg)
![Design and Development of a Hybrid UAV
xiii | P a g e
ME5308 – Major Group Project
Figure 91 EDF unit weight vs Thrust Capability ..................................................... 141
Figure 92 STOL mission current comparison for the initial and final endurance
calculations ............................................................................................................ 148
Figure 93 VTOL mission current draw comparison for the initial and final endurance
calculations. ........................................................................................................... 148
Figure 94 General control scheme of the UAV [87]................................................ 151
Figure 95 ArduPilot Mega 2.6 from 3D Robotics .................................................... 152
Figure 96 Schematic of MPU-6000. ....................................................................... 152
Figure 97 Block diagram of tri-copter control include 2 gain values [57]. ............... 153
Figure 98 Control allocation by a controller on a tri-copter..................................... 153
Figure 99 Example of Cascade Control.................................................................. 154
Figure 100 Cascaded PID used by APM [88]......................................................... 154
Figure 101 MaxBotix XL MaxSonarEZL0. .............................................................. 155
Figure 102 Sonar EM Noise reduction modification. .............................................. 156
Figure 103 APM 2.6 anatomy................................................................................. 156
Figure 104 Phases of flight during the transition maneuver from hover to horizontal
flight........................................................................................................................ 158
Figure 105 Cantilever Load Testing Arrangement.................................................. 160
Figure 106 Cantilever Physical Stress Test Results Graph.................................... 161
Figure 107 Three Point Physical Stress Test Results Graph ................................. 162
Figure 108 Three Point Physical Stress Test Results Graph ................................. 163
Figure 109 demonstration of carbon fiber rod deflection with cantilever point loading
............................................................................................................................... 165
Figure 110: Experimental setup of the test conducted (left) and a drawing of the
component (right) ................................................................................................... 167
Figure 111: Load vs Tensile extension for the 10mm diameter hole ...................... 168
Figure 112: Load vs Tensile extension for the 20mm diameter hole. ..................... 169
Figure 113 Thrust bench and NI High USB carrier used for motor characterisation.
............................................................................................................................... 171
Figure 114 EDF Mount for thrust bench. ................................................................ 171
Figure 115 Motor Mount for thrust bench. .............................................................. 171
Figure 116 80A ESC Turnigy Superbrain............................................................... 172
Figure 117 Turnigy KV-RPM Meter. ....................................................................... 172
Figure 118 National Instruments Hi-Speed USB Carrier. ....................................... 172](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-14-320.jpg)
![Design and Development of a Hybrid UAV
xiv | P a g e
ME5308 – Major Group Project
Figure 119 Turnigy 4000 mAh LiPO Battery (6s). .................................................. 172
Figure 120 PWM changing the angle of a dc motor [95]. ....................................... 173
Figure 121 Sample calibration curve for the test bench. ........................................ 174
Figure 122 Numeric Loading for EDF and trend line. ............................................. 177
Figure 123 Thrust Results for the EDF................................................................... 178
Figure 124 Thrust efficiency of two and three bladed propellers [96]. .................... 179
Figure 125 VTOL Motor test with different propellers............................................. 180
Figure 126 Current Draw of the motor for any given thrust. ................................... 181
Figure 127 Laser cutting the aft EDF bulkhead ...................................................... 187
Figure 128 Fuselage during initial Epoxy resin stage of construction (left), tilting
Propeller mount (Right) .......................................................................................... 188
Figure 129 rear view of the front Bulkhead displaying the nose gear mechanism.. 189
Figure 130 drilling axle holes on the non-vertical mounting plate of the carbon fiber
Landing gear .......................................................................................................... 191
Figure 131 Rear landing gear assembly................................................................. 192
Figure 132 Fuselage structure with back-up rear undercarriage (left), Nose gear
(right)...................................................................................................................... 192
Figure 133: Schematic of assembly of the aluminum VTOL motor mounts............ 193
Figure 134: Load tests conducted on the P400 ABS plastic (left) and the 3mm (right)
plywood motor mounts. .......................................................................................... 196
Figure 135 EDF Mount to the fuselage, Side view (left), top view (right)................ 197
Figure 136 Reinforced rear landing gear mount..................................................... 203
Figure 137 Strengthened Retro-fit Nose Landing Gear.......................................... 204
Figure 138 Second flight test ground roll demonstration ........................................ 205
Figure 139 Second flight test tip stall demonstration.............................................. 206
Figure 140 Second flight test landing stall demonstration ...................................... 207
Figure 141 UAV in Tri-copter mode........................................................................ 209
Figure 142 V-n Diagram and Gust Loading graph.................................................. 211
Figure 144 Roskam Constraint Analysis ................................................................ 228
Figure 145 To obtain for Step 4 in Table 17 [49]....................................... 231
Figure 146 To obtain for Step 5 in Table 17 [49].................................. 231](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-15-320.jpg)
![Design and Development of a Hybrid UAV
xv | P a g e
ME5308 – Major Group Project
List of Tables
Table 1: Sketch of concept design idea.................................................................... 16
Table 2: Key parameters of individual concept design ............................................. 17
Table 3 Key parameters of individual proposal ........................................................ 18
Table 4 Individual proposal by Bennie Mwiinga ....................................................... 20
Table 5 Individual Concept Design........................................................................... 28
Table 6 Individual Concept #4.................................................................................. 31
Table 7 Typical Aircraft Parameters. [26] ................................................................. 33
Table 8 House of Quality table, How’s vs How’s ...................................................... 36
Table 9 Group Concepts .......................................................................................... 38
Table 10 Constraint Analysis Equations, obtained from Mattingly et All [27]............ 45
Table 11 Constraint Analysis Parameters. ............................................................... 46
Table 12 Different Wing Geometry Design Aspects ................................................. 49
Table 13 Wing Geometry Parameters...................................................................... 50
Table 14: Effects of changes in tail volume coefficients ........................................... 61
Table 15 Drag Components of the aircraft for cruise, 22.2 m/s. ............................... 69
Table 16: Time to achieve specific bank angles...................................................... 71
Table 17 Process to attain the lift curve slopes of the wing and the deflected control
surface. ..................................................................................................................... 75
Table 18 Parameters and Results............................................................................ 76
Table 19 Lift variation with control surface deflection............................................... 77
Table 20: Horizontal and vertical tail design details.................................................. 82
Table 21: Static and dynamic stability requirements. [40] ........................................ 83
Table 22: Methods of determining the location of neutral point [38] [53] .................. 85
Table 23: Control Surface Functions........................................................................ 89
Table 24: Rudder deflection required during various landing at various crosswind
velocities. ................................................................................................................. 95
Table 25 Rudder and elevator curve slope results using ESDU method, to be used in
the control surface sizing.......................................................................................... 96
Table 26 showing properties of similar thickness plywood material strength [72] .. 125
Table 27 properties comparison of foam core wing reinforced with carbon fibre spars
to balsawood ribbed structure reinforced with carbon fibre spars [77] ................... 128
Table 28 Battery properties for a suitable range of products [86]........................... 146](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-16-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
4 | P a g e
ME5308 – Major Group Project
2. Literature Review
2.1. State of The Art Technology
The idea of an aerial vehicle that can perform both VTOL like a helicopter as well as
STOL like a fixed wing aircraft is not a new one. However limitations in technology as
well as the complexity of the associated control system has prevented widespread
development of these types of vehicles especially those with an autonomous nature.
Recently a handful of fully autonomous hybrids have been unveiled some are still
prototypes while others are fully functioning production models. Below are some
examples of these state of the art V/STOL aircraft. Being able to hybridize rotorcraft
and fixed wing aircraft provides the opportunity for further applications of UAV/S in
roles that would normally be exclusive to either one or the other.
The mini panther is a smaller version of its larger relative ‘The Panther’ and weighs
12 kg (shown in figure 1). It is however the newest Iteration in the family to date. It
has 3 propeller motors, the front two motors tilt upwards. In conjunction with the aft
propeller; that is permanently in the normal position relative to the aircraft, the mini-
panther is able to perform vertical take-off and transition into cruise, as well as
transition from cruise to stable hover flight. The third motor on the aft section of the
fuselage acts as the third arm of a tri-copter when the front two are tilted, this allows
for yaw control in hover and vertical flight.
Figure 1 The IAI Mini Panther in level cruise flight [1] .
Another approach to producing a VTOL aircraft is the V-Bat from MLB (shown in
figure 2). It is an alternate solution to a VTOL UAV as it is designed as a sitter type
aircraft. As a sitter type aircraft the V-bat begins its flight in a vertical position on the
ground and then hovers to a predetermined altitude [2]. At said altitude it is able to](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-24-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
5 | P a g e
ME5308 – Major Group Project
transition autonomously from hover to cruise and vice versa. Developed with funding
from DARPA, the military version also includes a 6 foot extending arm to pick up
objects whilst at hover close to ground level [3]. This UAV is capable of flight up to
15,000 feet altitude, with a maximum endurance of 10 hours.
Figure 2 Sitter type UAV the V-Bat from MLB [4].
Similar to the MLB V-Bat are systems that only use a ducted body design (Figure 3).
These systems are sitter-type VTOL that also have a number of rotors (normally 4)
placed in an X or + formation similar to a quad rotor. The rotors and control surface
are all enclosed within a duct body. These duct type UAV are able to transition from
hover to horizontal flight by tilting themselves forward and increasing and vectoring
the thrust generated by one motor. An example of this design is the Santos Lab
Orbis which currently uses a hydrogen fuel cell and has a span of 3.8m [5].
Figure 3 The Orbis from Santos Labs in hover [6]
It is well known that rotor craft are able to perform VTOL in the most efficient
manner. By hybridizing a rotor craft with a fixed wing craft the benefits of both types
of vehicles can be maintained. Latitude Engineering, a small drone company from](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-25-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
6 | P a g e
ME5308 – Major Group Project
Tucson, Arizona USA has developed such a hybrid [7]. Calling it the Hybrid Quad
rotor (HQ) (shown in Figure 4 below) it is simply a hybrid between a quad rotor and a
fixed wing aircraft. It weighs 27 Kg and has 4 electric motors that it uses to hover and
1 gas powered motor mounted on the aft of the aircraft to provide thrust for forward
flight. The HQ is still in development but Latitude Engineering has been able to
maintain a hover and transition to forward flight.
Figure 4 The Latitude Engineering HQ hybrid Prototype [8].
Another project that has taken the same approach as Latitude Engineering is the
Wing copter V13CH project by Jonathan Hesselbarth. The Wing copter (shown in
Figure 5) also hybridizes a fixed wing and a quad rotor, however, the Wing copter
utilizes its 4 electric brushless motors to produce thrust in forward flight and in VTOL.
Utilizing a set of swivel arms; on which the four motors are mounted, the Wing copter
is able to perform transition from hover to horizontal forward flight and vice versa.
This a novel solution that also requires a control system that can keep the aircraft
stable while transition is being performed. The transition performed by the
Wingcopter however not an automated one is and is instead manually controlled
(with some assistance from on board flight control). By utilizing the rotary heads on a
radio control transmitter, the controller is able to vary the tilt of the four motors and
transition into horizontal flight.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-26-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
7 | P a g e
ME5308 – Major Group Project
Figure 5 The Wingcopter V13CH in VTOL mode.
Similarly employing a tilting mechanism to achieve VTOL is the Hammerhead
developed by David Howe & Lyndon Caine of Advanced VTOL Technologies [9]
(shown in Figure 6). It has a canard that helps improve the UAVs stall characteristics
as well as the ability to thrust vector in order to limit pitch divergence [9].The
hammerhead employs twin counter rotating electric rotors stationed on a tilting stub
wing assembly which AVT claims “minimises pitch, roll and yaw coupling” [9]. The
hammerhead is capable of performing either STOL or VTOL. In STOL mode the tilt
wing stub is positioned such that the rotors produce a thrust propelling the UAV
forward. In VTOL mode the tilt wing stub is positioned vertically allowing the
hammerhead to take off and hover like a helicopter.
Figure 6 Advanced VTOL Technologies' Hammerhead [10]
Another example of hybrid VTOL design is the Bell Eagle Eye UAV (see Figure 7).
Developed by Bell Helicopter – Textron, Texas USA. It is a Tilt rotor UAV capable of
VTOL which it achieves by tilting its nacelles in the appropriate direction to either
perform VTOL or STOL. The Bell Eagle Eye has a payload weight of 90kg, a
maximum speed of 200kt and an endurance of 8 hours [11] [12]. It also utilizes an
automated flight control system to assist in performing transition.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-27-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
8 | P a g e
ME5308 – Major Group Project
Figure 7 Bell Eagle Eye Tiltrotor UAV [11].
Use of tilting mechanisms can also be found in other hybrid VTOL aircraft such as
the QTW-UAV by Chiba University, Japan (see Figure 8). The QTW-UAV uses 4
rotors placed on tilting wings that swivel to the vertical position to gain altitude and
then swivel forward to transition to horizontal flight. It utilizes 4 electric rotors giving it
a payload weight of 5kg, endurance of 15 min and a max speed of 81kt [13].
Figure 8 QTW-UAV developed by Chiba University, Japan [14].](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-28-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
9 | P a g e
ME5308 – Major Group Project
2.2. Market Analysis
In order to design a vehicle fit for purpose, market research was undertaken to find a
suitable starting point to work to. Typical dimensions and functionalities of existing
real world UAVs were used to compare the sizes the aircraft under development
should be close to. Below is a diagram showing a triangular relationship between
wing span, total length, and Max Take-off weight with a logarithmic scale.
Figure 9 Graph showing basic relation between small UAVs
AV RQ 11 B Raven [15] Bayraktar Mini UAS [16]](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-29-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
10 | P a g e
ME5308 – Major Group Project
AV Wasp III [17] Innocon Micro Falcon [18]
The graph on Figure 9 Graph showing basic relation between small UAVs represents
information gathered on similar sized small UAV platforms. This was used during the
early stages of initial geometric sizing to ensure the values that were being
calculated for the aircraft were within a set industry trend. In essence it was an early
rudimentary matching plot to define design tendencies for wingspan, vehicle length,
and Maximum Take-Off weight. Both the first converged group concept as well as
the current project aircraft design are listed on the plot. An important performance
parameter to note is that the cruise velocities were not included in analysis as values
could not be found for all vehicles looked at. However when averaged out along
other smaller UAVs, typical cruise velocities were around 20 .
Once the vehicle size had been constricted to a hypothetical box, performance
characteristics were then researched for a broad range of UAVs. Figure 10below
was sourced out from a thesis from MIT, displaying the relationship between
Maximum take-off weight and endurance in hours for a wide variety of UAVs. The
theoretical project aircraft would lie between 1 kg to 10 Kg along the bottom axis.
This range on the plot has a trend of a maximum endurance of 1-2 hours.
Considering the vehicle in question would be a hybrid, the added dead weights, the
use of all-electronic propulsion as well as a vertical flight mission segment which
would drain a higher amount than the usual battery draw during straight level un-
accelerated flight would all impact this maximum endurance value. As a result it
would be expected to be significantly lower in reality.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-30-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
11 | P a g e
ME5308 – Major Group Project
Figure 10 Maximum Endurance vs. Maximum Take-off weight for a range of UAVs [19]
This market research of current UAVs in service was invaluable in producing
information to use as a guideline to the aircraft design process. After every main
phase was completed values of the aircrafts performance, and sizing was checked
with current vehicles such as these. From this initial selection of fixed wing UAVs,
the variety was narrowed down to include specialized aircraft which were applicable
to similar mission profiles, which is to say a fixed wing UAV that has VTOL
functionality.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-31-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
12 | P a g e
ME5308 – Major Group Project
3. Requirements
3.1. Regulations
Every country has its own aviation regulations. The Civil Aviation (CAA) Authority
states that any UAV exceeding a weight of 7 kg need to be certified or approved.
Due to this, the main priority of the aircraft was its weight. It was decided that the
aircraft’s maximum take-off weight was kept under 6.5 kg and this was ensured
throughout the design process.
However, the CAP 722 and CAP 393 Air Navigation Order states that aircraft that
weigh less than 7 kg should also follow some regulations depending on whether they
are being used for commercial purposes or not. The regulations for aircraft with a
mass of less than 7 kg states that the aircraft should abide by appropriate
operational constraints in order to ensure public safety. The regulations are based on
the flying operation being conducted and the potential risks to any third party.
General principles for UAV operations outside segregated airspace should follow an
approved “detect and avoid” system and avoid crowded areas. It is also important for
the aircraft to not fly beyond the visual line of sight. The CAP 722 also states that the
aircraft should be flown such that the pilot controlling it can take manual control at
any point of time and fly the aircraft out of danger. It also states that the aircraft
should not be flown above 400 feet at any point of time. Further details of regulations
applicable to this UAV can be found in the CAP 722 and CAP 393 of the Civil
Aviation Authority Air Navigation Order. These regulations were kept in mind
throughout the build of this UAV. [20] [21]](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-32-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
17 | P a g e
ME5308 – Major Group Project
and have a smaller wetted area which would result in drag reduction. The V-tail
configuration also uses fewer control surfaces compared to a conventional tail.
These control surfaces are called ruddervators and are a combination of rudder and
elevators. [22]
The use of winglets was also considered in the concept for a number of reasons.
Winglets are small wing-like lifting surfaces that are fitted at the tip of the wings for
the purpose of reducing the trailing-vortex drag. As a result, this would increase the
lift generated on the aircraft. [23]
The materials considered for this design were a combination of foam and carbon
composites (main airframe) which are both lightweight materials that can take high
loads.
Category Abbinaya T Jagannathan
Mission Type Reconnaissance/ Surveillance
Environment Outdoor
Design Type Modular
Modular Options Camera
TO (Take-Off Type) V/STOL
L (Landing) V/STOL
Powerplant 1× Push Propeller & 2 × Rotor Integrated in Wings
Wing Medium/High Fixed Wing
Tail V-Tail
Airframe Foam & Carbon Composite
Landing Gear Fixed
Endurance(Prospective) 60 min
Altitude 50-100 m
Glide Capability (Inc. Design) Yes
Radio Controlled Back up Yes
Autonomous Yes
Table 2: Key parameters of individual concept design](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-37-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
21 | P a g e
ME5308 – Major Group Project
Figure 14 Individual Proposal Concept by Bennie Mwiinga.
In order to achieve these requirements a V/STOL type system is proposed. V/STOL
would allow for the UAV to be deployed in any terrain or environment without the
requirement of having a prebuilt runway. The V/STOL system in this proposal is
achieved by having two ducted fans mounted on the wing tips that are able to tilt for
vertical and horizontal flight. This approach was taken by the Doak Aircraft Company
in developing their Doak VZ-4 (Figure 15) in 1958.
Figure 15 The Doak VZ-4 by Doak Aircraft Company [24].
An additional ducted fan is then placed on the aft of the aircraft to provide additional
forward thrust for STOL and transition to horizontal flight. The use of ducted fans
may seem to be the wrong choice due to the lower efficiency when compared to a
purely fan based propulsion. As can be seen in equation 4 Area (A) if increased will
increase the amount of force created. A prop or fan has the advantage that it can
have a wide area and move more mass of air and thus create more force. A ducted
fan however has a smaller area and to move the same amount of air as a prop or fan
it has to increase the rate at which it accelerates the air while moving a smaller
amount.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-41-320.jpg)
![Design and Development of a Hybrid UAV
Brett M.
23 | P a g e
ME5308 – Major Group Project
Brett McMahon
Figure 16 Blown Flight Control Concept
This design is inspired to some extent by the DEMON concept demonstrator [25]
developed by BAe Systems, Cranfield and other universities which used jets of air to
control the aircraft in flight. Using low profile wing tip fans, the aircraft will be able to
perform roll manoeuvres using pulses of air from the appropriate fan. Short take off
performance would be possible using both fans together to provide lift in addition to
that provided by the main wing. The aircraft would be propelled through the air using
a large pushing motor of sufficient power (with reserve) to achieve a desired flight
speed. The aircraft would be all electrically powered for relative simplicity and safety
compared to petrol powered motors and their required ancillaries. Two batteries are
In-wing motors,
provide short
take off and
airbourne roll
control
Camera
provides visual
target ‘hit’ data
Rod and slot
construction
method, removable
wings if needed
Internal components
laid out with respect
to a selected Centre
of Gravity position
Internal avionics components
mounted on aluminium cruciform,
outer surfaces of foam or vacuum
formed plastic shell pieces
Infra-red range
finder provides
altitude data to
flight controller](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-43-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
33 | P a g e
ME5308 – Major Group Project
Figure 20 Concept Design Sketch.
To achieve good aerodynamic performance and the main mission aim, short take
off/landing, the concept to should have a similar look to that of a glider with high
aspect ratio, minimal weight and streamlined. RC trainer aircraft were also taken into
consideration since they are supposed to be easy to handle, which would benefit the
autonomous nature of the aircraft.
Table 7 shows typical design aspects for a trainer aircraft:
Trainer Aircraft Glider
Wingspan (b) 152 cm 152 cm
AR 6-7 8-10
Overall Length 127 cm 102 cm
Wing Area (S) 0.4216 m2
0.323 m2
Flying Weight 1.81 Kg 0.454 Kg
Wing Loading (W/S) 59 N/m2
20 N/m2
Table 7 Typical Aircraft Parameters. [26]](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-53-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
44 | P a g e
ME5308 – Major Group Project
4.2.2. Aircraft Sizing: Constraint Analysis
The first and most important procedure in a new aircraft design is to perform a
constraint analysis and produce the constraint analysis graph. This graph will provide
essential information for the initial design, the two main values are: wing loading and
the thrust/weight ratio. With initial weight estimation, an accurate sizing of the wing
area and minimum thrust required can be obtained, knowing that they will meet all
performance requirements, limitations and regulations mentioned previously.
The general methodology to start the constraint analysis is by summarising the
forces, or components of the forces that arise in different attitudes of the aircraft.
Figure 23 shows the two generic free body diagrams were most the equations can
be derived from.
a)
b)
Figure 23 Force Diagrams: a) Forces on a climbing aircraft, b) Forces on aircraft at constant bank angle
[27].
The force equilibrium equation may be obtained from Figure 23; with further
assumptions at different aircraft attitudes the general equation may be simplified and
rearranged to make thrust to weight ratio the subject. Table 10 shows all the
equations used in the constraint analysis with their corresponding assumptions.
Attitude Assumptions Final Equation
Cruise](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-64-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
45 | P a g e
ME5308 – Major Group Project
Table 10 Constraint Analysis Equations, obtained from Mattingly et All [27].
The final parameters used in these equations are all defined in Table11.
Constraint Analysis Parameters Obtained from…
Take-Off Velocity (m/s) 16.82 Calculated in spread sheet
Cruise Velocity (m/s) 22.23
Stall Velocity (m/s) 14.02
Dynamic Pressure TO (qto)
(kg/ms2
)
173.38
Dynamic Pressure cruise
(qcruise) (kg/ms2
)
303.82
Nmax (g’s) 3 Performance Requirements
AR 8.85 Calculated in spread sheet
Oswald factor ( e ) 0.95 Calculated in wing geometry section.
Fig. 6 in reference [28].
LA ground roll (SLA) (m) 52 Calculated in spread sheet
TO ground roll (STO) (m) 42 Calculated in spread sheet
Maneuver/Turn
at Constant
Altitude
Take-Off
Given values of
Landing
Given values of
Constant Speed
Climb
TO Climb
Stall at Cruise
( )](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-65-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
47 | P a g e
ME5308 – Major Group Project
The initial values for the constraint analysis were obtained from Dr. J. Roskam and
D. P. Raymer [29, 30]. These estimates are meant to be for conceptual design of a
commercial airline aircraft, hence not being optimised for a UAV. These estimate and
assume some regulations and performance requirements that do not apply to the
design of this aircraft, and therefore can only be used as a rough guideline. The
initial constraint analysis, following Dr. J. Roskams methodology can be seen in
Figure 144 in Appendix A.
Table 11 shows the final parameters that were used in the constraint analysis shown
above. All of these parameters are linked via a Microsoft excel spread sheet that
updates the original estimated values automatically once the user has performed a
later step in the design of the aircraft. An example of this are the drag coefficients
and drag constants, these were updated once the drag prediction model was
completed on a later stage of the design process.
The constraint analysis delimits the range of thrust to weight ratios and wing loading
values. As shown in Figure 24 the design space is limited by different attitudes of
the aircraft. Wing loading is restricted by the stall speed and the landing
configuration (the main factor being the landing distance). The thrust to weight ratios
are only limited on the lower side of the design space by the climb attitude, take-off
distance, 3g manoeuvre and the cruise conditions.
The two optimal configurations or the aircraft are shown by the blue arrows on the
graph, these are the positions of lowest thrust to weight, and hence less engine
thrust, that the aircraft can have taking into consideration the performance limits and
regulations [31].
The initial aim of the UAV being designed is to be positioned on the right blue arrow,
were the final configuration is shown. This combination of T/W and W/S is better
than the other optimal design point because the aircraft has a larger wing loading, or
in simple terms a smaller wing with greater forces being excreted on it. Accounting
that modern materials have great yield strength and the estimated weight of the
aircraft should not exceed 6kg, a greater wing loading is readily achievable.
The final configuration of the aircraft has a wing loading of 130.2 N/m2
and a thrust to
weight ratio of 0.5.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-67-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
49 | P a g e
ME5308 – Major Group Project
Sweep, Twist, Dihedral/Anhedral and Aspect Ratio [29]. Some examples are shown
in Table 12 below.
Wing Characteristic Effect
Straight
Taper Affects the lift distribution,
tip loading, induced drag
and wing stalling
characteristics.
Sweep Delays the mach drag rise
at transonic speeds.
Twist Alters wing tip stall
characteristics. Most
common being “Washout”
to delay tip stall.
Dihedral Increase roll stability
Wing
Placement
Affects stability and internal
structural layout. [32]
Table 12 Different Wing Geometry Design Aspects
Finally, the Aspect Ratio (AR) defines the slenderness of the wing. The equation
below shows how it can be obtained. It has a great effect on the induced drag that
the wing experiences during flight.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-69-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
51 | P a g e
ME5308 – Major Group Project
depending on the wing characteristics; as described by M. Niƫă and D. Scholz [28]. A
lifting line theory based method has been used because the wing does not have
sweep or twist. The method uses the taper ratio and aspect ratio of the wing to
determine the induced drag factor ( ), which is introduced in the following equation
to obtain the Oswald efficiency factor:
Figure 26 shows the lifting line theory relationship to obtain the induced drag factor
( ). With the values from Table 13, , obtaining an Oswald efficiency factor of
.
Figure 26 Induced Drag Factor Vs. Taper and Aspect Ratio [28].
To further reduce the wing tip vortex detrimental effect on the wing, the induced drag
may be reduced by the addition of winglets. These act as a fence between the high
and low pressure at the wings tips reducing the magnitude of the vortices and the
downwash effect they have over the wing span. A simple way to understand the
effect they have on the induced drag would be by incrementing the span and hence
increasing the aspect ratio of the wing [33]. The induced drag equation in Figure 26
shows that aspect ratio is inversely proportional to the aspect ratio. The positive
effects of winglets on the current UAV design would be minor. The addition of
winglets would further complicate the manufacturing of the wings and have a weight](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-71-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
52 | P a g e
ME5308 – Major Group Project
penalty which could not be tolerated taking into account the, already high, weight
estimation.
With the wing design concluded the final lifting and pitching characteristics of the
wing may be determined, these will later on affect the tail size and its placement.
Aerofoil Selection
The Lift Coefficient is related to the profile section chosen. The profile section mainly
varies in relation to these parameters: Camber, thickness to chord (t/c) ratio, and
position of maximum thickness. The higher the camber and t/c ratio the higher the lift
will be [29]. Figure 27 shows the typical cambered aerofoil section.
Figure 27 Typical Cambered Aerofoil [31]
It was initially intended to have a high lift low Reynolds number aerofoil,
characterised by a greater camber. After some investigation, [34], the Selig S1223
was chosen with a corresponding cruise velocity of 13.9m/s, but was then discarded
due to the high moment coefficient that would require a relatively big tail and a long
tail arm.
Consequently an analysis of a variety of aerofoils was made to compare mainly the
lifting characteristics against the moment coefficients. The outcome pursued in the
analysis was to have the greatest lifting coefficient with the lowest moment possible
to reduce the tail size and arm; which would also reduce the structural weight of the
aircraft. Figure 28 below shows the negative correlation between Cl and Cm, which is
previously expected. Highlighted in red is the Selig 1223 aerofoil and in yellow the
final aerofoil, NACA 63(2)-A015. There is a very big difference in both the lifting
characteristics and the moment characteristics of these aerofoils; this comes at the
expense, looking back at the lift equation, of higher cruise velocity.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-72-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
53 | P a g e
ME5308 – Major Group Project
Figure 28 Lift Coefficient Vs. Moment Coefficient Analysis of different profiles.
Lowering the lift coefficient also increases the take-off speed of the aircraft in an
STOL (standard take-off and landing) mission. This would not represent a serious
problem for the design since additional high lift devices may be used at landing and
take-off. It was a general consensus of the team to use the wing control surfaces as
flaps and ailerons to provide extra lift in certain flight segments.
The UAV will spend most of its fly time cruising in a reconnaissance role or
surveillance mission. In order to improve its efficiency careful attention has to be set
on the wing incident angle. When analysing the variety of profiles, the cruise angle of
attack has been equated to the angle of attack where the wing generates the
minimum lift to drag ratio. Only the lift to drag of the wing has been taking into
consideration because the aircrafts minimum drag angle has been assumed to be
zero degrees. At this angle the aircraft has the minimum front cross sectional area
and hence produced the lowest pressure (form) drag with an unchangeable amount
of skin friction drag [31].
Lift to drag is a very important performance and efficiency indicator which dictates
other performance factors such as endurance or influences the take-off weight of the
aircraft. The higher the lift to drag the longer the aircraft will fly. This is demonstrated
in the Breguet range equation, where the rage is proportional to the lift to drag ratio.
In terms of an electric vehicle, at cruise the thrust is equivalent to the drag which is
Initial Profile
S1223
Final Profile
NACA63(2) A-015
-0.7
-0.6
-0.5
-0.4
-0.3
-0.2
-0.1
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
WingMomentCoefficientatCruiseCm
Lift Coefficient Cl
Lift Coefficeint Vs. Moment Coefficent at Cruise](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-73-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
54 | P a g e
ME5308 – Major Group Project
equal to the motor current draw and hence the battery size [35, 36]. These factors
have a heavy influence on the aircrafts payload capacity which can also be linked to
operational costs of the aircraft.
Figure 29 Initial and Final profile Comparison.
Figure 29 above shows a comparison of the initial and the final aerofoils. The
difference in lifting capability is clearly seen by the difference in camber and the
difference in moment of the wing is also shown by the S1223 highly asymmetrical
profile compared to the NACA 63(2)-A015. The NACA profile has a thickness to
chord ratio of 15% whilst the Selig has 12.1%, this would partially explain the lower
drag to lift ratio of the Selig aerofoil.
One of the main reasons why the S1223 was chosen at the beginning of the design
process was due to its high profile lift to drag ratio of almost 55, but when changed to
the NACA 63(2)-A015 this value was reduced to 46.The development of why the
aerofoil has changed is discussed further in the stability section.
The analysis of the aerofoils taken into consideration for the UAV is shown in
Appendix A Table 37 “Additional Aerofoil Analysis”. It is important to note that all
values have been obtained with the aid of XFLR5 using the viscous vortex lattice
method at an estimated aircraft velocity of 20m/s. Another method used to analyse
the wing properties was the panel method. Table 38 in Appendix A shows the
difference in values from VLM to Panel and there is a very small difference that does
not exceed 1.2% in any of the calculated values of lift, drag and moment coefficients.
VLM has been chosen because obtains converging results for a wider range of
angles of attack.
Results for the final wing and aerofoil configuration at an angle of attack of 4° (cruise
condition).](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-74-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
55 | P a g e
ME5308 – Major Group Project
XFLR5 Setup - Viscous, Vortex Lattice Method at 20m/s.
(a)
(b)
(c)
Figure 30 XFLR results for the final wing configuration. (a) Moment Force and chord wise lift distribution.
(b) Spanwise lift distribution. (c) ISO view of lift and lift distribution.
Figure 30 shows the lift distribution of the aerofoil in the chord wise and spanwise
directions. It may be seen that the spanwise distribution is not exactly uniform like
that on an elliptical wing, but there is some resemblance. The optimal taper ratio to
imitate an elliptical wing would be about 0.3 [35]. This has a very low amount of
induced drag but poor stalling characteristics and hence a large amount of washout
would be needed to delay tip stall. Initially a value of taper of 0.3 was chosen, but to
maintain the wing area obtained in the constraint analysis the root chord would have
to be enormous or the wing span would have to increase over the 2m limit set by the
group. To find the balance between washout (negative twist) and the lift distribution
characteristics the taper ratio has to be increased. The final value is 0.8 because no
twist is designed into the wing due to manufacturing purposes, and it is more
important to reduce the stalling characteristics rather than reduce the induced drag.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-75-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
57 | P a g e
ME5308 – Major Group Project
maximum camber, maximum thickness, Reynolds number, aerofoil roughness and
Mach number lift increase [36]. This is a very complex method hence the non-linear
region being ignored in the lift curve slope. Similar aerofoil wind tunnel data suggest
that the linear range ends around 10 degrees with the maximum lift coefficient being
at about 15 degrees angle of attack [37]. XFLR5 suggest similar values for the 2D
profile, but cannot predict the non-linear region in the 3D wing setup. It only
produces the curve slope seen above (a).
Figure 31 (b) is the drag polar of the wing. It may be observed that the drag
generated by the wing is very small, and will be seen in the drag section that the
contribution of the wing is very small relative to the whole aircraft. The maximum lift
to drag ratio may be obtained by drawing a line though the origin so that it becomes
a tangent to the polar. This obtains the smallest value of CD with the highest CL. A
more accurate way to determine the maximum lift to drag ratio is by obtaining graph
(c). It demonstrated that the wing incident angle has to be around degrees to
optimise the wings performance in cruise.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-77-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
58 | P a g e
ME5308 – Major Group Project
Tail Sizing
The solution to the empennage design was analysed from two perspectives; the tail
sizing required for the STOL mission and the tail requirements needed to be able to
achieve the VTOL mission. The tail sizing procedure used for the STOL mission is
the same as that used in conventional aircraft by using standard procedure which
are illustrated by Jan Roskam, Mohammad Sadraey and D. Raymer. [38] [39] [22].
Given below in figure 32 is the tail design procedure. The tail design procedure itself
is an iterative process.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-78-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
59 | P a g e
ME5308 – Major Group Project
Figure 32: Tail design procedure as illustrated by Mohammad Sadraey. [40]
The tail configuration
The current aircraft configuration is such that it is propelled forward by the EDF
installed at the rear of the fuselage. Since this is the case, designing a conventional
tail would have a lower effectiveness due to interference between the EDF flow and](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-79-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
61 | P a g e
ME5308 – Major Group Project
Table 14 below illustrates the effects of changes in tail volume coefficients in aircraft.
Volume
Coefficient
Aircraft Stability Structural Weight
High High High
Low Low Low
Table 14: Effects of changes in tail volume coefficients
Using a very low horizontal tail volume coefficient ( ) would make the aircraft’s the
pitch behaviour very sensitive to the CG location. This would imply poor gust
resistance and therefore result in difficult pitch control. When the vertical tail volume
coefficient ( ) is too low, the aircraft will tend to oscillate along the vertical axis
inducing a dutch roll. This would make the directional control of the aircraft more
difficult. [38].
The horizontal and vertical tail volume coefficients for the designed aircraft are 0.56
and 0.05 respectively.
Optimum tail arm and tail plan form area:
The tail arm and the tail area are the two basic parameters of the tail that are
correlated to each other. The lift generated by the tail is dictated by the tail area. The
tail arm works as an arm for the pitching moment of the tail. The moments generated
by the tail is calculated by multiplying its lift force with the tail arm. The tail arm can
either be long or short as long as it is balanced with a suitable tail area. The tail arm
used for this aircraft is 0.8 meter.
Aircraft trim plays a very important role in aircraft to operate safe flight. The aircraft
trim must be maintained about the lateral(x), longitudinal (y) and directional (z) axes.
An aircraft is considered to be at trim when the summation of the forces about all
three directions equals zero. The horizontal tail is responsible for maintaining
longitudinal trim and the vertical tail is responsible for maintaining directional trim.
Section 161 of PART 23 published by the Federal Aviation Regulations (FAR) states
the requirements of an aircrafts trimmed condition. The aircraft must maintain
longitudinal trim under conditions such as climb, cruise, descent and approach. The](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-81-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
63 | P a g e
ME5308 – Major Group Project
Drag
Drag is one of the major forces acting on the aircraft in flight. In an ideal case drag
would be equivalent to zero, but in reality air flowing around a body will always
create an opposing force. It is a key parameter that has subsequent impact in other
aircraft aspects. It is used in the aircraft design phase to determine the ceiling, thrust
needed and performance of the aircraft. Figure 33 shows the decomposition of the
total drag. The most important components are Lift Induced, Skin friction and Form
Drag [31] [41].
Figure 33 Total Drag Decomposition.
For the purpose of this aircraft, the wave drag may be ignored because no shocks
will arise at the maximum speed of the aircraft. Form the previous market analysis it
was determined that the cruise speed of the aircraft would be in a range of 15 to 20
m/s. At standard atmospheric conditions this results in a Mach number of 0.06, at
which compressibility effects, such as shock generation and wave drag, will not be
an issue and hence may be ignored. Interference Drag can also be ignored for the
same reasons. It is the result of the boundary layers or streamlines of different
components of the aircraft interfering with each other to produce high velocity and
hence a normal shock. It usually occurs at the joint of the fuselage and wing or the
horizontal and vertical stabilisers [31].
Total Drag
Profile Drag
Viscous Drag
Skin Friction Form Drag
Miscellaneous
(Wave,
Interference
etc.)
Lift Induced
Drag](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-83-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
64 | P a g e
ME5308 – Major Group Project
Miscellaneous refers to those small factors that at this stage of the project lack
importance and don’t affect the accuracy of the final drag estimation.
Profile Drag
The profile drag is a combination of the Skin Friction and Shape Factor (SF) of the
aircraft. It is one of the main constituents of the total drag of the aircraft, hence the
importance of an accurate estimation at this point. The minimum profile drag
coefficient can be estimated using below.
To calculate the minimum profile drag force we insert the profile drag coefficient into
a derivation of the lift/drag equation seen below:
Skin Friction Drag
This is the drag due to wall shear stresses; in basic terms, the force created by a
viscous flow over an object. A good approximation for this term is the 1/7 Power Law
derived by Theodore Van Karman [42]:](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-84-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
65 | P a g e
ME5308 – Major Group Project
Form Drag
This is drag dependent upon the shape of the body and the pressure differences
created from an incoming flow. It is determined with a parameter called Shape
Factor. There is two general equations that determine the SF, one for thin bodies
and another for bluffed bodies. A thin body is defined by the thickness to chord ( )
or diameter over length ( ) of the body being below 30% [43] [44]. These are the
equations used to determine the SF:
( ) ( )
Induced Drag
This component of drag arises from the pressure differences above and below the
wing. The pressure difference cause wing tip vortices, which induce a downwash on
the rest of the wing creating an adverse effect called, induced drag. This is directly](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-85-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
66 | P a g e
ME5308 – Major Group Project
related to the lift generated by the wing, its Aspect Ratio (AR) and shape of the wing.
Below is the equation to calculate induced drag [41] [31]:
Total Drag Vs. Velocity Curve
The Drag vs. Velocity curve can be drawn up with the aid of Microsoft excel to solve
the above equations for a variety of free stream velocities. All the parameters
required in the equations have been either inserted or calculated in the spread sheet,
under a different section. This back reference makes it very helpful to update all the
calculations when there are any speed changes or any component changes size,
etc. Figure 34 below shows the total drag vs. velocity curve:
Figure 34 Drag velocity curve.
y = 8E-05x4 - 0.0075x3 + 0.2583x2 - 3.7915x + 23.283
0.000
1.000
2.000
3.000
4.000
5.000
6.000
7.000
8.000
5.00 10.00 15.00 20.00 25.00 30.00
Drag(N)
Velocity (m/s)
Drag Vs. Velocity
Induced D
Total P D.
Total D.
Vstall
Poly. (Total D.)](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-86-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
67 | P a g e
ME5308 – Major Group Project
The blue curve represents the induced drag, the red curve is the profile drag and the
green line is the summation of both drags making the Total Drag. For optimal
performance the ideal cruise speed is that where the least total drag is obtained, for
our aircraft it should be around 15 m/s or slightly higher to maintain speed stability.
This speed is definitely within our stall speed, as shown by the vertical line light blue
line. The total drag curve has been fitted with a trend line and the corresponding
equation is shown on Figure 34. If the drag at a specific airspeed is required, replace
it in the x term in the equation to obtain the drag force at that speed.
An important factor is mentioned above, Speed Stability. Imagine if the aircraft is
flying to the right of the minimum drag speed at a set thrust and altitude. If the speed
of the aircraft is increased due to any sort of disturbance, the drag will increase and
slow down the aircraft. In the case where the airspeed is reduce by a disturbance;
the drag will decrease, speeding up the aircraft, giving it speed stability. Flying at a
lower speed than the minimum drag would result in a continuous increase of speed
due to a disturbance, which is undesirable. To maintain speed stability it is crucial to
fly above the minimum drag [45]. Hence the aircrafts cruise speed was initially
estimated to be around 20 m/s.
CD Vs. CL Polar
The same process was used to obtain Figure 35 shown below.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-87-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
68 | P a g e
ME5308 – Major Group Project
Figure 35 CD Vs. CL Polar for the wing and the aircraft.
In comparison to the drag polar of the wing this is an estimation of the corresponding
drag of the whole aircraft with respect the lift of the wing. The trend line on the
aircrafts drag polar in Figure 35 is second order meaning that the drag calculated
corresponds to the quadratic drag estimation method. This method has a highly
accuracy compared to wind tunnel data for lift coefficients in the range of ±1.2 units.
Depending on the position of the minimum drag, the curve will vary exponentially at
higher CL values [31].
Below Table 15 resuming the skin friction coefficient (Cf), induced drag coefficient
(CDI), profile drag coefficient (CDP) and the total drag of the aircraft with a final
refined cruise speed of 22.2 m/s. it shows the decomposition in terms of the following
aircraft components: Body, Wings, Empennage and Landing Gear. An assumption
made in the landing gear is that it is made of a cylinder and a straight flat plate,
corresponding to the wheels and the landing gear arms respectively.
y = 0.0181x2 + 0.0028x + 0.0129
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
0.16
0 0.5 1 1.5 2
DragCoefficeint(CD)
Lift Coefficeint (CL)
CD vs CL Polar
Aircraft
wing](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-88-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
69 | P a g e
ME5308 – Major Group Project
Note: The drag coefficients with a “ * ” before the name suggest an external value
has been used, [46].
Cruise Drag Components
Body Cf 0.0035
Body CDI 0
Body CDP 0.0084
Total Body Drag 0.0411
Wing Cf 0.0046
Wing CDI 0.0070
Wing CDP 0.0118
Total Wing Drag 3.2410
Empennage Cf 0.0052
Empennage CDI 0.0032
Empennage CDP 0.0127
Total Empennage Drag 0.9865
*LG Cylinder CD 1.05
LG Cylinder Drag 0.0320
*LG Flat Plate CD 1.27
LG Flat Plate Drag 3.5499
Total LG Drag 3.5820
Zero Lift Drag Coefficient (CD0) 0.0248
Total Profile Drag (CDP) 7.8506
Total Induced Drag (CDI) 1.0778
Total Drag 8.9533
Table 15 Drag Components of the aircraft for cruise, 22.2 m/s.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-89-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
70 | P a g e
ME5308 – Major Group Project
Control Surface Sizing: Wing
The UAV designed has ailerons and flaps on the wing elevator and rudder on the
empennage. This section will cover the design process for the control surfaces on
the wings.
For sake of simplicity in manufacture the ailerons and flaps were designed in
combination as flaperons. Flaperons are devices that can function as flaps as well as
ailerons.
Ailerons:
The design procedure of the control surfaces is an iterative process. Therefore an
excel sheet was formulated in order to calculate the required size of the ailerons.
First, the initial parameters such as chord, span and maximum deflection angles
were decided based on typical values that have been used for similar aircraft. Then,
step by step lists of calculations were carried out to determine the performance
characteristics of the ailerons to check whether or not they meet the requirements.
The aileron sizing method was done based on the approach used by Mohammad
Sadraey. The excel sheet was formulated such that the time taken to achieve a
desired bank angle is calculated. The results were matched with typical values of
time taken for similar category aircraft to see how they compare.
Given below is the steps used to estimate the responsiveness of the ailerons.
1. Estimation of initial geometry of aileron and maximum deflection angle.
2. Calculation of the aileron rolling moment coefficient, (1/rad):
[ ( )]
3. Calculation of the aircraft rolling moment coefficient, when aileron is deflected
with maximum deflection:
4. Determination of rolling moment, (Nm):](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-90-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
72 | P a g e
ME5308 – Major Group Project
Flaps:
Flaps are control surfaces mounted on a wing in order to increase the lift generated.
The use of high lift devices are beneficial especially in manoeuvres where the
aircraft’s velocity is reduced. Deflecting the flaps increases the camber of the wing
and therefore increase the lift generated by it. For this particular UAV designed, flaps
are to be retracted during take-off and landing.
The use of flaps during take-off will help the aircraft take-off at a shorter runway
distance. This feature will be advantageous especially when the runway surface has
a higher friction. Using flaps during landing will reduce the aircraft’s stall speed and
allow a slower and steeper approach. Flaps deflected at higher angles also increase
drag and this will help reduce the runway distance for landing.
The increment in lift coefficient at different flap deflection was obtained and checked
to determine whether it was suitable to fulfil the task [47] [40].](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-92-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
73 | P a g e
ME5308 – Major Group Project
Lift Curve Slope Numerical Prediction: Wing and Aileron/Flaps
The lift curve slope is the variation of the lift coefficient with the free stream incident
angle of the lifting surface [31]. This is strictly necessary in order to calculate the
effect that a deflected control surface has on the lift. There are a series of methods
for obtaining these values for specific control surface sizes and deflections; this
especially helpful in preliminary design of wings and control surfaces.
The first and most accurate method is to make a model of the wing section with the
control surface and test it in a wind tunnel. This is an extensive and costly process,
especially if the design needs to be optimised and a series of models have to be
made to obtain the correct control surface sizes.
When resources are limited there are other methods that will provide good
estimations. Vortex Lattice Method (VLM), used by XFLR5, can model the flow over
a profile and the wing obtaining the lift curve slopes. There are other available
resources such as online databases that provide the curve slopes of specific
aerofoils, but the reliability of the source may be questioned. Another numerical
method is using ESDU sheets (Engineering Science Data Unit), they provide
numerical solutions that interpolate, non-linearly, between existing experimental data
correlations for specific problem.
Figure 36 below shows a comparison of the different methods used to predict the lift
curve slopes for the wing design of the final UAV configuration. The first turquoise
line shows the 2D polar obtained from XFLR5 by simple anlayis of the profile. The
orange line is the 3D result obtained using lifting line theory in XFLR5. The light blue
line is the corresponiding 3D result for VLM and finaly the dark blue line represents
the ESDU three dimesional wing lift curve slope.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-93-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
74 | P a g e
ME5308 – Major Group Project
Figure 36 Comparison of the Lift Curve Slopes using different predicting methods: Online
database, XFLR5 and ESDU sheets.
Since ESDU is the most reliable resource compared to a free online program such
as XFLR5, then ESDU have been taken as the point of reference to obtain the lifting
characteristics of the wing and to know the effect that the control surfaces will have.
ESDU also provides the lowest gradient, hence taking it as a reference will
overestimate the control surface sizing rather than underestimate and then not being
able to fly adequately.
Some ESDU sheets used were W.01.01.05 [48] to calculate the lift curve slope of the
wing and C.01.01.03 [49] to calculate the lift curve slope of the deflected control
surface. Table 17 shows the steps that were taken to obtain the lift curve slopes for
the wing and the control surface (aileron or flaps), so that the sizing could be
completed. This is an iterative process between the initial control surface size, the
initial curve slope and a repetition until the final corresponding values are obtained.
Step Goal Additional Notes
1 Wing: Obtain the relationship
between the curve slope and the
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
-3 -1 1 3 5 7
LiftCoefficent
Angle of Attack α (degrees)
CL vs. Alpha
2D XFLR5
Polar
3D XFLR5
Polar
XFLR5 3D
Result (VLM)
CL Wing
(ESDU)](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-94-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
75 | P a g e
ME5308 – Major Group Project
theoretical curve slope, .
2 Wing: Obtain the theoretical slope
of lift for 2D aerofoil in
incompressible and inviscid flow,
.
3 Wing: Obtain the 2D
incompressible flow curve slope,
.
4 Control Surface: Obtain the
theoretical slope of lift for 2D
control surface in incompressible
and inviscid flow .
From Figure 1 in ESDU 01.01.03 or
Figure 145 in Appendix A.
5 Control Surface: Obtain the
relationship between the curve
slope and the theoretical curve
slope, .
From Figure 2 in ESDU 01.01.03 or
Figure 146 in Appendix A.
6 Control Surface: Obtain the 2D
incompressible flow curve slope,
.
7 Wing & Control Surface:
Extrapolate into 3D accounting for
compressibility effects.
√
8 Correct the angle of control
surface deflection [31].
9 Correct the zero-lift angle of
attack.
Should be equivalent to the 2D curve
slope zero-lift angle of attack.
10 Obtain the lift coefficient of the
wing and deflected control surface
by the following equation:
Table 17 Process to attain the lift curve slopes of the wing and the deflected control surface.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-95-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
76 | P a g e
ME5308 – Major Group Project
Parameters used for the curve slope prediction and Results
Key Parameters Curve Slopes Results
20 (a1)0/(a1)0T 0.669
Reynolds number 300000 (a1)0T 6.997 1/rad
X-t/c 0.75 (a1)0 4.679 1/rad
Thickness to chord ratio 0.15
Wing incidence angle
(degrees)
4 (a1)0 -3D 3.982 1/rad
Cruise Velocity (m/s) 22.200
Speed of Sound at Sea Level
(m/s)
340 (a2)0T 3.750 1/rad
Mach (M) 0.066 (a2)0/(a2)0T 0.335
AR 8.85 (a2)0 1.256 1/rad
e 0.95
(a2)0 -3D 1.202 1/rad
Correction Angle (degrees) Aileron Deflection Angle (degrees)
Delta 15 30 45
-1.23 Delta
Effective
11.25 22.5 33.75
Table 18 Parameters and Results.
Table 18 shows the all the parameters necessary to predict the theoretical curve
slope for the wing, and the corresponding relationship of . These values are
used in the equations for Steps 1 and 2 of the process.
is the trailing edge angle that is obtained from the 2D aerofoil profile. Figure 37
below describes how to obtain it. is the chord wise location were boundary layer
transition takes place. This value has been estimated to occur at around 75% as
discussed by K. Laurence in a technical paper of 6-series aerofoil investigation. [50]
Figure 37 How to obtain Trailing Edge Angle .
Using the equation in the final step of Table 17 the curve slope may be predicted as
shown in Table 18 and Figure 38 . This method allows the lift difference of the
ailerons or flaps to be determined for specific deflection angles. For a variety of](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-96-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
78 | P a g e
ME5308 – Major Group Project
4.2.4. Centre of Gravity
The Centre of Gravity (CG) is vital to aircraft characteristics in flight as it affects key
stability parameters. Alongside the weight, it was also estimated and updated
constantly in a bid to more accurately predict where it would be on the aircraft in
order to be able to balance out internal components to achieve the optimum
condition and reduce or avoid any extra dead weights that would need to be
implemented to balance out the aircraft.
Once the weights had been initially calculated the first thing to do; in parallel to
aerodynamic investigations conducted by the group, was to ascertain an initial value
for center of gravity. The datum was set to the UAV’s assumed nose location. Initial
sizing was determined with the use of the findings from market analysis, and the
fuselage outline was drawn out in scale. From here the wing leading edge was
established and using Raymer’s initial estimate of center of gravity as a percentage
of MAC (15-25%) [51], the components were laid out in a manner which brought the
center of gravity close to this specified point and the center of gravity was calculated
using simple moment calculations:
= ( ) ( )
The exact desired position was later better defined in the static margin calculations
for stability in flight. The CG of the geometry of the aircraft was estimated using the
assumption of point loads on each separate section of the geometry which moved
the calculated value to the aft. With this in mind the battery compartment of the UAV
was given special consideration to allow any movement forward to adjust the CG
back to the initial theoretical optimum. Throughout the design process as details
were altered or refined, this estimate for the optimum CG point was kept fixed by the
movement of all the components within the aircraft through the use of an excel
spreadsheet.
The CG point was made a design requirement due to the stability required of the tri-
copter platform which would be embedded within the aircraft. The Vertical Take-Off
propellers all had to be equidistant from the CG point to balance out the thrust
required for hover to be 1/3 of the vehicle weight per motor. This also had the benefit
of making it easier to fly for stability reasons. This is commonly referred to as the
golden triangle of stability, due to the equilateral triangle created by the position of](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-98-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
83 | P a g e
ME5308 – Major Group Project
return back to its original flight path when disturbed by factors such as gusts. The
stability of an aircraft can be analysed as static stability and dynamic stability. The
stability of an aircraft is measured about the lateral, longitudinal and directional axes.
Given below in Table 21 are some of the requirements that the aircraft needs to
meet and typical values in order to be stable. The table also shows the stability
derivatives that influence the stability and what their typical values should be. These
values of stability derivatives were used as a guideline when designing the UAV.
Stability
requirement
Stability derivative and function Typical values
in (1/rad)
Static longitudinal
stability
Rate of change of pitching
moment coefficient with respect to
AoA
-0.3 to -1.5
Dynamic
longitudinal
stability
, Rate of change of pitching
moment coefficient with respect to
pitch rate
-5 to -40
Static directional
stability
, Rate of change of yawing
moment coefficient with respect to
sideslip angle β
+0.05 to +0.4
Dynamic
directional
stabilty
, Rate of change of yawing
moment with respect to yaw rate
-0.1 to -1
Table 21: Static and dynamic stability requirements. [40]
The values of stability derivatives and has to be negative in order for the
aircraft to be statically longitudinally stable. These values are highly influenced by
the design of the horizontal tail of the aircraft. The values of and are highly
influenced by the design of the vertical tail. The value of has to be positive for
the aircraft to be statically directionally stable and has to be negative for it to
have strong stabilizing effect on the aircraft. [40]](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-103-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
84 | P a g e
ME5308 – Major Group Project
Figure 42: Graphs indicating the derivatives and for stable and instable aircraft conditions.
Static margin and neutral point
There are two key points located on an aircraft that determine it degree of pitch
stability. These are the Centre of Gravity (CG) and the Neutral Point (NP). The CG is
the point where the weight of the aircraft acts and the NP is the point where the
pitching moment does not vary with variation in the aircrafts angle of attack [52].
Static Margin (SM) is a parameter that measures the degree of pitch stability by
taking into account the position of the CG and NP [52]. The equation below shows
this relationship:
Even though a high positive static margin such as 0.5 would make the aircraft very
stable, it is not really preferable. This is because a high static margin would result in
the aircraft having a sluggish response to elevator pitch. Therefore a smaller positive
static margin of 0.135 was selected for this UAV. The neutral point of the aircraft is
dependent on the aircraft’s geometry and the CG of the aircraft is dependent on
point loads of the aircraft. Two methods were used in order to estimate the aircraft’s](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-104-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
85 | P a g e
ME5308 – Major Group Project
neutral point. The neutral point of the aircraft was determined with the following
equations:
Result
Method 1 ⁄
⁄
( )
Method 2
( )
( )
Table 22: Methods of determining the location of neutral point [38] [53]
Method 1 above uses the geometric dimensions of the aircraft to calculate the
location of the NP. Method 2 depends on the curve slope of the wing and tail and the
downwash angle to estimate the NP and assumes that = 0. Even though both
results give very similar results, the result from the first method were used as it quite
straight forward and does make have any assumptions.
Since the location of the Neutral Point is determined, and the desired Static Margin is
known, components were positioned on the UAV such that the desired static margin
is obtained. Currently the aircraft has a static margin of approximately 13.5 %.
However, changing position of components (such as avionics) in the UAV will make
the CG vary and allow the static margin to be changed for specific flight mission in
the future.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-105-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
89 | P a g e
ME5308 – Major Group Project
Control
Flight control surfaces are used to control the aircrafts attitude and direction. They
are mainly considered as primary and secondary surfaces. Like stability, the aircrafts
control can also be evaluated as lateral, longitudinal and directional control. The
lateral control of the aircraft is governed by the aileron. The sizing of the aileron is
covered in the previous section of this report. The elevator on the horizontal tail is
responsible for the longitudinal controllability of the aircraft and the rudder on the
vertical tail is responsible for the directional controllability of the aircraft. The design
sizing of the elevator and rudder will be covered in the upcoming section.
Primary surfaces compromise of the ailerons, elevator and rudder. Table 23 below
illustrates the functioning of the different primary control surfaces.
Primary Control
Surface
Aircraft
movement
Axes of
rotation
Type of
stability
Ailerons Roll Longitudinal Lateral
Elevator Pitch Lateral Longitudinal
Rudder Yaw Vertical Directional
Table 23: Control Surface Functions.
Control mechanisms such as flaps, spoilers, slats and trim controls are typical
secondary control surfaces. They are used to enhance the performance
characteristics of aircraft and alleviate the forces exerted on the controls. [54]](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-109-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
90 | P a g e
ME5308 – Major Group Project
Control Surface Sizing: Rudder and Elevator
Elevator
The elevator is a primary control surface that is placed on the trailing edge of the
horizontal stabiliser. The elevator controls the pitching moment of the aircraft. The
design of the elevator design itself is an iterative process like the design process for
the other control surfaces on the aircraft. Initial design parameters are first
determined. The design parameters are then used to calculate non-dimensional
derivatives in order to check the effectiveness of the elevator design. For this
particular UAV, the use of the elevator would be most crucial during the take-off
climb. Microsoft excel was used to formulate a programme in order to find out the
total elevator deflection required at maximum climb during take-off. The size of the
elevator was changed if the deflection required exceeded the maximum deflection
possible.
The elevator deflects in two directions. The downward deflection of an elevator is
referred to as positive deflection ad the upward deflection of the elevator is referred
as negative deflection. A negative elevator deflection causes the aircraft’s nose to
pitch up and a positive elevator deflection causes the aircraft’s nose to pitch down.
Various parameters such as the elevator effectiveness parameter, increment in lift
coefficient with elevator deflection, elevator effectiveness derivatives and aircraft
static longitudinal stability derivatives were calculated to check if the amount of
deflection required is within the limit of the maximum elevator deflection. The results
obtained were also checked to see if the horizontal stabiliser configuration would
stall or not in a critical manoeuvre such as climb [40].
Given below are the steps used to determine the effectiveness of the elevator.
1. Basic design parameters such as elevator planform area (Se), elevator chord (ce),
elevator span (be) and elevator maximum positive and negative deflection were
determined. These values were initially estimated based on typical values used for
general aviation aircraft. The maximum elevator deflection was limited to 20° in order
to prevent flow separation and stall of the horizontal tail.
2. The Lift at take-off was calculated. ⁄](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-110-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
92 | P a g e
ME5308 – Major Group Project
Figure 46: control surface effectiveness parameter vs. control surface to lifting surface chord ratio. [40]
9. The lift coefficient available from the horizontal tail with maximum deflection was
then determined. This was done with the help of ESDU sheet. The available
horizontal tail lift coefficient was then compared to with the desired lift coefficient
calculated in step 4 above. It is required that the available lift coefficient is slightly
higher than the desired one. Insufficient lift available would make the elevator
unacceptable and require the elevator to be redesigned.
10. The next step is to determine the deflection required by the elevator. To know
this, it is important to calculate the elevator effectiveness
derivatives and the aircraft’s static longitudinal stability derivative
are first determined.
[ ] [ ]
11. The elevator deflection required to maintain longitudinal trim when the aircraft
flies at its maximum velocity with most aft CG location was then determined.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-112-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
93 | P a g e
ME5308 – Major Group Project
[ ]
The elevator deflection required was found to be less than the maximum possible
elevator deflection making the elevator design acceptable. Since the elevator design
was acceptable, a few other checks were done to ensure that the horizontal tail does
not stall during take-off rotation.
Rudder
The rudder is a primary control surface placed in the trailing edge of a vertical tail. It
enables the pilot to have directional control on the aircraft. The use of rudder is
mainly implemented in conditions such as adverse yaw, aircraft spin recovery,
coordinated turns or crosswind landings.
The rudder works by generating differential lift on each direction depending on the
side of deflection. This works by changing the camber in the symmetric aerofoil with
the deflection. The rudder lift (side force) generated is at a distance (optimum arm
and this generates a torque on the aircraft and which causes the aircraft to rotate
about its CG. [55]
The main requirements of rudder in the current mission profile would be to enable
the aircraft to land safely with crosswinds. Hence, the aircraft’s rudder design was
done by mainly taking into account the requirements for crosswind landings.
Parameters such as aircraft sideslip angle, stability derivatives and control
derivatives where calculated to obtain the required crab angle and rudder deflection
for safe cross-wind landing [40].
Given below is the steps used for rudder design.
1. Basic design parameters such as rudder area, rudder chord, rudder span and
maximum rudder deflection were determined.
2. Determination of some velocities such as maximum crosswind velocity and
total velocity during landing with presence of crosswind.
√](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-113-320.jpg)
![Design and Development of a Hybrid UAV
Abbinaya T.J.
94 | P a g e
ME5308 – Major Group Project
3. The side force of the aircraft, produced by the crosswind is then determined.
The force generated is dependent on the projected side view, of the aircraft. Since
the rudder deflection is the same towards both sides, a crosswind from the right was
assumed. This would generate a positive sideslip angle. A side drag coefficient,
of 0.6 was assumed because of the fuselage having a cylindrical shape.
⁄
4. The aircraft’s sideslip angle was determined.
[ ]
5. The aircraft sideslip derivatives were then determined.
[ ]
[ ]
in the equation is very dependent on the shape of the fuselage and its projected
side. It represents the contribution of the fuselage to the aircraft side slip
derivative .The contribution of the fuselage to the directional static stability tends
to be negative. The typical values used for in an aircraft are between 0.65 and
0.85. [40]
The aircraft sideslip derivative, is usually determined using wind tunnel testing.
However for purpose of calculation and lack of time, the value was estimated with
calculations. in the equation is very dependent on the shape of the fuselage and
its projected side. It represents the contribution of the fuselage to the aircraft side slip
derivative .The contribution of the fuselage to the directional static stability tends
to be positive. The typical values used for in an aircraft are between 1.3 and 1.4.
[40]
6. The aircraft control derivatives were also determined.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-114-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
98 | P a g e
ME5308 – Major Group Project
CG Envelope
The center of gravity envelope refers to a visual representation of the limitations for
the center of gravity depending on different total vehicle weights. Diagrams can
display longitudinal, lateral as well as vertical envelopes [56]. For the purpose of this
report, and prototype aircraft, the longitudinal envelope was studied to ensure
stability on different mission loads and weights. The position of the center of gravity
within an aircraft is limited by the maximum force created by the tail, as-well as the
neutral point. These give the forward and aft limits respectively [47]. There are two
differences between the project vehicle cg envelope diagram and a normal aircraft
cg envelope. First is the additional limit visible on the actual diagram. This is the
legal weight restriction of 7.0 kg maximum aircraft weight, and it acts as a cut-off
point on the diagram as can be seen below. The second difference to traditional
envelopes is the fact that throughout flight due to the electric nature of the UAV, the
weight does not change between take-off to landing.
Figure 49 Longitudinal CG Envelope for Project vehicle](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-118-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
100 | P a g e
ME5308 – Major Group Project
4.2.6. Stability: Vertical Take-Off and Landing
In addition to the stability of an aircraft being implemented in the design, the
structure had to allow for the cg to be placed in such a way as to allow the tri-copter
aspect of the aircraft stable. This stability is known as the golden triangle, which
refers to a point which is equidistant from all the motors.
Two of the propellers in the tri-copter are in a counter rotating pair, cancelling out
any yawing moments created by the spin of the propellers. However since the third
propeller doesn’t have a countering partner, it will have a yaw effect on the aircraft
which if left to its own devices would make it uncontrollable. To counter this moment,
the tail arm motor which in this case is the propeller mounted on the nose of the
aircraft has a tilting mechanism to cancel out any yaw induced by its propeller. It also
helps to steer the tri-copter by providing additional yaw control during flight.
The design of the UAV is such that it hybridises a fixed wing and tri-copter. In this
section the VTOL aspect of the design is discussed in the context of the tri-copter as
it will be primarily responsible for providing the lift to achieve vertical lift off when
VTOL is being attempted. How the stability of the tri-copter aspect of the design is
achieved is also discussed as well as important theory behind rotorcraft relative to
the VTOL system.
The tri-copter utilises 3 rotors to achieve lift and perform manoeuvres. A model of the
tri-copter can be created when regarding the tri-copter as a rigid body [57] [58].
Using modelling methods described in [58] a base model of the tri-copter can be
created as shown in Figure 50. From the model we can then derive Force, Moment
and Kinematic equations [57] [58] as shown in Equations 4.2.6.1, 4.2.6.2 and
4.2.6.3. In accordance to [57] [58] we assume that external forces in relative axes
( ) and moments (L, M, N) are acting on the CG, this will produce translational
velocities (u, v, w), rotational velocities (p, q, r) as well as rotational angles ( )
and rotational inertias ( ).](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-120-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
102 | P a g e
ME5308 – Major Group Project
Pitch & Roll Control
It is important to understand how a tri-copter is able to perform attitude adjustments
such as pitch, roll and yaw. As the tri-copter does not have control surfaces like that
of a fixed wing aircraft it has to use its rotors to perform attitude adjustment. This is
achieved by adjusting the angular velocity and thus the RPM of a rotor. Shown in
Figure 51 it is observed that in order to achieve pitch Rotor 1’s angular velocity (1)
has to be greater than that of Rotor 2 (2) and Rotor 3 (3) while 2 and 3 are
equal to each other (1 > 2 = 3) [57]. In Figure 52 it is also shown that in order to
achieve CW roll 3 has to be less than 1 which is less than 2 (2 > 1 > 3) [57].
Conversely to achieve CCW roll, as shown in Figure 53, 3 has to be greater than
1 which is greater than 2 (2 < 1 < 3) [57].
Figure 51 Pitch up by using Rotor 1.
Figure 52 Roll in the Clockwise direction.
Figure 53 Roll in the Counter Clockwise direction.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-122-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
103 | P a g e
ME5308 – Major Group Project
Yaw Control
As shown in Figure 50 the tri-copter has 3 rotors setup in such a way that 2 rotate
counter clockwise and one clockwise. Due to the fact that there are two coupled
rotors rotating in the same direction they have a sum yawing effect on the tri-copter.
Thus establishing yaw authority is important as it is important that the tri-copter be
able to track correctly when command is given to move forward, backwards, left,
right, up or down. This is achieved by tilting Rotor 1 (see Figure 54) at an angle of α
referred to as the “tilting angle” [57], in doing so the yawing moment caused due to
the coupled CW rotors; in this case 1 & 3, is cancelled [57].
In [57] and [58] it is shown that the tilting angle has 3 DOF and is essential during
hover and movement in order to maintain an appropriate amount of accuracy [57].
Using the model developed and shown in Figure 50 it is possible to derive force and
moment equations that show the importance and effect α has on the force and
moments produced by the tri-copter. These equations are shown in matrix form in
Equations 4.2.6.6 and 4.2.6.7. Additionally, deriving Force (F) and Torque () in
regards to angular velocity () is important and can be used for each individual rotor
using Equations 4.2.6.4 and 4.2.6.5. where k and kt are the torque and thrust
coefficients respectively.
Figure 54 Yaw authority of a tri-copter.
Where (4.2.6.4)
Where (4.2.6.5)](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-123-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
104 | P a g e
ME5308 – Major Group Project
So: ⃗ [ ] (4.2.6.6)
So: ⃗⃗⃗ [ ] (4.2.6.7)
To analyse how the tri-copter is able to perform manoeuvres such as hover, vertical
climb and forward flight we can look at a single rotor and use the equations for each
rotor.
Hover
Figure 55 Mass Flow of air through rotor in hover.
In hover a rotorcraft will normally be attempting to produce enough lifting force (in
this case thrust) to keep the weight of the craft aloft in the air or just off the ground;
provided phenomena such as ground effect are negated. If the aircraft is assumed to
not be accelerating in any axis (x, y, z) then there is no production of dynamic thrust
and instead only static thrust is considered. Additionally assuming momentum is
conserved we can derive equation 4.2.6.8 from equation 4.1.1.1b:
̇ (4.2.6.8)
With the aforementioned assumptions of static thrust, no acceleration (no
movement) and negation of ground effect we can assume that ‘the airflow far above
and the rotor are zero while in a hover’ [59] as shown in Figure 55. Thus the](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-124-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
105 | P a g e
ME5308 – Major Group Project
difference between the velocity below (v) the rotor and far above (v = 0) is equal to
the velocity below (v) this means that net mass flow through the rotor [59] is:
̇ (4.2.6.9)
As mentioned, when the propeller rotates through the air it will induce a lifting force
as well as an induced velocity ( ) as shown in Figure 16, thus equation (4.2.6.9) can
be re-written as [59]:
Where = w
∴
(4.2.6.10)
Additionally if considering Figure 55 to be a closed system work done by the rotor
and work done by the rotor wake would be the same. Where the work done by the
wake would be the total change in kinetic energy in the wake [59]. Thus we can
derive equations for both work done by the rotor ( ) and work done by the wake
( ):
(4.2.6.11)
̇ (4.2.6.12)
However
∴ = ⇨ (4.2.6.13)
Now we can subst. 4.2.6.13 into 4.2.6.10
(4.2.6.14)
Rearranging (4.2.6.14) for
√ (4.2.6.15)
If the UAV was initially at ground level and when powered up rose to some altitude at
which it hovered there would be some form of drag on the UAV as the air flowed
over the UAV. Using the drag equation (4.2.6.16); where ( ) is drag coefficient and
( ) is plan-form area, we can then determine that for the UAV to maintain a hover
and remain in equilibrium, the thrust produced by the rotor most equal the weight of
the UAV while overcoming the effect of drag(vertical drag) [59]. This will produce](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-125-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
106 | P a g e
ME5308 – Major Group Project
equation (4.2.6.17). It is also possible to determine the power needed to achieve the
hover if we ignore power-loses [59] this is shown in equation (4.2.6.18).
(4.2.6.16)
(4.2.6.17)
(4.2.6.18)
When all 3 rotors are in use in order to achieve hover they must all be running at the
same rpm and therefore be producing the same amount of thrust as shown in Figure
56 this way there will be no production of pitch or roll. However yaw would still be
problematic but easily overcome by having rotor 1 tilt to cancel the adverse yaw and
run at an rpm that produced a vertical component of thrust equal to that of the other
two none tilting rotors.
Figure 56 Altitude Hold (Hover) with all 3 rotors.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-126-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
107 | P a g e
ME5308 – Major Group Project
Vertical Climb
Figure 57 Mass Flow of air through rotor in vertical climb.
In a climb it is it important to take into account the vertical climb velocity ( ) as
shown in Figure 57. The model is exactly the same as the hover model however now
we account for [59]. This is done by adding and then substituting in to
equations 4.2.6.9 and 4.2.6.10:
̇ (4.2.6.19)
(4.2.6.20)
Also work will be the same so we use 4.2.6.13 and substitute it into 4.2.6.20:
(4.2.6.21)
From equation 4.2.6.21 we can gain an expression for the induced velocity :
√( ) (4.2.6.22)
Similarly the vertical drag experienced in the climb has to account for the vertical
climb velocity ( ) additionally we account for the area outside the rotor wake ( ):
(4.2.6.23)](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-127-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
109 | P a g e
ME5308 – Major Group Project
Forward Flight
Figure 59 Flow of air through the rotor in forward flight.
In forward flight we have a number of factors at work and the complexity of analysis
is increased due to the fact that local airspeed will vary depending on what section of
the rotor is being analysed [59]. In Figure 59 it is shown how the wake is developed
when the rotor is pitched at an angle of attack ( ) where it is considered negative
when pitching down and positive when pitching up [59]. We derive that:
(4.2.6.25)
To better understand how forward flight works for the rotor craft we have to consider
the azimuth angle () this is the angle the blade of the propeller traverses when
rotating as shown in Figure 60(x-y axis). We then consider the propeller to be a disc
(rotor disc) when in the x-z axis.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-129-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
113 | P a g e
ME5308 – Major Group Project
4.2.7. Structures
Fuselage
Fuselage design is a key part of the aircraft design process. It is this structure which
will need to be capable of holding all the mission payloads and is required to take
any loads expected within the mission profile. Typically these Loads which are
experienced in flight arise in the form of five main loads from the wing during gusts,
turbulence or general aircraft manoeuvres, Landing gear loads from ground impact
and induced load from the thrust provider, when it is attached to the fuselage [60].
These are:
- Tension
- Compression
- Torsion
- Shear
- Bending
A single component of the structure can be subjected to multiple forces in one
instance [32]. There is usually an additional load due to cabin pressurization which
can be ignored for the UAV in the case of the project vehicle. For the aircraft in
question, an additional fuselage load comes from the tri-copter mission segment
where the fuselage would experience additional bending forces from being
suspended between 3 motors at hover.
To be able to support and resist the aforementioned loads, currently three main
types of fuselage design exist. These are the monocoque, semi monocoque and
truss structured fuselage (warren truss) [47].
Monocoque designs consist of a series of formers wrapped in a load bearing skin.
The issue with a monocoque design is in finding a material strong enough to sustain
the stresses of the fuselage without any connecting members between the
bulkheads or formers within the structure itself. As the structural integrity of the skin
is key to the whole structure strength in the event of a tear in one location, the whole
fuselage loses a significant amount of rigidity and strength.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-133-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
114 | P a g e
ME5308 – Major Group Project
Figure 64 Monocoque fuselage design [61]
The truss structured fuselage is common in older and smaller types of aircraft, for
example typically micro-light aircraft such as the sky ranger [62] and older bi-
plane/tri-planes such as the Fokker DR-1 [32].The truss faded after heavy use during
the First World War due to breakthroughs with technologies regarding monoplane
designs as well as the advent of semi monocoque techniques which reduced the
weight of the structure.
Figure 65 Truss fuselage structure [32]
Semi monocoque fuselage design mimics a full monocoque design. However it
bypasses issues usually found in the latter with the relative weight of the structural
components by incorporating more structural members between the formers as seen
in the figure below.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-134-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
115 | P a g e
ME5308 – Major Group Project
Figure 66 Semi-monocoque Fuselage [32]
Of the aforementioned methods the most widely used is the semi monocoque design
method. This encompasses the rigidity of a skin but with the bulkheads connected by
structural longerons and stringers to help take the load along the fuselage from the
forces during flight. These additional structures prevent tension and compression
stresses from causing a potentially fatal bending on the fuselage [63].
Good examples within UAVs of a semi monocoque structure consists of a wide
range of vehicles varying in size from the very large Global Hawk, a medium sized
Falco, to the smaller scan eagle. The fuselage compositions consist of load bearing
frames in conjunction with a load bearing skin, cutaway sections for the above
mentioned aircraft are shown below.
Figure 67 Global Hawk Cutaway [64]
Above can be seen the cut-out showing the fuselage of the Global hawk. This semi
monocoque design can be seen to have bulkheads in the form of formers connected
with structural longeron beams along the outside corners of the formers. The](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-135-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
116 | P a g e
ME5308 – Major Group Project
payload of avionics is stored within the structure, along the empty space created by
the hollowed out formers.
Figure 68 Falco Cutaway diagram [64]
The Falco structure is similar to the global hawk however it is a lot smaller in size, so
there are less formers and longerons along the internal structure. Various bulkheads
create compartments within the fuselage which is held by four beams forming a
rectangular frame at the bottom for heavy landing loads. Although this UAV has a
fixed undercarriage the wheels have been faired to reduce any extra drag created by
the components themselves.
In contrast to the above two the Scan Eagle has a particularly interesting structure.
The fuselage bulkheads and frames all come together within the skin to form
modular compartments which appear to be easy to get to for maintenance work as
well as configuration of payloads between missions. This adaptability of the structure
can be seen in the cutaway figure shown below.
Figure 69 Cutaway of the ScanEagle [64]](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-136-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
117 | P a g e
ME5308 – Major Group Project
Where-as some follow a more classical design, the Demon UAV follows the scan
eagle in having a distinct fuselage structure. It consists of formers linked together to
create a solid structure underneath the carbon fibre skin without the aid of additional
longerons and stringers within, it was not completely a true monocoque, more
inbetween monocoque and semi-monocoque as can be seen by the incomplete
structure shown on the figure below.
Figure 70 Bonding in progress of the Demon UAV composite structure [65]
The final source of inspiration and guidance on what the structure of the project
aircraft should be originated from smaller, cheaper UAV/FPV (First Person View)
Aircraft originating in the R/C market. These other lightweight UAVs include the
Bormatec Maja [66] and the RV JET [67]. The Bormatec fuselage consisted of a soft
foam encasing a supporting structure in the form of corrogated plastic plating running
the length of the fuselage where as the RV Jet resembles the scan eagle in its
modular composition with a monocoque twist within the fuselage design. The
fuselage comes in two halves ready made with formers within the structure with the
wings being able to slot in, with allowance for extensions along two carbon rod spars
running through the whole aircraft. Unorthodox designs built around specific vehicle
requirements allowed for the general types of fuselage studied to be narrowed down
and modified to cater for the particular design restrictions required by the project
aircraft.
Due to the heavy industry tendency to design and implement a semi monocoque
fuselage, this format became the selected structure type for the prototype being
developed. This led to the fuselage structure that consists of 3 main load bearing
carbon rods of 10mm diameter, with Plywood bulkheads of varying thickness spread
along vital areas of the fuselage. This was then covered by a surface skin of foam](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-137-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
118 | P a g e
ME5308 – Major Group Project
board to give it additional rigidity as-well as provide a light covering for the fuselage
section.
The main three spars within the fuselage were laid out in a triangular set up
essentially borrowing elements of the truss fuselage structural design, the figure
below shows the reasons for the design decision. Any landing loads transferred
along the bottom rod and landing gear bulkhead is transferred between all three
members through compression and tensile forces. This set-up helps the structure
absorb any potential hard landing loads.
Figure 71 Loading on a triangular structure [68]
The shape of the bulkheads was also key in designing an inherently strong structure.
These were heavily influenced by the use of arches in the civil engineering sector
[69]. Essentially the job of the bulkheads was to provide adequate load paths in the
form of the interconnecting material, mimicking the characteristics of a triangular
truss component for any compressive and tensile loading experienced upon landing
from horizontal flight. The Bulkheads were designed so that the load bearing
characteristics of a circular fuselage remained in the form of arches on the bottom
and top, whilst decreasing the overall surface area and volume of the equivalent
circular fuselage cross section. Figure 72 shows a computer image of the skeletal
structure.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-138-320.jpg)
![Design and Development of a Hybrid UAV
Brett M.
120 | P a g e
ME5308 – Major Group Project
Landing Gear
Brett
The landing gear is the intermediary between the aircraft main structures and the
ground. A typical landing gear system includes legs (also known as struts), wheels
and brakes or even skids or floats if the aircraft is destined to operate from snow and
ice fields or on water. The landing gear could be either fixed or retractable but at a
weight and cost penalty. The landing gear must be designed big enough to support
the aircraft’s weight but light enough as to not compromise the handling qualities of
the aircraft.
The design selected for the aircraft is what’s known as the tricycle configuration [70]
with a single nose strut and two rear main struts. Large diameter wheels have been
used to reduce rolling resistance and improve the aircraft’s handling over rough
ground. The tricycle configuration was selected because it is structurally efficient,
provides good control characteristics (for ground manoeuvring) and can be built or
sourced relatively cheaply and easily.
In establishing the size of the landing gear, considerations must be made for the
static and the dynamic loads the gear is required to cope with. Static loads are that
when the aircraft is not moving and only the weight of the aircraft under the force of
gravity is present. Dynamic loads are additional forces acting as a result of
movement of the aircraft over rough ground, for example where the wheels are
moving erratically as the result of a rough take-off or landing field surface. Mounting
brackets and points suitable for use to attach the bought in landing gear equipment
were almost entirely defined and tested using Computer Aided Design (CAD)
packages to size and validate their layout. Mounting bracket definition and results of
several stress test simulations are shown in Section 4.2.8.
The location of the landing gear system is usually determined by a load sharing
scheme [71] whereby a proportion of the aircraft’s total mass is supported by the
front strut and the rest by the rear struts. A graphical representation of the method
used is given by Figure 73 and the calculated position of the landing gear is given by
Equations 4.2.7.1 to 4.2.7.3.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-140-320.jpg)
![Design and Development of a Hybrid UAV
Brett M.
121 | P a g e
ME5308 – Major Group Project
The rear wheels act as the pivot on which the aircraft rotates during take-off. This
load sharing uses the weight of the aircraft to aid in the front wheel’s ability to steer
the aircraft when the aircraft is taxiing. Too much weight over the front gear would
likely mean that the aircraft’s horizontal tail would struggle to produce the force
needed to rotate the aircraft on the rear wheels to achieve take-off. For this reason,
the mounting bracket was designed to allow variable positioning of the rear landing
gear placement to achieve the best take-off rotation performance.
Figure 73 Landing Gear Positioning for Proper Weight Distribution [71]
To determine the placement of the landing gear, prior knowledge of the position of
the front landing gear and the aircraft’s theoretical centre of gravity (CG) are
required. The front landing gear strut was to be mounted onto the front fuselage
bulkhead, the position of which from the nose of the aircraft was known from a
master CAD layout of the aircraft and measured at 290mm. From the master CG
calculation table, a distance of 710mm was determined when measured from the
nose. Thus the distance between the front landing gear strut and the CG follows
Equation 1: Front Landing Gear to Aircraft CG Dimension
Equation 2: Total length Between Front and Rear Main Landing Gear
Equation 3: Distance to Rear Main Landing Gear from Aircraft CG](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-141-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
125 | P a g e
ME5308 – Major Group Project
The simulation was modelled with three fixed supports acting on where the rods
would connect to the bulkhead, with a load of 120 N acting upwards into the
assembly. This was to simulate the rigidity of the skeletal structure through the
bulkheads on a hard landing. It is important to note that any bending shown below is
excessively exaggerated by the program. When results of the maximum forces are
compared to typical values of plywood material properties, the material is more than
capable of withstanding such loads being exerted upon it. The maximum principle
stress experienced by the material is 0.213 MPa. From this maximum value the
weakest material properties of plywood were researched to assess where the failure
of the material would occur. Below is a figure showing typical values.
7.5 mm Canadian Aspen
Plywood
9.5 mm Canadian Aspen
Plywood
Orientation of
applied force
to face grain
0° 90° 0° 90°
Planar Shear
Strength
(MPa)
0.72 0.72 0.55 0.72
Table 26 showing properties of similar thickness plywood material strength [72]
The strength shown is the strength of the material under a shear forces through the
plane which is the direction the load of the landing traverses through the bulkhead.
These values stay roughly the same although other material strength properties
increase with an increase in the thickness of material [72]. As a result it can be seen
that the material can easily sustain a heavy landing of a 2 g load, with the plate along
the bottom taking the majority of the load. In this manner if anything were to fail it
would be the plate. In addition to this “throw-away” component which can be easily
replaced, the epoxy joint would fracture further absorbing the load of the landing
protecting the important structural bulkhead of the fuselage.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-145-320.jpg)
![Design and Development of a Hybrid UAV
Arturs D.
126 | P a g e
ME5308 – Major Group Project
Wings
Aircraft wings are the main part of the aircraft since they generate lift which allows
aircraft to take-off and maintain altitude. A wing can have different profiles which
could determine the complexity of the structure but mainly affects the performance of
the aircraft. In order to determine the aircraft structure it is important to know the
loads and forces to which the aircraft and the wing specifically will be subjected. First
of all the wing holds the weight of the whole aircraft so it would be subjected to
aircraft weight as well as additional lift forces for altitude gain, aircraft acceleration
and the moments about the wing. There are five major types of stresses to which the
aircrafts are subjected: tension, compression, torsion, shear and bending. Bending is
a combination of tension and compression where, tension occurs on the outside of
the band and compression on the inside. The wing is usually subjected to some
torsion and bending which is tension and compression. The wing would be subjected
to compression all through the flight where for example the leading edge meets the
airflow and stagnation point occurs where the air flow exerts a force on the leading
edge of the wing exerting a compressive force on it [73].
One of the main factors which have to be taken into consideration when designing a
wing structure would be a Load Factor - n.
Where L is the lift force (N) exerted on the aircraft and W (N) is the weight of the
aircraft. Usually highest forces exerted on the aircraft are at manoeuvres such as
sharp turns and the aircraft has to be designed to be able to withstand these forces.
The regulations state that the aircraft has to be able to withstand maximum load
factor of 3.8 and the minimal requirement is 2 if it can be proven through probability
analysis that the load factor would not be exceeded. Similarly for negative load factor
the values are -1.5 and -0.5 with proof [74]. These are one of the strictest regulations
which have to be met for a UAV in order to be certified. Since this is not an aerobatic
aircraft and was not designed to do sharp manoeuvres the limit load factor which it
needs to be designed for, is 3 including the safety factor.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-146-320.jpg)
![Design and Development of a Hybrid UAV
Arturs D.
127 | P a g e
ME5308 – Major Group Project
Figure 77 demonstration of typical wing structure [75]
Generally aircraft wing consists of 4 main components demonstrated in figure 1. Ribs
playing the role of bulkheads which determine the aerofoil shape and sustain the
compressive force, they position chord wise and determine the internal structure of
the wing. The ribs are joined together by stringers which have little load bearing
purposes. The core load bearing component would be spars. They prevent bending
and torsion, carry most of the load and usually connect the wing to the fuselage.
Spars could be in different shaped beams such as I-beams or cylindrical beams. The
final component of the structure would be the outer skin of the wing which mainly
protects the aircraft from the external influences and gives it the final shape. It is
important for the skin to be smooth to keep the skin friction drag to a minimum and
solid to maintain the wing profile [73].
The aircraft wing requires o be light and strong and this design allows the majority of
the wing to be hollow. Such wing structure used by commercial aircrafts as well as
military, UAVs and hobby RC planes. There are some differences such as materials
and some minor components however, the principle is the same.
Majority of homebuilt RC planes use balsawood ribs and stringers for the internal
structure of the aircraft since they are quick and easy to manufacture (by laser
cutting) and they are very light. Such structures are usually covered with plastic film](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-147-320.jpg)
![Design and Development of a Hybrid UAV
Arturs D.
128 | P a g e
ME5308 – Major Group Project
which is heat shrank around the structure. It is difficult to use other materials for skin
which use adhesive since there is little surface are to attach it to. One of the
disadvantages using this type of structure is that between the ribs the film sags and
changes the profile shape which affects the performance of the wing. In order to
decrease this effect the distance between the ribs has to be decreased by
introducing more ribs which increase the weight of the whole structure. The VTOL
aspect of the aircraft introduced a lot of additional weight due to 3 additional motors,
large battery to power them and other avionics which increases the wing loading.
Using different materials such as plywood would increase the weight of the structure
dramatically. Another option which is often used by RC planes manufacturers is
using low density foam for the wing core such as “Transall C-160” for an example
[76]. Foam core replaces the ribs and stringers by introducing whole wing shaped
piece of foam. Further analysis of foam and balsawood properties were analysed
and the data in table 1 was used as a guideline for decision making.
Table 27 properties comparison of foam core wing reinforced with carbon fibre spars to balsawood
ribbed structure reinforced with carbon fibre spars [77]
Expanded Polystyrene Foam (EPS) and Expanded Polypropylene Foam (EPP) are
both suitable for wing manufacturing and both are being used however they have
different properties and different densities. EPS foam is a lot lighter that EPP but a
lot less rigid therefore, when using this foam it has become a standard procedure to
reinforce the leading and trailing edges of the wing by the foam wing manufacturers
[78]. Reinforcement is usually done with wood such as balsa or obeche. It is a lot
easier to work with foam since it has large surface area which allows broader range
of materials for the skin. Various materials were considered for outer skin of the wing
such as: carbon fibre polymer lamination, fiberglass polymer lamination, plastic film
heat shrink and wood veneer. Fibreglass and carbon fibre reinforced epoxies have](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-148-320.jpg)
![Design and Development of a Hybrid UAV
Arturs D.
129 | P a g e
ME5308 – Major Group Project
great properties and relatively low weight however the weight is still larger than wood
and requires complex manufacturing methods such as Vacuum Bagging [79]. After
considering pros and cons it was decided to use a thin layer of obeche veneer to add
rigidity to the wind and heat shrink a thin layer of plastic film on top to add smooth
surface finish. [80]
In order to reinforce the wing, different types of spar structures were considered such
as: monospar, multispar and box beam. It was decided to use cylindrical carbon fibre
reinforced multispar system with two beams due to, mechanical properties, wide
availability and ease of use. Using rodss as spars fits well with the rest of the aircraft
skeletal structure which is mainly consists of carbon fibre reinforced rods. Especially
designed for this project rod-to-rod connections were used for connecting wings to
the boom tail and wings to fuselage where a perpendicular cross fitting connection
were used. The fuselage rod slots into the connection through the bottom elongated,
rod slot and both spars from the wing slot through two top, perpendicular hoops
attached to the slot. Very similar wing to fuselage connection type is used for
commercial “RVJET” flying wing by RangeVideo [81]. Both carbon fibre reinforced
resin rods and the connections are described in greater detail later in the report as
well as the results of the tests carried out to prove the capability to sustain the
exerted loads and forces.
Obeche Wing veneer
Obeche timber comes from African Triplochiton Scleroxylon tree. Most of aircraft
foam wings are veneered with a thin layer of obeche since they add rigidity to the
wing however, it is easy to work with since it is not too rigid and can be shaped
around the wing. It has very low density almost like balsa wood but better properties
for wing veneering. It works very well with glues and can easily be glued on to the
foam. Smooth surface finish is easier to achieve that with balsa wood. To further
improve surface finish the wing was covered with plastic film using heat shrinking
technique which improves the properties further and protects from moisture and
other external influences.
Sourced pieces
Main sourced parts that were obtained for the project are carbon fibre reinforced
resin rod. Rods mainly connected major sections of the aircraft using rod](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-149-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
143 | P a g e
ME5308 – Major Group Project
Motor and Propeller Selection
Selecting the motor and propeller combination for the VTOL aspect of the UAV is
important. Ensuring that the propeller chosen allows for the motor to perform at its
rated output ensures performance of the motor and prop are optimal [82]. In doing so
it also helps to ensure that when there is a mass flow of air the dynamic thrust being
produced is enough to keep the aircraft aloft. While the static thrust must be enough
to lift the UAV.
There are two methods that can be used, one uses the RPM of the motor to match
the propeller (RPM method) and the other uses T/W as a baseline in the selection of
the motor and then attempts to match propeller rpm to that of the motor (T/W
method). Both methods were used and results varied (different props showed as
being suitable) these props would later be tested to determine which ones could be
used and gave the required performance (see section 6.2.2).
RPM Method
First a motor is selected and its maximum RPM is determined. In our case using
electric DC motors this is determined from equation 4.2.9.1:
(4.2.9.1)
Next the motor’s ideal RPM is determined. This done be selecting a multiplier which
can vary depending on motor type however in the case of DC motors this is normally
0.25 [83]. This results in equation (4.2.9.2)
(4.2.9.2)
Once max and ideal RPM are known we determine max current (I-max), operational
current (I-op), no load current (I-noload) and internal resistance. These are normally
available from manufacturers of the motor, component distributors or in the manuals
that come with the motor. Once these are known copper loss and iron loss are
determined using equations 4.2.9.3 and 4.2.9.4:
(4.2.9.3)](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-163-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
144 | P a g e
ME5308 – Major Group Project
(4.2.9.4)
Now Power in and Power Out can be determined using equations 4.2.9.5 and
4.2.9.6:
(4.2.9.5)
(4.2.9.6)
Having obtained our Pin and Pout we can determine the efficiency of the motor ( )
using equation 4.2.6.7 and then determine the produced power in the propeller using
equation 4.2.6.8:
(4.2.9.7)
(4.2.9.8)
After acquiring the propeller power we are able to determine the ideal prop RPM
(equation 4.2.9.9) and in turn find out the differences between the Ideal rpm of the
prop and motor as well as between the max rpms. Propeller data such as Kp (prop.
constant) and PF (power factor) was acquired from [84].
(( ) ( )) (4.2.9.9)
Once the ideal RPM of the prop is known it is subtracted from the Ideal RPM of the
motor (Delta Ideal). The ideal RPM of the prop is also subtracted from the max RPM
of the motor (Delta Max). In order to complete the matching numerous propellers are
used in the process the criteria used to select the appropriate propeller is:
1. A high ideal propeller RPM
2. A value of Delta Ideal close to zero
3. A high value of Delta Max
Using this method it was found that using the Turnigy 3648-1450 motor with the
2 Bladed APC Slow Fly Series 9x7.5. A table of the process and results can be
found on the CD attached with this report in the excel spreadsheet Propeller Analysis
> RPM Method.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-164-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
145 | P a g e
ME5308 – Major Group Project
T/W method
The T/W (Thrust to weight) method is the second method that was used to attempt to
match viable propellers to the DC motor which was to be used for VTOL. In this
method the T/W ratio is the desired design point. The VTOL aspect of the aircraft
requires a T/W ratio of 1.2+ [85]. This is then used as the criteria to select if a
propeller/motor combination will be suited to the task.
First the Maximum Thrust the motor can produce is acquired, this was found from
the manufacturer provided datasheet of the motor. Once the max motor thrust is
known the ideal thrust is determined. Ideal thrust is assumed to be the amount of
thrust you require the motor + propeller to totally produce in order to lift the weight of
the UAV. This will then produce the T/W that the motor will produce.
Similar to the RPM Method the ideal and max RPM of the motor are determined.
Once these are known the ideal RPM of the propeller is determined using equation
4.2.9.10:
( ) (
√
) (4.2.9.10)
After determining the ideal RPM of the propeller it is then compare to the ideal RPM
of the motor. The criteria used is that the motor chosen must be able to produce a
T/W ratio that is adequate for the design. This process is repeat until a motor and
propeller have close ideal RPMs. Once the two criteria have been met the propeller
and motor are assumed to be matched.
When using this method it was found that the ideal RPM of the motor was 8048 and
the ideal RPM of the propeller was 8054. This paired the Turnigy 3648 motor with
the APC Electric E –Series 9x9 Propellers (3 Bladed). A table of the process and
results can be found on the CD attached with this report in the excel spreadsheet
Propeller Analysis > Hybrid Method.
Battery Selection
As the required propulsion units have a high amp drawn, the battery selected had to
be capable of powering the vehicle for both mission profiles without needing](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-165-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
146 | P a g e
ME5308 – Major Group Project
recharging. A LiPo platform was chosen due to its high power to weight ratio. From
initial estimates minimum capacity required was that of a 6 cell battery with
9000mAh, and the maximum 11000mAh.
Discharge (C) Capacity (mAh) Weight (kg) Price Excluding
Shipping (£)
100 9000 1.278 216.99
120 10900 1.371 266.59
100 12000 1.65 297.59
Table 28 Battery properties for a suitable range of products [86]
On the above table is listed the batteries which were available and fit the
requirements of the mission profiles. As the required performance was so specific,
the only power plants which fit the description were sourced overseas from the USA.
From this list, budgetary restrictions as-well as weight limitations further narrowed
down the choice to the lightest option, the 9000 mAh battery.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-166-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
151 | P a g e
ME5308 – Major Group Project
on-board avionics systems.
Airspeed
Sensor
Needed to provide the flight control
unit with accurate airspeed readings.
3DRobotics Airspeed
Sensor
Radio
Receiver
Needed to allow the UAV to be
manually controlled by a human pilot
using a Radio Transmitter
Futaba 8 channel
Receiver 2.4 GHz
Table 30 Necessary Avionics Components for the UAV.
These components are available from many sources online and in some cases
locally allowing for rapid procurement in case of the need to replace damaged
components.
Main Control Scheme and sensor array
As the UAV is in flight or is given an input by a pilot the Flight control system is
responsible for interpreting those inputs and ensuring that the response of the UAV
is what is expected. A general control scheme to better represent the UAV and its
avionics is presented in Figure 94. As can be seen in Figure 94 the microcontroller
sends the input task to the servos (actuators) and this information is used to induce a
change in attitude, speed, performance of the UAV. This change is then fed back to
the micro controller where it is processed and used to make adjustments if the
desired input has not been met [87].
Figure 94 General control scheme of the UAV [87].
The Micro controller used by the UAV is the APM 2.6 shown in Figure 95. It is based
off the Arduino 2560 Mega micro controller. This version of the microcontroller](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-171-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
152 | P a g e
ME5308 – Major Group Project
however comes with an on-board sensor array. This sensor array includes a
barometric pressure sensor for altitude and a 6 DOF Accelerometer/Gyro for attitude
sensing.
Figure 95 ArduPilot Mega 2.6 from 3D Robotics
The MPU-6000 on-board Accelerometer/Gyro allows for the APM to detect the
current attitude of the UAV. It utilises a digital motion processing unit (Shown in the
MPU’s schematic in Figure 96) that fuses the information from the accelerometer
(which measure acceleration in each direction) and the gyros (which measure the
angular velocity).
This creates quaternions as opposed to Euler angles for pitch, roll and yaw [88].
Quaternions are easier to interpolate and have a smaller memory footprint than Euler
rotation matrices angle [89], [90] this is done by use of a Direct Cosine Matrix (DCM)
or a Kalman filter that allows us to fuse the two signals from the gyros and
accelerometer [88]. This allows for the FCS to have a much faster awareness of
what its orientation is.
Figure 96 Schematic of MPU-6000.
The microcontroller is able to have a GPS/Compass module connected via an inter-
integrated-circuit (I2C) connection adding additional sensing functionality to the
Micro controller. Being based off the Arduino Mega means the APM also has
analogue and digital pin inputs and outputs.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-172-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
153 | P a g e
ME5308 – Major Group Project
On the analogue pins devices such as the Sonar and Airspeed sensor are connected
as they provide analogue data such as pressure difference and sound delay to
produce digital information used by the control.
A control scheme can be made specifically to the tri-copter aspect of the UAV. The
normal inputs in this case are longitudinal (δ-long), lateral (δ-lat) and yaw (δ-yaw)
[57]. We can develop a block diagram as shown in Figure 97
Figure 97 Block diagram of tri-copter control include 2 gain values [57].
Once the longitudinal (δ-long), lateral (δ-lat) and yaw (δ-yaw) are read the controller
can interpret them into angular velocities of the rotors and the angle of attack
necessary the UAV needs in yaw this is shown in Figure 98.
Figure 98 Control allocation by a controller on a tri-copter.
PID Controller
In-order to dictate to the UAV how to behave in terms of manoeuvres we have to
send signals to control surfaces and motors as to how much rpm they must be at or
angle they must deflect at and at what rate. To do this a Proportional Integral
Derivative (PID) controller is implemented. The APM 2.6 however uses a cascaded](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-173-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
154 | P a g e
ME5308 – Major Group Project
PID structure. A cascaded PID controller is used since it is able to achieve smooth
tracking with fast disturbance rejection [91]. An example of cascade control is shown
in Figure 97.
Figure 99 Example of Cascade Control.
The Cascaded PID used by APM however is shown in Figure 100. There are two
PIDs one is to stabilise which will take the current actual angle from the sensors and
the desired angle from the pilot output a rotational rate into a rate PID. At the rate
PID the rotational rate from the gyro is also taken into account and a final out is sent
to the motor.
Figure 100 Cascaded PID used by APM [88].
In order to get an output that is satisfactory the PIDs will need to be tuned. Tuning of
the PIDs can be done by using the Mission-Planner software that APM uses. PID
tuning will require ground and flight testing to determine what amount of Kp, Ki, Kd
are needed for best performance
Related software and Full system schematic
As mentioned before the APM uses an open source software package called
Mission-Planner. It is used to upload the appropriate type of firmware onto the APM.
The two firmware that were used for this project were ArduCopter and ArduPlane.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-174-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
158 | P a g e
ME5308 – Major Group Project
Manual Transition
Although the project goals are specifically to perform an STOL mission, and perform
a VTOL mission separately, the manner in which these systems have been
implemented into the aircraft allows for a very basic form of manual transition in-
between the two in the near future, to avoid having to take-off in rough conditions the
framework exists but has not been tested. Shown on the figure below is the
manoeuvre split into its basic 3 elements when the controls have been mapped with
a three position switch routed onto channel 5 of the transmitter and receiver. This
method incorporates the dual APM set-up with an additional relay Arduino board to
control the distribution of the inputs from the receiver between the other two micro
controllers. The code within the relay board was created with the help of another
student whose background knowledge of coding with Arduino originated from an
obstacle avoidance project where a different operational problem was solved utilizing
similar methods [92].
Figure 104 Phases of flight during the transition maneuver from hover to horizontal flight
The first phase of the manoeuvre is with the three position switch in its bottom
position, where the on-board relay is routing all controls directly to the Arducopter
installed APM, which operates under the stabilize mode of the firmware. Here the
aircraft must climb to a safe altitude of at least 20 m via manual input from the pilot.
At this point the second phase can be initiated. When the pilot moves the three
position switch onto the central position, it causes the relay board to send a signal to
the Arducopter APM changing the mode from stabilize into Altitude hold. In this state
the tri-copter will not resist any translation of motion on the same altitude and can
drift but will retain its pitch an height. The switch change also allows the relay to send
all the inputs on the receiver for flight to the Arduplane APM. In this way the aircraft](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-178-320.jpg)
![Design and Development of a Hybrid UAV
Camilo V.
159 | P a g e
ME5308 – Major Group Project
hovers autonomously maintaining its altitude, whilst the pilot has manual control of
all STOL related systems. The pilot will proceed to throttle the EDF unit up and start
gaining momentum. As the wings start to generate lift the tri copter will reduce thrust
to a minimum as less and less is required to keep it at the same altitude. The last
phase of the manoeuvre is with the three position switch at its last position. Here the
tri-copter firmware will go into acrobatic mode, with all propellers powered down to
idle. This is only to occur once the aircraft has gained sufficient speed where the
wings are generating enough lift to prevent the aircraft from stalling. As with the
previous phase, all the controls are routed to the control surfaces and the EDF
propulsion unit.
Technically the same method can be applied for landing vertically from horizontal
flight, with the procedures reversed. The only difference between transitioning from
hover to horizontal flight and vice versa is the small issue that upon entering stabilize
mode on the Arducopter APM from altitude hold, the throttle will automatically
change to match the setting that it currently is at on the transmitter [93]. This is an
issue as when stopping into altitude hold, the throttle will be at its minimum value so
the EDF isn't providing any more forward momentum. As such the altitude must be
sufficient that when the Arducopter firmware switches into stabilize mode, the drop in
altitude from the change in thrust can be compensated for to regain control and
continue in a manual controlled descent to land. It is important to note that the
endurance of the vehicle will be severely affected and will be substantially shorter
than its STOL only mission profile as it will be limited by the increased power
consumption of all motors drawing power from the on-board battery.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-179-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
170 | P a g e
ME5308 – Major Group Project
6.2. Motor Characterisation
It is important to characterise the rotors and EDF that are used on the UAV. This is
done due to the fact that linear behaviour of these components can vary due to many
factors such as production defects or electromagnetic interference (EMI). The motors
used in both cases are brushless and utilise electronic commutation [94]. The
rotation of the motors is created the permanent magnetic rotors chasing a revolving
magnetic field which is induced by the current in stator windings inside the motor
[94]. Using PWM signals that are either on or off the magnetic field varies causing
the drum to rotate. The motor will thus accelerate or decelerate depending on the
current density changes caused by the changing PWM state [94].
Because of this use of magnetic fields EMI can cause behaviour and performance of
the motor to be non-linear at certain PWM duty cycles and it is important to know
what range of PWM signal allows for linear behaviour and thus performance. Once
this region is known the amount of thrust produced within the linear region is
analysed in order to ensure the required thrust does not lie within the nonlinear
region which is normally at very low and very high PWM duty cycle. By doing a
characterization test we are able to check and verify that the motor and propeller
combinations assumed in section 4.2.9 are viable.
6.2.1. Set-Up
In order to perform motor characterisation the thrust bench in the aerospace
engineering was used. It consisted of pivoted arm with a pulley that allowed for the
loading of weights. The motor to be tested is mounted on the top of the arm such
that it creates a moment that activates the force sensor. The bottom of the arm
connects to a force sensor which will send data to a computer.
On the computer a LabView code is used which gives us numeric data that is then
used to determine the thrust and characterise the motor. The apparatus is shown in
Figure 113.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-190-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
172 | P a g e
ME5308 – Major Group Project
Figure 116 80A ESC Turnigy Superbrain.
Figure 117 Turnigy KV-RPM Meter.
Figure 118 National Instruments Hi-Speed USB
Carrier.
Figure 119 Turnigy 4000 mAh LiPO Battery (6s).
Motor Testing Code
To run the motors the microcontroller that was to be used for flight control on the
UAV was used (APM 2.6). Since it is based on the Arduino AT2560 Mega board a
custom code had to be written to allow for selectable PWM signals to be sent to the
motor being tested. The custom code called “Manual_Input_MotorTest” is included in
the report Appendix A.
The code used the Arduino servo library which actually uses PPM signals which are
then converted to PWM since the width of the pulse determines the position or angle
the motor will rotate through [95] as shown in Figure 120.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-192-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
173 | P a g e
ME5308 – Major Group Project
Figure 120 PWM changing the angle of a dc motor [95].
Since an ESC was being used the motor would have to be armed. To accomplish
this a function was made called void_arm that would send a high signal to the ESC
while the battery’s positive terminal was not connected. Once the program displayed
“Connect battery” the positive terminal would be connected to the ESC and the
motor would be armed. It should be noted that when performing the EDF test the
ESC it was using during tests was the 80A ESC rather than the 100A ESC it was
going to be coupled with on the UAV. This was due to the 100A ESC not being able
to be armed using this method.
Once the motor is armed a loop statement is run that allows the tester to input a
PWM signal that would run the motor. Due to the risk and hazard associated with
this type of testing a fail-safe was implemented in the code that allowed for the motor
to be stopped at any point in case of an emergency. Another fail-safe was
implemented to prevent PWM values that were higher than 100 as to prevent
damage to the motors or mistakes such as mistyped values being reCGnised.
Code was also implemented to slowly step down the running speed of the motor this
would allow for a controlled stop at the end of every test run.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-193-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
179 | P a g e
ME5308 – Major Group Project
6.2.5. Motor Results
The final propellers considered for the VTOL motor test were the two bladed APC
6x6 and the three bladed E-series slow fly 9x7.5 These are two and three bladed
respectively. As shown in Figure 124 the three bladed propeller should be more
efficient for thrust load coefficient over 0.5, but the current thrust load coefficient is
almost 1.5. According to the experimental results of J. Dang and H. Laheij [96] the
two bladed propeller should provide more thrust for the same amount for the same
input signal.
Figure 124 Thrust efficiency of two and three bladed propellers [96].
Due to the nature of VTOL and the yawing moments generated when the propellers
spin, the tricopter should have two motor spinning in one direction and the other one
in the opposite. Since it was very difficult to find the appropriate counter rotating
propellers, then pusher and tractor props where used. They are the same as counter
rotating propellers, but have a different name.
All three motors where used in the testing phase with the three different ESCs. The
corresponding propellers tested are:
3 Bladed Pusher Propeller
Pusher Propeller
Tractor Propeller](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-199-320.jpg)
![Design and Development of a Hybrid UAV
Bennie M.
199 | P a g e
ME5308 – Major Group Project
The live stream board had to position so that it would not cause any radio frequency
interference (RFI) or EMI with either the radio receiver or the GPS/Compass uBlock
which was mounted on the wing.
GPS & Compass installation
The GPS/Compass uBlock was mounted on the wing. This was due to it
needing to be exposed to the environment to get a GPS lock and maintain said lock.
This also was done to prevent EMI caused by the EMF produced by DC current
sources such as the PDH. The uBlock module is held in place with pvc tape.
Sonar Installation
The ultrasonic range finder (sonar) was also placed in the bottom mid-section
of the fuselage this allowed for the sonar to be directed directly to the ground. Due to
EMI it was necessary to modify the sonar to reduce this effect by using a 100uF
Capacitor, a 10Ohm resistor, and shielded cabling. This solution was recommend by
the manufacturer of the ultrasonic range finder MaxBotix [97]. At this position the
sonar would also be fair away from any air turbulence created by the rotors which
could cause acoustic or mechanical noise that could result in inaccurate readings.
Flight Control System & Air speed sensor Installation
The flight control system (APM) was installed in the rear section of the fuselage
under the wings. A mounting tray was fabricated for it and a male and female Velcro
strip was used to secure the APM into its position. The tray was placed in an
elevated position in the fuselage to avoid as much as possible the EMI from the PDH
cables running at the bottom of the fuselage.
The Airspeed sensor came in three parts, the Pitot tube, silicon tubes, and airspeed
sensing chip. The Pitot tube was connected to one and of the silicon tubes and a
custom mounting brace was made to keep the Pitot tube outside of the boundary
layer when the UAV was in flight. The tubes were then connected to the airspeed
chip which was placed inside the fuselage on the top mounting tray. This was then
connected to the APM via a servo cable.
ESC Installation and calibration
ESCs are known to produce high levels of heat and EMF. For this reason the
ESCs had to be positioned; where possible, far away from sensitive systems. Two](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-219-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
218 | P a g e
ME5308 – Major Group Project
13. Bibliography
[1] “Flight Global,” 26 August 2011. [Online]. Available:
http://www.unmanned.co.uk/unmanned-vehicles-news/unmanned-aerial-
vehicles-uav-news/iai-completes-development-work-on-mini-panther-uas/.
[Accessed 19 Oct 2013].
[2] H. Stone and G. Clarke, “T-Wing VTOL UAV: Technology Deomnstrator,” The
University of Sydney, Sydney, 2002.
[3] R. Gallagher, “Coming Soon to a Sky Near You: a Pentagon-Backed "Bat
Drone",” 15 March 2013. [Online]. Available:
http://www.slate.com/blogs/future_tense/2013/03/15/v_bat_drone_from_mlb_c
ompany_can_conduct_urban_surveillance_wildlife_monitoring.html.
[4] MLB Company, “Products,” 2013. [Online]. Available:
http://spyplanes.com/products-v-bat/. [Accessed 19 Oct 2013].
[5] S. Trimble, “Flight Global,” Reed Business Information, 10 April 2013. [Online].
Available: http://www.flightglobal.com/news/articles/laad-santos-lab-unveils-
new-uavs-signs-insitu-pact-384491/. [Accessed 4 October 2013].
[6] T. Skinner, “Shepard Media,” Shepard Press Limited, 11 April 2013. [Online].
Available: http://www.shephardmedia.com/news/uv-online/laad-2013-santos-
lab-unveils-vtol-uav/. [Accessed 19 October 2013].
[7] E. Schechler, “Popular Mechanics,” Hearst Communication INc., 15 August
2013. [Online]. Available:
http://www.popularmechanics.com/technology/aviation/news/this-unmanned-
plane-can-land-on-a-helo-pad-15812116. [Accessed 15 October 2013].
[8] B. Coxworth, “ Gizmag Aircraft,” 20 August 2013. [Online]. Available:
http://www.gizmag.com/hybrid-quadrotor-hq-uav/28767/. [Accessed 19 Oct
2013].](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-238-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
219 | P a g e
ME5308 – Major Group Project
[9] Advanced VTOL Technologies, “AVT Hammerhead V2,” AVT , 2013.
[10] Advanced VTOL Technologies, “Hammerhead VTOL Aircraft,” 2013. [Online].
Available: http://www.avtolt.com/products/hammerhead.php. [Accessed 19 Oct
2013].
[11] Naval Technology, “Projects,” 2013. [Online]. Available: http://www.naval-
technology.com/projects/belleagleeyeuav/. [Accessed 21 October 2013].
[12] Federation of American Scientists, “FAS Intelligence Resource Program,” 27
November 1999. [Online]. Available:
http://www.fas.org/irp/program/collect/eagle-eye.htm. [Accessed 19 October
2013].
[13] R. Coppinger, “Flight Global UAVs,” 20 March 2008. [Online]. Available:
http://www.flightglobal.com/news/articles/fully-autonomous-quad-tilt-wing-uav-
makes-maiden-flight-222373/. [Accessed 21 October 2013].
[14] Chiba University Nonami Laboratory, “Unmanned Aerial Vehicle Gallery,” 2013.
[Online]. Available: http://mec2.tm.chiba-
u.jp/~nonami/english/research/uav/gal.html. [Accessed 2013].
[15] AV AeroVironment, “AV AeroVironment, Raven,” AeroVironment, [Online].
Available: http://www.avinc.com/uas/small_uas/raven/.
[16] BAYKAR, “BAYKAR Mini UAS,” BAYKAR, [Online]. Available:
http://www.baykarmakina.com/en-EN/category/10/Systems/1/bayraktar-mini-
iha.aspx.
[17] N. Mckeegan, “BATMAV System with AeroVironment's Wasp III MIcro Air
Vehicle achieves full rate production,” Gizmag, 8 January 2008. [Online].
Available: http://www.gizmag.com/aerovironment-wasp-iii-micro-air-
vehicle/8605/.
[18] INNOCON, “INNOCON Micro Falcon Brochure,” INNOCON Ltd.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-239-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
220 | P a g e
ME5308 – Major Group Project
[19] J. T. VanderMey, “A tilt rotor UAV for long endurance operations in remote
environments,” Massachusetts Institute of Technology, 2011.
[20] Civil Aviation Authority, “CAP 722- Unmanned Aircraft System Operations in
UK Airspace - Guidance,” West Sussex, The Stationery Office (TSO), 2012.
[21] Civil Aviation Authority, “Air Navigation: The Order and the Regulations,”
London, TSO (The Stationery Offi ce), 2014.
[22] D. P. Raymer, Aircraft Design a Conceptual Approach, 5th Revised edition
edition ed., American Institute of Aeronautics & Astronautics, 2012.
[23] Transonic Aerodynamics Committee, “Aerodynamic principles of winglets -
ESDU 98013,” Engineering Sciences Data Unit (ESDU), 1998.
[24] San Diego Air & Space Museum, “San Diego Air & Space Museum Archives,”
U.S. Army, 1958. [Online]. Available:
http://www.flickr.com/photos/sdasmarchives/4564654232/. [Accessed 2013].
[25] S. Nathan, “Flaviir project trials new forms of wing command,” The Engineer,
12 July 2010. [Online]. Available: http://www.theengineer.co.uk/in-
depth/analysis/flaviir-project-trials-new-forms-of-wing-
command/1003464.article. [Accessed 10 October 2013].
[26] M. James, “Nextcraft.com,” 2012. [Online]. Available:
http://nextcraft.com/aircraftdesignconcepts.html.
[27] W. H. H. D. T. P. Jack D. Mattingly, Aircraft Engine Design 2nd Edition,
Washington D.C: American Institute of Aeronautics and Astronomics Inc.,
2003.
[28] D. S. M. Niƫă, “Estimating The Oswald Efficiency Factor From Basic Aircraft
Geometrical Parameters,” Hamburg University, Hamburg, Germany, 2012.
[29] D. P. Raymer, Aircraft Design: A Conceptual Approach, Washington D.C.:
American Institute of Aeronautics and Astronautics Inc., 1989.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-240-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
221 | P a g e
ME5308 – Major Group Project
[30] D. J. Roskam, Airplane Design: Part I Preliminary Sizing of Airplanes, Kansas
USA: DARCorp, 2005.
[31] S. Gudmundsson, General Aviation Aircraft Design Applied Methods and
Procedures, Oxford: Elsevier, 2013.
[32] FAA, “Aviation Maintenance Technician Handbook - Airframe,” United States
Department of Transportation, Federal Aviation Administration, Airman Testing
Standards Branch, Oklahoma, USA, 2012.
[33] L. Dimitri, “Numerical Investigation of Blended Winglet Effects on Wing
Performances,” Aalborg Univerisity, Aalborg, Denmark, 2008.
[34] J. J. a. S. M. S. Guglielmo, “High-Lift Low Reynolds Number Airfoil Design,”
Journal of Aircraft, vol. 34, no. 1, January-February 1997.
[35] D. J. Roskam, Airplane Design: Part VII Determination of Stability, Control and
Performance Characteristics, Kansas USA: DARCorp, 2006.
[36] D. J. Roskam, Airplane Design: Part VI Preliminary Calculation of
Aerodynamics, Thrust and Power Characteristics, Kansas USA: DARCorp,
2004.
[37] P. F. J. J. I. A. Christian Bak, “Wind Tunnel Test of the NACA 63-415 and a
Modifie NACA 63-415 Airfoil,” Risø National Laboratory, Roskilde, Denmark,
2000.
[38] MIT, “www.mit.edu,” 6 April 2006. [Online]. Available:
http://ocw.mit.edu/courses/aeronautics-and-astronautics/16-01-unified-
engineering-i-ii-iii-iv-fall-2005-spring-2006/systems-labs-06/spl8.pdf. [Accessed
October 2013].
[39] A. K. Kundu, Aircraft Design, vol. 1, Cambridge: Cambridge University Press,
2010, pp. 180-185.
[40] M. Sadraey, “Chapter 12: Design of Control Surfaces,” in Aircraft Design: A](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-241-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
222 | P a g e
ME5308 – Major Group Project
Systems Engineering Approach, Wiley Publications, 2012.
[41] D. M. Jabbal, Advanced Aerodynamics Note Lecture 2 "Vortex Lattice
Numerical Method", 2013.
[42] D. Y. Liu, “MIT Marine Hydrodynamics Lecture 19,” 2009. [Online]. Available:
http://web.mit.edu/2.20/www/lectures/lec19/lecture19.pdf..
[43] R. Hendrickson, D. Roman and D. Rajkovic, “Drag: An Introduction,” 2006.
[Online]. Available:
http://www.dept.aoe.vt.edu/~mason/Mason_f/ConfigAeroDrag.pdf.
[44] D. MacWilkinson, W. Blackerby and J. Paterson, “Correlation of full-scale drag
predictions with flight measurements on the C-141A aircraft. Phase II wind
tunnel test, analysis and prediction techniques. Volume 1 – Drag Predictions,
Wind tunnel data analysis and correlation,” 1974.
[45] P. M. Cavcar, “Minimum Drag Speed,” Eskisehir, Turkey, 2004.
[46] D. S. Miklosovic, Drag on basic shapes, 2004.
[47] S. Gudmundsson, General Aviation Aircraft Design: Applied Methods and
Procedures, Elsevier, 2013.
[48] ESDU, “SLOPE OF LIFT CURVE FOR TWO-DIMENSIONAL FLOW,” ESDU,
1955.
[49] ESDU, “RATE OF CHANGE OF LIFT COEFFICIENT WITH CONTROL
DEFLECTION IN INCOMPRESSIBLE TWO-DIMENSIONAL FLOW,” ESDU,
1956.
[50] L. J. W. J. B. Laurence K, “The effects of variations in reynolds number
between 3 to 25 million upon the aerodyamic characteristics of a number of
NACA 6 series airfoil sections.,” Virginia, USA, 1950.
[51] D. P. Raymer, Aircraft Design: A Conceptual Approach, AIAA Education
Series, 2012.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-242-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
223 | P a g e
ME5308 – Major Group Project
[52] “xflr5-About stability analysis using XFLR5,” A. Deperrois, November 2010.
[Online]. Available:
http://www.xflr5.com/docs/XFLR5_and_Stability_analysis.pdf.
[53] “ Longitudinal stability and control,” [Online]. Available: www.aerostudents.com.
[54] Federal Aviation Administration, “Pilot's Handbook of Aeronautical Knowledge,”
in Chapter 05: Flight Controls , Washington, Federal Aviation Administration
(FAA).
[55] National Aeronautics and Space Administration, “Vertical stabilizer - rudder,”
March 2014. [Online]. Available: http://www.grc.nasa.gov/WWW/k-
12/airplane/rud.html. [Accessed 13 September 2010].
[56] G. Shpati, “Aircraft CG Envelopes,” 17 September 2011. [Online]. Available:
http://www.sawe.ca/download/tech2011/Aircraft%20CG%20Envelopes.pdf.
[Accessed 10 February 2014].
[57] D.-W. Yoo, H.-D. Oh, D.-Y. Won and M.-J. Tahk, “Dynamic Modeling and
Stablization Techniques for Tri-Rotor Unmanned Aerial Vehicles,” vol. 11(3),
no. 167, 2010.
[58] B. Mettler, Identification Modeling and Characteristics of Miniature Rotorcraft,
Boston: Kluwer Academic Publishers, 2003.
[59] D. Moell and J. Nordin, “VTOL UAV - A Concept Study,” Linkopings Universitet,
Linkoping, 2008.
[60] J. Roskam, Airplane Design Part III: Layout Design of Cockpit, Fuselage, Wing
and Empennage: Cutaways and Inboard Profiles, Darcorporation, 1986.
[61] F. O. P. P. G. School, “Aircraft Structure - An introduction to major airplane
components,” [Online]. Available: http://www.free-online-private-pilot-ground-
school.com/aircraft-structure.html. [Accessed 15 February 2014].
[62] F. A. Ltd., “SkyRanger Construction,” Flylight Airsports Ltd., [Online]. Available:
http://www.flylight.co.uk/skyranger/construction.htm. [Accessed 1 February](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-243-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
224 | P a g e
ME5308 – Major Group Project
2014].
[63] “Chapter 4: Basic Aircraft Construction,” [Online]. Available:
http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0
CCsQFjAA&url=http%3A%2F%2Fwww.ae.metu.edu.tr%2F~ae362%2FAircraft
%2520Basic%2520Construction.pdf&ei=XAoiU6OyD-
Gc0QXSrIHwCg&usg=AFQjCNHXUWY-7e3WbznllYgtQ5Q399N83g.
[64] Flight Global , “Flight Global Cutaways,” [Online]. Available:
http://www.flightglobal.com/cutaways/. [Accessed 20 November 2013].
[65] J. Fielding, C. Lawson, R. Pires and G. Monterzino, “Development of the
Demon technolody demonstrator UAV,” in 27th International Congress of the
Aeronautical Sciences, 2010.
[66] BormaTec, “UAV MAJA,” Bormatec Unmanned Vehicles, [Online]. Available:
http://www.bormatec.com/index.php?option=com_content&view=article&id=78
&Itemid=59. [Accessed 10 October 2013].
[67] Range Video, “RV Jet Airframe Manual,” Range Video, 2012.
[68] A. Daviel, “Building a Popc,” March 2004. [Online]. Available:
http://andrew.triumf.ca/andrew/popsicle-bridge/ . [Accessed 10 December
2013].
[69] “Basis of Structural Design: Course 2 Structural Action: cables and arches,”
[Online]. Available: http://www.ct.upt.ro/users/AurelStratan/bsd%5Cc02-
bsd.pdf. [Accessed 20 October 2013].
[70] D. P. Raymer, Aircraft Design: A Conceptual Approach, Reston, Virginia:
American Institute of Aeronautics and Astronautics Inc., 2006.
[71] R. S. e. a. S. Brandt, Intorduction to Aeronautics: A Design Perspective,
Reston: American Institute of Aeronautics and Astronautics Inc, 2004, pp. 350-
354.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-244-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
225 | P a g e
ME5308 – Major Group Project
[72] canply.org, “CANPLY Plywood Design Fundamentals,” 2012. [Online].
Available: www.tolko.com/documents/plywood_designfund.pdf. [Accessed 20
January 2014].
[73] Federal Aviation Administration, “Aviation Maintenance Technician Handbook,
Chapter 1: Aircraft Structures,” Federal Aviation Administration, 2013.
[74] Civil Aviation Safety Authority, “DESIGN STANDARDS: UNMANNED AERIAL
VEHICLES - AEROPLANES,” Civil Aviation Safety Authority, 2000.
[75] Federal Aviation Administration, “Chapter 02: Aircraft Structure,” in Pilot's
Handbook of Aeronautical Knowledge, Federal Aviation Administration, 2013.
[76] Nitroplanes.com, “Nitroplanes.com's,” Nitroplanes.com's , [Online]. Available:
http://www.nitroplanes.com/transall160.html. [Accessed 18 03 2014].
[77] G. Landolfo, “AERODYNAMIC AND STRUCTURAL DESIGN OF A SMALL
NONPLANAR WING UAV,” UNIVERSITY OF DAYTON, Dayton, Ohio, 2008.
[78] BillKits, “billkits.com/,” BillKits, 2012. [Online]. Available:
http://www.billkits.com/. [Accessed 18 03 2014].
[79] D. W. Song, “Composite materials lecture notes part 3 Manufacturing
Processing,” Brunel University, London, 2013.
[80] Federal Aviation, “Chapter 3: Aircraft Fabric Covering,” in Aviation Maintenance
Technician Handbook - Airframe, Federal Aviation, 2013.
[81] RangeVideo, “RangeVideo.com,” RangeVideo, 2013. [Online]. Available:
http://www.rangevideo.com/en/18-rvjet. [Accessed 18 03 2014].
[82] J. Woodward, “Matching Engine and Propeller,” The University of Michigan,
1973.
[83] MIT, “DC Motor/ Propeller Matching,” MIT, 2005.
[84] “Aircraft Datasheet,” 2013. [Online]. Available: http://aircraft-
world.com/prod_datasheets/hp/emeter/hp-propconstants.htm. [Accessed 2 Nov](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-245-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
226 | P a g e
ME5308 – Major Group Project
2013].
[85] D. Raymer, Aircraft Design: A Conceptual Approach, AIAA, 2012.
[86] MaxAmps, “MaxAmps,” MaxAmps, [Online]. Available: www.maxamps.com.
[Accessed 10 November 2013].
[87] E. Cetinsoy, S. Dikyar, C. Hancer, E. Oner, E. Sirimoglu, M. Unel and M. Aksit,
“Design and contstruction of a novel quad tilt-wing UAV,” vol. 26 Nov 2012,
2012.
[88] G. Owen, “How to build your own quadcopter flight controller,” 2013. [Online].
Available: http://ghowen.me/build-your-own-quadcopter-autopilot/. [Accessed
10 October 2013].
[89] J. Gauterin, “StackOverlow,” 9 December 2009. [Online]. Available:
http://stackoverflow.com/questions/1840314/when-do-i-need-to-use-
quaternions. [Accessed 10 October 2013].
[90] B. Horn, Some Notes on Unit Quaternions and Rotation, MIT, 2001.
[91] Mathworks, “Tuning Multi-Loop control systems,” 2014. [Online]. Available:
http://www.mathworks.co.uk/help/robust/examples/tuning-multi-loop-control-
systems.html. [Accessed 2014].
[92] D. Heath-Caldwell, “Relay code,” 2014.
[93] AdruCopter, “Adrucopter- Instructions,” [Online]. Available:
http://copter.ardupilot.com. [Accessed 2 March 2014].
[94] G. Hunt, “How to select a DC Motor,” Micro Motion Solutions, 2013.
[95] Aniss1001, “Adruino Hardware Interfereence,” 2009. [Online]. Available:
http://forum.arduino.cc/index.php/topic,14146.0.html. [Accessed Dec 2013].
[96] J. Dang and H. Laheij, “Hydrodynamic Aspects of Steerable Thrusters: Session
Thrusters/Propulsion,” Wartsila propulsion Netherlands BV, Drunen,](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-246-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
227 | P a g e
ME5308 – Major Group Project
Netherlands, 2004.
[97] MaxBotix, “MaxSonar Operation on a multicopter,” MaxBotix, Dec 2013.
[Online]. Available: http://www.maxbotix.com/articles/067.htm. [Accessed Jan
2014].
[98] S. Shrouder, “DEMON UAV - FLYING WITHOUT FLAPS,” BAE Systems,
[Online]. Available:
http://www.baesystems.com/product/BAES_051903/demon-uav---flying-
without-
flaps.?_afrLoop=162667540574000&_afrWindowMode=0&_afrWindowId=null&
baeSessionId=1pRTTbnQpDRkwNHLrVhCfXQvbyJJ5npWKxYp3Cxpylh1sG1n
5ktS!-909532306#%40%3F_afrWindowId%3Dnull%26baeSession. [Accessed
15 October 2013].
[99] D. J. Roskam, Airplane Design: Part III Layout of Cockpit, Fuselage, Wing and
Emepennage, Kansas USA: DARCorp, 2002.
[100] D. J. Roskam, Airplane Design: Part II Preliminary Configuration Design and
Integration of Propulsion Systems, Kansas USA: DarCorp, 2004.
[101] D. J. Roskam, Airplane Design: Part IV Layout of Landing Gear and Systems,
Kansas USA: DARCorp, 2004.
[102] D. J. Roskam, Airplane Design: Part V Component Weight Estimation, Kansas
USA: DARCorp, 2003.
[103] D. J. Roskam, Airplane Design: Component Weight Estimation, Kansas USA:
DRACorp, 2003.
[104] DARCorporation, Advanced Aerodynamics Analysis Help, Kansas USA:
DARCorp, 2013.](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-247-320.jpg)
![Design and Development of a Hybrid UAV
Carlos C.M.
231 | P a g e
ME5308 – Major Group Project
Figure 144 To obtain for Step 4 in Table 17 [49]
Figure 145 To obtain for Step 5 in Table 17 [49].](https://image.slidesharecdn.com/ba17c8f4-d1ff-4655-87ff-8d640ca3ddec-150317042234-conversion-gate01/85/Design-and-Development-of-a-Hybrid-UAV-251-320.jpg)
Design and Development of a Hybrid UAV
- 1. Design and Development of a Hybrid UAV BRUNEL UNIVERSITY Design and Development of a Hybrid UAV Abbinaya T Jagannathan (1037583) Arturs Dubovojs (1008780) Bennie Mwiinga (1020511) Brett McMahon (0813201) Carlos B. Calles Marin (1018922) Camilo Vergara (1010295) Primary Supervisor: Prof. Ibrahim Esat Secondary Supervisor: Dr Mark Jabbal Word Count: 45,854
- 2. Design and Development of a Hybrid UAV Arthur D. i | P a g e ME5308 – Major Group Project Abstract The report of the project describes various design stages in detail as it was carried out from conceptual design stage all the way to the final aircraft testing. It describes the unique concept of fixed wing aircraft hybridised with tri-copter into a hybrid UAV. The report describes how the configuration of such aircraft was achieved through careful design stages, build and implementation, testing and further improvements and suggestions.
- 3. Design and Development of a Hybrid UAV Brett M. Acknowledgements The group would like to thank our supervisors and the lab technicians for their understanding, advice and assistance during the design and build of the aircraft. We owe a big thank you to lecturers Ibi Esat and Mark Jabbal for taking the time to meet with the team on a weekly basis to discuss problems and solutions during the aircraft’s design. We thank the lab technicians, in particular Kevin Robinson in the aerospace lab for his endless advice and assistance and good humour in building the aircraft. Additionally we thank the technicians Keith Withers, Steven Riley and Chris Ellis for their assistance in manufacturing and testing many of the specialised components of the aircraft, often to short deadlines. We thank Dr Alvin Gatto for his advice and expertise in preparing for and performing the flight testing of the aircraft and finally, thank you to the many external part suppliers for their effort in delivering the parts needed to build the aircraft.
- 4. Design and Development of a Hybrid UAV iii | P a g e ME5308 – Major Group Project Statement of Relative Contributions Every contribution to this report has been clearly marked in the header of each page with the author’s first name and initials of the surname. I confirm that all work presented is original and where other sources or reports have been referred to in the text have been referenced appropriately. Abbinaya T Jagannathan Bennie Mwiinga Carlos B Calles Marin Arturs Dubovojs Camilo Vergara Brett McMahon Shown on the table below is a personal statement of each individual’s contribution and role during the project Name Responsibilities Abbinaya T Jagannathan In the first term I started getting involved with initial geometric and performance requirements of the aircraft and later started to move into aircraft tail sizing, aircraft stability and control surface sizing. During the second term, I was involved with design and testing of components and connections required for the aircraft. I was also involved with the build manufacture of the fuselage and other aircraft components. Overall, I was working mainly on the theoretical side in the first term and more towards the practical build in the second term Arturs Dubovojs
- 5. Design and Development of a Hybrid UAV iv | P a g e ME5308 – Major Group Project At the beginning of the project I have extensively contributed to different aspects of the design process by selecting and analyzing key aircraft parameters and researching various topics during the preliminary design phase such as suitable motors selection and combination of aircraft stability for tri-copter and fixed wing configuration. During the detail design phase, I have designed the tail of the aircraft as well as contributed to the rod to rod connections. I further Contributed by finding suitable manufacturers of the sourced parts and handled the orders as well as other segments building and testing of the aircraft. Bennie Mwiinga In the first term I was involved in the concept and preliminary aircraft design of the aircraft. I also concentrated on the research and procurement of flight control and avionics systems that would be on the UAV so that it would to be fully functional. I was involved in initial sizing focusing on the VTOL propulsion system and what would be needed to achieve a stable VTOL UAV. In the second term I was mainly focused on the preparation and installation of avionics and the FCS onto the UAV. Additionally working on developing computer testing codes needed for component tests such as Motor Characterization. Brett McMahon
- 6. Design and Development of a Hybrid UAV v | P a g e ME5308 – Major Group Project During the design and build phases of the project, I assisted in creating virtual models of the aircraft and its components using 3D CAD software including Autodesk’s Inventor, AutoCAD and Dassault Systems’ Solidworks. During the second term, I assisted with hands on building. Carlos B Calles Marin During the first term I was involved in the initial sizing and development of the aerodynamics of the aircraft, mainly focusing on the wing parameters, drag estimations and performance calculations. As the project evolved I helped with the motor selection and detail design of connectors and wing. As the group leader it was also my job to manage the other team members making sure there was active communication between design phases. Camilo Vergara During early stages of the project I was involved in the initial geometric sizing of the aircraft and components through the use of Computer Aided Design (CAD) software. As the design progressed I became responsible for all aspects of the aircraft fuselage design, in addition to estimation of weight and center of gravity. Throughout the whole of term two I was heavily involved with the build of the UAV, and ultimately also helped out with the avionics towards the end.
- 7. Design and Development of a Hybrid UAV vi | P a g e ME5308 – Major Group Project Contents Abstract....................................................................................................................... i Acknowledgements .....................................................................................................ii Statement of Relative Contributions...........................................................................iii List of Figure .............................................................................................................. x List of Tables.............................................................................................................xv Nomenclature..........................................................................................................xvii 1. Introduction ......................................................................................................... 1 1.1. Motivation...................................................................................................... 1 1.2. Project Description........................................................................................ 3 2. Literature Review ................................................................................................ 4 2.1. State of The Art Technology.......................................................................... 4 2.2. Market Analysis............................................................................................. 9 3. Requirements.................................................................................................... 12 3.1. Regulations ................................................................................................. 12 3.2. Aims and Objectives ................................................................................... 13 Aim.................................................................................................................... 13 Objectives ......................................................................................................... 13 3.3. Mission Profile............................................................................................. 14 4. Design Process................................................................................................. 16 4.1. Concept Design........................................................................................... 16 4.1.1. Individual Proposals ............................................................................. 16 4.1.2. Quality Function Deployment................................................................ 35 4.1.3. Group Concept ..................................................................................... 37 4.2. Preliminary Design ...................................................................................... 41 4.2.1. Weights................................................................................................. 43 4.2.2. Aircraft Sizing: Constraint Analysis....................................................... 44 4.2.3. Aerodynamics....................................................................................... 48 Lift Equation.......................................................................................................... 48 Wing Geometry..................................................................................................... 48 Aerofoil Selection.................................................................................................. 52 Tail Sizing............................................................................................................. 58 The tail configuration......................................................................................... 59
- 8. Design and Development of a Hybrid UAV vii | P a g e ME5308 – Major Group Project Volume Coefficients: ......................................................................................... 60 Optimum tail arm and tail plan form area: ......................................................... 61 Tail Aerofoil: ...................................................................................................... 62 Drag...................................................................................................................... 63 Control Surface Sizing: Wing ............................................................................ 70 Lift Curve Slope Numerical Prediction: Wing and Aileron/Flaps........................ 73 4.2.4. Centre of Gravity .................................................................................. 78 4.2.5. Stability and Control: Standard Take-Off and Landing ......................... 81 4.2.6. Stability: Vertical Take-Off and Landing ............................................. 100 4.2.7. Structures ........................................................................................... 113 4.2.8. Computer Aided Design and Technical Drawings .............................. 131 4.2.9. Propulsion........................................................................................... 139 5. Avionics and Flight Control ............................................................................. 150 Components ....................................................................................................... 150 Main Control Scheme and sensor array ............................................................. 151 PID Controller ..................................................................................................... 153 Related software and Full system schematic...................................................... 154 Sonar and Noise Reduction................................................................................ 155 APM Anatomy..................................................................................................... 156 Manual Transition ............................................................................................... 158 6. Component Testing......................................................................................... 160 6.1. Stress Tests .............................................................................................. 160 6.1.1. Rods ................................................................................................... 160 6.1.2. Connections........................................................................................ 166 6.2. Motor Characterisation.............................................................................. 170 6.2.1. Set-Up ................................................................................................ 170 6.2.2. Calibration of Equipment .................................................................... 174 6.2.3. Procedure for Testing ......................................................................... 174 6.2.4. EDF Results ....................................................................................... 177 6.2.5. Motor Results ..................................................................................... 179 7. Build & Manufacturing Methods & Materials in Chronological Order............... 182 7.1. Logistics .................................................................................................... 182 7.1.2................................................................................................................ 183
- 9. Design and Development of a Hybrid UAV viii | P a g e ME5308 – Major Group Project Key materials used.......................................................................................... 184 7.2. Fuselage ................................................................................................... 187 7.3. Connections .............................................................................................. 193 7.4. Wings & Tail.............................................................................................. 194 7.5. Propulsion ................................................................................................. 195 7.5.1. VTOL Propeller motors Mount............................................................ 195 7.5.2. Lander 90 mm EDF Mount ................................................................. 197 7.6. Avionics..................................................................................................... 198 Servo Installation............................................................................................. 198 Camera, Live stream & OSD Installation......................................................... 198 GPS & Compass installation ........................................................................... 199 Sonar Installation ............................................................................................ 199 Flight Control System & Air speed sensor Installation..................................... 199 ESC Installation and calibration ...................................................................... 199 Battery and Power Distribution Harness (PDH) Installation ............................ 200 Telemetry (MAVlink) Installation...................................................................... 200 8. After Build Testing........................................................................................... 201 8.1. Flight Tests................................................................................................ 201 8.1.1. Horizontal Flight Test 1....................................................................... 201 Horizontal Flight Test 2....................................................................................... 205 8.1.2. Vertical Flight Tests ............................................................................ 208 9. V-n Diagram.................................................................................................... 210 10. Budget.......................................................................................................... 212 11. Conclusion.................................................................................................... 214 12. Improvements and Further Research........................................................... 215 Landing Gear...................................................................................................... 215 STOL Propulsion ................................................................................................ 215 Power Plant Upgrade.......................................................................................... 215 Transmitter/Camera Range ................................................................................ 216 Aircraft Systems.................................................................................................. 216 Material upgrade................................................................................................. 216 Transition attempt............................................................................................... 216 Autonomous Flight capabilities ........................................................................... 216
- 10. Design and Development of a Hybrid UAV ix | P a g e ME5308 – Major Group Project Tilt rotors............................................................................................................. 217 Ultimate load testing on wings and fuselage....................................................... 217 Stress testing of 3D printed parts ....................................................................... 217 13. Bibliography.................................................................................................. 218 14. Appendices................................................................................................... 228 14.1. Appendix A – Technical Details ............................................................. 228 Roskam Constraint Analysis ........................................................................... 228 Wing Profile Analysis: Additional Aerofoils...................................................... 229 Comparison between the two analysis methods in XFLR5 ............................. 230 Results for Vortex Lattice Method (VLM) and the Panel Method with a percentage comparison................................................................................... 230 Motor Test Code ............................................................................................. 235 14.2. Appendix B – Project Plan and Management ........................................ 237 14.2.1. Gantt Chart...................................................................................... 237 14.2.2. Logistics .......................................................................................... 239
- 11. Design and Development of a Hybrid UAV x | P a g e ME5308 – Major Group Project List of Figure Figure 1 The IAI Mini Panther in level cruise flight [1] . .............................................. 4 Figure 2 Sitter type UAV the V-Bat from MLB [4]. ...................................................... 5 Figure 3 The Orbis from Santos Labs in hover [6]...................................................... 5 Figure 4 The Latitude Engineering HQ hybrid Prototype [8]....................................... 6 Figure 5 The Wingcopter V13CH in VTOL mode. ...................................................... 7 Figure 6 Advanced VTOL Technologies' Hammerhead [10] ...................................... 7 Figure 7 Bell Eagle Eye Tiltrotor UAV [11]. ................................................................ 8 Figure 8 QTW-UAV developed by Chiba University, Japan [14]. ............................... 8 Figure 9 Graph showing basic relation between small UAVs..................................... 9 Figure 10 Maximum Endurance vs. Maximum Take-off weight for a range of UAVs [19] ........................................................................................................................... 11 Figure 11: The STOL mission profile........................................................................ 14 Figure 12: The VTOL mission profile........................................................................ 15 Figure 13 approximate sketch of the concept idea................................................... 18 Figure 14 Individual Proposal Concept by Bennie Mwiinga...................................... 21 Figure 15 The Doak VZ-4 by Doak Aircraft Company [24]. ...................................... 21 Figure 16 Blown Flight Control Concept................................................................... 23 Figure 17Tube and Tray Fuselage Concept............................................................. 25 Figure 18 Tube and Tray Fuselage as Used by Hobby Flyers ................................. 26 Figure 19 Concept Sketch of an initial idea.............................................................. 29 Figure 20 Concept Design Sketch............................................................................ 33 Figure 21 A quality function deployment (QFD) Matrix............................................. 35 Figure 22 Flow Diagram showing Design Stages..................................................... 42 Figure 23 Force Diagrams: a) Forces on a climbing aircraft, b) Forces on aircraft at constant bank angle [27]. ......................................................................................... 44 Figure 24 Constraint Analysis for the UAV............................................................... 46 Figure 25 Wing Geometry, Note: dimensions in millimetres..................................... 50 Figure 26 Induced Drag Factor Vs. Taper and Aspect Ratio [28]............................. 51 Figure 27 Typical Cambered Aerofoil [31]................................................................ 52 Figure 28 Lift Coefficient Vs. Moment Coefficient Analysis of different profiles........ 53 Figure 29 Initial and Final profile Comparison. ......................................................... 54
- 12. Design and Development of a Hybrid UAV xi | P a g e ME5308 – Major Group Project Figure 30 XFLR results for the final wing configuration. (a) Moment Force and chord wise lift distribution. (b) Spanwise lift distribution. (c) ISO view of lift and lift distribution................................................................................................................ 55 Figure 31 Polars: (a) Variation of Lift coefficient with AoA. (b) Variation of the drag coefficient with lift coefficient. (c) Variation of lift to drag ratio with AoA. .................. 56 Figure 32: Tail design procedure as illustrated by Mohammad Sadraey. [40].......... 59 Figure 33 Total Drag Decomposition........................................................................ 63 Figure 34 Drag velocity curve................................................................................... 66 Figure 35 CD Vs. CL Polar for the wing and the aircraft........................................... 68 Figure 36 Comparison of the Lift Curve Slopes using different predicting methods: Online database, XFLR5 and ESDU sheets............................................................. 74 Figure 37 How to obtain Trailing Edge Angle ...................................................... 76 Figure 38 Wing Curve slopes with control surface deflections. ................................ 77 Figure 39 Force balance kit to acquire aircraft CG location...................................... 79 Figure 40 Front load with dual dead weight batteries and back-up 4000 mah main battery ...................................................................................................................... 80 Figure 41: Tail incidence angle vs. Moments generated. ......................................... 82 Figure 42: Graphs indicating the derivatives and for stable and instable aircraft conditions. .................................................................................................... 84 Figure 43 Wing and tail forces.................................................................................. 86 Figure 44 Statically Stable and Unstable pitching moment curves........................... 87 Figure 45 Final aircraft CG Lift configuration............................................................ 88 Figure 46: control surface effectiveness parameter vs. control surface to lifting surface chord ratio. [40]............................................................................................ 92 Figure 47 Shows the rudder curve slope with deflection angles of ±20 degrees...... 97 Figure 48 Shows the elevator curve slope with deflection angles of ±20 degrees. .. 97 Figure 49 Longitudinal CG Envelope for Project vehicle .......................................... 98 Figure 50 Tri-copter configuration with reference axes. ......................................... 101 Figure 51 Pitch up by using Rotor 1. ...................................................................... 102 Figure 52 Roll in the Clockwise direction................................................................ 102 Figure 53 Roll in the Counter Clockwise direction.................................................. 102 Figure 54 Yaw authority of a tri-copter. .................................................................. 103 Figure 55 Mass Flow of air through rotor in hover.................................................. 104 Figure 56 Altitude Hold (Hover) with all 3 rotors..................................................... 106
- 13. Design and Development of a Hybrid UAV xii | P a g e ME5308 – Major Group Project Figure 57 Mass Flow of air through rotor in vertical climb. ..................................... 107 Figure 58 A level vertical climb by the tri-copter..................................................... 108 Figure 59 Flow of air through the rotor in forward flight.......................................... 109 Figure 60 Rotor Disc showing Azimuth angle......................................................... 110 Figure 61 Full model of UAV at a hover. ................................................................ 111 Figure 62 Full model of UAV in transition............................................................... 111 Figure 63 UAV model in full horizontal flight........................................................... 112 Figure 64 Monocoque fuselage design [61] ........................................................... 114 Figure 65 Truss fuselage structure [32].................................................................. 114 Figure 66 Semi-monocoque Fuselage [32] ............................................................ 115 Figure 67 Global Hawk Cutaway [64]..................................................................... 115 Figure 68 Falco Cutaway diagram [64]................................................................... 116 Figure 69 Cutaway of the ScanEagle [64].............................................................. 116 Figure 70 Bonding in progress of the Demon UAV composite structure [65] ......... 117 Figure 71 Loading on a triangular structure [68]..................................................... 118 Figure 72Skeletal frame of the fuselage................................................................. 119 Figure 73 Landing Gear Positioning for Proper Weight Distribution [71] ................ 121 Figure 74 Moveable Landing Gear Concept........................................................... 122 Figure 75 ABS Landing Gear Mount - Broken During Aircraft Assembly................ 123 Figure 76 ANSYS principle stress analysis on bulkhead displaying key on the left 124 Figure 77 demonstration of typical wing structure [75] ........................................... 127 Figure 78 Single Spar Wing Connection ................................................................ 131 Figure 79 Double Spar Wing Connection............................................................... 132 Figure 80 Moveable Landing Gear Mount.............................................................. 133 Figure 81 Computational Stress Test Result for Basic Landing Gear Mount ......... 134 Figure 82 Computational Stress Test Result for Lightweight Landing Gear Mount 134 Figure 83 Nose Vertical Lift Fan Skeletal Structure................................................ 135 Figure 84 Detail View of the Tongue and Groove Assembly Method ..................... 135 Figure 85 Initial fuselage concept........................................................................... 136 Figure 86 First Full Group CAD Aircraft Design ..................................................... 136 Figure 87 Structure and connections of various components within the aircraft..... 137 Figure 88 Top, front and side views of the final CAD model................................... 138 Figure 89 T/W vs Maximum amp draw................................................................... 140 Figure 90 T/W vs EDF price ................................................................................... 141
- 14. Design and Development of a Hybrid UAV xiii | P a g e ME5308 – Major Group Project Figure 91 EDF unit weight vs Thrust Capability ..................................................... 141 Figure 92 STOL mission current comparison for the initial and final endurance calculations ............................................................................................................ 148 Figure 93 VTOL mission current draw comparison for the initial and final endurance calculations. ........................................................................................................... 148 Figure 94 General control scheme of the UAV [87]................................................ 151 Figure 95 ArduPilot Mega 2.6 from 3D Robotics .................................................... 152 Figure 96 Schematic of MPU-6000. ....................................................................... 152 Figure 97 Block diagram of tri-copter control include 2 gain values [57]. ............... 153 Figure 98 Control allocation by a controller on a tri-copter..................................... 153 Figure 99 Example of Cascade Control.................................................................. 154 Figure 100 Cascaded PID used by APM [88]......................................................... 154 Figure 101 MaxBotix XL MaxSonarEZL0. .............................................................. 155 Figure 102 Sonar EM Noise reduction modification. .............................................. 156 Figure 103 APM 2.6 anatomy................................................................................. 156 Figure 104 Phases of flight during the transition maneuver from hover to horizontal flight........................................................................................................................ 158 Figure 105 Cantilever Load Testing Arrangement.................................................. 160 Figure 106 Cantilever Physical Stress Test Results Graph.................................... 161 Figure 107 Three Point Physical Stress Test Results Graph ................................. 162 Figure 108 Three Point Physical Stress Test Results Graph ................................. 163 Figure 109 demonstration of carbon fiber rod deflection with cantilever point loading ............................................................................................................................... 165 Figure 110: Experimental setup of the test conducted (left) and a drawing of the component (right) ................................................................................................... 167 Figure 111: Load vs Tensile extension for the 10mm diameter hole ...................... 168 Figure 112: Load vs Tensile extension for the 20mm diameter hole. ..................... 169 Figure 113 Thrust bench and NI High USB carrier used for motor characterisation. ............................................................................................................................... 171 Figure 114 EDF Mount for thrust bench. ................................................................ 171 Figure 115 Motor Mount for thrust bench. .............................................................. 171 Figure 116 80A ESC Turnigy Superbrain............................................................... 172 Figure 117 Turnigy KV-RPM Meter. ....................................................................... 172 Figure 118 National Instruments Hi-Speed USB Carrier. ....................................... 172
- 15. Design and Development of a Hybrid UAV xiv | P a g e ME5308 – Major Group Project Figure 119 Turnigy 4000 mAh LiPO Battery (6s). .................................................. 172 Figure 120 PWM changing the angle of a dc motor [95]. ....................................... 173 Figure 121 Sample calibration curve for the test bench. ........................................ 174 Figure 122 Numeric Loading for EDF and trend line. ............................................. 177 Figure 123 Thrust Results for the EDF................................................................... 178 Figure 124 Thrust efficiency of two and three bladed propellers [96]. .................... 179 Figure 125 VTOL Motor test with different propellers............................................. 180 Figure 126 Current Draw of the motor for any given thrust. ................................... 181 Figure 127 Laser cutting the aft EDF bulkhead ...................................................... 187 Figure 128 Fuselage during initial Epoxy resin stage of construction (left), tilting Propeller mount (Right) .......................................................................................... 188 Figure 129 rear view of the front Bulkhead displaying the nose gear mechanism.. 189 Figure 130 drilling axle holes on the non-vertical mounting plate of the carbon fiber Landing gear .......................................................................................................... 191 Figure 131 Rear landing gear assembly................................................................. 192 Figure 132 Fuselage structure with back-up rear undercarriage (left), Nose gear (right)...................................................................................................................... 192 Figure 133: Schematic of assembly of the aluminum VTOL motor mounts............ 193 Figure 134: Load tests conducted on the P400 ABS plastic (left) and the 3mm (right) plywood motor mounts. .......................................................................................... 196 Figure 135 EDF Mount to the fuselage, Side view (left), top view (right)................ 197 Figure 136 Reinforced rear landing gear mount..................................................... 203 Figure 137 Strengthened Retro-fit Nose Landing Gear.......................................... 204 Figure 138 Second flight test ground roll demonstration ........................................ 205 Figure 139 Second flight test tip stall demonstration.............................................. 206 Figure 140 Second flight test landing stall demonstration ...................................... 207 Figure 141 UAV in Tri-copter mode........................................................................ 209 Figure 142 V-n Diagram and Gust Loading graph.................................................. 211 Figure 144 Roskam Constraint Analysis ................................................................ 228 Figure 145 To obtain for Step 4 in Table 17 [49]....................................... 231 Figure 146 To obtain for Step 5 in Table 17 [49].................................. 231
- 16. Design and Development of a Hybrid UAV xv | P a g e ME5308 – Major Group Project List of Tables Table 1: Sketch of concept design idea.................................................................... 16 Table 2: Key parameters of individual concept design ............................................. 17 Table 3 Key parameters of individual proposal ........................................................ 18 Table 4 Individual proposal by Bennie Mwiinga ....................................................... 20 Table 5 Individual Concept Design........................................................................... 28 Table 6 Individual Concept #4.................................................................................. 31 Table 7 Typical Aircraft Parameters. [26] ................................................................. 33 Table 8 House of Quality table, How’s vs How’s ...................................................... 36 Table 9 Group Concepts .......................................................................................... 38 Table 10 Constraint Analysis Equations, obtained from Mattingly et All [27]............ 45 Table 11 Constraint Analysis Parameters. ............................................................... 46 Table 12 Different Wing Geometry Design Aspects ................................................. 49 Table 13 Wing Geometry Parameters...................................................................... 50 Table 14: Effects of changes in tail volume coefficients ........................................... 61 Table 15 Drag Components of the aircraft for cruise, 22.2 m/s. ............................... 69 Table 16: Time to achieve specific bank angles...................................................... 71 Table 17 Process to attain the lift curve slopes of the wing and the deflected control surface. ..................................................................................................................... 75 Table 18 Parameters and Results............................................................................ 76 Table 19 Lift variation with control surface deflection............................................... 77 Table 20: Horizontal and vertical tail design details.................................................. 82 Table 21: Static and dynamic stability requirements. [40] ........................................ 83 Table 22: Methods of determining the location of neutral point [38] [53] .................. 85 Table 23: Control Surface Functions........................................................................ 89 Table 24: Rudder deflection required during various landing at various crosswind velocities. ................................................................................................................. 95 Table 25 Rudder and elevator curve slope results using ESDU method, to be used in the control surface sizing.......................................................................................... 96 Table 26 showing properties of similar thickness plywood material strength [72] .. 125 Table 27 properties comparison of foam core wing reinforced with carbon fibre spars to balsawood ribbed structure reinforced with carbon fibre spars [77] ................... 128 Table 28 Battery properties for a suitable range of products [86]........................... 146
- 17. Design and Development of a Hybrid UAV xvi | P a g e ME5308 – Major Group Project Table 29 VTOL and STOL endurance.................................................................... 149 Table 30 Necessary Avionics Components for the UAV. ....................................... 151 Table 31 APM Anatomy Glossary. ......................................................................... 157 Table 32: Results obtained from the stress test conducted on the 3D printed component. ............................................................................................................ 168 Table 33 Testing Procedure for Motor Test............................................................ 176 Table 34 List of Suppliers and any comments surrounding orders and components delivered................................................................................................................. 183 Table 35 Mid-Project Budget.................................................................................. 212 Table 36 Final-Project Budget................................................................................ 213 Table 37 Wing profile: Additional Analysis ............................................................. 229 Table 38 VLM and Panel Method Result comparison from XFLR5 ........................ 230 Table 39 STOL Mission profile and current specifications...................................... 232 Table 40 VTOL Mission profile and current specifications...................................... 232
- 18. Design and Development of a Hybrid UAV xvii | P a g e ME5308 – Major Group Project Nomenclature VTOL - Vertical Take-Off and Landing STOL - Standard Take-off and Landing EDF - Electronic Ducted Fan MAC - Mean Aerodynamic Chord CAD Computer Aided Design AR - Aspect Ratio Reynolds Number - Weight Force - Mass - Air density - Aspect Ratio of Horizontal Tail - Volume Coefficient of horizontal tail - Volume Coefficient of Vertical tail - Area of Horizontal Tail - Area of Vertical Tail - Optimum arm of the Horizontal Tail - Optimum arm of the Vertical Tail - Area of Wing - Centre of Gravity - Static longitudinal stability - Dynamic longitudinal stability - Static directional stability - Dynamic directional stability - Location of Neutral Point - Location of Centre of Gravity - Location of Aerodynamic Centre - Static Margin - Efficiency of stabiliser - Wing Curve Slope - Horizontal tail Curve Slope - Vertical tail Curve Slope - Aircraft static longitudinal Stability Derivative - elevator effectiveness directive – control surface chord effectiveness parameter - Wing Root Chord - Wing Tip Chord - Inboard location of ailerons - Outboard location of ailerons - Aileron deflection - Aircraft rolling moment coefficient - Approach Velocity - Rolling Moment - Steady state roll rate - Wing Area
- 19. Design and Development of a Hybrid UAV xviii | P a g e ME5308 – Major Group Project - Horizontal Tail Area - Vertical Tail Area Se - Elevator Area ce – Elevator Chord be - Elevator Span – Induced Drag - Bank Angle - Second moment of area ̇ - Steady State Roll Rate - Lift at Take-off - Rotational Velocity - Moments about the aerodynamic centre - Horizontal tail curve slope - Aircraft Lift coefficient at take-off - Maximum profile lift coefficient. – Wing angle of attack - Angle of attack - downwash angle - Horizontal tail incidence angle - Angle of attack horizontal tail - Elevator chord effectiveness parameter - Elevator deflection - Elevator effectiveness derivative - Elevator effectiveness derivative - Elevator effectiveness derivative - Static longitudinal stability derivative - Distance between aerodynamic centre and main landing gear - Distance between centre of gravity and main landing gear -Curve slope of wing-fuselage combination - Vertical distance between thrust provider and centre of gravity - Lift coefficient at cruise incidence angle - Lift coefficient at zero wing incidence angle - Crosswind velocity - Aircraft side force due to crosswind - Sideslip angle - aircraft sideslip derivative - aircraft sideslip derivative - Vertical tail lift curve slope - Aircraft control derivative - Efficiency of vertical tail - Vertical tail side wash gradient - Rudder deflection - Aircraft crab angle during crosswind landing - Centre of aircraft side projected area - Aircraft centre of gravity - Aerofoil training edge angle
- 20. Design and Development of a Hybrid UAV xix | P a g e ME5308 – Major Group Project - Slope of lift-coefficient curve with incidence for two-dimensional aerofoil in incompressible flow - Theoretical slope of lift-coefficient curve with incidence for two- dimensional aerofoil in inviscid, incompressible flow - Slope of lift-coefficient curve with control deflection for two- dimensional aerofoil in incompressible flow - Theoretical slope of lift-coefficient curve with control deflection for two-dimensional aerofoil in inviscid, incompressible flow
- 21. Design and Development of a Hybrid UAV Arturs D. 1 | P a g e ME5308 – Major Group Project 1. Introduction 1.1. Motivation The UAV industry is developing rapidly and currently is a very popular topic due to the broad variety of applications of this technology. This increase in popularity creates higher demands in the field and calls for constant technological advance. There have been a number of projects which have involved designing, building and programming of unmanned aerial, some of these types of projects concentrated on developing autonomous flight and obstacle avoidance techniques. Usually such projects concentrate on one aspect since it is very time consuming especially as a university project where time is very limited and not all of it be dedicated to a project. Combining few of such aspect together is a lot harder and challenging due to time limitations and limited resources. Airports and aircraft carriers take up a lot of space due to lengthy runways which, creates some problems finding the airfields launching aircrafts even for home built RC planes. On the other hand, fixed wing aircrafts are very efficient for distance travelling and staying in the air longer comparing to rotor crafts. After individual concept designs have been proposed, the group selected a collaborated idea and it was decided to design, build and program a hybridised UAV of VTOL aircraft and fixed wing aircraft. This idea combines both concepts and enables using the benefits of both. After further research into hybrid UAVs, the decision was to develop a combination of tri-copter with fixed wing aircraft. At the time there was a quad-copter hybridised with a fixed wing aircraft however, a tri-copter has not been done before. Quad-copter combined with a fixed wing aircraft has at least 5 thrust generators unless it is a tilt rotor where, tri-copter has one less motor which can decrease overall weight of the vehicle. Such design could be developed further to improve specifications and achieve better performance and parameters than the existing UAVs on the market. A project like this have not been done at Brunel University previously therefore, success of this project would be a great achievement for the university and could even attract publicity and improve university rating.
- 22. Design and Development of a Hybrid UAV Arturs D. 2 | P a g e ME5308 – Major Group Project The project’s success could give a big contribution to university’s teaching curriculum regarding UAVs and improve it for future students. Due to tight time constraints of this project there will be a lot of room for improvement of this project, further development and expansion therefore, this project could be used as a dissertation topic for future years for individuals as well as groups. Once the UAV is fully ready it could be used as a learning platform for students about UAVs. At last, a project like this is a great way to apply the knowledge gained through 4 years of university where theory is applied to a real life problem to which the solution is yet to be found. Not only it is a way to apply the knowledge but also, there is a lot to be learned during the course of the project, aspects which have not been covered during the course of education. Besides the application of theoretical knowledge it allows to compare the theoretical input to outcome of the result and feasibility of theory in practice. Most important such project would allow each group member to carry out self-assessment and evaluate what they have achieved over the 4 year period of the course.
- 23. Design and Development of a Hybrid UAV Arturs D. 3 | P a g e ME5308 – Major Group Project 1.2. Project Description Initially the project idea was to design and build a UAV however, each group member had a different idea and view of the project. After proposing individual concepts, a combined idea based on individual inputs of the group members was carried forward, to design a fixed wing aircraft with short take-off/landing (STOL) ability as well as a vertical take-off/landing (VTOL) capability. Further decisions were made to design an electrical vehicle rather than using gas/fuel due to health and safety regulations and limiting time constraints. The fixed wing part of the aircraft is straight forward, conventional concept which have been used for almost a century now however, for VTOL is there were few considerations such as the number of rotors and their configuration. The optimum tri-copter configuration was selected to decrease the stability complexity. Also, tilt rotor configuration was excluded due to its increased complexity with additional servos and mechanics for tilt mechanisms. So the final decided concept design was of a fixed wing aircraft with one thrust generator for horizontal flight combined with 3 vertical motors (tri-copter configuration) for VTOL. The final, fully developed aircraft was planned to have the option of programmable, fully autonomous flight which does not need external, manual inputs to operate as well as a remote control capabilities. The aircraft required to be equipped with a camera allowing live streamed video to the user for surveillance purposes. However, due to very limited time constraints of this project, the realistic objectives had to be decided which involved designing and building the aircraft with functioning tri-copter configuration as well as the fixed wing, horizontal flight configuration. The aircraft had to be remote controllable for both configurations. If the main objectives are achieve, the secondary, optional objectives such as transition segment between VTOL and horizontal flight can be worked on. If the main objectives are achieved the project would be considered successful.
- 24. Design and Development of a Hybrid UAV Bennie M. 4 | P a g e ME5308 – Major Group Project 2. Literature Review 2.1. State of The Art Technology The idea of an aerial vehicle that can perform both VTOL like a helicopter as well as STOL like a fixed wing aircraft is not a new one. However limitations in technology as well as the complexity of the associated control system has prevented widespread development of these types of vehicles especially those with an autonomous nature. Recently a handful of fully autonomous hybrids have been unveiled some are still prototypes while others are fully functioning production models. Below are some examples of these state of the art V/STOL aircraft. Being able to hybridize rotorcraft and fixed wing aircraft provides the opportunity for further applications of UAV/S in roles that would normally be exclusive to either one or the other. The mini panther is a smaller version of its larger relative ‘The Panther’ and weighs 12 kg (shown in figure 1). It is however the newest Iteration in the family to date. It has 3 propeller motors, the front two motors tilt upwards. In conjunction with the aft propeller; that is permanently in the normal position relative to the aircraft, the mini- panther is able to perform vertical take-off and transition into cruise, as well as transition from cruise to stable hover flight. The third motor on the aft section of the fuselage acts as the third arm of a tri-copter when the front two are tilted, this allows for yaw control in hover and vertical flight. Figure 1 The IAI Mini Panther in level cruise flight [1] . Another approach to producing a VTOL aircraft is the V-Bat from MLB (shown in figure 2). It is an alternate solution to a VTOL UAV as it is designed as a sitter type aircraft. As a sitter type aircraft the V-bat begins its flight in a vertical position on the ground and then hovers to a predetermined altitude [2]. At said altitude it is able to
- 25. Design and Development of a Hybrid UAV Bennie M. 5 | P a g e ME5308 – Major Group Project transition autonomously from hover to cruise and vice versa. Developed with funding from DARPA, the military version also includes a 6 foot extending arm to pick up objects whilst at hover close to ground level [3]. This UAV is capable of flight up to 15,000 feet altitude, with a maximum endurance of 10 hours. Figure 2 Sitter type UAV the V-Bat from MLB [4]. Similar to the MLB V-Bat are systems that only use a ducted body design (Figure 3). These systems are sitter-type VTOL that also have a number of rotors (normally 4) placed in an X or + formation similar to a quad rotor. The rotors and control surface are all enclosed within a duct body. These duct type UAV are able to transition from hover to horizontal flight by tilting themselves forward and increasing and vectoring the thrust generated by one motor. An example of this design is the Santos Lab Orbis which currently uses a hydrogen fuel cell and has a span of 3.8m [5]. Figure 3 The Orbis from Santos Labs in hover [6] It is well known that rotor craft are able to perform VTOL in the most efficient manner. By hybridizing a rotor craft with a fixed wing craft the benefits of both types of vehicles can be maintained. Latitude Engineering, a small drone company from
- 26. Design and Development of a Hybrid UAV Bennie M. 6 | P a g e ME5308 – Major Group Project Tucson, Arizona USA has developed such a hybrid [7]. Calling it the Hybrid Quad rotor (HQ) (shown in Figure 4 below) it is simply a hybrid between a quad rotor and a fixed wing aircraft. It weighs 27 Kg and has 4 electric motors that it uses to hover and 1 gas powered motor mounted on the aft of the aircraft to provide thrust for forward flight. The HQ is still in development but Latitude Engineering has been able to maintain a hover and transition to forward flight. Figure 4 The Latitude Engineering HQ hybrid Prototype [8]. Another project that has taken the same approach as Latitude Engineering is the Wing copter V13CH project by Jonathan Hesselbarth. The Wing copter (shown in Figure 5) also hybridizes a fixed wing and a quad rotor, however, the Wing copter utilizes its 4 electric brushless motors to produce thrust in forward flight and in VTOL. Utilizing a set of swivel arms; on which the four motors are mounted, the Wing copter is able to perform transition from hover to horizontal forward flight and vice versa. This a novel solution that also requires a control system that can keep the aircraft stable while transition is being performed. The transition performed by the Wingcopter however not an automated one is and is instead manually controlled (with some assistance from on board flight control). By utilizing the rotary heads on a radio control transmitter, the controller is able to vary the tilt of the four motors and transition into horizontal flight.
- 27. Design and Development of a Hybrid UAV Bennie M. 7 | P a g e ME5308 – Major Group Project Figure 5 The Wingcopter V13CH in VTOL mode. Similarly employing a tilting mechanism to achieve VTOL is the Hammerhead developed by David Howe & Lyndon Caine of Advanced VTOL Technologies [9] (shown in Figure 6). It has a canard that helps improve the UAVs stall characteristics as well as the ability to thrust vector in order to limit pitch divergence [9].The hammerhead employs twin counter rotating electric rotors stationed on a tilting stub wing assembly which AVT claims “minimises pitch, roll and yaw coupling” [9]. The hammerhead is capable of performing either STOL or VTOL. In STOL mode the tilt wing stub is positioned such that the rotors produce a thrust propelling the UAV forward. In VTOL mode the tilt wing stub is positioned vertically allowing the hammerhead to take off and hover like a helicopter. Figure 6 Advanced VTOL Technologies' Hammerhead [10] Another example of hybrid VTOL design is the Bell Eagle Eye UAV (see Figure 7). Developed by Bell Helicopter – Textron, Texas USA. It is a Tilt rotor UAV capable of VTOL which it achieves by tilting its nacelles in the appropriate direction to either perform VTOL or STOL. The Bell Eagle Eye has a payload weight of 90kg, a maximum speed of 200kt and an endurance of 8 hours [11] [12]. It also utilizes an automated flight control system to assist in performing transition.
- 28. Design and Development of a Hybrid UAV Bennie M. 8 | P a g e ME5308 – Major Group Project Figure 7 Bell Eagle Eye Tiltrotor UAV [11]. Use of tilting mechanisms can also be found in other hybrid VTOL aircraft such as the QTW-UAV by Chiba University, Japan (see Figure 8). The QTW-UAV uses 4 rotors placed on tilting wings that swivel to the vertical position to gain altitude and then swivel forward to transition to horizontal flight. It utilizes 4 electric rotors giving it a payload weight of 5kg, endurance of 15 min and a max speed of 81kt [13]. Figure 8 QTW-UAV developed by Chiba University, Japan [14].
- 29. Design and Development of a Hybrid UAV Camilo V. 9 | P a g e ME5308 – Major Group Project 2.2. Market Analysis In order to design a vehicle fit for purpose, market research was undertaken to find a suitable starting point to work to. Typical dimensions and functionalities of existing real world UAVs were used to compare the sizes the aircraft under development should be close to. Below is a diagram showing a triangular relationship between wing span, total length, and Max Take-off weight with a logarithmic scale. Figure 9 Graph showing basic relation between small UAVs AV RQ 11 B Raven [15] Bayraktar Mini UAS [16]
- 30. Design and Development of a Hybrid UAV Camilo V. 10 | P a g e ME5308 – Major Group Project AV Wasp III [17] Innocon Micro Falcon [18] The graph on Figure 9 Graph showing basic relation between small UAVs represents information gathered on similar sized small UAV platforms. This was used during the early stages of initial geometric sizing to ensure the values that were being calculated for the aircraft were within a set industry trend. In essence it was an early rudimentary matching plot to define design tendencies for wingspan, vehicle length, and Maximum Take-Off weight. Both the first converged group concept as well as the current project aircraft design are listed on the plot. An important performance parameter to note is that the cruise velocities were not included in analysis as values could not be found for all vehicles looked at. However when averaged out along other smaller UAVs, typical cruise velocities were around 20 . Once the vehicle size had been constricted to a hypothetical box, performance characteristics were then researched for a broad range of UAVs. Figure 10below was sourced out from a thesis from MIT, displaying the relationship between Maximum take-off weight and endurance in hours for a wide variety of UAVs. The theoretical project aircraft would lie between 1 kg to 10 Kg along the bottom axis. This range on the plot has a trend of a maximum endurance of 1-2 hours. Considering the vehicle in question would be a hybrid, the added dead weights, the use of all-electronic propulsion as well as a vertical flight mission segment which would drain a higher amount than the usual battery draw during straight level un- accelerated flight would all impact this maximum endurance value. As a result it would be expected to be significantly lower in reality.
- 31. Design and Development of a Hybrid UAV Camilo V. 11 | P a g e ME5308 – Major Group Project Figure 10 Maximum Endurance vs. Maximum Take-off weight for a range of UAVs [19] This market research of current UAVs in service was invaluable in producing information to use as a guideline to the aircraft design process. After every main phase was completed values of the aircrafts performance, and sizing was checked with current vehicles such as these. From this initial selection of fixed wing UAVs, the variety was narrowed down to include specialized aircraft which were applicable to similar mission profiles, which is to say a fixed wing UAV that has VTOL functionality.
- 32. Design and Development of a Hybrid UAV Abbinaya T.J. 12 | P a g e ME5308 – Major Group Project 3. Requirements 3.1. Regulations Every country has its own aviation regulations. The Civil Aviation (CAA) Authority states that any UAV exceeding a weight of 7 kg need to be certified or approved. Due to this, the main priority of the aircraft was its weight. It was decided that the aircraft’s maximum take-off weight was kept under 6.5 kg and this was ensured throughout the design process. However, the CAP 722 and CAP 393 Air Navigation Order states that aircraft that weigh less than 7 kg should also follow some regulations depending on whether they are being used for commercial purposes or not. The regulations for aircraft with a mass of less than 7 kg states that the aircraft should abide by appropriate operational constraints in order to ensure public safety. The regulations are based on the flying operation being conducted and the potential risks to any third party. General principles for UAV operations outside segregated airspace should follow an approved “detect and avoid” system and avoid crowded areas. It is also important for the aircraft to not fly beyond the visual line of sight. The CAP 722 also states that the aircraft should be flown such that the pilot controlling it can take manual control at any point of time and fly the aircraft out of danger. It also states that the aircraft should not be flown above 400 feet at any point of time. Further details of regulations applicable to this UAV can be found in the CAP 722 and CAP 393 of the Civil Aviation Authority Air Navigation Order. These regulations were kept in mind throughout the build of this UAV. [20] [21]
- 33. Design and Development of a Hybrid UAV Arturs D. 13 | P a g e ME5308 – Major Group Project 3.2. Aims and Objectives Aim To design, manufacture and test an aircraft of fixed wing configuration hybridised with a tri-copter. Horizontal flight capabilities of the aircraft have to be demonstrated as well as the VTOL capability using a remote control transmitter. . Objectives 1. To use aircraft design techniques and approaches to design a fixed wing tri- copter hybrid aircraft 2. To test and prove the suitability of load critical components of the aircraft 3. To build and test the final selected UAV design 4. To program the aircraft and have the avionics ready for both horizontal and vertical flight 5. Carry out a demonstrative flight for each of the configurations of the UAV
- 34. Design and Development of a Hybrid UAV Abbinaya T.J. 14 | P a g e ME5308 – Major Group Project 3.3. Mission Profile The purpose of the UAV is to be able to perform a surveillance and reconnaissance role. This requires the aircraft to be equipped with a camera and live stream capabilities. The main concept of this project is to design an aircraft that is capable of a Vertical Take-off & Landing as well as a Short take-off and landing (V/STOL). This would allow for the UAV to be launched from different environments where a runway is not available such as urban areas with limited airspace or launches from the sea. Due to the tight time schedule and the complexity of the project the transition between vertical and horizontal flight is not a priority for the project. Initially the UAV has to be able to perform a VTOL mission: take off vertically, climb, hover, climb further, hover at new altitude and descend to land. For the STOL mission it has to perform a separate mission profile where the aircraft has to: take-off, climb, cruise, loiter, descend and land. It is expected that parts of the mission profile segments are performed autonomously by an on-board autopilot. Figure 11 and 12 below illustrate the mission profiles allocated for the STOL and VTOL flight. The calculations of the mission profiles were done based on the current drawn from the batteries to be used. The total time of UAV operation for the STOL mission should be at least 9 minutes. The UAV should be able to operate for around 3 minutes when performing the VTOL mission. Figure 11: The STOL mission profile. 0 5 10 15 20 25 30 35 0 100 200 300 400 500 600 Altitude(m) Time (s) STOL mission profile
- 35. Design and Development of a Hybrid UAV Abbinaya T.J. 15 | P a g e ME5308 – Major Group Project Figure 12: The VTOL mission profile. Even though the transition is not a set priority for the project at the moment, it is most likely to be attempted once the VTOL and STOL missions have been successfully completed. In the case of the transition being attempted, the aircraft should be able to operate for around 4 minutes. In order to design and develop the UAV an appropriate design procedure must be performed. Since the project is aerospace orientated, the avionics and electronics to be used by the UAV are to be off-the-shelf components that are marginally modified to assist in the completion of the project objectives. The aircraft should be safe to operate, undergo a safety assessment and meet minimum requirements for this type of aircraft. The UAV is to be equipped with a flight control system capable of autonomous flight and manoeuvres. However, for the purposes of build and testing the UAV will be remotely controlled by a human pilot via a radio receiver and transmitter controller. 0 2 4 6 8 10 12 14 16 0 10 20 30 40 50 60 70 80 90 Altitude(m) Time (s) VTOL mission profile
- 36. Design and Development of a Hybrid UAV Abbinaya T.J. 16 | P a g e ME5308 – Major Group Project 4. Design Process 4.1. Concept Design 4.1.1. Individual Proposals Abbinaya T Jagannathan Table 1: Sketch of concept design idea The aim of this design concept is to have an autonomous UAV that is suitable for reconnaissance and surveillance purposes. The objective of this UAV design was to have an aircraft that is aerodynamically sound and also attempt to achieve a vertical/short take-off or landing. A push propeller was to be used at the fuselage to provide the main forward thrust. The integrated motors at the wing are meant to provide vertical thrust. Since the design consists of only 2 vertical thrust providers, the feasibility of VTOL is uncertain and if this is the case then, the goal is to achieve a short take-off by using the vertical thrust providers. The fuselage in this design is shaped as an aerofoil to have a body that is able to contribute to the lift force produced by the aircraft. An aerofoil with a high thickness to chord ratio is to be used for the fuselage. The figure above illustrates an idea of the UAV proposal. From the figure it can be seen that the other key aspects of this concept are the V-tail and the use of winglets. The V-tail was selected mainly because of the push propeller to be used at the rear end of the fuselage. Other reasons for V-tail selection are that it has a smaller size therefore; it will be lighter
- 37. Design and Development of a Hybrid UAV Abbinaya T.J. 17 | P a g e ME5308 – Major Group Project and have a smaller wetted area which would result in drag reduction. The V-tail configuration also uses fewer control surfaces compared to a conventional tail. These control surfaces are called ruddervators and are a combination of rudder and elevators. [22] The use of winglets was also considered in the concept for a number of reasons. Winglets are small wing-like lifting surfaces that are fitted at the tip of the wings for the purpose of reducing the trailing-vortex drag. As a result, this would increase the lift generated on the aircraft. [23] The materials considered for this design were a combination of foam and carbon composites (main airframe) which are both lightweight materials that can take high loads. Category Abbinaya T Jagannathan Mission Type Reconnaissance/ Surveillance Environment Outdoor Design Type Modular Modular Options Camera TO (Take-Off Type) V/STOL L (Landing) V/STOL Powerplant 1× Push Propeller & 2 × Rotor Integrated in Wings Wing Medium/High Fixed Wing Tail V-Tail Airframe Foam & Carbon Composite Landing Gear Fixed Endurance(Prospective) 60 min Altitude 50-100 m Glide Capability (Inc. Design) Yes Radio Controlled Back up Yes Autonomous Yes Table 2: Key parameters of individual concept design
- 38. Design and Development of a Hybrid UAV Arturs D. 18 | P a g e ME5308 – Major Group Project Arturs Dubovojs Figure 13 approximate sketch of the concept idea Category Arturs Dubovojs Mission Type Reconnaissance/ Payload delivery Environment Outdoor Design Type Modular Modular Options Cargo/Camera Take-Off Type Catapult Landing Parachute/STOL Power plant 1x Push Motor Wing High Fixed Wing/Joint Wing Tail Conventional/ Tailless (Joint Wing) Airframe Foam Landing Gear Fixed Endurance (Prospective) 2 hours Altitude 100m Glide Capability (Inc. Design) Yes Radio Controlled Back up Yes Autonomous Yes Table 3 Key parameters of individual proposal Initial idea was to Design and build a high endurance UAV with recon and payload delivery capabilities for outdoor use. The aircraft had to be able to carry a live
- 39. Design and Development of a Hybrid UAV Arturs D. 19 | P a g e ME5308 – Major Group Project streaming camera and a small payload. It has to be designed to be able to take off and land in a standard manner as well as been optimised for catapult mechanism and a parachute for emergency, vertical landing. Take-off and landing would require a landing gear, for weight reduction and less complexity a fixed landing gear would have been used. Aircraft required being equipped with only one push/pull propeller either on the nose or top of the fuselage with an electric motor capable of providing enough thrust. The aircraft would have to be able to carry out autonomous flight as well as a radio controlled option. After the basic mission and guidelines been set an investigation into wing types was carried out. A joint wing configuration was selected for the aircraft because of its ease of implementing it to a small scale UAV in comparison to a full scale aircraft. Since the aircraft should have a catapult mechanism it should be relatively small and compact for ease of deploying in unequipped circumstances, joint wing configuration reduces the required wing span. Since the joint wing configuration is relatively new invention it has not been implemented on many full scale aircrafts and there is mainly research being carried out on using it on smaller scale, high altitude UAVs. The joints between the wings would act as winglets, reducing induced drag. The option of having a tail additionally to the joint wing was still available. A tail would improve the aircrafts manoeuvrability by adding yaw capability which joint wing aircrafts lack. Also by making a separate control surface – tail, the second wing would become a lifting surface therefore, would generate more lift.
- 40. Design and Development of a Hybrid UAV Bennie M. 20 | P a g e ME5308 – Major Group Project Bennie Mwiinga Bennie Mwiinga Mission Type Recon/Surveillance Environment Outdoor & Urban Design Type Modular Modular Options Camera TO (Take-Off Type) V/STOL L (Landing) V/STOL Power-plant 3x Ducted Fans Wing Mid Fixed Wing Tail H-Tail Airframe Carbon Composite/Foam Landing Gear Fixed Endurance(Prospective) 180 Mins Altitude 60+ m Glide Capability (Inc. Design) No Radio Controlled Back up Yes Autonomous Yes Table 4 Individual proposal by Bennie Mwiinga The current military and commercial applications of UAV/S has increased in the past 14 years at an exponential rate. Also the environments in which these systems are expected to operate have changed and outdoor operations require novel design solutions in order to accomplish requirements such as quick deployment, long range and endurance. Military operations are more frequently being carried out in urban environments where close quarter combat is conduct. The ability to have a UAV that can be used in such an environment would assist troops in such operations by being able to be deployed on the spot to conduct reconnaissance and surveillance. This proposal is shown in Figure 14.
- 41. Design and Development of a Hybrid UAV Bennie M. 21 | P a g e ME5308 – Major Group Project Figure 14 Individual Proposal Concept by Bennie Mwiinga. In order to achieve these requirements a V/STOL type system is proposed. V/STOL would allow for the UAV to be deployed in any terrain or environment without the requirement of having a prebuilt runway. The V/STOL system in this proposal is achieved by having two ducted fans mounted on the wing tips that are able to tilt for vertical and horizontal flight. This approach was taken by the Doak Aircraft Company in developing their Doak VZ-4 (Figure 15) in 1958. Figure 15 The Doak VZ-4 by Doak Aircraft Company [24]. An additional ducted fan is then placed on the aft of the aircraft to provide additional forward thrust for STOL and transition to horizontal flight. The use of ducted fans may seem to be the wrong choice due to the lower efficiency when compared to a purely fan based propulsion. As can be seen in equation 4 Area (A) if increased will increase the amount of force created. A prop or fan has the advantage that it can have a wide area and move more mass of air and thus create more force. A ducted fan however has a smaller area and to move the same amount of air as a prop or fan it has to increase the rate at which it accelerates the air while moving a smaller amount.
- 42. Design and Development of a Hybrid UAV Bennie M. 22 | P a g e ME5308 – Major Group Project (4.1.1.1b) Where (4.1.1.2b) ∴ (4.1.1.3b) Where (4.1.1.4b) (4.1.1.5b) At the time the Doak was engineered this statement would be true, however, modern day ducted fans (electric) are designed and engineered to operate at higher efficiencies as loses are reduced by designing more aerodynamic shrouds and utilizing more efficient electric motors. These new generation of ducted fans have been used by many entities ranging from RC Hobbyists to UAVs developed by companies like Honeywell and Boeing. These ducted fans utilise higher efficiency motors capable of high rpm and specially designed props to produce high level performance. A modular type design would be used in this proposal. It would also allow for the UAV to service other sectors such as agricultural monitoring and scientific research, as the UAV would be capable of being outfitted with varying types of payloads such as FLIR or other EVS.
- 43. Design and Development of a Hybrid UAV Brett M. 23 | P a g e ME5308 – Major Group Project Brett McMahon Figure 16 Blown Flight Control Concept This design is inspired to some extent by the DEMON concept demonstrator [25] developed by BAe Systems, Cranfield and other universities which used jets of air to control the aircraft in flight. Using low profile wing tip fans, the aircraft will be able to perform roll manoeuvres using pulses of air from the appropriate fan. Short take off performance would be possible using both fans together to provide lift in addition to that provided by the main wing. The aircraft would be propelled through the air using a large pushing motor of sufficient power (with reserve) to achieve a desired flight speed. The aircraft would be all electrically powered for relative simplicity and safety compared to petrol powered motors and their required ancillaries. Two batteries are In-wing motors, provide short take off and airbourne roll control Camera provides visual target ‘hit’ data Rod and slot construction method, removable wings if needed Internal components laid out with respect to a selected Centre of Gravity position Internal avionics components mounted on aluminium cruciform, outer surfaces of foam or vacuum formed plastic shell pieces Infra-red range finder provides altitude data to flight controller
- 44. Design and Development of a Hybrid UAV Brett M. 24 | P a g e ME5308 – Major Group Project envisaged for the aircraft, one large capacity battery used for the forward thrust provider and wing tip motors while a second smaller battery would independently supply the flight control systems. The primary potential benefits of the wing tip fan arrangement is the level of simplicity that can be achieved over the otherwise complex arrangements associated with moving flight control surfaces, whilst also allowing for some moderate weight savings. The wings will be designed so they can provide the lift required with minimal drag at moderate flight speeds. Sizing of the aircraft must be sufficient such that the body can adequately house all the avionic components as well as provide a small degree of flexibility for adjustments (moving or addition of components) which might be needed during refinement. The wings must be able to generate more than enough lift required to support the aircraft’s mass, with a margin of safety for gusting conditions or lack of performance from the propelling motor. The flight control system will most likely be based on the Ardupilot series of control boards available at many remote control hobby shops. The benefit of these control systems is their general availability, relatively cheap price and extensive programming support from the remote control operator community. Power distribution, relays, wiring and programming must be defined separately and accounted for in the budget to be defined later. Materials will be determined that possess sufficient strength for the specified application whilst having minimal weight. All materials must be readily available from local suppliers and be sourced at the lowest price possible. Construction techniques used must be such that the overall component weights are kept as low as possible. Processes and tools required for fabrication and assembly must be simple and readily available through the university workshops or specially bought in by ourselves. Total aircraft cost must not exceed departmental budget constraints to be discussed with the project supervisors. After discussions within the group, the main problem with this aircraft concept was the general lack of elevator and tail sections required for pitch and yaw control. Roll control is managed primarily by the wing motors but additional aileron control surfaces may need to be added for higher speed flight, to be determined as the design progressed.
- 45. Design and Development of a Hybrid UAV Brett M. 25 | P a g e ME5308 – Major Group Project Conceptual Fuselage Design The fuselage is the main body of the aircraft. Depending on the aircraft layout, the fuselage is responsible for housing fuel, weapon stores, cargo, avionics equipment and passengers. For the UAV aircraft, the fuselage will house all of the electronic parts such as the flight control boards, navigation systems, cameras and radio receivers as well as the batteries. In addition to housing of the internal parts, the fuselage must also provide the fixing structure for the aircraft’s other component parts such as the wings, motors and the landing gear. The fuselage concept shown below in Figure 17 uses a thin walled outer skin and a removable avionics tray, which slides along runners extending the full length of the fuselage. This approach has previously been used by remote control aircraft builders as shown by Figure 18. The avionics tray allows for very tight packaging of the internal electronics and batteries of the aircraft and would be fully removable for servicing and adjustment, save for a few connections to the aircraft control surfaces such as ailerons, elevators, rudder and the motors. Figure 17Tube and Tray Fuselage Concept
- 46. Design and Development of a Hybrid UAV Brett M. 26 | P a g e ME5308 – Major Group Project Figure 18 Tube and Tray Fuselage as Used by Hobby Flyers Dependant on material availability and price, the thin walled (approximately 0.5 to 1mm thick) fuselage material of plastic or aluminium would allow for the required strength and low weight but could be further reinforced with the addition of stringers running top and bottom of the fuselage (along with the metal runners for the avionics tray located along the mid-line). This simple construction with easy access via the tray arrangement could be made almost entirely from off the shelf parts and materials. The wing box junction would have to be specially made to encompass the mid wing section. This junction would then be mated to the fuselage tubular section using fore and aft fastening wedges. Screws would be inserted through the fuselage skin and into the wedges, holding it (and therefore the wing box section) firmly in place. Toward the nose of the aircraft, a swivelling fan arrangement would be fixed between two bulkheads allowing the fan to be adjusted using a dedicated servo mechanism to counteract unwanted yaw from the other lift fans when the aircraft is in vertical flight. Further ahead of the swivelling fan arrangement would be a clear plastic nose cone, inside which the video camera and GPS receiver could be mounted, providing a clear view ahead and upward. The nose section would likely be a two part assembly made from either specially ordered injection moulded acrylic or made on campus using a vacuum forming method. This concept however was not used for the final aircraft as the required plastic or metal thin walled, large diameter fuselage tubing could not be sourced at reasonable cost. A case was therefore made for a bespoke fuselage design which could be
- 47. Design and Development of a Hybrid UAV Brett M. 27 | P a g e ME5308 – Major Group Project tailored to the changing requirements as the design and the parts specification evolved.
- 48. Design and Development of a Hybrid UAV Camilo V. 28 | P a g e ME5308 – Major Group Project Camilo Vergara Category Camilo Mission Type Recon/Multi-role Environment Outdoor Design Type Modular Modular Options Cargo/Camera TO (Take-Off Type) V/STOL L (Landing) V/STOL Powerplant 2x Tilt EDF & 1x Fixed VTOL EDF Wing Mid/High Fixed Wing Tail Boom tail Airframe Metal/Foam Landing Gear Fixed Endurance(Prospective) 90-120 Mins Altitude 100m Glide Capability (Inc. Design) Powered Radio Controlled Back up Yes Autonomous Yes Table 5 Individual Concept Design Every UAV design incorporates various design solutions and ideas as well as a set level of autonomy that is dictated by its mission parameters and operating environment. For the purpose of exploring different approaches to the design problem presented, various types of configurations were looked at. The potential for a VTOL system on a fixed wing reconnaissance drone is significant. Not only would it eliminate the need for a runway and be easy to retrieve which essentially makes it deployable from any location, but from an intelligence aspect, it would be able to do what no other normal type of fixed wing UAV could do, which is stop and hover mid-air over a point of interest allowing for a detailed inspection of
- 49. Design and Development of a Hybrid UAV Camilo V. 29 | P a g e ME5308 – Major Group Project the area ahead instead of having to perform circuits or flyby’s around a target. In order to achieve this, a particular solution was researched. This initial VTOL system came in the form of a tilt rotor design, which has its origins from the V-22 Osprey. Upon First glance there seems to be a very select few UAV’s currently on the market with this type of technology, the biggest being the Bell 'Eagle eye' which was a true representative of the twin tilt rotor Osprey. There are a couple more variations to mention, the first being Israeli Aerospace Industries ‘Panther’ VTOL UAV utilizing a 3 motor design with 2 being mounted on the wings, and a third around the rear section of the fuselage. Another worth mentioning is a prototype VTOL aircraft called the ‘phantom swift’ by Boeing which incorporated 4 ducted fans, two being located at opposite wing tips, and two being incorporated into the fuselage itself. Below is a diagram of the initial concept inspired from existing real world solutions. Figure 19 Concept Sketch of an initial idea The challenge of such a design would include aspects like the complexity of the control system integrated into the autonomous nature of the UAV. From a design perspective, several considerations must be taken into account for an aircraft of this nature. with regard to the power-plants themselves, this would include the position of the rotors from the longitudinal center of gravity, the connections between the servos and motors themselves, the physical connections to the wing or fuselage depending on the motors location, and the individual propeller rotation direction. The power output of the motors utilized would be a design aspect, due to the extra weight of the
- 50. Design and Development of a Hybrid UAV Camilo V. 30 | P a g e ME5308 – Major Group Project control system for a tilt rotor design, the motors selected must have enough power to lift the aircraft vertically, as well as perform well at horizontal flight. Electronic Ducted Fan (EDF) systems have rarely been used for VTOL hybrid applications; so the aircraft would be experimental by nature.
- 51. Design and Development of a Hybrid UAV Carlos C.M. 31 | P a g e ME5308 – Major Group Project Carlos Calles Marin Category Carlos Mission Type Recon/Surveillance Environment Outdoor Design Type Modular Modular Options Cargo/Camera TO (Take-Off Type) STOL L (Landing) STOL Power plant 1x Push Motor Wing High Fixed Wing Tail Boom Tail Airframe Balsawood Landing Gear Fixed Endurance(Prospective) 60 min Altitude 100 m Glide Capability (Inc. Design) Yes Radio Controlled Back up Yes Autonomous Yes Table 6 Individual Concept #4 The initial idea for the concept was very conservative. The initial requirements for the design were “very short landing and take-off or hand launched” and some aspect of autonomous behaviour. From the design point of view these are the different configurations considered: 1. Type of wing – High, Medium or Low 2. Power plant – Tractor, Pusher or both 3. Tail Type – V-Tail, Standard or Boom Tail 4. Wing – Sweep, Taper, Dihedral, Wash-in/out A high wing was chosen because the aircraft had to be possibly hand launched, which means that it needs good stability at low speeds until it reaches cruise. High mounted wings have better lateral stability than medium or low mounted. Considering the power plant the pusher configuration was chosen to improve the aerodynamic performance. The flow behind the propeller no longer has to flow over the wings, which would be the case with a normal tractor power plant. There are some situations were both types are used in to increase the thrust provided, acting in
- 52. Design and Development of a Hybrid UAV Carlos C.M. 32 | P a g e ME5308 – Major Group Project line with the centre of gravity. In this case it would not be necessary to have so much thrust. Regarding the type of tail needed V-Tail was quite interesting, reducing the amount of drag produced by the tail. For the pusher propeller configuration chosen a Boom- Tail is required to correctly place the motor. This doesn’t discard the V tail, but it changes it into inverted V. the problem with V tail is that it requires more expert knowledge of coding to have autonomous behaviour. The standard Boom Tail was chosen to avoid any control problems. In terms of wing design, sweep would not be an option because it is intended for high speed flight and this aircraft would fly at relatively small speeds. Taper would increment our performance, by reducing wing tip vortex downwash effect. The optimal wing shape would be elliptical to have a uniform span wise distribution of lift with the lowest induced drag possible. Due to its hard manufacturing process the elliptical wing shape can be approximated by a straight tapered wing, with a taper ratio of 0.3-0.4, hence it would be desirable to have a taper ratio around those values. The drawback of uniform lift distribution is that stall is reached evenly throughout the wing plan form; therefore washout would have to be considered to have more margin for error. Dihedral would increase the stability, but decrease the effective span of the aircraft, therefore it is not going not be part of the concept. To control the aircraft autonomously some readily available micro processing computers were thought of. There are two options which are Arduino and Raspberry Pi, these are open source platforms with widely available codes that can perform as an autopilot for the aircraft. Figure 20 is a representation of the initial concept where the taper, boom tail and pusher propeller can be observed.
- 53. Design and Development of a Hybrid UAV Carlos C.M. 33 | P a g e ME5308 – Major Group Project Figure 20 Concept Design Sketch. To achieve good aerodynamic performance and the main mission aim, short take off/landing, the concept to should have a similar look to that of a glider with high aspect ratio, minimal weight and streamlined. RC trainer aircraft were also taken into consideration since they are supposed to be easy to handle, which would benefit the autonomous nature of the aircraft. Table 7 shows typical design aspects for a trainer aircraft: Trainer Aircraft Glider Wingspan (b) 152 cm 152 cm AR 6-7 8-10 Overall Length 127 cm 102 cm Wing Area (S) 0.4216 m2 0.323 m2 Flying Weight 1.81 Kg 0.454 Kg Wing Loading (W/S) 59 N/m2 20 N/m2 Table 7 Typical Aircraft Parameters. [26]
- 54. Design and Development of a Hybrid UAV Carlos C.M. 34 | P a g e ME5308 – Major Group Project To have an idea of what speeds the aircraft would be flying at an initial estimate of the weights was made as a group effort, and came to the conclusion that the aircraft would weight about 3.7 Kg. √ It can be shown that the aircraft would need a velocity ( ) of with a wing area ( ) of to fly. Using the Aspect Ratio formula the span can be determined. √ The wingspan comes to be around .
- 55. Design and Development of a Hybrid UAV Arturs D. 35 | P a g e ME5308 – Major Group Project 4.1.2. Quality Function Deployment Figure 21 A quality function deployment (QFD) Matrix Figure 21 demonstrates House of Quality, How’s vs How’s which demonstrates the importance of different parameters in terms of percentage as well as the importance in relation to other parameters. Main two parameters were determined to be the Electrical efficiency and Hover Capability. Hover capability if one of the main objectives of the project therefore it is one of the main parameter on the other hand if the system is not efficient enough the current will be drawn very rapidly during hover mode since there are 3 motors that would be operating at the same time therefore, it is essential for the electrical system to be efficient otherwise there would not be enough electrical power to fulfil the mission profile. The third most important parameter for Table 8 is the overall weight of the aircraft for the same reason the previous one. The lower the weight of the aircraft the less current it draws, the less power required to operate it which leads to improve in efficiency. Those are three main aspects of the aircraft which were concentrated the most on during the design and the built phase of the project.
- 56. Design and Development of a Hybrid UAV Arturs D. 36 | P a g e ME5308 – Major Group Project Nevertheless, the other parameters of the aircraft are very important and failure to reCGnise that could lead to unsuccessful project. Aircraft Attribute Score Importance (%) Relative Importance (%) Electrical Efficiency 84.33 100.00 15.99 Hover Capability 71.00 84.19 13.46 Weight 65.67 77.87 12.45 Cruise Speed 41.67 49.41 7.90 Reliability 40.78 48.35 7.73 Range 38.78 45.98 7.35 Drag 37.44 44.40 7.10 Manufacturing Costs 29.44 34.91 5.58 Easy to operate 23.44 27.80 4.45 STOL Distance 23.00 27.27 4.36 Noise 22.11 26.22 4.19 Rate of Climb 17.67 20.95 3.35 Max g-loading 12.11 14.36 2.30 Maneuverability 8.56 10.14 1.62 High quality image 6.11 7.25 1.16 Operation beyond line of sight 5.22 6.19 0.99 Table 8 House of Quality table, How’s vs How’s
- 57. Design and Development of a Hybrid UAV Abbinaya T.J. 37 | P a g e ME5308 – Major Group Project 4.1.3. Group Concept Category Initial Group Concept Design Final Group Concept Design Mission Type Reconnaissance/ Surveillance Reconnaissance/ Surveillance Environment Outdoor Outdoor Design Type Modular Modular Modular Options Camera Camera Take-Off Type V/STOL V/STOL Landing V/STOL V/STOL Power plant 1x STOL EDF & 3x VTOL EDFs 1x STOL EDF & 3x VTOL Propeller Motors Wing High Fixed Wing High Fixed Wing Tail H-Tail Boom Tail Airframe Balsa Ply wood, Foam, Carbon Composites and Balsa Landing Gear Fixed Fixed Endurance (Prospective) 30 min 30 min (depending on motor thrust test) Altitude 40 - 60m 40 - 60m Glide Capability (Inc. Design) Yes Yes Radio Controlled Back up Yes Yes Autonomous Yes Yes
- 58. Design and Development of a Hybrid UAV Abbinaya T.J. 38 | P a g e ME5308 – Major Group Project Pictures Initial Group Concept Design Final Group Concept Design Table 9 Group Concepts
- 59. Design and Development of a Hybrid UAV Abbinaya T.J. 39 | P a g e ME5308 – Major Group Project Table 9 outlines the basic ideas of the group concept designs. After discussion and consideration of the various ideas proposed an initial group concept design was confirmed. The final concept design was evolved with changes being made to design in order to make initial design more feasible. When comparing the sketches of the initial group concept and the final group concept a lot of differences can be noticed. One such change made is the change in placement of the VTOL motors from the wing tips to the behind the wings. The VTOL motors were initially placed at the wing tips and then moved to be integrated into the wing. This change was made in order to reduce the loads on the wing tips as well as to reduce the moment arm that could possibly flutter and generate some moments on the wing. This change was also opted because it could possibly reduce the drag generated on the aircraft. However, this configuration of the VTOL motors being integrated in to the wing was again changed to be placed behind the wings just as in the current concept. This was done due to some stability issues that came up with the VTOL tricopter system. Another major change in concept design was made with material selection. Balsa was selected for the initial concept design as it is a traditional material used in small UAV and remote controlled planes. This changed when in depth research was conducted on various other materials. With the knowledge gained from research, the traditionally used balsa was replaced with other modern materials and composites. A lot of other changes with design had also been made to the aircraft before deciding on the final concept because of conflicting design issues between the systems required for the VTOL mission and the systems required for the STOL mission. Successful completion of the given design concept should provide an UAV that is capable of doing a vertical take-off and landing as well as a standard take-off and landing. The propulsion system available for VTOL is a tri-copter made of propeller motors. The tri-copter is designed such that there are two counter rotating motors behind each wing and one motor at the nose of the aircraft. The motor at the aircrafts nose would be equipped with a tilt mechanism in order to counteract the yaw force. The thrust for the normal flight would be provided by an Electric Ducted Fan (EDF). The EDF would be placed at the aft of the fuselage in order to provide push propulsion.
- 60. Design and Development of a Hybrid UAV Abbinaya T.J. 40 | P a g e ME5308 – Major Group Project Other aspects of the design such as the wing and tail are kept fairly simple with no dihedral, twist or sweep. A boom tail was selected mainly to keep it from interfering with the EDF placed at the aft of the fuselage. Even though the concept design had been decided an open minded was kept during the phase of the design and manufacture to cope with unanticipated problems that could arise.
- 61. Design and Development of a Hybrid UAV Carlos C.M. 41 | P a g e ME5308 – Major Group Project 4.2. Preliminary Design Once there is a concept idea, the project may move forward into the preliminary design phase. It consists of making the concept idea reality, all the requirements and limits are applied in the aircraft/copter design calculations to obtain a unique outcome to complete the objectives stated. The preliminary design was decomposed into smaller subsections, allowing different group member to focus on individual tasks and work more efficiently. These subsections, along with the rest of the project, may be seen in the workflow diagram below, Figure 22
- 62. Design and Development of a Hybrid UAV Carlos C.M. 42 | P a g e ME5308 – Major Group Project Figure 22 Flow Diagram showing Design Stages Requirement s Market Analysis Technology Concept Design Initial weight estimation Flight control and avionics Aerodynamics Wing Geometry Initial Tail Sizing Propulsion Initial Layout Initial Costing Preliminary Design Initail CG estimation Performance Check Wing Aerofoil Selection Final Wing Design Final Tail Sizing Control Surface Design Lift Curve for Control Surfaces Fuselage Design Landing Gear Optimal CG and NP Motor and Propeller Selection Stability and Control Frozen Design Preliminary Budget and Costs Detail Design and Build Structure Technology Implementation and Sourcing CAD Model Material and Equipment Logistics Experiments and Testing Final Design Configuration Fabrication and Assembly Proof of Concept Ground Testing Flight Test Performance and Results Design Optimisation Final Optimised Design
- 63. Design and Development of a Hybrid UAV Camilo V. 43 | P a g e ME5308 – Major Group Project 4.2.1. Weights Weights estimation is an important aspect as it directly affects the lift required for flight at cruise, take-off and landing as well as dictating ground roll for the latter two mission segments. The heavier the vehicle gets, the faster it needs to go with the same wing to maintain steady flight. In addition to horizontal flight, special consideration was maintained for the vertical flight mission profile, as if the weight was incorrect, then there would be a possibility that the sized motors would not have sufficient power to VTOL. As such the weight was closely monitored and kept constantly up to date throughout the whole design process. To estimate the initial weight of the UAV a components list was made. This was done in parallel with the initial budget seen in the Budget section of the report. It allowed for an initial layout to be done early in the development process of the systems intended to fly on the aircraft. To simulate structural weight, assumptions on volume and initial size were made. Point loads were created to represent the different components of the structure. The density values of the materials chosen were then used to get a rough idea of the mass each segment of the structure would add to the total weight. This factored in calculations of all structural components, from the boom tail rods, to the foam wings. In addition to the structure, all the avionics and propulsion weights were taken into account to get as accurate an estimate for the vehicle weight as possible. Upon completion of the construction of the aircraft, the structure weight was compared with the initial estimated value. Final weight of the vehicle itself devoid of components and motors was 2.75 kg; the original value was 2.35 kg. However the third indicated weight used as a guideline was from the computer aided design model. This value was 3.13 kg. The reasons behind the initial value being almost half a kilo lighter than the actual structure weight was due to the detail undergone at that point in time. The initial weight estimation is from the conceptual stage where the basic components of the structure were outlined and accounted for. The computer aided design weight value was off by 380 g due to the slight differences in the materials assigned to the components from those available in the software and to those actually used in the actual model.
- 64. Design and Development of a Hybrid UAV Carlos C.M. 44 | P a g e ME5308 – Major Group Project 4.2.2. Aircraft Sizing: Constraint Analysis The first and most important procedure in a new aircraft design is to perform a constraint analysis and produce the constraint analysis graph. This graph will provide essential information for the initial design, the two main values are: wing loading and the thrust/weight ratio. With initial weight estimation, an accurate sizing of the wing area and minimum thrust required can be obtained, knowing that they will meet all performance requirements, limitations and regulations mentioned previously. The general methodology to start the constraint analysis is by summarising the forces, or components of the forces that arise in different attitudes of the aircraft. Figure 23 shows the two generic free body diagrams were most the equations can be derived from. a) b) Figure 23 Force Diagrams: a) Forces on a climbing aircraft, b) Forces on aircraft at constant bank angle [27]. The force equilibrium equation may be obtained from Figure 23; with further assumptions at different aircraft attitudes the general equation may be simplified and rearranged to make thrust to weight ratio the subject. Table 10 shows all the equations used in the constraint analysis with their corresponding assumptions. Attitude Assumptions Final Equation Cruise
- 65. Design and Development of a Hybrid UAV Carlos C.M. 45 | P a g e ME5308 – Major Group Project Table 10 Constraint Analysis Equations, obtained from Mattingly et All [27]. The final parameters used in these equations are all defined in Table11. Constraint Analysis Parameters Obtained from… Take-Off Velocity (m/s) 16.82 Calculated in spread sheet Cruise Velocity (m/s) 22.23 Stall Velocity (m/s) 14.02 Dynamic Pressure TO (qto) (kg/ms2 ) 173.38 Dynamic Pressure cruise (qcruise) (kg/ms2 ) 303.82 Nmax (g’s) 3 Performance Requirements AR 8.85 Calculated in spread sheet Oswald factor ( e ) 0.95 Calculated in wing geometry section. Fig. 6 in reference [28]. LA ground roll (SLA) (m) 52 Calculated in spread sheet TO ground roll (STO) (m) 42 Calculated in spread sheet Maneuver/Turn at Constant Altitude Take-Off Given values of Landing Given values of Constant Speed Climb TO Climb Stall at Cruise ( )
- 66. Design and Development of a Hybrid UAV Carlos C.M. 46 | P a g e ME5308 – Major Group Project Density ( ) (kg/m3 ) 1.225 Standard Atmosphere density at sea level 15°C *Ground Friction ( ) 0.5 Estimated For Rubber and Wet Concrete CL cruise 0.4287 Corresponding for aerofoil at cruise( 4°) CL max Climb 0.8713 Corresponding for aerofoil at cruise( 7°) Gravitational Acc. ( ) (m/s2 ) 9.81 Standard gravitational acceleration value *Weight Lapse ( ) 1 Constant at 1 because there is no change of weight during flight (electrical aircraft) *Thrust Lapse ( ) 0.9 Limited Ceiling to 400ft. by regulations. Total Drag Coefficient (CD) 0.025 Calculated in spread sheet Zero Lift Drag Coefficient (CD0) 0.0217 Calculated in spread sheet *External Drag Coefficient (CDr) 0.02 Estimated for extra drag producing factors. Weight (Kg) 6 Calculated in spread sheet Wing Area (S) (m2 ) 0.45 Calculated in spread sheet *Vertical Velocity ( ) (m/s) 2 Group decision, estimate from market analysis k1 0.0181 Calculated in spread sheet k2 0.0028 Calculated in spread sheet Table 11 Constraint Analysis Parameters. With the aid of Microsoft excel, all of these equations can be plotted for a series of wing loading values to obtain the corresponding thrust to weight ratios. The constraint analysis graph can be obtained, see Figure 24 Figure 24 Constraint Analysis for the UAV Design Space 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 50 100 150 Trust-to-Weightratio(T/W) Wing Loading (W/S) (N/m2) Mattingly et al Constraint Analysis Manoeuvre 3g's Cruise Take Off Final Configuration Landing Constant Speed Climb Take Off Climb Stall
- 67. Design and Development of a Hybrid UAV Carlos C.M. 47 | P a g e ME5308 – Major Group Project The initial values for the constraint analysis were obtained from Dr. J. Roskam and D. P. Raymer [29, 30]. These estimates are meant to be for conceptual design of a commercial airline aircraft, hence not being optimised for a UAV. These estimate and assume some regulations and performance requirements that do not apply to the design of this aircraft, and therefore can only be used as a rough guideline. The initial constraint analysis, following Dr. J. Roskams methodology can be seen in Figure 144 in Appendix A. Table 11 shows the final parameters that were used in the constraint analysis shown above. All of these parameters are linked via a Microsoft excel spread sheet that updates the original estimated values automatically once the user has performed a later step in the design of the aircraft. An example of this are the drag coefficients and drag constants, these were updated once the drag prediction model was completed on a later stage of the design process. The constraint analysis delimits the range of thrust to weight ratios and wing loading values. As shown in Figure 24 the design space is limited by different attitudes of the aircraft. Wing loading is restricted by the stall speed and the landing configuration (the main factor being the landing distance). The thrust to weight ratios are only limited on the lower side of the design space by the climb attitude, take-off distance, 3g manoeuvre and the cruise conditions. The two optimal configurations or the aircraft are shown by the blue arrows on the graph, these are the positions of lowest thrust to weight, and hence less engine thrust, that the aircraft can have taking into consideration the performance limits and regulations [31]. The initial aim of the UAV being designed is to be positioned on the right blue arrow, were the final configuration is shown. This combination of T/W and W/S is better than the other optimal design point because the aircraft has a larger wing loading, or in simple terms a smaller wing with greater forces being excreted on it. Accounting that modern materials have great yield strength and the estimated weight of the aircraft should not exceed 6kg, a greater wing loading is readily achievable. The final configuration of the aircraft has a wing loading of 130.2 N/m2 and a thrust to weight ratio of 0.5.
- 68. Design and Development of a Hybrid UAV Carlos C.M. 48 | P a g e ME5308 – Major Group Project 4.2.3. Aerodynamics Aerodynamics is the study of the flow of air about a body and forces produced in the process. The ability to modify these forces to ones requirements is the aim of this section. The aircraft has to be able to lift its weight and produce minimal opposing force to obtain good efficiency and performance. Lift Equation To define a good geometrical design of the wing and an appropriate profile selection it is important to know which factors are required. The first place to look is in the lift equation, seen below: There are four aspects that affect the lifting capability of the wing, these are the wing profile (or the lift coefficient term), the wing area, aircraft speed and air properties that affect the air density. The air density varies continuously and therefore is an independent variable in this equation. The ISA (International Standard Atmosphere) value at a sea level temperature of 15°C has been taken for all the calculations; . The aircraft speed has been initially estimated to be around 20 m/s from the market analysis, but may still vary depending on differences in weight (or lift required) or practical discrepancies with theoretical values. The wing area has already been determined in the constraint analysis and has a value of 0.45m2 . The shape of the wing can vary and affect performance of the aircraft in many ways, and will therefore be further discussed below. Finally, the aerofoil selection will affect the lift coefficient and other parameters that affect the stability of the aircraft, such as moment coefficients. Wing Geometry The area of the wing has to be 0.45m2 . There is a variation of wing configurations that will satisfy the wing area stated, but different configurations will affect the performance of the wing. These geometrical aspects are: Taper Ratio, Leading Edge
- 69. Design and Development of a Hybrid UAV Carlos C.M. 49 | P a g e ME5308 – Major Group Project Sweep, Twist, Dihedral/Anhedral and Aspect Ratio [29]. Some examples are shown in Table 12 below. Wing Characteristic Effect Straight Taper Affects the lift distribution, tip loading, induced drag and wing stalling characteristics. Sweep Delays the mach drag rise at transonic speeds. Twist Alters wing tip stall characteristics. Most common being “Washout” to delay tip stall. Dihedral Increase roll stability Wing Placement Affects stability and internal structural layout. [32] Table 12 Different Wing Geometry Design Aspects Finally, the Aspect Ratio (AR) defines the slenderness of the wing. The equation below shows how it can be obtained. It has a great effect on the induced drag that the wing experiences during flight.
- 70. Design and Development of a Hybrid UAV Carlos C.M. 50 | P a g e ME5308 – Major Group Project Taking into consideration these wing characteristics and their effects on the performance the UAV would greatly benefit of inherent stability and a resemblance of an elliptical wing lift distribution to reduce drag effects. To achieve these goals the following setup was initially chosen: high wing, slight taper, with a few degrees of washout and dihedral. There would be no need of sweep due to the low cruise velocity. After choosing the design configuration of the wing, the performance of the aircraft will change considerably therefore it is recommended to re-iterate and optimise the wing once the rest of the design of the aircraft has been completed. The final configuration of the UAVs wing can be seen in Figure 25 and Table 13. Figure 25 Wing Geometry, Note: dimensions in millimetres Wing Parameters S (m2 ) 0.45 Length (m) 2 Chord Tip (m) 0.2 Chord Root (m) 0.25 Taper ratio (Λ) 0.8 MAC Wing (m) 0.226 Aspect Ratio 8.85 Oswald Efficiency Factor 0.95 Table 13 Wing Geometry Parameters There are differences from the initially chosen set of parameters. The twist and dihedral were sparred because of manufacturing purposes, the twist would make the wing too hard to cut out of foam and the dihedral would not allow a main and secondary spar to run straight thought the aircraft; instead it would involve additional weight to attach them separately on a complex wing box setup. The Oswald efficiency parameter is an indicator that suggests how alike the wing geometry resembles the optimal characteristics of an elliptical wing. It is an important factor estimating induced drag. There are many numerical methods to estimate it
- 71. Design and Development of a Hybrid UAV Carlos C.M. 51 | P a g e ME5308 – Major Group Project depending on the wing characteristics; as described by M. Niƫă and D. Scholz [28]. A lifting line theory based method has been used because the wing does not have sweep or twist. The method uses the taper ratio and aspect ratio of the wing to determine the induced drag factor ( ), which is introduced in the following equation to obtain the Oswald efficiency factor: Figure 26 shows the lifting line theory relationship to obtain the induced drag factor ( ). With the values from Table 13, , obtaining an Oswald efficiency factor of . Figure 26 Induced Drag Factor Vs. Taper and Aspect Ratio [28]. To further reduce the wing tip vortex detrimental effect on the wing, the induced drag may be reduced by the addition of winglets. These act as a fence between the high and low pressure at the wings tips reducing the magnitude of the vortices and the downwash effect they have over the wing span. A simple way to understand the effect they have on the induced drag would be by incrementing the span and hence increasing the aspect ratio of the wing [33]. The induced drag equation in Figure 26 shows that aspect ratio is inversely proportional to the aspect ratio. The positive effects of winglets on the current UAV design would be minor. The addition of winglets would further complicate the manufacturing of the wings and have a weight
- 72. Design and Development of a Hybrid UAV Carlos C.M. 52 | P a g e ME5308 – Major Group Project penalty which could not be tolerated taking into account the, already high, weight estimation. With the wing design concluded the final lifting and pitching characteristics of the wing may be determined, these will later on affect the tail size and its placement. Aerofoil Selection The Lift Coefficient is related to the profile section chosen. The profile section mainly varies in relation to these parameters: Camber, thickness to chord (t/c) ratio, and position of maximum thickness. The higher the camber and t/c ratio the higher the lift will be [29]. Figure 27 shows the typical cambered aerofoil section. Figure 27 Typical Cambered Aerofoil [31] It was initially intended to have a high lift low Reynolds number aerofoil, characterised by a greater camber. After some investigation, [34], the Selig S1223 was chosen with a corresponding cruise velocity of 13.9m/s, but was then discarded due to the high moment coefficient that would require a relatively big tail and a long tail arm. Consequently an analysis of a variety of aerofoils was made to compare mainly the lifting characteristics against the moment coefficients. The outcome pursued in the analysis was to have the greatest lifting coefficient with the lowest moment possible to reduce the tail size and arm; which would also reduce the structural weight of the aircraft. Figure 28 below shows the negative correlation between Cl and Cm, which is previously expected. Highlighted in red is the Selig 1223 aerofoil and in yellow the final aerofoil, NACA 63(2)-A015. There is a very big difference in both the lifting characteristics and the moment characteristics of these aerofoils; this comes at the expense, looking back at the lift equation, of higher cruise velocity.
- 73. Design and Development of a Hybrid UAV Carlos C.M. 53 | P a g e ME5308 – Major Group Project Figure 28 Lift Coefficient Vs. Moment Coefficient Analysis of different profiles. Lowering the lift coefficient also increases the take-off speed of the aircraft in an STOL (standard take-off and landing) mission. This would not represent a serious problem for the design since additional high lift devices may be used at landing and take-off. It was a general consensus of the team to use the wing control surfaces as flaps and ailerons to provide extra lift in certain flight segments. The UAV will spend most of its fly time cruising in a reconnaissance role or surveillance mission. In order to improve its efficiency careful attention has to be set on the wing incident angle. When analysing the variety of profiles, the cruise angle of attack has been equated to the angle of attack where the wing generates the minimum lift to drag ratio. Only the lift to drag of the wing has been taking into consideration because the aircrafts minimum drag angle has been assumed to be zero degrees. At this angle the aircraft has the minimum front cross sectional area and hence produced the lowest pressure (form) drag with an unchangeable amount of skin friction drag [31]. Lift to drag is a very important performance and efficiency indicator which dictates other performance factors such as endurance or influences the take-off weight of the aircraft. The higher the lift to drag the longer the aircraft will fly. This is demonstrated in the Breguet range equation, where the rage is proportional to the lift to drag ratio. In terms of an electric vehicle, at cruise the thrust is equivalent to the drag which is Initial Profile S1223 Final Profile NACA63(2) A-015 -0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 WingMomentCoefficientatCruiseCm Lift Coefficient Cl Lift Coefficeint Vs. Moment Coefficent at Cruise
- 74. Design and Development of a Hybrid UAV Carlos C.M. 54 | P a g e ME5308 – Major Group Project equal to the motor current draw and hence the battery size [35, 36]. These factors have a heavy influence on the aircrafts payload capacity which can also be linked to operational costs of the aircraft. Figure 29 Initial and Final profile Comparison. Figure 29 above shows a comparison of the initial and the final aerofoils. The difference in lifting capability is clearly seen by the difference in camber and the difference in moment of the wing is also shown by the S1223 highly asymmetrical profile compared to the NACA 63(2)-A015. The NACA profile has a thickness to chord ratio of 15% whilst the Selig has 12.1%, this would partially explain the lower drag to lift ratio of the Selig aerofoil. One of the main reasons why the S1223 was chosen at the beginning of the design process was due to its high profile lift to drag ratio of almost 55, but when changed to the NACA 63(2)-A015 this value was reduced to 46.The development of why the aerofoil has changed is discussed further in the stability section. The analysis of the aerofoils taken into consideration for the UAV is shown in Appendix A Table 37 “Additional Aerofoil Analysis”. It is important to note that all values have been obtained with the aid of XFLR5 using the viscous vortex lattice method at an estimated aircraft velocity of 20m/s. Another method used to analyse the wing properties was the panel method. Table 38 in Appendix A shows the difference in values from VLM to Panel and there is a very small difference that does not exceed 1.2% in any of the calculated values of lift, drag and moment coefficients. VLM has been chosen because obtains converging results for a wider range of angles of attack. Results for the final wing and aerofoil configuration at an angle of attack of 4° (cruise condition).
- 75. Design and Development of a Hybrid UAV Carlos C.M. 55 | P a g e ME5308 – Major Group Project XFLR5 Setup - Viscous, Vortex Lattice Method at 20m/s. (a) (b) (c) Figure 30 XFLR results for the final wing configuration. (a) Moment Force and chord wise lift distribution. (b) Spanwise lift distribution. (c) ISO view of lift and lift distribution. Figure 30 shows the lift distribution of the aerofoil in the chord wise and spanwise directions. It may be seen that the spanwise distribution is not exactly uniform like that on an elliptical wing, but there is some resemblance. The optimal taper ratio to imitate an elliptical wing would be about 0.3 [35]. This has a very low amount of induced drag but poor stalling characteristics and hence a large amount of washout would be needed to delay tip stall. Initially a value of taper of 0.3 was chosen, but to maintain the wing area obtained in the constraint analysis the root chord would have to be enormous or the wing span would have to increase over the 2m limit set by the group. To find the balance between washout (negative twist) and the lift distribution characteristics the taper ratio has to be increased. The final value is 0.8 because no twist is designed into the wing due to manufacturing purposes, and it is more important to reduce the stalling characteristics rather than reduce the induced drag.
- 76. Design and Development of a Hybrid UAV Carlos C.M. 56 | P a g e ME5308 – Major Group Project (a) (b) (c) Figure 31 Polars: (a) Variation of Lift coefficient with AoA. (b) Variation of the drag coefficient with lift coefficient. (c) Variation of lift to drag ratio with AoA. The polars above describe the wing lifting and drag properties. The wing curve slope (a) shows the variation of lift. Only the section with linear variation is shows because the models do not predict very well were the non-linear segment starts or ends. Dr. Jan Roskam has a generalised numerical method that suggest how to estimate the non-linear region, it decomposes the lift into contributions from the geometry, -0.5 0 0.5 1 1.5 -2 0 2 4 6 8 10 LiftCoefficient(CL) Angle of Attack α (degrees) CL vs Angle of Attack Curve 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0 0.2 0.4 0.6 0.8 1 1.2 DragCoefficeint(CD) Lift Coefficeint (CL) CD vs CL Polar of the Wing -10 0 10 20 30 -2 0 2 4 6 8 10 12 CL/CD Angle of Attack α (degrees) CL/CD vs Angle of Attack
- 77. Design and Development of a Hybrid UAV Carlos C.M. 57 | P a g e ME5308 – Major Group Project maximum camber, maximum thickness, Reynolds number, aerofoil roughness and Mach number lift increase [36]. This is a very complex method hence the non-linear region being ignored in the lift curve slope. Similar aerofoil wind tunnel data suggest that the linear range ends around 10 degrees with the maximum lift coefficient being at about 15 degrees angle of attack [37]. XFLR5 suggest similar values for the 2D profile, but cannot predict the non-linear region in the 3D wing setup. It only produces the curve slope seen above (a). Figure 31 (b) is the drag polar of the wing. It may be observed that the drag generated by the wing is very small, and will be seen in the drag section that the contribution of the wing is very small relative to the whole aircraft. The maximum lift to drag ratio may be obtained by drawing a line though the origin so that it becomes a tangent to the polar. This obtains the smallest value of CD with the highest CL. A more accurate way to determine the maximum lift to drag ratio is by obtaining graph (c). It demonstrated that the wing incident angle has to be around degrees to optimise the wings performance in cruise.
- 78. Design and Development of a Hybrid UAV Abbinaya T.J. 58 | P a g e ME5308 – Major Group Project Tail Sizing The solution to the empennage design was analysed from two perspectives; the tail sizing required for the STOL mission and the tail requirements needed to be able to achieve the VTOL mission. The tail sizing procedure used for the STOL mission is the same as that used in conventional aircraft by using standard procedure which are illustrated by Jan Roskam, Mohammad Sadraey and D. Raymer. [38] [39] [22]. Given below in figure 32 is the tail design procedure. The tail design procedure itself is an iterative process.
- 79. Design and Development of a Hybrid UAV Abbinaya T.J. 59 | P a g e ME5308 – Major Group Project Figure 32: Tail design procedure as illustrated by Mohammad Sadraey. [40] The tail configuration The current aircraft configuration is such that it is propelled forward by the EDF installed at the rear of the fuselage. Since this is the case, designing a conventional tail would have a lower effectiveness due to interference between the EDF flow and
- 80. Design and Development of a Hybrid UAV Abbinaya T.J. 60 | P a g e ME5308 – Major Group Project the tail. The solution to this problem was a boom mounted H-tail. The H-tail comprises of 2 vertical tails with a horizontal tail running in between the vertical tails just like the letter “H”. This tail configuration would be beneficial especially because the horizontal tail and vertical tails would not be influenced by the wake of the EDF. Other advantages include better lateral control due to shorter vertical tail span and improved efficiency of the horizontal tail due to the vertical tail acting as end plates. The chosen tail configuration would also allow for a smaller fuselage length due to the presence of the booms. Even though the H-tail has 2 vertical tails, the design process will only consider one vertical tail. The designed vertical tail will be then split as two tails at the end. Volume Coefficients: The stability and control of an airplane are mainly dictated by their tail planes which provide longitudinal and lateral stability. Hence they are often referred to the stabilisers of the aircraft. Volume coefficients are parameters that measure the effectiveness of the horizontal and vertical stabilisers. Ability to select volume coefficients for new aircraft design comes only with experience. Therefore, the volume coefficients used for this aircraft were selected based on typical values used for aircraft with similar mission. The geometry of the horizontal and vertical tails is both dependent on their respective volume coefficients. Given below are the equations for the horizontal ( ) and vertical ( ) tail volume coefficients.
- 81. Design and Development of a Hybrid UAV Abbinaya T.J. 61 | P a g e ME5308 – Major Group Project Table 14 below illustrates the effects of changes in tail volume coefficients in aircraft. Volume Coefficient Aircraft Stability Structural Weight High High High Low Low Low Table 14: Effects of changes in tail volume coefficients Using a very low horizontal tail volume coefficient ( ) would make the aircraft’s the pitch behaviour very sensitive to the CG location. This would imply poor gust resistance and therefore result in difficult pitch control. When the vertical tail volume coefficient ( ) is too low, the aircraft will tend to oscillate along the vertical axis inducing a dutch roll. This would make the directional control of the aircraft more difficult. [38]. The horizontal and vertical tail volume coefficients for the designed aircraft are 0.56 and 0.05 respectively. Optimum tail arm and tail plan form area: The tail arm and the tail area are the two basic parameters of the tail that are correlated to each other. The lift generated by the tail is dictated by the tail area. The tail arm works as an arm for the pitching moment of the tail. The moments generated by the tail is calculated by multiplying its lift force with the tail arm. The tail arm can either be long or short as long as it is balanced with a suitable tail area. The tail arm used for this aircraft is 0.8 meter. Aircraft trim plays a very important role in aircraft to operate safe flight. The aircraft trim must be maintained about the lateral(x), longitudinal (y) and directional (z) axes. An aircraft is considered to be at trim when the summation of the forces about all three directions equals zero. The horizontal tail is responsible for maintaining longitudinal trim and the vertical tail is responsible for maintaining directional trim. Section 161 of PART 23 published by the Federal Aviation Regulations (FAR) states the requirements of an aircrafts trimmed condition. The aircraft must maintain longitudinal trim under conditions such as climb, cruise, descent and approach. The
- 82. Design and Development of a Hybrid UAV Abbinaya T.J. 62 | P a g e ME5308 – Major Group Project horizontal tail was designed in order to balance the longitudinal moment of the wings lift about the aircraft’s CG and the wing aerodynamic pitching moment. The main function of the vertical tail is to generate a yawing moment in order to balance the moment generated by engines. It plays an important role especially during one engine inoperative conditions in multi-engine aircraft. Since the current aircraft is to be symmetric about the xz plane and one engine operative conditions do not apply, more emphasis was given to the horizontal tail design. Tail Aerofoil: Usually symmetrical aerofoil profiles are used in tail design. The sum of the pitching moments about the aircraft’s CG was found to be negative. Therefore the tail was mounted at a negative incidence angle in order to generate downward lift and counteract the moments. The commonly used NACA 0012 profile was selected for the tail as it is quite simple to manufacture.
- 83. Design and Development of a Hybrid UAV Carlos C.M. 63 | P a g e ME5308 – Major Group Project Drag Drag is one of the major forces acting on the aircraft in flight. In an ideal case drag would be equivalent to zero, but in reality air flowing around a body will always create an opposing force. It is a key parameter that has subsequent impact in other aircraft aspects. It is used in the aircraft design phase to determine the ceiling, thrust needed and performance of the aircraft. Figure 33 shows the decomposition of the total drag. The most important components are Lift Induced, Skin friction and Form Drag [31] [41]. Figure 33 Total Drag Decomposition. For the purpose of this aircraft, the wave drag may be ignored because no shocks will arise at the maximum speed of the aircraft. Form the previous market analysis it was determined that the cruise speed of the aircraft would be in a range of 15 to 20 m/s. At standard atmospheric conditions this results in a Mach number of 0.06, at which compressibility effects, such as shock generation and wave drag, will not be an issue and hence may be ignored. Interference Drag can also be ignored for the same reasons. It is the result of the boundary layers or streamlines of different components of the aircraft interfering with each other to produce high velocity and hence a normal shock. It usually occurs at the joint of the fuselage and wing or the horizontal and vertical stabilisers [31]. Total Drag Profile Drag Viscous Drag Skin Friction Form Drag Miscellaneous (Wave, Interference etc.) Lift Induced Drag
- 84. Design and Development of a Hybrid UAV Carlos C.M. 64 | P a g e ME5308 – Major Group Project Miscellaneous refers to those small factors that at this stage of the project lack importance and don’t affect the accuracy of the final drag estimation. Profile Drag The profile drag is a combination of the Skin Friction and Shape Factor (SF) of the aircraft. It is one of the main constituents of the total drag of the aircraft, hence the importance of an accurate estimation at this point. The minimum profile drag coefficient can be estimated using below. To calculate the minimum profile drag force we insert the profile drag coefficient into a derivation of the lift/drag equation seen below: Skin Friction Drag This is the drag due to wall shear stresses; in basic terms, the force created by a viscous flow over an object. A good approximation for this term is the 1/7 Power Law derived by Theodore Van Karman [42]:
- 85. Design and Development of a Hybrid UAV Carlos C.M. 65 | P a g e ME5308 – Major Group Project Form Drag This is drag dependent upon the shape of the body and the pressure differences created from an incoming flow. It is determined with a parameter called Shape Factor. There is two general equations that determine the SF, one for thin bodies and another for bluffed bodies. A thin body is defined by the thickness to chord ( ) or diameter over length ( ) of the body being below 30% [43] [44]. These are the equations used to determine the SF: ( ) ( ) Induced Drag This component of drag arises from the pressure differences above and below the wing. The pressure difference cause wing tip vortices, which induce a downwash on the rest of the wing creating an adverse effect called, induced drag. This is directly
- 86. Design and Development of a Hybrid UAV Carlos C.M. 66 | P a g e ME5308 – Major Group Project related to the lift generated by the wing, its Aspect Ratio (AR) and shape of the wing. Below is the equation to calculate induced drag [41] [31]: Total Drag Vs. Velocity Curve The Drag vs. Velocity curve can be drawn up with the aid of Microsoft excel to solve the above equations for a variety of free stream velocities. All the parameters required in the equations have been either inserted or calculated in the spread sheet, under a different section. This back reference makes it very helpful to update all the calculations when there are any speed changes or any component changes size, etc. Figure 34 below shows the total drag vs. velocity curve: Figure 34 Drag velocity curve. y = 8E-05x4 - 0.0075x3 + 0.2583x2 - 3.7915x + 23.283 0.000 1.000 2.000 3.000 4.000 5.000 6.000 7.000 8.000 5.00 10.00 15.00 20.00 25.00 30.00 Drag(N) Velocity (m/s) Drag Vs. Velocity Induced D Total P D. Total D. Vstall Poly. (Total D.)
- 87. Design and Development of a Hybrid UAV Carlos C.M. 67 | P a g e ME5308 – Major Group Project The blue curve represents the induced drag, the red curve is the profile drag and the green line is the summation of both drags making the Total Drag. For optimal performance the ideal cruise speed is that where the least total drag is obtained, for our aircraft it should be around 15 m/s or slightly higher to maintain speed stability. This speed is definitely within our stall speed, as shown by the vertical line light blue line. The total drag curve has been fitted with a trend line and the corresponding equation is shown on Figure 34. If the drag at a specific airspeed is required, replace it in the x term in the equation to obtain the drag force at that speed. An important factor is mentioned above, Speed Stability. Imagine if the aircraft is flying to the right of the minimum drag speed at a set thrust and altitude. If the speed of the aircraft is increased due to any sort of disturbance, the drag will increase and slow down the aircraft. In the case where the airspeed is reduce by a disturbance; the drag will decrease, speeding up the aircraft, giving it speed stability. Flying at a lower speed than the minimum drag would result in a continuous increase of speed due to a disturbance, which is undesirable. To maintain speed stability it is crucial to fly above the minimum drag [45]. Hence the aircrafts cruise speed was initially estimated to be around 20 m/s. CD Vs. CL Polar The same process was used to obtain Figure 35 shown below.
- 88. Design and Development of a Hybrid UAV Carlos C.M. 68 | P a g e ME5308 – Major Group Project Figure 35 CD Vs. CL Polar for the wing and the aircraft. In comparison to the drag polar of the wing this is an estimation of the corresponding drag of the whole aircraft with respect the lift of the wing. The trend line on the aircrafts drag polar in Figure 35 is second order meaning that the drag calculated corresponds to the quadratic drag estimation method. This method has a highly accuracy compared to wind tunnel data for lift coefficients in the range of ±1.2 units. Depending on the position of the minimum drag, the curve will vary exponentially at higher CL values [31]. Below Table 15 resuming the skin friction coefficient (Cf), induced drag coefficient (CDI), profile drag coefficient (CDP) and the total drag of the aircraft with a final refined cruise speed of 22.2 m/s. it shows the decomposition in terms of the following aircraft components: Body, Wings, Empennage and Landing Gear. An assumption made in the landing gear is that it is made of a cylinder and a straight flat plate, corresponding to the wheels and the landing gear arms respectively. y = 0.0181x2 + 0.0028x + 0.0129 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0 0.5 1 1.5 2 DragCoefficeint(CD) Lift Coefficeint (CL) CD vs CL Polar Aircraft wing
- 89. Design and Development of a Hybrid UAV Carlos C.M. 69 | P a g e ME5308 – Major Group Project Note: The drag coefficients with a “ * ” before the name suggest an external value has been used, [46]. Cruise Drag Components Body Cf 0.0035 Body CDI 0 Body CDP 0.0084 Total Body Drag 0.0411 Wing Cf 0.0046 Wing CDI 0.0070 Wing CDP 0.0118 Total Wing Drag 3.2410 Empennage Cf 0.0052 Empennage CDI 0.0032 Empennage CDP 0.0127 Total Empennage Drag 0.9865 *LG Cylinder CD 1.05 LG Cylinder Drag 0.0320 *LG Flat Plate CD 1.27 LG Flat Plate Drag 3.5499 Total LG Drag 3.5820 Zero Lift Drag Coefficient (CD0) 0.0248 Total Profile Drag (CDP) 7.8506 Total Induced Drag (CDI) 1.0778 Total Drag 8.9533 Table 15 Drag Components of the aircraft for cruise, 22.2 m/s.
- 90. Design and Development of a Hybrid UAV Abbinaya T.J. 70 | P a g e ME5308 – Major Group Project Control Surface Sizing: Wing The UAV designed has ailerons and flaps on the wing elevator and rudder on the empennage. This section will cover the design process for the control surfaces on the wings. For sake of simplicity in manufacture the ailerons and flaps were designed in combination as flaperons. Flaperons are devices that can function as flaps as well as ailerons. Ailerons: The design procedure of the control surfaces is an iterative process. Therefore an excel sheet was formulated in order to calculate the required size of the ailerons. First, the initial parameters such as chord, span and maximum deflection angles were decided based on typical values that have been used for similar aircraft. Then, step by step lists of calculations were carried out to determine the performance characteristics of the ailerons to check whether or not they meet the requirements. The aileron sizing method was done based on the approach used by Mohammad Sadraey. The excel sheet was formulated such that the time taken to achieve a desired bank angle is calculated. The results were matched with typical values of time taken for similar category aircraft to see how they compare. Given below is the steps used to estimate the responsiveness of the ailerons. 1. Estimation of initial geometry of aileron and maximum deflection angle. 2. Calculation of the aileron rolling moment coefficient, (1/rad): [ ( )] 3. Calculation of the aircraft rolling moment coefficient, when aileron is deflected with maximum deflection: 4. Determination of rolling moment, (Nm):
- 91. Design and Development of a Hybrid UAV Abbinaya T.J. 71 | P a g e ME5308 – Major Group Project 5. Calculation of Steady State roll rate, (rad/sec): √ ( ) 6. Calculation of bank angle, (rad) at which the aircraft achieves steady state roll rate: ( ) 7. Calculation of the aircrafts rate of roll rate, ̇ (rad/sec2 ) produced by the ailerons rolling moment until the aircraft reaches the steady state roll rate: ̇ 8. The time taken, (sec) to achieve the desired bank angle at maximum deflection is calculated: √ ̇ Given in Table 16 is the time it takes for the aircraft to achieve a specified bank angle at a maximum deflection of 22.5 degrees. Bank angle (degrees) Time to achieve bank angle (seconds) 30° 0.812 45° 0.994 60° 1.148 Table 16: Time to achieve specific bank angles
- 92. Design and Development of a Hybrid UAV Abbinaya T.J. 72 | P a g e ME5308 – Major Group Project Flaps: Flaps are control surfaces mounted on a wing in order to increase the lift generated. The use of high lift devices are beneficial especially in manoeuvres where the aircraft’s velocity is reduced. Deflecting the flaps increases the camber of the wing and therefore increase the lift generated by it. For this particular UAV designed, flaps are to be retracted during take-off and landing. The use of flaps during take-off will help the aircraft take-off at a shorter runway distance. This feature will be advantageous especially when the runway surface has a higher friction. Using flaps during landing will reduce the aircraft’s stall speed and allow a slower and steeper approach. Flaps deflected at higher angles also increase drag and this will help reduce the runway distance for landing. The increment in lift coefficient at different flap deflection was obtained and checked to determine whether it was suitable to fulfil the task [47] [40].
- 93. Design and Development of a Hybrid UAV Carlos C.M. 73 | P a g e ME5308 – Major Group Project Lift Curve Slope Numerical Prediction: Wing and Aileron/Flaps The lift curve slope is the variation of the lift coefficient with the free stream incident angle of the lifting surface [31]. This is strictly necessary in order to calculate the effect that a deflected control surface has on the lift. There are a series of methods for obtaining these values for specific control surface sizes and deflections; this especially helpful in preliminary design of wings and control surfaces. The first and most accurate method is to make a model of the wing section with the control surface and test it in a wind tunnel. This is an extensive and costly process, especially if the design needs to be optimised and a series of models have to be made to obtain the correct control surface sizes. When resources are limited there are other methods that will provide good estimations. Vortex Lattice Method (VLM), used by XFLR5, can model the flow over a profile and the wing obtaining the lift curve slopes. There are other available resources such as online databases that provide the curve slopes of specific aerofoils, but the reliability of the source may be questioned. Another numerical method is using ESDU sheets (Engineering Science Data Unit), they provide numerical solutions that interpolate, non-linearly, between existing experimental data correlations for specific problem. Figure 36 below shows a comparison of the different methods used to predict the lift curve slopes for the wing design of the final UAV configuration. The first turquoise line shows the 2D polar obtained from XFLR5 by simple anlayis of the profile. The orange line is the 3D result obtained using lifting line theory in XFLR5. The light blue line is the corresponiding 3D result for VLM and finaly the dark blue line represents the ESDU three dimesional wing lift curve slope.
- 94. Design and Development of a Hybrid UAV Carlos C.M. 74 | P a g e ME5308 – Major Group Project Figure 36 Comparison of the Lift Curve Slopes using different predicting methods: Online database, XFLR5 and ESDU sheets. Since ESDU is the most reliable resource compared to a free online program such as XFLR5, then ESDU have been taken as the point of reference to obtain the lifting characteristics of the wing and to know the effect that the control surfaces will have. ESDU also provides the lowest gradient, hence taking it as a reference will overestimate the control surface sizing rather than underestimate and then not being able to fly adequately. Some ESDU sheets used were W.01.01.05 [48] to calculate the lift curve slope of the wing and C.01.01.03 [49] to calculate the lift curve slope of the deflected control surface. Table 17 shows the steps that were taken to obtain the lift curve slopes for the wing and the control surface (aileron or flaps), so that the sizing could be completed. This is an iterative process between the initial control surface size, the initial curve slope and a repetition until the final corresponding values are obtained. Step Goal Additional Notes 1 Wing: Obtain the relationship between the curve slope and the -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 -3 -1 1 3 5 7 LiftCoefficent Angle of Attack α (degrees) CL vs. Alpha 2D XFLR5 Polar 3D XFLR5 Polar XFLR5 3D Result (VLM) CL Wing (ESDU)
- 95. Design and Development of a Hybrid UAV Carlos C.M. 75 | P a g e ME5308 – Major Group Project theoretical curve slope, . 2 Wing: Obtain the theoretical slope of lift for 2D aerofoil in incompressible and inviscid flow, . 3 Wing: Obtain the 2D incompressible flow curve slope, . 4 Control Surface: Obtain the theoretical slope of lift for 2D control surface in incompressible and inviscid flow . From Figure 1 in ESDU 01.01.03 or Figure 145 in Appendix A. 5 Control Surface: Obtain the relationship between the curve slope and the theoretical curve slope, . From Figure 2 in ESDU 01.01.03 or Figure 146 in Appendix A. 6 Control Surface: Obtain the 2D incompressible flow curve slope, . 7 Wing & Control Surface: Extrapolate into 3D accounting for compressibility effects. √ 8 Correct the angle of control surface deflection [31]. 9 Correct the zero-lift angle of attack. Should be equivalent to the 2D curve slope zero-lift angle of attack. 10 Obtain the lift coefficient of the wing and deflected control surface by the following equation: Table 17 Process to attain the lift curve slopes of the wing and the deflected control surface.
- 96. Design and Development of a Hybrid UAV Carlos C.M. 76 | P a g e ME5308 – Major Group Project Parameters used for the curve slope prediction and Results Key Parameters Curve Slopes Results 20 (a1)0/(a1)0T 0.669 Reynolds number 300000 (a1)0T 6.997 1/rad X-t/c 0.75 (a1)0 4.679 1/rad Thickness to chord ratio 0.15 Wing incidence angle (degrees) 4 (a1)0 -3D 3.982 1/rad Cruise Velocity (m/s) 22.200 Speed of Sound at Sea Level (m/s) 340 (a2)0T 3.750 1/rad Mach (M) 0.066 (a2)0/(a2)0T 0.335 AR 8.85 (a2)0 1.256 1/rad e 0.95 (a2)0 -3D 1.202 1/rad Correction Angle (degrees) Aileron Deflection Angle (degrees) Delta 15 30 45 -1.23 Delta Effective 11.25 22.5 33.75 Table 18 Parameters and Results. Table 18 shows the all the parameters necessary to predict the theoretical curve slope for the wing, and the corresponding relationship of . These values are used in the equations for Steps 1 and 2 of the process. is the trailing edge angle that is obtained from the 2D aerofoil profile. Figure 37 below describes how to obtain it. is the chord wise location were boundary layer transition takes place. This value has been estimated to occur at around 75% as discussed by K. Laurence in a technical paper of 6-series aerofoil investigation. [50] Figure 37 How to obtain Trailing Edge Angle . Using the equation in the final step of Table 17 the curve slope may be predicted as shown in Table 18 and Figure 38 . This method allows the lift difference of the ailerons or flaps to be determined for specific deflection angles. For a variety of
- 97. Design and Development of a Hybrid UAV Carlos C.M. 77 | P a g e ME5308 – Major Group Project angles of attack the deflections of the control surface of 15, 30 and 45 was calculated and plotted in Figure 38. Figure 38 Wing Curve slopes with control surface deflections. Deflection 15 30 45 Lift Increment 0.236 0.472 0.708 Table 19 Lift variation with control surface deflection. If the control surfaces are being used as flaps then the increment will be equal to that stipulated in the Table above. As high lift theory suggests with flaps deployed the lift curve slope of the wing will be displaced upward, as seen in Figure 38. If they are used as ailerons then the lifts of the wing at the sections were the ailerons are placed will increment or decrease depending if the aileron deflects up or down by the amount shown above. The lift difference will then provide the rolling motion to the aircraft. -0.400 -0.200 0.000 0.200 0.400 0.600 0.800 1.000 1.200 1.400 -2 0 2 4 6 LiftCoefficent Angle Of Attack α (degrees) CL vs. Alpha Control Surface Defelction CL Wing (ESDU) CL (CS-15deg) CL (CS-30deg) CL (CS-45deg)
- 98. Design and Development of a Hybrid UAV Camilo V. 78 | P a g e ME5308 – Major Group Project 4.2.4. Centre of Gravity The Centre of Gravity (CG) is vital to aircraft characteristics in flight as it affects key stability parameters. Alongside the weight, it was also estimated and updated constantly in a bid to more accurately predict where it would be on the aircraft in order to be able to balance out internal components to achieve the optimum condition and reduce or avoid any extra dead weights that would need to be implemented to balance out the aircraft. Once the weights had been initially calculated the first thing to do; in parallel to aerodynamic investigations conducted by the group, was to ascertain an initial value for center of gravity. The datum was set to the UAV’s assumed nose location. Initial sizing was determined with the use of the findings from market analysis, and the fuselage outline was drawn out in scale. From here the wing leading edge was established and using Raymer’s initial estimate of center of gravity as a percentage of MAC (15-25%) [51], the components were laid out in a manner which brought the center of gravity close to this specified point and the center of gravity was calculated using simple moment calculations: = ( ) ( ) The exact desired position was later better defined in the static margin calculations for stability in flight. The CG of the geometry of the aircraft was estimated using the assumption of point loads on each separate section of the geometry which moved the calculated value to the aft. With this in mind the battery compartment of the UAV was given special consideration to allow any movement forward to adjust the CG back to the initial theoretical optimum. Throughout the design process as details were altered or refined, this estimate for the optimum CG point was kept fixed by the movement of all the components within the aircraft through the use of an excel spreadsheet. The CG point was made a design requirement due to the stability required of the tri- copter platform which would be embedded within the aircraft. The Vertical Take-Off propellers all had to be equidistant from the CG point to balance out the thrust required for hover to be 1/3 of the vehicle weight per motor. This also had the benefit of making it easier to fly for stability reasons. This is commonly referred to as the golden triangle of stability, due to the equilateral triangle created by the position of
- 99. Design and Development of a Hybrid UAV Camilo V. 79 | P a g e ME5308 – Major Group Project the motors around the CG. This restriction also had an impact on the minimum fuselage length forward of the wing, as it could only be as short as the distance from the body propeller to the CG. After build the CG location was found experimentally in the aero lab using force balancing apparatus which when attached to the wings of the aircraft could be modified to pivot it around a specified point anywhere under the wing. The aircraft would then remain in place if perfectly balanced, or would pivot around and tilt towards where the true location of the real vehicle tended to. Below is a figure displaying the apparatus in use. Figure 39 Force balance kit to acquire aircraft CG location Due to an earlier issue that the group experienced with the main battery, two different configurations of the aircraft were developed to allow for testing whilst a secondary main battery arrived: The first was with a 4000 mAh 6s powering just the main horizontal thrust provider for a standard aircraft mission profile. In addition two 3900 3s Li-Po's were placed on-board to offset the left over mass that would come from the main 9000 mAh battery for CG balancing. This gave the unique opportunity to link the batteries together into series and creates a back-up power source to allow for an extra flight once the 4000 mAh battery was depleted, or the ability to run the APM with one of the additional dead weight batteries.
- 100. Design and Development of a Hybrid UAV Camilo V. 80 | P a g e ME5308 – Major Group Project Figure 40 Front load with dual dead weight batteries and back-up 4000 mah main battery The second was the original designed load with the 9000 mah present. In this configuration there would be enough discharge and battery capacity to run all VTOL and normal flight propulsion at the same time. From the initial balancing of the aircraft using estimated locations for the components the following discrepancies were found between theoretical estimate, and actual real value: Structure= Estimate: 70.0 cm Actual: 79.0 cm Complete Aircraft = Estimate: 71.5 cm Actual: 75.0 cm The values shown above were taken from a datum of 29cm forward of the front fuselage Bulkhead. to put it contextually, the CG position had to be 10.5 cm from the leading edge of the wing for stability reasons, however once the vehicle was fully loaded it was 14 cm. From here the internal components of both configurations were adjusted to provide for the CG point required by design.
- 101. Design and Development of a Hybrid UAV Abbinaya T.J. 81 | P a g e ME5308 – Major Group Project 4.2.5. Stability and Control: Standard Take-Off and Landing The main functions of the tail in a typical aircraft are to satisfy the conditions of longitudinal and directional: i. Trim ii. Stability iii. Control. In the initial stages of design, the horizontal and vertical tails are designed to satisfy the requirements for longitudinal and directional trim. Conditions to satisfy the requirements for longitudinal and directional stability & control are covered in later stages of design. Aircraft trim: Figure 41 shows a graph of tail incidence angles against the moments generated for different plan form area. The XFLR 5 software was used to estimate the lift generated by the various tail sizes at different angles of attack. This graph was used to study the pitching moments generated about the CG at various angles in order to select the most suitable tail size and required tail setting angle. From the graph in Figure 41 it can be seen that the 0.8 × 0.15m and the 0.6 × 0.12m generate the required moments for cruise and climb at smaller incidence angles. Since the results from both these designs are similar, the one with a lower area was selected. A smaller area was preferred as it would be beneficial with weight reduction. The NACA 0012 profile with a 0.6m span and 0.12m chord was finally selected for the horizontal tail. The graph shows that the selected horizontal tail should be at an angle of -0.7° at cruise conditions and at -4.8° for maximum climb conditions. The incidence angle was selected to be -1 degree as cruise covers majority of the flight mission. The horizontal tail will comprise of an elevator which will deflect in order to generate a higher lift and therefore higher moment at climb or descent conditions.
- 102. Design and Development of a Hybrid UAV Abbinaya T.J. 82 | P a g e ME5308 – Major Group Project Figure 41: Tail incidence angle vs. Moments generated. Table 20 below shows the details of the selected tail geometry of the horizontal and vertical tail. Horizontal tail Vertical tail Volume coefficient (Vh) 0.56 0.05 Tail arm (Lh) 0.8m 0.8m Area 0.072m2 0.056m2 Aerofoil NACA 0012 NACA 0012 Span 0.6m 0.19 Mean Aerodynamic Chord 0.12m 0.1 ̇ Aspect Ratio 5 0.64 Taper Ratio 1 0.5 Setting Angle -1 deg 0 deg Table 20: Horizontal and vertical tail design details As already mentioned, the stability of an aircraft is mainly governed by the tail. The tail is also responsible for control of the aircraft to some extent due to the presence of the elevator horizontal tail and the rudder on the vertical tail. Stability Section 173, 177 and 181 of PART 23 published by the Federal Aviation Authority (FAR) states the requirements that a general aviation aircraft should meet in order for it to be stable. The stability of an aircraft is defined by the ability of an aircraft to
- 103. Design and Development of a Hybrid UAV Abbinaya T.J. 83 | P a g e ME5308 – Major Group Project return back to its original flight path when disturbed by factors such as gusts. The stability of an aircraft can be analysed as static stability and dynamic stability. The stability of an aircraft is measured about the lateral, longitudinal and directional axes. Given below in Table 21 are some of the requirements that the aircraft needs to meet and typical values in order to be stable. The table also shows the stability derivatives that influence the stability and what their typical values should be. These values of stability derivatives were used as a guideline when designing the UAV. Stability requirement Stability derivative and function Typical values in (1/rad) Static longitudinal stability Rate of change of pitching moment coefficient with respect to AoA -0.3 to -1.5 Dynamic longitudinal stability , Rate of change of pitching moment coefficient with respect to pitch rate -5 to -40 Static directional stability , Rate of change of yawing moment coefficient with respect to sideslip angle β +0.05 to +0.4 Dynamic directional stabilty , Rate of change of yawing moment with respect to yaw rate -0.1 to -1 Table 21: Static and dynamic stability requirements. [40] The values of stability derivatives and has to be negative in order for the aircraft to be statically longitudinally stable. These values are highly influenced by the design of the horizontal tail of the aircraft. The values of and are highly influenced by the design of the vertical tail. The value of has to be positive for the aircraft to be statically directionally stable and has to be negative for it to have strong stabilizing effect on the aircraft. [40]
- 104. Design and Development of a Hybrid UAV Abbinaya T.J. 84 | P a g e ME5308 – Major Group Project Figure 42: Graphs indicating the derivatives and for stable and instable aircraft conditions. Static margin and neutral point There are two key points located on an aircraft that determine it degree of pitch stability. These are the Centre of Gravity (CG) and the Neutral Point (NP). The CG is the point where the weight of the aircraft acts and the NP is the point where the pitching moment does not vary with variation in the aircrafts angle of attack [52]. Static Margin (SM) is a parameter that measures the degree of pitch stability by taking into account the position of the CG and NP [52]. The equation below shows this relationship: Even though a high positive static margin such as 0.5 would make the aircraft very stable, it is not really preferable. This is because a high static margin would result in the aircraft having a sluggish response to elevator pitch. Therefore a smaller positive static margin of 0.135 was selected for this UAV. The neutral point of the aircraft is dependent on the aircraft’s geometry and the CG of the aircraft is dependent on point loads of the aircraft. Two methods were used in order to estimate the aircraft’s
- 105. Design and Development of a Hybrid UAV Abbinaya T.J. 85 | P a g e ME5308 – Major Group Project neutral point. The neutral point of the aircraft was determined with the following equations: Result Method 1 ⁄ ⁄ ( ) Method 2 ( ) ( ) Table 22: Methods of determining the location of neutral point [38] [53] Method 1 above uses the geometric dimensions of the aircraft to calculate the location of the NP. Method 2 depends on the curve slope of the wing and tail and the downwash angle to estimate the NP and assumes that = 0. Even though both results give very similar results, the result from the first method were used as it quite straight forward and does make have any assumptions. Since the location of the Neutral Point is determined, and the desired Static Margin is known, components were positioned on the UAV such that the desired static margin is obtained. Currently the aircraft has a static margin of approximately 13.5 %. However, changing position of components (such as avionics) in the UAV will make the CG vary and allow the static margin to be changed for specific flight mission in the future.
- 106. Design and Development of a Hybrid UAV Carlos C.M. 86 | P a g e ME5308 – Major Group Project Aircraft Moments It is fundamental in static longitudinal stability that the moments of the aircraft are balanced. To balance the aircraft in flight and during manoeuvres it is essential that the tail can produce the sufficient forces to counteract moment increments due to an increase in angle of attack of the incoming flow or have the ability to move the centre of gravity around the aircraft so that it has more flexibility in terms of loading. Figure 43 shows the forces on the aircraft with a negative lifting tail, were the force and dimensionless moment equilibrium equations seen below may be derived form. Figure 43 Wing and tail forces. An essential requirement for the aircraft to be stable is that the pitching moment varies negatively with the lift numerically represented by .This is represented graphically in Figure 44.
- 107. Design and Development of a Hybrid UAV Carlos C.M. 87 | P a g e ME5308 – Major Group Project Figure 44 Statically Stable and Unstable pitching moment curves. When the aircraft is statically stable it has a nose down tendency. If the aircraft slows down, lift will be reduced therefore the aircraft will pitch down gaining speed and hence rapidly gaining lift that will make the nose pitch back up. In the initial UAV configuration the wing lift, tail lift and centre of gravity are distributed as seen in the example Figure 43. The Centre of gravity is in front of the wings aerodynamic centre (AC). This creates a counter-clockwise moment from the wings lift, in addition to the wing moment force. The tails functionality is to counteract these moments, but if they add up, as happens in this configuration, the negative(downward) force the tail produces has to be very large and the tail size has to be very large or the tail arm has to be very long. Balancing forces in the vertical direction results in a wing lift being a lot higher than rather than the lift being equal to the weight force. To avoid this problem, for the final configuration of the aircraft the centre of gravity has been moved behind the wings AC. Now the moments from the wing lift and the moment force of the wing oppose each other, partially cancelling out a component of the force that the tail has to provide to maintain the aircraft attitude. This allows the tail size to be smaller and at the same time the negative tail force is lower, which balancing the forces in the vertical components makes the wings necessary lift smaller, almost equal to the aircraft weight.
- 108. Design and Development of a Hybrid UAV Carlos C.M. 88 | P a g e ME5308 – Major Group Project Figure 45 Final aircraft CG Lift configuration. The only limitation is that the centre of gravity has to be maintained in front of the neutral point of the aircraft, producing a positive static margin.
- 109. Design and Development of a Hybrid UAV Abbinaya T.J. 89 | P a g e ME5308 – Major Group Project Control Flight control surfaces are used to control the aircrafts attitude and direction. They are mainly considered as primary and secondary surfaces. Like stability, the aircrafts control can also be evaluated as lateral, longitudinal and directional control. The lateral control of the aircraft is governed by the aileron. The sizing of the aileron is covered in the previous section of this report. The elevator on the horizontal tail is responsible for the longitudinal controllability of the aircraft and the rudder on the vertical tail is responsible for the directional controllability of the aircraft. The design sizing of the elevator and rudder will be covered in the upcoming section. Primary surfaces compromise of the ailerons, elevator and rudder. Table 23 below illustrates the functioning of the different primary control surfaces. Primary Control Surface Aircraft movement Axes of rotation Type of stability Ailerons Roll Longitudinal Lateral Elevator Pitch Lateral Longitudinal Rudder Yaw Vertical Directional Table 23: Control Surface Functions. Control mechanisms such as flaps, spoilers, slats and trim controls are typical secondary control surfaces. They are used to enhance the performance characteristics of aircraft and alleviate the forces exerted on the controls. [54]
- 110. Design and Development of a Hybrid UAV Abbinaya T.J. 90 | P a g e ME5308 – Major Group Project Control Surface Sizing: Rudder and Elevator Elevator The elevator is a primary control surface that is placed on the trailing edge of the horizontal stabiliser. The elevator controls the pitching moment of the aircraft. The design of the elevator design itself is an iterative process like the design process for the other control surfaces on the aircraft. Initial design parameters are first determined. The design parameters are then used to calculate non-dimensional derivatives in order to check the effectiveness of the elevator design. For this particular UAV, the use of the elevator would be most crucial during the take-off climb. Microsoft excel was used to formulate a programme in order to find out the total elevator deflection required at maximum climb during take-off. The size of the elevator was changed if the deflection required exceeded the maximum deflection possible. The elevator deflects in two directions. The downward deflection of an elevator is referred to as positive deflection ad the upward deflection of the elevator is referred as negative deflection. A negative elevator deflection causes the aircraft’s nose to pitch up and a positive elevator deflection causes the aircraft’s nose to pitch down. Various parameters such as the elevator effectiveness parameter, increment in lift coefficient with elevator deflection, elevator effectiveness derivatives and aircraft static longitudinal stability derivatives were calculated to check if the amount of deflection required is within the limit of the maximum elevator deflection. The results obtained were also checked to see if the horizontal stabiliser configuration would stall or not in a critical manoeuvre such as climb [40]. Given below are the steps used to determine the effectiveness of the elevator. 1. Basic design parameters such as elevator planform area (Se), elevator chord (ce), elevator span (be) and elevator maximum positive and negative deflection were determined. These values were initially estimated based on typical values used for general aviation aircraft. The maximum elevator deflection was limited to 20° in order to prevent flow separation and stall of the horizontal tail. 2. The Lift at take-off was calculated. ⁄
- 111. Design and Development of a Hybrid UAV Abbinaya T.J. 91 | P a g e ME5308 – Major Group Project 3. Moments about the wing-fuselage aerodynamic centre were determined. ⁄ ̅ 4. The lift coefficient desired from the tail is calculated. 5. Assuming that the angle of attack of the wing at take-off is the same as the wing incidence angle, the downwash effect, is determined. Where, And 6. The angle of attack of the horizontal tail at take-off rotation was calculated. 7. The effectiveness of the elevator angle of attack was then determined. This was determined by assuming using maximum elevator deflection. ⁄ 8. The elevator to horizontal tail chord ratio was then determined. This was done by reading of from Figure 46.
- 112. Design and Development of a Hybrid UAV Abbinaya T.J. 92 | P a g e ME5308 – Major Group Project Figure 46: control surface effectiveness parameter vs. control surface to lifting surface chord ratio. [40] 9. The lift coefficient available from the horizontal tail with maximum deflection was then determined. This was done with the help of ESDU sheet. The available horizontal tail lift coefficient was then compared to with the desired lift coefficient calculated in step 4 above. It is required that the available lift coefficient is slightly higher than the desired one. Insufficient lift available would make the elevator unacceptable and require the elevator to be redesigned. 10. The next step is to determine the deflection required by the elevator. To know this, it is important to calculate the elevator effectiveness derivatives and the aircraft’s static longitudinal stability derivative are first determined. [ ] [ ] 11. The elevator deflection required to maintain longitudinal trim when the aircraft flies at its maximum velocity with most aft CG location was then determined.
- 113. Design and Development of a Hybrid UAV Abbinaya T.J. 93 | P a g e ME5308 – Major Group Project [ ] The elevator deflection required was found to be less than the maximum possible elevator deflection making the elevator design acceptable. Since the elevator design was acceptable, a few other checks were done to ensure that the horizontal tail does not stall during take-off rotation. Rudder The rudder is a primary control surface placed in the trailing edge of a vertical tail. It enables the pilot to have directional control on the aircraft. The use of rudder is mainly implemented in conditions such as adverse yaw, aircraft spin recovery, coordinated turns or crosswind landings. The rudder works by generating differential lift on each direction depending on the side of deflection. This works by changing the camber in the symmetric aerofoil with the deflection. The rudder lift (side force) generated is at a distance (optimum arm and this generates a torque on the aircraft and which causes the aircraft to rotate about its CG. [55] The main requirements of rudder in the current mission profile would be to enable the aircraft to land safely with crosswinds. Hence, the aircraft’s rudder design was done by mainly taking into account the requirements for crosswind landings. Parameters such as aircraft sideslip angle, stability derivatives and control derivatives where calculated to obtain the required crab angle and rudder deflection for safe cross-wind landing [40]. Given below is the steps used for rudder design. 1. Basic design parameters such as rudder area, rudder chord, rudder span and maximum rudder deflection were determined. 2. Determination of some velocities such as maximum crosswind velocity and total velocity during landing with presence of crosswind. √
- 114. Design and Development of a Hybrid UAV Abbinaya T.J. 94 | P a g e ME5308 – Major Group Project 3. The side force of the aircraft, produced by the crosswind is then determined. The force generated is dependent on the projected side view, of the aircraft. Since the rudder deflection is the same towards both sides, a crosswind from the right was assumed. This would generate a positive sideslip angle. A side drag coefficient, of 0.6 was assumed because of the fuselage having a cylindrical shape. ⁄ 4. The aircraft’s sideslip angle was determined. [ ] 5. The aircraft sideslip derivatives were then determined. [ ] [ ] in the equation is very dependent on the shape of the fuselage and its projected side. It represents the contribution of the fuselage to the aircraft side slip derivative .The contribution of the fuselage to the directional static stability tends to be negative. The typical values used for in an aircraft are between 0.65 and 0.85. [40] The aircraft sideslip derivative, is usually determined using wind tunnel testing. However for purpose of calculation and lack of time, the value was estimated with calculations. in the equation is very dependent on the shape of the fuselage and its projected side. It represents the contribution of the fuselage to the aircraft side slip derivative .The contribution of the fuselage to the directional static stability tends to be positive. The typical values used for in an aircraft are between 1.3 and 1.4. [40] 6. The aircraft control derivatives were also determined.
- 115. Design and Development of a Hybrid UAV Abbinaya T.J. 95 | P a g e ME5308 – Major Group Project above is the angle of attack effectiveness of the rudder. This is determined by reading of from the graph in Figure 46 for the chord size selected. This figure can be referred to for the design sizing of all of the control surfaces in order to find its effectiveness. 7. An excel sheet was formulated such that the concerned parameters are calculated and equations are solved simultaneously to give the crab angle, and rudder deflection required at a particular crosswind velocity. The simultaneous equations solved were: ⁄ ( ) ( ) ⁄ ( ) The crab angle is the angle between the aircraft’s centre line and the runway centre line when making a crosswind landing. Given below in Table 24 are the details of rudder deflection required at various cross wind velocities. Cross Wind Velocity (knots) Crab angle required Rudder deflection required 5 12.37° 3.78° 10 27.95° 11.45° 15 43.46° 17.82° 20 56.95° 20.66° 25 67.86° 20.26° Table 24: Rudder deflection required during various landing at various crosswind velocities.
- 116. Design and Development of a Hybrid UAV Carlos C.M. 96 | P a g e ME5308 – Major Group Project Lift Curve Slopes Numerical Prediction: Rudder and Elevator The process behind the calculations for the curve slope prediction of the elevator and rudder is the same as the methodology used for the wing control surfaces, resumed in Table 17. First the properties of the stabilizers have to be determined, and then the properties of the control surface with an initial estimate of their size and finally the effect of their deflection can be obtained and reiterate to optimise the control surface size. The table with all the results for the rudder and elevator may be seen below. Curve Slope Results Rudder Elevator units (a1)0/(a1)0T 0.642 0.642 (a1)0T 6.854 6.854 1/rad (a1)0 4.401 4.401 1/rad (a1)0 -3D 1.340 3.404 1/rad (a2)0T 4.525 4.525 1/rad (a2)0/(a2)0T 0.345 0.345 (a2)0 1.561 1.561 1/rad (a2)0 -3D 0.863 1.416 1/rad Table 25 Rudder and elevator curve slope results using ESDU method, to be used in the control surface sizing. Note: No angle correction is needed because the empennage has a symmetrical aerofoil NACA0012, meaning that the lift curve slope goes through the origin, or has a zero-lift angle of attack of zero degrees. -0.4 -0.3 -0.2 -0.1 0.0 0.1 0.2 0.3 0.4 -6 -4 -2 0 2 4 6 LiftCoefficent Angle of Attack α (degrees) CL vs. Alpha Rudder Defelction CL ESDU (CS 0deg) CL (CS- -20deg) CL (CS- -10deg) CL (CS-10deg) CL (CS-20deg)
- 117. Design and Development of a Hybrid UAV Carlos C.M. 97 | P a g e ME5308 – Major Group Project Figure 47 Shows the rudder curve slope with deflection angles of ±20 degrees. Figure 48 Shows the elevator curve slope with deflection angles of ±20 degrees. Figures 47 and 48 show the effect of the control surface deflection of the horizontal and vertical stabilisers. The rudder has an increment in the lift coefficient of ±0.12 per 10degrees deflection. The Elevator has a variation in the lift coefficient of about ±0.25 per 10 degrees deflection. -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 -8 -6 -4 -2 0 2 4 6 8 LiftCoefficent Angle of Attack α (degrees) CL vs. Alpha Elevator Defelction CL (CS -20deg) Up CL (CS 0deg) CL (CS 20deg) Down
- 118. Design and Development of a Hybrid UAV Camilo V. 98 | P a g e ME5308 – Major Group Project CG Envelope The center of gravity envelope refers to a visual representation of the limitations for the center of gravity depending on different total vehicle weights. Diagrams can display longitudinal, lateral as well as vertical envelopes [56]. For the purpose of this report, and prototype aircraft, the longitudinal envelope was studied to ensure stability on different mission loads and weights. The position of the center of gravity within an aircraft is limited by the maximum force created by the tail, as-well as the neutral point. These give the forward and aft limits respectively [47]. There are two differences between the project vehicle cg envelope diagram and a normal aircraft cg envelope. First is the additional limit visible on the actual diagram. This is the legal weight restriction of 7.0 kg maximum aircraft weight, and it acts as a cut-off point on the diagram as can be seen below. The second difference to traditional envelopes is the fact that throughout flight due to the electric nature of the UAV, the weight does not change between take-off to landing. Figure 49 Longitudinal CG Envelope for Project vehicle
- 119. Design and Development of a Hybrid UAV Camilo V. 99 | P a g e ME5308 – Major Group Project Usually structural considerations such as maximum Landing gear load depending on the balance between all the components of the undercarriage are incorporated into the envelope diagram. However it was not a limiting factor in the case of the project aircraft due to the loads restricted by legal reasons, hence it was not put into the CG Envelope.
- 120. Design and Development of a Hybrid UAV Bennie M. 100 | P a g e ME5308 – Major Group Project 4.2.6. Stability: Vertical Take-Off and Landing In addition to the stability of an aircraft being implemented in the design, the structure had to allow for the cg to be placed in such a way as to allow the tri-copter aspect of the aircraft stable. This stability is known as the golden triangle, which refers to a point which is equidistant from all the motors. Two of the propellers in the tri-copter are in a counter rotating pair, cancelling out any yawing moments created by the spin of the propellers. However since the third propeller doesn’t have a countering partner, it will have a yaw effect on the aircraft which if left to its own devices would make it uncontrollable. To counter this moment, the tail arm motor which in this case is the propeller mounted on the nose of the aircraft has a tilting mechanism to cancel out any yaw induced by its propeller. It also helps to steer the tri-copter by providing additional yaw control during flight. The design of the UAV is such that it hybridises a fixed wing and tri-copter. In this section the VTOL aspect of the design is discussed in the context of the tri-copter as it will be primarily responsible for providing the lift to achieve vertical lift off when VTOL is being attempted. How the stability of the tri-copter aspect of the design is achieved is also discussed as well as important theory behind rotorcraft relative to the VTOL system. The tri-copter utilises 3 rotors to achieve lift and perform manoeuvres. A model of the tri-copter can be created when regarding the tri-copter as a rigid body [57] [58]. Using modelling methods described in [58] a base model of the tri-copter can be created as shown in Figure 50. From the model we can then derive Force, Moment and Kinematic equations [57] [58] as shown in Equations 4.2.6.1, 4.2.6.2 and 4.2.6.3. In accordance to [57] [58] we assume that external forces in relative axes ( ) and moments (L, M, N) are acting on the CG, this will produce translational velocities (u, v, w), rotational velocities (p, q, r) as well as rotational angles ( ) and rotational inertias ( ).
- 121. Design and Development of a Hybrid UAV Bennie M. 101 | P a g e ME5308 – Major Group Project Figure 50 Tri-copter configuration with reference axes. Derived Force Equations ̇ ̇ ̇ (4.2.6.1) Derived Moment Equations ̇ ( ) ̇ ̇ ( ) (4.2.6.2) Derived Kinematics Equation ̇ ̇ ̇ (4.2.6.3)
- 122. Design and Development of a Hybrid UAV Bennie M. 102 | P a g e ME5308 – Major Group Project Pitch & Roll Control It is important to understand how a tri-copter is able to perform attitude adjustments such as pitch, roll and yaw. As the tri-copter does not have control surfaces like that of a fixed wing aircraft it has to use its rotors to perform attitude adjustment. This is achieved by adjusting the angular velocity and thus the RPM of a rotor. Shown in Figure 51 it is observed that in order to achieve pitch Rotor 1’s angular velocity (1) has to be greater than that of Rotor 2 (2) and Rotor 3 (3) while 2 and 3 are equal to each other (1 > 2 = 3) [57]. In Figure 52 it is also shown that in order to achieve CW roll 3 has to be less than 1 which is less than 2 (2 > 1 > 3) [57]. Conversely to achieve CCW roll, as shown in Figure 53, 3 has to be greater than 1 which is greater than 2 (2 < 1 < 3) [57]. Figure 51 Pitch up by using Rotor 1. Figure 52 Roll in the Clockwise direction. Figure 53 Roll in the Counter Clockwise direction.
- 123. Design and Development of a Hybrid UAV Bennie M. 103 | P a g e ME5308 – Major Group Project Yaw Control As shown in Figure 50 the tri-copter has 3 rotors setup in such a way that 2 rotate counter clockwise and one clockwise. Due to the fact that there are two coupled rotors rotating in the same direction they have a sum yawing effect on the tri-copter. Thus establishing yaw authority is important as it is important that the tri-copter be able to track correctly when command is given to move forward, backwards, left, right, up or down. This is achieved by tilting Rotor 1 (see Figure 54) at an angle of α referred to as the “tilting angle” [57], in doing so the yawing moment caused due to the coupled CW rotors; in this case 1 & 3, is cancelled [57]. In [57] and [58] it is shown that the tilting angle has 3 DOF and is essential during hover and movement in order to maintain an appropriate amount of accuracy [57]. Using the model developed and shown in Figure 50 it is possible to derive force and moment equations that show the importance and effect α has on the force and moments produced by the tri-copter. These equations are shown in matrix form in Equations 4.2.6.6 and 4.2.6.7. Additionally, deriving Force (F) and Torque () in regards to angular velocity () is important and can be used for each individual rotor using Equations 4.2.6.4 and 4.2.6.5. where k and kt are the torque and thrust coefficients respectively. Figure 54 Yaw authority of a tri-copter. Where (4.2.6.4) Where (4.2.6.5)
- 124. Design and Development of a Hybrid UAV Bennie M. 104 | P a g e ME5308 – Major Group Project So: ⃗ [ ] (4.2.6.6) So: ⃗⃗⃗ [ ] (4.2.6.7) To analyse how the tri-copter is able to perform manoeuvres such as hover, vertical climb and forward flight we can look at a single rotor and use the equations for each rotor. Hover Figure 55 Mass Flow of air through rotor in hover. In hover a rotorcraft will normally be attempting to produce enough lifting force (in this case thrust) to keep the weight of the craft aloft in the air or just off the ground; provided phenomena such as ground effect are negated. If the aircraft is assumed to not be accelerating in any axis (x, y, z) then there is no production of dynamic thrust and instead only static thrust is considered. Additionally assuming momentum is conserved we can derive equation 4.2.6.8 from equation 4.1.1.1b: ̇ (4.2.6.8) With the aforementioned assumptions of static thrust, no acceleration (no movement) and negation of ground effect we can assume that ‘the airflow far above and the rotor are zero while in a hover’ [59] as shown in Figure 55. Thus the
- 125. Design and Development of a Hybrid UAV Bennie M. 105 | P a g e ME5308 – Major Group Project difference between the velocity below (v) the rotor and far above (v = 0) is equal to the velocity below (v) this means that net mass flow through the rotor [59] is: ̇ (4.2.6.9) As mentioned, when the propeller rotates through the air it will induce a lifting force as well as an induced velocity ( ) as shown in Figure 16, thus equation (4.2.6.9) can be re-written as [59]: Where = w ∴ (4.2.6.10) Additionally if considering Figure 55 to be a closed system work done by the rotor and work done by the rotor wake would be the same. Where the work done by the wake would be the total change in kinetic energy in the wake [59]. Thus we can derive equations for both work done by the rotor ( ) and work done by the wake ( ): (4.2.6.11) ̇ (4.2.6.12) However ∴ = ⇨ (4.2.6.13) Now we can subst. 4.2.6.13 into 4.2.6.10 (4.2.6.14) Rearranging (4.2.6.14) for √ (4.2.6.15) If the UAV was initially at ground level and when powered up rose to some altitude at which it hovered there would be some form of drag on the UAV as the air flowed over the UAV. Using the drag equation (4.2.6.16); where ( ) is drag coefficient and ( ) is plan-form area, we can then determine that for the UAV to maintain a hover and remain in equilibrium, the thrust produced by the rotor most equal the weight of the UAV while overcoming the effect of drag(vertical drag) [59]. This will produce
- 126. Design and Development of a Hybrid UAV Bennie M. 106 | P a g e ME5308 – Major Group Project equation (4.2.6.17). It is also possible to determine the power needed to achieve the hover if we ignore power-loses [59] this is shown in equation (4.2.6.18). (4.2.6.16) (4.2.6.17) (4.2.6.18) When all 3 rotors are in use in order to achieve hover they must all be running at the same rpm and therefore be producing the same amount of thrust as shown in Figure 56 this way there will be no production of pitch or roll. However yaw would still be problematic but easily overcome by having rotor 1 tilt to cancel the adverse yaw and run at an rpm that produced a vertical component of thrust equal to that of the other two none tilting rotors. Figure 56 Altitude Hold (Hover) with all 3 rotors.
- 127. Design and Development of a Hybrid UAV Bennie M. 107 | P a g e ME5308 – Major Group Project Vertical Climb Figure 57 Mass Flow of air through rotor in vertical climb. In a climb it is it important to take into account the vertical climb velocity ( ) as shown in Figure 57. The model is exactly the same as the hover model however now we account for [59]. This is done by adding and then substituting in to equations 4.2.6.9 and 4.2.6.10: ̇ (4.2.6.19) (4.2.6.20) Also work will be the same so we use 4.2.6.13 and substitute it into 4.2.6.20: (4.2.6.21) From equation 4.2.6.21 we can gain an expression for the induced velocity : √( ) (4.2.6.22) Similarly the vertical drag experienced in the climb has to account for the vertical climb velocity ( ) additionally we account for the area outside the rotor wake ( ): (4.2.6.23)
- 128. Design and Development of a Hybrid UAV Bennie M. 108 | P a g e ME5308 – Major Group Project In-order to climb the amount of thrust required is intuitively higher therefore the amount of power required increases as well this can be derived by accounting for in equation 4.2.6.18: (4.2.6.24) If the tri-copter is attempting a vertical climb and maintains a level attitude (no pitching or rolling) then we can assume that the thrust produced by all 3 rotors has to be greater than the weight of the UAV as shown in Figure 58 below. Figure 58 A level vertical climb by the tri-copter.
- 129. Design and Development of a Hybrid UAV Bennie M. 109 | P a g e ME5308 – Major Group Project Forward Flight Figure 59 Flow of air through the rotor in forward flight. In forward flight we have a number of factors at work and the complexity of analysis is increased due to the fact that local airspeed will vary depending on what section of the rotor is being analysed [59]. In Figure 59 it is shown how the wake is developed when the rotor is pitched at an angle of attack ( ) where it is considered negative when pitching down and positive when pitching up [59]. We derive that: (4.2.6.25) To better understand how forward flight works for the rotor craft we have to consider the azimuth angle () this is the angle the blade of the propeller traverses when rotating as shown in Figure 60(x-y axis). We then consider the propeller to be a disc (rotor disc) when in the x-z axis.
- 130. Design and Development of a Hybrid UAV Bennie M. 110 | P a g e ME5308 – Major Group Project Figure 60 Rotor Disc showing Azimuth angle. Similar to hover and vertical climb we can derive mass flow rate for the rotor when the tri-copter is in forward flight. We assume that the resultant velocity (U) at the rotor disk is equal to the velocity in front of the rotor disk (U= ): ̇ (4.2.6.26) ∴ (4.2.6.27) ∴ √ (4.2.6.28) Now substituting 4.2.6.28 into 4.2.6.26 ̇ √ (4.2.6.29) It is now possible to derive an equation for thrust when in forward flight: √ (4.2.6.30) Transition As explained in the beginning of section 4.2.6 the UAV is a hybrid between a fixed wing aircraft and a tri-copter rotorcraft. In order to achieve VTOL three rotors are to be used to provide the vertical thrust required to lift the UAV to a hover altitude. In order to accomplish STOL an EDF is used to produce high forward thrust when attempting to transition from hover (an altitude hold) to horizontal flight. Shown in
- 131. Design and Development of a Hybrid UAV Bennie M. 111 | P a g e ME5308 – Major Group Project Figure 61 is model of the complete UAV setup consisting of the EDF and three rotors. Figure 61 Full model of UAV at a hover. Once the UAV is hovering the EDF is then activated to produce forward thrust (see Figure 62). It is necessary for the EDF to produce enough thrust to allow the UAVs wings to generate lift to keep the UAV in flight. As the UAV approaches cruise the appropriate velocity the three rotors are powered down proportionally to the increase of thrust produce produced by the EDF. Figure 62 Full model of UAV in transition.
- 132. Design and Development of a Hybrid UAV Bennie M. 112 | P a g e ME5308 – Major Group Project When the wings of the aircraft a producing enough lift to keep the UAV aloft the rotors are fully powered down and only the EDF is producing thrust when required by pilot input. This is when the UAV is considered to be in full horizontal flight as shown in Figure 63. Figure 63 UAV model in full horizontal flight.
- 133. Design and Development of a Hybrid UAV Camilo V. 113 | P a g e ME5308 – Major Group Project 4.2.7. Structures Fuselage Fuselage design is a key part of the aircraft design process. It is this structure which will need to be capable of holding all the mission payloads and is required to take any loads expected within the mission profile. Typically these Loads which are experienced in flight arise in the form of five main loads from the wing during gusts, turbulence or general aircraft manoeuvres, Landing gear loads from ground impact and induced load from the thrust provider, when it is attached to the fuselage [60]. These are: - Tension - Compression - Torsion - Shear - Bending A single component of the structure can be subjected to multiple forces in one instance [32]. There is usually an additional load due to cabin pressurization which can be ignored for the UAV in the case of the project vehicle. For the aircraft in question, an additional fuselage load comes from the tri-copter mission segment where the fuselage would experience additional bending forces from being suspended between 3 motors at hover. To be able to support and resist the aforementioned loads, currently three main types of fuselage design exist. These are the monocoque, semi monocoque and truss structured fuselage (warren truss) [47]. Monocoque designs consist of a series of formers wrapped in a load bearing skin. The issue with a monocoque design is in finding a material strong enough to sustain the stresses of the fuselage without any connecting members between the bulkheads or formers within the structure itself. As the structural integrity of the skin is key to the whole structure strength in the event of a tear in one location, the whole fuselage loses a significant amount of rigidity and strength.
- 134. Design and Development of a Hybrid UAV Camilo V. 114 | P a g e ME5308 – Major Group Project Figure 64 Monocoque fuselage design [61] The truss structured fuselage is common in older and smaller types of aircraft, for example typically micro-light aircraft such as the sky ranger [62] and older bi- plane/tri-planes such as the Fokker DR-1 [32].The truss faded after heavy use during the First World War due to breakthroughs with technologies regarding monoplane designs as well as the advent of semi monocoque techniques which reduced the weight of the structure. Figure 65 Truss fuselage structure [32] Semi monocoque fuselage design mimics a full monocoque design. However it bypasses issues usually found in the latter with the relative weight of the structural components by incorporating more structural members between the formers as seen in the figure below.
- 135. Design and Development of a Hybrid UAV Camilo V. 115 | P a g e ME5308 – Major Group Project Figure 66 Semi-monocoque Fuselage [32] Of the aforementioned methods the most widely used is the semi monocoque design method. This encompasses the rigidity of a skin but with the bulkheads connected by structural longerons and stringers to help take the load along the fuselage from the forces during flight. These additional structures prevent tension and compression stresses from causing a potentially fatal bending on the fuselage [63]. Good examples within UAVs of a semi monocoque structure consists of a wide range of vehicles varying in size from the very large Global Hawk, a medium sized Falco, to the smaller scan eagle. The fuselage compositions consist of load bearing frames in conjunction with a load bearing skin, cutaway sections for the above mentioned aircraft are shown below. Figure 67 Global Hawk Cutaway [64] Above can be seen the cut-out showing the fuselage of the Global hawk. This semi monocoque design can be seen to have bulkheads in the form of formers connected with structural longeron beams along the outside corners of the formers. The
- 136. Design and Development of a Hybrid UAV Camilo V. 116 | P a g e ME5308 – Major Group Project payload of avionics is stored within the structure, along the empty space created by the hollowed out formers. Figure 68 Falco Cutaway diagram [64] The Falco structure is similar to the global hawk however it is a lot smaller in size, so there are less formers and longerons along the internal structure. Various bulkheads create compartments within the fuselage which is held by four beams forming a rectangular frame at the bottom for heavy landing loads. Although this UAV has a fixed undercarriage the wheels have been faired to reduce any extra drag created by the components themselves. In contrast to the above two the Scan Eagle has a particularly interesting structure. The fuselage bulkheads and frames all come together within the skin to form modular compartments which appear to be easy to get to for maintenance work as well as configuration of payloads between missions. This adaptability of the structure can be seen in the cutaway figure shown below. Figure 69 Cutaway of the ScanEagle [64]
- 137. Design and Development of a Hybrid UAV Camilo V. 117 | P a g e ME5308 – Major Group Project Where-as some follow a more classical design, the Demon UAV follows the scan eagle in having a distinct fuselage structure. It consists of formers linked together to create a solid structure underneath the carbon fibre skin without the aid of additional longerons and stringers within, it was not completely a true monocoque, more inbetween monocoque and semi-monocoque as can be seen by the incomplete structure shown on the figure below. Figure 70 Bonding in progress of the Demon UAV composite structure [65] The final source of inspiration and guidance on what the structure of the project aircraft should be originated from smaller, cheaper UAV/FPV (First Person View) Aircraft originating in the R/C market. These other lightweight UAVs include the Bormatec Maja [66] and the RV JET [67]. The Bormatec fuselage consisted of a soft foam encasing a supporting structure in the form of corrogated plastic plating running the length of the fuselage where as the RV Jet resembles the scan eagle in its modular composition with a monocoque twist within the fuselage design. The fuselage comes in two halves ready made with formers within the structure with the wings being able to slot in, with allowance for extensions along two carbon rod spars running through the whole aircraft. Unorthodox designs built around specific vehicle requirements allowed for the general types of fuselage studied to be narrowed down and modified to cater for the particular design restrictions required by the project aircraft. Due to the heavy industry tendency to design and implement a semi monocoque fuselage, this format became the selected structure type for the prototype being developed. This led to the fuselage structure that consists of 3 main load bearing carbon rods of 10mm diameter, with Plywood bulkheads of varying thickness spread along vital areas of the fuselage. This was then covered by a surface skin of foam
- 138. Design and Development of a Hybrid UAV Camilo V. 118 | P a g e ME5308 – Major Group Project board to give it additional rigidity as-well as provide a light covering for the fuselage section. The main three spars within the fuselage were laid out in a triangular set up essentially borrowing elements of the truss fuselage structural design, the figure below shows the reasons for the design decision. Any landing loads transferred along the bottom rod and landing gear bulkhead is transferred between all three members through compression and tensile forces. This set-up helps the structure absorb any potential hard landing loads. Figure 71 Loading on a triangular structure [68] The shape of the bulkheads was also key in designing an inherently strong structure. These were heavily influenced by the use of arches in the civil engineering sector [69]. Essentially the job of the bulkheads was to provide adequate load paths in the form of the interconnecting material, mimicking the characteristics of a triangular truss component for any compressive and tensile loading experienced upon landing from horizontal flight. The Bulkheads were designed so that the load bearing characteristics of a circular fuselage remained in the form of arches on the bottom and top, whilst decreasing the overall surface area and volume of the equivalent circular fuselage cross section. Figure 72 shows a computer image of the skeletal structure.
- 139. Design and Development of a Hybrid UAV Camilo V. 119 | P a g e ME5308 – Major Group Project Figure 72Skeletal frame of the fuselage Consideration was taken with regards to where the rod which would attach the body propeller would be paced in order to allow for sufficient tilt of the mechanism without striking the top two fuselage carbon rods. Surrounding the skeletal structure of the fuselage is a 3mm thick layer of foam board. It was implemented to provide a rigid body and protection of the internal electrical components from the elements. Foam board is a material which takes the concept of sandwiching a layer of a soft core material in-between two harder components to create an overall stronger substance. In this case it is a thin pliable polystyrene board
- 140. Design and Development of a Hybrid UAV Brett M. 120 | P a g e ME5308 – Major Group Project Landing Gear Brett The landing gear is the intermediary between the aircraft main structures and the ground. A typical landing gear system includes legs (also known as struts), wheels and brakes or even skids or floats if the aircraft is destined to operate from snow and ice fields or on water. The landing gear could be either fixed or retractable but at a weight and cost penalty. The landing gear must be designed big enough to support the aircraft’s weight but light enough as to not compromise the handling qualities of the aircraft. The design selected for the aircraft is what’s known as the tricycle configuration [70] with a single nose strut and two rear main struts. Large diameter wheels have been used to reduce rolling resistance and improve the aircraft’s handling over rough ground. The tricycle configuration was selected because it is structurally efficient, provides good control characteristics (for ground manoeuvring) and can be built or sourced relatively cheaply and easily. In establishing the size of the landing gear, considerations must be made for the static and the dynamic loads the gear is required to cope with. Static loads are that when the aircraft is not moving and only the weight of the aircraft under the force of gravity is present. Dynamic loads are additional forces acting as a result of movement of the aircraft over rough ground, for example where the wheels are moving erratically as the result of a rough take-off or landing field surface. Mounting brackets and points suitable for use to attach the bought in landing gear equipment were almost entirely defined and tested using Computer Aided Design (CAD) packages to size and validate their layout. Mounting bracket definition and results of several stress test simulations are shown in Section 4.2.8. The location of the landing gear system is usually determined by a load sharing scheme [71] whereby a proportion of the aircraft’s total mass is supported by the front strut and the rest by the rear struts. A graphical representation of the method used is given by Figure 73 and the calculated position of the landing gear is given by Equations 4.2.7.1 to 4.2.7.3.
- 141. Design and Development of a Hybrid UAV Brett M. 121 | P a g e ME5308 – Major Group Project The rear wheels act as the pivot on which the aircraft rotates during take-off. This load sharing uses the weight of the aircraft to aid in the front wheel’s ability to steer the aircraft when the aircraft is taxiing. Too much weight over the front gear would likely mean that the aircraft’s horizontal tail would struggle to produce the force needed to rotate the aircraft on the rear wheels to achieve take-off. For this reason, the mounting bracket was designed to allow variable positioning of the rear landing gear placement to achieve the best take-off rotation performance. Figure 73 Landing Gear Positioning for Proper Weight Distribution [71] To determine the placement of the landing gear, prior knowledge of the position of the front landing gear and the aircraft’s theoretical centre of gravity (CG) are required. The front landing gear strut was to be mounted onto the front fuselage bulkhead, the position of which from the nose of the aircraft was known from a master CAD layout of the aircraft and measured at 290mm. From the master CG calculation table, a distance of 710mm was determined when measured from the nose. Thus the distance between the front landing gear strut and the CG follows Equation 1: Front Landing Gear to Aircraft CG Dimension Equation 2: Total length Between Front and Rear Main Landing Gear Equation 3: Distance to Rear Main Landing Gear from Aircraft CG
- 142. Design and Development of a Hybrid UAV Brett M. 122 | P a g e ME5308 – Major Group Project From Equation 4.2.7.3 we now have a location at which to place the rear main landing gear behind the aircraft’s CG. The weight sharing front to rear can be checked physically once the aircraft is assembled. Figure 74 Moveable Landing Gear Concept Material selection for use in a landing gear system must meet the criteria already specified namely lightweight and high strength characteristics as well as being readily available and cheap. Many off-the-shelf landing gear stems are now made from carbon fibre for this reason and it is this material that was used where possible, such as the rear main struts. Where carbon fibre was not available, aluminium can used as an alternative. The mounting bracket will be fabricated using 3D printed ABS because of its lightweight and high strength characteristic and the additive manufacturing process allows for good dimensional accuracy and repeatability if and when the mount requires redesign. This landing gear system design did not make the final build because doubts had been raised on the ability of the anisotropic ABS material to manage the dynamic loading the landing gear might experience during take-off or landing on a rough field. During the build in fact, the bracket broke as shown in Figure 75. A redesign of the bracket or the re-orientation of the ABS layers was not possible in the time window remaining. Nose Section Swivelling nose fan Forward bulkhead 710mm 290mm CG position 74.1mm Front landing gear position
- 143. Design and Development of a Hybrid UAV Brett M. 123 | P a g e ME5308 – Major Group Project Figure 75 ABS Landing Gear Mount - Broken During Aircraft Assembly
- 144. Design and Development of a Hybrid UAV Camilo V. 124 | P a g e ME5308 – Major Group Project Camilo The rear landing gear concept of a moveable rig was in development for a long time, and when it was showcased to the group, the component failed under very low loading. Due to the uncertainties leading up to this point regarding the main rear landing gear, a back-up design was devised. This consisted of a simple plate attachment using a separate carbon fibre strut from the primary one which had been purchased abroad. In order to ensure that the load bearing aspect of the aircraft was sound, the bulkheads and back-up landing mount were tested via simulation to see if they would be able to cope with transferring the load between the rods without breaking. FEA Analysis was run using the geometry imported from Solid works into the Static Structural module of ANSYS Workbench 14.0. Below can be seen a screen shot of the moment of greatest loading (principle stress) on the landing gear bulkhead with the plate connection between the aft landing gear and the bulkhead during a hard landing equivalent of a 2 g force. Included in the assembly assessed is the mount for the landing gear in order to greater displays the load bearing capacity of the structure as a whole. It can be seen that the plywood plate along the bottom takes the brunt of the stress, with compression and tension occurring along the bottom and top two rod connections taking some of the load of the landing, as per the figure displaying the triangular loads in the above section. Figure 76 ANSYS principle stress analysis on bulkhead displaying key on the left
- 145. Design and Development of a Hybrid UAV Camilo V. 125 | P a g e ME5308 – Major Group Project The simulation was modelled with three fixed supports acting on where the rods would connect to the bulkhead, with a load of 120 N acting upwards into the assembly. This was to simulate the rigidity of the skeletal structure through the bulkheads on a hard landing. It is important to note that any bending shown below is excessively exaggerated by the program. When results of the maximum forces are compared to typical values of plywood material properties, the material is more than capable of withstanding such loads being exerted upon it. The maximum principle stress experienced by the material is 0.213 MPa. From this maximum value the weakest material properties of plywood were researched to assess where the failure of the material would occur. Below is a figure showing typical values. 7.5 mm Canadian Aspen Plywood 9.5 mm Canadian Aspen Plywood Orientation of applied force to face grain 0° 90° 0° 90° Planar Shear Strength (MPa) 0.72 0.72 0.55 0.72 Table 26 showing properties of similar thickness plywood material strength [72] The strength shown is the strength of the material under a shear forces through the plane which is the direction the load of the landing traverses through the bulkhead. These values stay roughly the same although other material strength properties increase with an increase in the thickness of material [72]. As a result it can be seen that the material can easily sustain a heavy landing of a 2 g load, with the plate along the bottom taking the majority of the load. In this manner if anything were to fail it would be the plate. In addition to this “throw-away” component which can be easily replaced, the epoxy joint would fracture further absorbing the load of the landing protecting the important structural bulkhead of the fuselage.
- 146. Design and Development of a Hybrid UAV Arturs D. 126 | P a g e ME5308 – Major Group Project Wings Aircraft wings are the main part of the aircraft since they generate lift which allows aircraft to take-off and maintain altitude. A wing can have different profiles which could determine the complexity of the structure but mainly affects the performance of the aircraft. In order to determine the aircraft structure it is important to know the loads and forces to which the aircraft and the wing specifically will be subjected. First of all the wing holds the weight of the whole aircraft so it would be subjected to aircraft weight as well as additional lift forces for altitude gain, aircraft acceleration and the moments about the wing. There are five major types of stresses to which the aircrafts are subjected: tension, compression, torsion, shear and bending. Bending is a combination of tension and compression where, tension occurs on the outside of the band and compression on the inside. The wing is usually subjected to some torsion and bending which is tension and compression. The wing would be subjected to compression all through the flight where for example the leading edge meets the airflow and stagnation point occurs where the air flow exerts a force on the leading edge of the wing exerting a compressive force on it [73]. One of the main factors which have to be taken into consideration when designing a wing structure would be a Load Factor - n. Where L is the lift force (N) exerted on the aircraft and W (N) is the weight of the aircraft. Usually highest forces exerted on the aircraft are at manoeuvres such as sharp turns and the aircraft has to be designed to be able to withstand these forces. The regulations state that the aircraft has to be able to withstand maximum load factor of 3.8 and the minimal requirement is 2 if it can be proven through probability analysis that the load factor would not be exceeded. Similarly for negative load factor the values are -1.5 and -0.5 with proof [74]. These are one of the strictest regulations which have to be met for a UAV in order to be certified. Since this is not an aerobatic aircraft and was not designed to do sharp manoeuvres the limit load factor which it needs to be designed for, is 3 including the safety factor.
- 147. Design and Development of a Hybrid UAV Arturs D. 127 | P a g e ME5308 – Major Group Project Figure 77 demonstration of typical wing structure [75] Generally aircraft wing consists of 4 main components demonstrated in figure 1. Ribs playing the role of bulkheads which determine the aerofoil shape and sustain the compressive force, they position chord wise and determine the internal structure of the wing. The ribs are joined together by stringers which have little load bearing purposes. The core load bearing component would be spars. They prevent bending and torsion, carry most of the load and usually connect the wing to the fuselage. Spars could be in different shaped beams such as I-beams or cylindrical beams. The final component of the structure would be the outer skin of the wing which mainly protects the aircraft from the external influences and gives it the final shape. It is important for the skin to be smooth to keep the skin friction drag to a minimum and solid to maintain the wing profile [73]. The aircraft wing requires o be light and strong and this design allows the majority of the wing to be hollow. Such wing structure used by commercial aircrafts as well as military, UAVs and hobby RC planes. There are some differences such as materials and some minor components however, the principle is the same. Majority of homebuilt RC planes use balsawood ribs and stringers for the internal structure of the aircraft since they are quick and easy to manufacture (by laser cutting) and they are very light. Such structures are usually covered with plastic film
- 148. Design and Development of a Hybrid UAV Arturs D. 128 | P a g e ME5308 – Major Group Project which is heat shrank around the structure. It is difficult to use other materials for skin which use adhesive since there is little surface are to attach it to. One of the disadvantages using this type of structure is that between the ribs the film sags and changes the profile shape which affects the performance of the wing. In order to decrease this effect the distance between the ribs has to be decreased by introducing more ribs which increase the weight of the whole structure. The VTOL aspect of the aircraft introduced a lot of additional weight due to 3 additional motors, large battery to power them and other avionics which increases the wing loading. Using different materials such as plywood would increase the weight of the structure dramatically. Another option which is often used by RC planes manufacturers is using low density foam for the wing core such as “Transall C-160” for an example [76]. Foam core replaces the ribs and stringers by introducing whole wing shaped piece of foam. Further analysis of foam and balsawood properties were analysed and the data in table 1 was used as a guideline for decision making. Table 27 properties comparison of foam core wing reinforced with carbon fibre spars to balsawood ribbed structure reinforced with carbon fibre spars [77] Expanded Polystyrene Foam (EPS) and Expanded Polypropylene Foam (EPP) are both suitable for wing manufacturing and both are being used however they have different properties and different densities. EPS foam is a lot lighter that EPP but a lot less rigid therefore, when using this foam it has become a standard procedure to reinforce the leading and trailing edges of the wing by the foam wing manufacturers [78]. Reinforcement is usually done with wood such as balsa or obeche. It is a lot easier to work with foam since it has large surface area which allows broader range of materials for the skin. Various materials were considered for outer skin of the wing such as: carbon fibre polymer lamination, fiberglass polymer lamination, plastic film heat shrink and wood veneer. Fibreglass and carbon fibre reinforced epoxies have
- 149. Design and Development of a Hybrid UAV Arturs D. 129 | P a g e ME5308 – Major Group Project great properties and relatively low weight however the weight is still larger than wood and requires complex manufacturing methods such as Vacuum Bagging [79]. After considering pros and cons it was decided to use a thin layer of obeche veneer to add rigidity to the wind and heat shrink a thin layer of plastic film on top to add smooth surface finish. [80] In order to reinforce the wing, different types of spar structures were considered such as: monospar, multispar and box beam. It was decided to use cylindrical carbon fibre reinforced multispar system with two beams due to, mechanical properties, wide availability and ease of use. Using rodss as spars fits well with the rest of the aircraft skeletal structure which is mainly consists of carbon fibre reinforced rods. Especially designed for this project rod-to-rod connections were used for connecting wings to the boom tail and wings to fuselage where a perpendicular cross fitting connection were used. The fuselage rod slots into the connection through the bottom elongated, rod slot and both spars from the wing slot through two top, perpendicular hoops attached to the slot. Very similar wing to fuselage connection type is used for commercial “RVJET” flying wing by RangeVideo [81]. Both carbon fibre reinforced resin rods and the connections are described in greater detail later in the report as well as the results of the tests carried out to prove the capability to sustain the exerted loads and forces. Obeche Wing veneer Obeche timber comes from African Triplochiton Scleroxylon tree. Most of aircraft foam wings are veneered with a thin layer of obeche since they add rigidity to the wing however, it is easy to work with since it is not too rigid and can be shaped around the wing. It has very low density almost like balsa wood but better properties for wing veneering. It works very well with glues and can easily be glued on to the foam. Smooth surface finish is easier to achieve that with balsa wood. To further improve surface finish the wing was covered with plastic film using heat shrinking technique which improves the properties further and protects from moisture and other external influences. Sourced pieces Main sourced parts that were obtained for the project are carbon fibre reinforced resin rod. Rods mainly connected major sections of the aircraft using rod
- 150. Design and Development of a Hybrid UAV Arturs D. 130 | P a g e ME5308 – Major Group Project connections. Rods were used as the main load bearing structure which reinforced the wings, the tail and held the whole fuselage together. For wing reinforcement and the boom tail rods were all 20mm outer diameter with 1mm wall thickness. Since the larger, 20mm diameter rods are main load bearing and will be exposed to higher forced, the braided type was selected for improved bending performance properties. The wings and the tail were also sourced externally by a wing manufacturer. The design of both the tail (two vertical stabilisers and one horizontal stabiliser) and wings were done by the team and the technical drawings were supplied to the manufacturer with material descriptions. The wing manufacturer used CNC hot wire technique to cut out the wing shape foam core. Then the foam core was manually veneered with thin layer of obeche, the control surfaces were cut out and both leading edge and the trailing edge were reinforced with balsa wood. Some procedure were carried out for the both vertical stabilisers and the horizontal stabilizer.
- 151. Design and Development of a Hybrid UAV Brett M. 131 | P a g e ME5308 – Major Group Project 4.2.8. Computer Aided Design and Technical Drawings Brett CAD is a very powerful tool which allows the designed to ‘flesh’ out ideas in a virtual environment. This allows the idea to be further defined and validated with the aid of in build Finite Element Analysis and Computation Fluid Dynamics packages without the cost associated with prototype building and testing. Below in Figures 78 and 79 are ideas that were investigated in order to join the boom tail to the wing structure whilst allowing the wing tips to be removable for transporting and servicing. At the time, the decision of whether to use a one or two spar arrangement was still in discussion and so both alternatives were explored. These joints used the idea of a solid plywood plate (of the same aerofoil section as the wing) bonded to the cut edge of the wing before the wing section was wrapped in a covering, permanently holding the plywood mounting plate in place. The fabricated aluminium plate which secured the tail boom would sit between the inner and outer wing sections and be bolted in place using nut and bolt fixings. The single spar would run continuously from the inner wing section, through the tail boom fixture and into the outer wing section through a hole bored through all three elements, preserving its strength that would otherwise be lost if the spar were cut to match the dimensions of the individual sections. Figure 78 Single Spar Wing Connection Inner wing section Tail boom fixture Outer wing tip section Single carbon rod spar
- 152. Design and Development of a Hybrid UAV Brett M. 132 | P a g e ME5308 – Major Group Project Similarly for the double spar arrangement, profiled plywood plates would provide the hard edge to which the nut and bolt fixing would butt up against, holding the parts firmly together. The tail boom fixture would first be bolted securely to the inner wing section before the outer wing tip section is fixed on the other side. A small access cut-out on the upper surfaces of the tail boom fixture and outer wing tip sections would allow access for spanners to get in to tighten the nuts onto the threaded end of the second spar. Figure 79 Double Spar Wing Connection Neither of these designs made it to final build as a more direct method of fixing the tail boom fixture was found which also simplified the fabrication process. With reference to Section 4.2.7, Figure 80 below shows the bracket designed to attach the bought in carbon fibre rear main landing gear struts to the aircraft fuselage rods. The bracket would be located between the aircraft landing gear struts and the lower carbon rod of the fuselage. Forces on take-off and landing would travel through the struts to the mount and into the carbon rod to be dissipated as flexing of the aircraft fuselage. In sizing the landing gear, the mating surface was made large enough to fully situate the flat edges of the landing gear struts and provide enough flat surface through which nut and bolt fasteners would pass to firmly fix the struts to the mount. The mount was also made tall enough to attach the long legs of the landing gear to the aircraft and have the aircraft rest in a flat and level position. Directly using the CAD Inner wing section Tail boom fixture Outer wing tip section First main carbon rod spar Second carbon rod spar Nut and bolt detail
- 153. Design and Development of a Hybrid UAV Brett M. 133 | P a g e ME5308 – Major Group Project definition of the mount, the part could be made on the university campus available 3D printing resources, the ABS material used offered enough strength and being lightweight enough to make a useful part. Figure 80 Moveable Landing Gear Mount Figures 81 and 82 show the landing gear mount being virtually stress tested to prove the design and optimise some of the dimensions such as web and mating surface thickness. The model seen here had been defined at full scale, using material properties like those the final part would be fabricated from (ABS plastic). Testing was performed with a load of 600N (Newton’s of force), ten times the static load of the aircraft to provide an estimate of the mounts behaviour as the aircraft was travelling over rough ground or while taking off or landing. The levels of force present here in the mount are reasonably low compared to the yield strength for ABS plastic meaning the part was not in danger of failing under these load conditions and therefore suitable for fabricating into a real part. Using test findings, revisions to the mount’s design were made to make the part more lightweight whilst still being capable of supporting the full load force. This very lightweight design is shown in Figure 82. Hole through which the lower fuselage carbon rod would pass Carbon fibre landing gear mating surface
- 154. Design and Development of a Hybrid UAV Brett M. 134 | P a g e ME5308 – Major Group Project Figure 81 Computational Stress Test Result for Basic Landing Gear Mount Figure 82 Computational Stress Test Result for Lightweight Landing Gear Mount The actual printed part whilst like the model in appearance unfortunately broke during assembly of the aircraft (see Figure 75) and an alternative solution was used. Figures 83 and 84 show the nose section of the aircraft designed and fabricated with a laser cut plywood construction in mind. This design allowed the nose to achieve the required bulk and strength to safely cover the rotating parts of the tilting vertical
- 155. Design and Development of a Hybrid UAV Brett M. 135 | P a g e ME5308 – Major Group Project lift fan with the least amount of weight. The joints of the assembly were designed using a tongue and groove method which gave the structure natural strength, aided by the addition of an epoxy resin glue. The whole construction was then covered in a shrink wrap plastic sheet to achieve a streamlined form as well as control the flow of air propelled by the lift motor when the aircraft is performing VTOL flight manoeuvers. Figure 83 Nose Vertical Lift Fan Skeletal Structure Figure 84 Detail View of the Tongue and Groove Assembly Method This nose section has been fully assembled but has remained unused while the rest of the aircraft is being developed and modified.
- 156. Design and Development of a Hybrid UAV Bennie M. & Camilo. V. 136 | P a g e ME5308 – Major Group Project Camilo Initially prior to the use of Solid works for a CAD model, the basic fuselage dimensions were drawn up on paper to scale, with a top and side view of the aircraft being designed. From this first drawing, components that would be housed on-board had their dimensions taken and the fuselage was sized to allow enough space for all of the systems required. From then onwards, the CAD model started to take form, with the figure below displaying the first fuselage concept early on. Figure 85 Initial fuselage concept From this as the design transferred from Conceptual into preliminary and detail, the Aircraft transformed with it as all the analyses, considerations, and calculations were covered by the group. One of the aforementioned is the power supply required on board for both mission profiles of the Vertical flight, and horizontal flight segments. This had a heavy influence on key geometric features which in addition to other advances in several other areas of the design allowed for the first full aircraft design to be drawn up on Solid works. This is shown below on Figure 86. Figure 86 First Full Group CAD Aircraft Design Further into the design process as components of the aircraft were finalized, they were drawn out on Solid works and added to the main aircraft assembly. With each iteration the model increased in complexity, as every connection was modelled, and
- 157. Design and Development of a Hybrid UAV Bennie M. & Camilo. V. 137 | P a g e ME5308 – Major Group Project designed. It can be easily shown in a later stage version of the aircraft assembly shown below in contrast with the first full aircraft design. Figure 87 Structure and connections of various components within the aircraft As the detail of the model decreased the materials listed were updated so as to provide realistic weights estimations on the final aircraft. This was then analysed with the original weights estimations and the alterations or differences between the two were noted. The CAD also allowed for a more refined estimation of where the structure CG would be, which helped in planning out the components layup to achieve the required CG for both tri-copter, and horizontal flight. The increase in level of detail also allowed for all the connections between the structural components to be planned out in advance with the aim to make manufacture and construction as quick and easy as possible to reduce the build time as much as possible. Below on the figure depicted are a series of pictures showing three view of the final CAD Model.
- 158. Design and Development of a Hybrid UAV Bennie M. & Camilo. V. 138 | P a g e ME5308 – Major Group Project Figure 88 Top, front and side views of the final CAD model
- 159. Design and Development of a Hybrid UAV Bennie M. & Camilo. V. 139 | P a g e ME5308 – Major Group Project 4.2.9. Propulsion As shown in section 4.2.6 it is important that the propulsion system be able to provide enough thrust to allow the UAV to perform as a VTOL and an STOL. TO ensure this is accomplished power plant sizing must be conducted to determine what propellers and motors when matched can produce the required thrust and finding a horizontal thrust generator capable of propelling the aircraft forwards on its own despite the weight penalty incurred by the 3 VTOL rotors. EDF Sizing The main propulsion chosen for horizontal flight was an Electric Ducted Fan (EDF) unit. It was chosen over the conventional propeller configuration due to the compactness of the entire unit, avoiding any issues there would have been with ground clearance and propeller strikes upon rotation of the aircraft on take-off. In addition there hadn't been any aircraft in the lab that have used this in the past so in the spirit of experimenting with new technology, it was selected. The EDF unit was sized according to static thrust, battery cell requirements on amp draw, weight and price. This is reflected on the study undergone to find the most suitable motor for the project vehicle. On the figures below can be seen a sample of the most appropriate EDFs being analysed in terms of performance. The trade study results from the excel sheet were implemented to narrow down options to a final EDF with respect to a main propulsion unit to be used in horizontal flight.
- 160. Design and Development of a Hybrid UAV Camilo. V. 140 | P a g e ME5308 – Major Group Project Figure 89 T/W vs Maximum amp draw It can be seen that the EDF which provides the highest vehicle T/W is the Lander 90 mm EDF. It has a lower maximum power consumption from the on-board battery both the Lander 70mm EDF, and the alloy DPS 78 mm series motor. At the same time there us a considerable jump in thrust produced from the rest of the EDFs. This jump in power is seen in the price comparison where although the 90 mm lander is the most powerful of the available EDFs, it comes with a higher price to its contemporaries. In addition to the extra price and power, comes an increase of weight of the actual EDF unit. This is shown in Figure 89 where the lander is almost 200 g heavier than the lightest alternatives. However the increase of weight when compared to its static thrust capacity is offset by a large margin. For example the Leopard L68 EDF weighs 280 g, with a thrust capacity of 1.91 kg. Whereas the Lander 90 mm on the other end of the spectrum weighs 440 g with a thrust capacity of 3 kg, more than a kilogram more thrust capacity for the extra 160 g of weight. 0 0.1 0.2 0.3 0.4 0.5 0.6 0 20 40 60 80 100 T/W(Weightof6Kg) Max Amp Draw (A) T/W vs Max Amp Draw Alloy DPS Series 68mm EDF unit with 2600kv Motor - 1280watt E-Flite EDF Delta-V (R) 32 80mm Unit + BL32 DF Brushless Motor, 2150Kv Leopard L68EDF-6B1-2550KV Ducted Fan System LANDER 7LEDFDPS76-1800Kv MOTOR FOR 6S LANDER 70MM EDF/10 BLADE METAL SPECIAL 2200Kv MOTOR 6S Alloy DPS series 78mm EDF with 2120kv Motor - 2000watt Lander 90 mm EDF
- 161. Design and Development of a Hybrid UAV Camilo. V. 141 | P a g e ME5308 – Major Group Project Figure 90 T/W vs EDF price Figure 91 EDF unit weight vs Thrust Capability Special consideration was given to EDFs which could be sourced out locally from the UK to avoid high prices of importing as-well as long shipping times. As such the 0 20 40 60 80 100 120 0 0.1 0.2 0.3 0.4 0.5 0.6 Price(£) Thrust to Weight Ratio (T/W) T/W vs Price Alloy DPS Series 68mm EDF unit with 2600kv Motor - 1280watt E-Flite EDF Delta-V (R) 32 80mm Unit + BL32 DF Brushless Motor, 2150Kv Leopard L68EDF-6B1-2550KV Ducted Fan System LANDER 7LEDFDPS76-1800Kv MOTOR FOR 6S LANDER 70MM EDF/10 BLADE METAL SPECIAL 2200Kv MOTOR 6S Alloy DPS series 78mm EDF with 2120kv Motor - 2000watt Lander 90 mm EDF 0 0.5 1 1.5 2 2.5 3 3.5 0 100 200 300 400 500 Thrust(kg) Motor weight (g) EDF Unit Weight vs Static Thrust Alloy DPS Series 68mm EDF unit with 2600kv Motor - 1280watt E-Flite EDF Delta-V (R) 32 80mm Unit + BL32 DF Brushless Motor, 2150Kv Leopard L68EDF-6B1-2550KV Ducted Fan System LANDER 7LEDFDPS76-1800Kv MOTOR FOR 6S LANDER 70MM EDF/10 BLADE METAL SPECIAL 2200Kv MOTOR 6S Alloy DPS series 78mm EDF with 2120kv Motor - 2000watt Lander 90 mm EDF
- 162. Design and Development of a Hybrid UAV Camilo. V. 142 | P a g e ME5308 – Major Group Project highest performing EDF was chosen as the main thrust provider; careful consideration was given to the extra cost over its alternatives, and allowed for in the budget due to its local availability. Ultimately if the UAV platform is to expand for future projects, the extra power the 90 mm EDF can provide will cater for any additional vehicle weight brought about by any modifications that bring the ultimate weight close to the 7 kg legal limit.
- 163. Design and Development of a Hybrid UAV Bennie M. 143 | P a g e ME5308 – Major Group Project Motor and Propeller Selection Selecting the motor and propeller combination for the VTOL aspect of the UAV is important. Ensuring that the propeller chosen allows for the motor to perform at its rated output ensures performance of the motor and prop are optimal [82]. In doing so it also helps to ensure that when there is a mass flow of air the dynamic thrust being produced is enough to keep the aircraft aloft. While the static thrust must be enough to lift the UAV. There are two methods that can be used, one uses the RPM of the motor to match the propeller (RPM method) and the other uses T/W as a baseline in the selection of the motor and then attempts to match propeller rpm to that of the motor (T/W method). Both methods were used and results varied (different props showed as being suitable) these props would later be tested to determine which ones could be used and gave the required performance (see section 6.2.2). RPM Method First a motor is selected and its maximum RPM is determined. In our case using electric DC motors this is determined from equation 4.2.9.1: (4.2.9.1) Next the motor’s ideal RPM is determined. This done be selecting a multiplier which can vary depending on motor type however in the case of DC motors this is normally 0.25 [83]. This results in equation (4.2.9.2) (4.2.9.2) Once max and ideal RPM are known we determine max current (I-max), operational current (I-op), no load current (I-noload) and internal resistance. These are normally available from manufacturers of the motor, component distributors or in the manuals that come with the motor. Once these are known copper loss and iron loss are determined using equations 4.2.9.3 and 4.2.9.4: (4.2.9.3)
- 164. Design and Development of a Hybrid UAV Bennie M. 144 | P a g e ME5308 – Major Group Project (4.2.9.4) Now Power in and Power Out can be determined using equations 4.2.9.5 and 4.2.9.6: (4.2.9.5) (4.2.9.6) Having obtained our Pin and Pout we can determine the efficiency of the motor ( ) using equation 4.2.6.7 and then determine the produced power in the propeller using equation 4.2.6.8: (4.2.9.7) (4.2.9.8) After acquiring the propeller power we are able to determine the ideal prop RPM (equation 4.2.9.9) and in turn find out the differences between the Ideal rpm of the prop and motor as well as between the max rpms. Propeller data such as Kp (prop. constant) and PF (power factor) was acquired from [84]. (( ) ( )) (4.2.9.9) Once the ideal RPM of the prop is known it is subtracted from the Ideal RPM of the motor (Delta Ideal). The ideal RPM of the prop is also subtracted from the max RPM of the motor (Delta Max). In order to complete the matching numerous propellers are used in the process the criteria used to select the appropriate propeller is: 1. A high ideal propeller RPM 2. A value of Delta Ideal close to zero 3. A high value of Delta Max Using this method it was found that using the Turnigy 3648-1450 motor with the 2 Bladed APC Slow Fly Series 9x7.5. A table of the process and results can be found on the CD attached with this report in the excel spreadsheet Propeller Analysis > RPM Method.
- 165. Design and Development of a Hybrid UAV Bennie M. 145 | P a g e ME5308 – Major Group Project T/W method The T/W (Thrust to weight) method is the second method that was used to attempt to match viable propellers to the DC motor which was to be used for VTOL. In this method the T/W ratio is the desired design point. The VTOL aspect of the aircraft requires a T/W ratio of 1.2+ [85]. This is then used as the criteria to select if a propeller/motor combination will be suited to the task. First the Maximum Thrust the motor can produce is acquired, this was found from the manufacturer provided datasheet of the motor. Once the max motor thrust is known the ideal thrust is determined. Ideal thrust is assumed to be the amount of thrust you require the motor + propeller to totally produce in order to lift the weight of the UAV. This will then produce the T/W that the motor will produce. Similar to the RPM Method the ideal and max RPM of the motor are determined. Once these are known the ideal RPM of the propeller is determined using equation 4.2.9.10: ( ) ( √ ) (4.2.9.10) After determining the ideal RPM of the propeller it is then compare to the ideal RPM of the motor. The criteria used is that the motor chosen must be able to produce a T/W ratio that is adequate for the design. This process is repeat until a motor and propeller have close ideal RPMs. Once the two criteria have been met the propeller and motor are assumed to be matched. When using this method it was found that the ideal RPM of the motor was 8048 and the ideal RPM of the propeller was 8054. This paired the Turnigy 3648 motor with the APC Electric E –Series 9x9 Propellers (3 Bladed). A table of the process and results can be found on the CD attached with this report in the excel spreadsheet Propeller Analysis > Hybrid Method. Battery Selection As the required propulsion units have a high amp drawn, the battery selected had to be capable of powering the vehicle for both mission profiles without needing
- 166. Design and Development of a Hybrid UAV Bennie M. 146 | P a g e ME5308 – Major Group Project recharging. A LiPo platform was chosen due to its high power to weight ratio. From initial estimates minimum capacity required was that of a 6 cell battery with 9000mAh, and the maximum 11000mAh. Discharge (C) Capacity (mAh) Weight (kg) Price Excluding Shipping (£) 100 9000 1.278 216.99 120 10900 1.371 266.59 100 12000 1.65 297.59 Table 28 Battery properties for a suitable range of products [86] On the above table is listed the batteries which were available and fit the requirements of the mission profiles. As the required performance was so specific, the only power plants which fit the description were sourced overseas from the USA. From this list, budgetary restrictions as-well as weight limitations further narrowed down the choice to the lightest option, the 9000 mAh battery.
- 167. Design and Development of a Hybrid UAV Carlos C.M. 147 | P a g e ME5308 – Major Group Project Endurance It is critical to know the amount of fly time the aircraft will have. The requirement is that the aircraft has to be able to complete both mission profiles, vertically take-off and normal flight, within one battery charge. An initial estimation of the endurance was made for both VTOL and STOL missions. Since no experimental data was available to know what current the motors draw at specific thrust levels, the maximum current draw of the motor was used. This would provide us with a comfortable overestimate of the minimum endurance of the battery. In order to calculate the amount of battery capacity the following equation is used: Figures 92 and 93 show and compare the initial estimate and final current draw for both missions. The final results were obtained from the motor test that will be discussed in the next section. During all the endurance calculations a safety margin of 20% was allowed in the Li- Po battery discharge, this is common practice so that the battery does not fall below the minimum voltage per cell. If this were to occur the battery would not be able to be recharged. Initially the STOL mission drew 71A for take-off and climb a about 45A for the rest of the mission until the motor current was shut for landing. This would provide a fly time of 460s (7 minutes 40seconds). In conjunction with the initial VTOL estimates the additional VTOL fly time would be 74s, using 50A current per motor. For this whole flight a total battery capacity of almost 11000mAh would be required, just as specified in the battery selection.
- 168. Design and Development of a Hybrid UAV Carlos C.M. 148 | P a g e ME5308 – Major Group Project Figure 92 STOL mission current comparison for the initial and final endurance calculations Figure 93 VTOL mission current draw comparison for the initial and final endurance calculations. Once the motor test were completed the actual current draw for the motors is known and therefore the Final VTOL and STOL mission profiles could be drawn up. The corresponding current draw at the specific mission segments may be applied and contrasted to the initial estimations, seen in Figures… and … above. As expected the initial current draws were high overestimates and reduced the fly time of the aircraft very noticeably. With a smaller 9000mAh battery a longer fly time may be obtained. For the STOL mission the cruise thrust was equated to the predicted drag of the whole aircraft. This would provide a realistic thrust and 0 10 20 30 40 50 60 70 0 100 200 300 400 500 600 CurrentDraw(A) Time (s) STOL Current Comparison Initial STOL Final STOL 0 20 40 60 80 100 120 140 160 0 50 100 150 200 CurrentDrawforallmotors(A) Time (s) VTOL Current Comparison Initial VTOL Final VTOL
- 169. Design and Development of a Hybrid UAV Carlos C.M. 149 | P a g e ME5308 – Major Group Project therefore motor current draw. The current draw was reduced from an initial estimate of 45A to 18A and the mission time incremented to over 9 minutes. The endurance could be greater for the STOL mission, but more battery capacity as dedicated to VTOL to be able to control the aircraft with a comfortable cushion of time. The updated current values were set for hover, climb and descent and the VTOL mission endurance was extended from 74 to 180 seconds. All the values and calculations are shown in Table 39 and 40 in Appendix A. It has already been shown that the two missions can be achieved with the current setup of a six cell 9000mAh battery. In the case were transition were to occur the test pilot had a request in order to be able to control the aircraft comfortably. He requested the aircraft to be able to power all motors, 3xVTOL and 1xSTOL, for five minutes. Again applying the battery safety margin and with the final currents obtained from the motor test, the aircraft should be able to have all four motors engaged for four and a half minutes, this was fed back to the test pilot who was content with the endurance achieved. Table 29 below shows the endurance obtained. VTOL & STOL VTOL hover current (A) 66 EDF climb current (A) 30 Total current (A) 96 Battery Capacity (mAh) 9000 Safety battery capacity (80%) 7200 time airborne (s) 270.000 time (min) 4.500 Table 29 VTOL and STOL endurance.
- 170. Design and Development of a Hybrid UAV Bennie M. 150 | P a g e ME5308 – Major Group Project 5. Avionics and Flight Control The UAV will need to have on aboard avionics in order to accomplish flight. From the deflection of a control surface to how the UAV responds in a disturbance, these are all dependent on having the right on board systems that provide the UAV with the required functionality to complete its mission and fulfil its design requirements. Components In order for the UAV to achieve the objectives set out in the project it was important to find components that were not exceedingly expensive but still would be of adequate quality and functionality needed for the UAV to be deemed a success. Component Requirement Choice Servos Needed for the actuation of control surfaces, the tilt mechanism and nose gear landing gear. Futaba Servos. Flight Controller Needed to provide a platform for both manual control and autonomous control. Preferably open source. ArduPilot Mega 2.6. Telemetry Needed to provide telemetry information as well as remote tracking and control of the UAV during flight and ground testing. 3DRobotics Telemetry Radio Kit. Camera & broadcast kit Needed to provide video feed from the UAV while in flight broadcast back to ground station. 3DRobotics Sony HAD CMOS Camera kit and OSD. GPS & Compass Needed to allow the flight control unit to establish position and direction. 3DRobotics GPS & Compass uBlock. Attitude sensor (gyro + accelerometer) Needed to allow the flight control unit to establish attitude such as pitch, roll and yaw, and feed this information to the flight controller to make corrections where needed. MPU-6000 Gyro/Accelerometer (On- board APM). Sonar Needed to allow the flight control unit to establish distance of the UAV away from obstacles such as the ground in hover and vertical climb in VTOL. MaxBotix EZL0 Ultrasonic Range Finder. Power Source Needed to provide electric power to all 3900 mAh LiPo Battery
- 171. Design and Development of a Hybrid UAV Bennie M. 151 | P a g e ME5308 – Major Group Project on-board avionics systems. Airspeed Sensor Needed to provide the flight control unit with accurate airspeed readings. 3DRobotics Airspeed Sensor Radio Receiver Needed to allow the UAV to be manually controlled by a human pilot using a Radio Transmitter Futaba 8 channel Receiver 2.4 GHz Table 30 Necessary Avionics Components for the UAV. These components are available from many sources online and in some cases locally allowing for rapid procurement in case of the need to replace damaged components. Main Control Scheme and sensor array As the UAV is in flight or is given an input by a pilot the Flight control system is responsible for interpreting those inputs and ensuring that the response of the UAV is what is expected. A general control scheme to better represent the UAV and its avionics is presented in Figure 94. As can be seen in Figure 94 the microcontroller sends the input task to the servos (actuators) and this information is used to induce a change in attitude, speed, performance of the UAV. This change is then fed back to the micro controller where it is processed and used to make adjustments if the desired input has not been met [87]. Figure 94 General control scheme of the UAV [87]. The Micro controller used by the UAV is the APM 2.6 shown in Figure 95. It is based off the Arduino 2560 Mega micro controller. This version of the microcontroller
- 172. Design and Development of a Hybrid UAV Bennie M. 152 | P a g e ME5308 – Major Group Project however comes with an on-board sensor array. This sensor array includes a barometric pressure sensor for altitude and a 6 DOF Accelerometer/Gyro for attitude sensing. Figure 95 ArduPilot Mega 2.6 from 3D Robotics The MPU-6000 on-board Accelerometer/Gyro allows for the APM to detect the current attitude of the UAV. It utilises a digital motion processing unit (Shown in the MPU’s schematic in Figure 96) that fuses the information from the accelerometer (which measure acceleration in each direction) and the gyros (which measure the angular velocity). This creates quaternions as opposed to Euler angles for pitch, roll and yaw [88]. Quaternions are easier to interpolate and have a smaller memory footprint than Euler rotation matrices angle [89], [90] this is done by use of a Direct Cosine Matrix (DCM) or a Kalman filter that allows us to fuse the two signals from the gyros and accelerometer [88]. This allows for the FCS to have a much faster awareness of what its orientation is. Figure 96 Schematic of MPU-6000. The microcontroller is able to have a GPS/Compass module connected via an inter- integrated-circuit (I2C) connection adding additional sensing functionality to the Micro controller. Being based off the Arduino Mega means the APM also has analogue and digital pin inputs and outputs.
- 173. Design and Development of a Hybrid UAV Bennie M. 153 | P a g e ME5308 – Major Group Project On the analogue pins devices such as the Sonar and Airspeed sensor are connected as they provide analogue data such as pressure difference and sound delay to produce digital information used by the control. A control scheme can be made specifically to the tri-copter aspect of the UAV. The normal inputs in this case are longitudinal (δ-long), lateral (δ-lat) and yaw (δ-yaw) [57]. We can develop a block diagram as shown in Figure 97 Figure 97 Block diagram of tri-copter control include 2 gain values [57]. Once the longitudinal (δ-long), lateral (δ-lat) and yaw (δ-yaw) are read the controller can interpret them into angular velocities of the rotors and the angle of attack necessary the UAV needs in yaw this is shown in Figure 98. Figure 98 Control allocation by a controller on a tri-copter. PID Controller In-order to dictate to the UAV how to behave in terms of manoeuvres we have to send signals to control surfaces and motors as to how much rpm they must be at or angle they must deflect at and at what rate. To do this a Proportional Integral Derivative (PID) controller is implemented. The APM 2.6 however uses a cascaded
- 174. Design and Development of a Hybrid UAV Bennie M. 154 | P a g e ME5308 – Major Group Project PID structure. A cascaded PID controller is used since it is able to achieve smooth tracking with fast disturbance rejection [91]. An example of cascade control is shown in Figure 97. Figure 99 Example of Cascade Control. The Cascaded PID used by APM however is shown in Figure 100. There are two PIDs one is to stabilise which will take the current actual angle from the sensors and the desired angle from the pilot output a rotational rate into a rate PID. At the rate PID the rotational rate from the gyro is also taken into account and a final out is sent to the motor. Figure 100 Cascaded PID used by APM [88]. In order to get an output that is satisfactory the PIDs will need to be tuned. Tuning of the PIDs can be done by using the Mission-Planner software that APM uses. PID tuning will require ground and flight testing to determine what amount of Kp, Ki, Kd are needed for best performance Related software and Full system schematic As mentioned before the APM uses an open source software package called Mission-Planner. It is used to upload the appropriate type of firmware onto the APM. The two firmware that were used for this project were ArduCopter and ArduPlane.
- 175. Design and Development of a Hybrid UAV Bennie M. 155 | P a g e ME5308 – Major Group Project When selecting ArduCopter a tri-copter setup is chosen as this is relative to the setup being used by the UAV. Once the required firmware is upload Mission-Planner can be used to calibrate all the sensors currently connected to the APM. MissionPlanner also has logging functionality that is used to monitor electromagnetic produced noise and telemetry. For autonomous missions the Mission Planner software is used to allocate way points and remotely control and monitor the UAV in flight. Full avionics schematics of the UAV set up (In STOL and VTOL) are attached in the Appendix A that show the wiring of the system for each setup. These schematics show a single APM in use for both setups. Sonar and Noise Reduction Figure 101 MaxBotix XL MaxSonarEZL0. The Ultra sonic range finder is a device that allows for the determination of distance between the source (the sonar) and an obstacle. It is often used for robotics and autonomous rovers to avoid obstacles in a maze. However it also has uses on a multi-rotor UAV. When used with on a multi-rotor UAV it allows for accurate altitude readings while close to the ground. The range finder that was chosen for the UAV was the MaxBotix EZLO (shown in Figure 101) that has a range of 10.68 m with a 1cm resolution making it viable for our purposes. It is more accurate than the on- board barometer that APM has for low altitudes as below 20m the pressure change is not much and this would mean the VTOL aspect of the project would encounter problems when trying to perform manoeuvres like altitude hold and steady vertical climb. In order to reduce EMI and other electromagnetic noise it was suggested by the manufacturer that the Sonar be modified using a 100uF capacitor, a 10 Ohm resistor and shielded jumper wires, the modification is shown in Figure 102. This was done
- 176. Design and Development of a Hybrid UAV Bennie M. 156 | P a g e ME5308 – Major Group Project and the sonar was then connected to the APM. To test if there had been a reduction in noise the test command was used. This is accessible in MissionPlanner > Terminal > Connect > test > sonar. The terminal window would output distance readings that; as long as the obstacle did not move, stayed constant. Figure 102 Sonar EM Noise reduction modification. APM Anatomy The APM has multiple inputs and outputs (buses) that enable it to be connected to a plethora of sensors. Figure 103 APM 2.6 anatomy. Part Name Description 1 Input Channels A Radio Receiver is connected to allow
- 177. Design and Development of a Hybrid UAV Bennie M. 157 | P a g e ME5308 – Major Group Project manual control input into the APM. It can also be used to send inputs into the APM from another device such as a controller. 2 Analogue I/O Channels Sensors are connected such as the Sonar and Airspeed sensor. It can also be used for connecting a camera actuated Gimbal. 3 Output Channels Devices such as servos and motors are connected once APM has processed an input it will output to the required channel. 4 Telemetry I2C bus This I2C buss is used to connect the 3DRobotics Telemetry radio. It can also be used to connect the OSD chip that allows for a HUD on a live feed from an on board camera. 5 JP1 connector The JP1 connecter is a jumper (by default not connected) that allows the user to let power from a device on the output channel (such as an ESC with a BEC) power the APM. This however is not advised if another power source is being used. 6 Power Module I2C Here the Power Module used to power the APM is connected. The Power module is a voltage regulator that will step down a maximum of 18v to a safe 5/6V that APM is rated for. 7 I2C bus This I2C bus can be used by any sensor with an I2C bus connector however it is normally used to connect the magnetometer that comes with e GPS/Compass uBlock module. 8 Reset Button This button reset the APM and is used to rerun the current code loaded on the APM or recalibrate the unit. 9 GPS I2C bus The GPS uBlock module is connected here 10 USB port (side-load) The APM has a side loading USB port. This is used to connect the APM to a computer. This is used by MissionPlanner to connect to the APM when not using the wireless telemetry (MavLink) to link wirelessly with the APM. Table 31 APM Anatomy Glossary.
- 178. Design and Development of a Hybrid UAV Camilo V. 158 | P a g e ME5308 – Major Group Project Manual Transition Although the project goals are specifically to perform an STOL mission, and perform a VTOL mission separately, the manner in which these systems have been implemented into the aircraft allows for a very basic form of manual transition in- between the two in the near future, to avoid having to take-off in rough conditions the framework exists but has not been tested. Shown on the figure below is the manoeuvre split into its basic 3 elements when the controls have been mapped with a three position switch routed onto channel 5 of the transmitter and receiver. This method incorporates the dual APM set-up with an additional relay Arduino board to control the distribution of the inputs from the receiver between the other two micro controllers. The code within the relay board was created with the help of another student whose background knowledge of coding with Arduino originated from an obstacle avoidance project where a different operational problem was solved utilizing similar methods [92]. Figure 104 Phases of flight during the transition maneuver from hover to horizontal flight The first phase of the manoeuvre is with the three position switch in its bottom position, where the on-board relay is routing all controls directly to the Arducopter installed APM, which operates under the stabilize mode of the firmware. Here the aircraft must climb to a safe altitude of at least 20 m via manual input from the pilot. At this point the second phase can be initiated. When the pilot moves the three position switch onto the central position, it causes the relay board to send a signal to the Arducopter APM changing the mode from stabilize into Altitude hold. In this state the tri-copter will not resist any translation of motion on the same altitude and can drift but will retain its pitch an height. The switch change also allows the relay to send all the inputs on the receiver for flight to the Arduplane APM. In this way the aircraft
- 179. Design and Development of a Hybrid UAV Camilo V. 159 | P a g e ME5308 – Major Group Project hovers autonomously maintaining its altitude, whilst the pilot has manual control of all STOL related systems. The pilot will proceed to throttle the EDF unit up and start gaining momentum. As the wings start to generate lift the tri copter will reduce thrust to a minimum as less and less is required to keep it at the same altitude. The last phase of the manoeuvre is with the three position switch at its last position. Here the tri-copter firmware will go into acrobatic mode, with all propellers powered down to idle. This is only to occur once the aircraft has gained sufficient speed where the wings are generating enough lift to prevent the aircraft from stalling. As with the previous phase, all the controls are routed to the control surfaces and the EDF propulsion unit. Technically the same method can be applied for landing vertically from horizontal flight, with the procedures reversed. The only difference between transitioning from hover to horizontal flight and vice versa is the small issue that upon entering stabilize mode on the Arducopter APM from altitude hold, the throttle will automatically change to match the setting that it currently is at on the transmitter [93]. This is an issue as when stopping into altitude hold, the throttle will be at its minimum value so the EDF isn't providing any more forward momentum. As such the altitude must be sufficient that when the Arducopter firmware switches into stabilize mode, the drop in altitude from the change in thrust can be compensated for to regain control and continue in a manual controlled descent to land. It is important to note that the endurance of the vehicle will be severely affected and will be substantially shorter than its STOL only mission profile as it will be limited by the increased power consumption of all motors drawing power from the on-board battery.
- 180. Design and Development of a Hybrid UAV Brett M. 160 | P a g e ME5308 – Major Group Project 6. Component Testing 6.1. Stress Tests 6.1.1. Rods First Rod Test Stress testing is the process of applying a load to a test piece and measuring its effect as a deformation of the part. The load can be applied in a number of ways to affect the piece differently for various conditions the piece might be found in an assembly. Stress testing of the carbon rods was undertaken prior to final specification with a view to appropriately size the rods to the task they were required to carry out once installed on the aircraft. Stress testing plays a crucial role in verifying manufacturer claims of material strength and toughness, statements that if taken only on face value could have a safety impact if the part were to subsequently fail during flight manoeuvres. The types of load application used in this round of testing were defined to simulate the conditions the rods would experience when used in the tail booms and the wing spars. For the tail booms which would be fixed at one end (at the wing) and support the load force generated by the tail at the other end, a cantilever arrangement emerges as shown by Figure 105 below. Figure 105 Cantilever Load Testing Arrangement During the test, the rod was secured at one end using a wooden block bolted to a rigid test frame. With no weight yet placed on the rod, a measurement was taken to mark a datum from which all subsequent measurements would be compared. Following definition of the datum, a hanger was secured using a clamp to the free
- 181. Design and Development of a Hybrid UAV Brett M. 161 | P a g e ME5308 – Major Group Project end of the rod, and incremental weights were added to the hangar. With each addition of weight the rod would deflect and a measurement was then taken. The difference between this measurement and the datum was the total deflection as a direct result of the weight being applied. The calculated deflection was then plotted against the applied load to create a deflection curve for carbon rods under a cantilever bending action and is shown in Figure 106. Figure 106 Cantilever Physical Stress Test Results Graph The curves shown by Figure 106 describe a near linear relationship between applied load and deflection which make estimations of deflection for a load not directly tested for to be made. Using this graph, the team selected the 20mm rod for the tail boom as it deflected the least at higher applied loads without looking unsightly. As a bonus, the weight differences between the 15mm and 20mm rods for a given length were comparable which made sense to select the stiffer rod without an additional weight penalty. For the wing spars which can be assumed to be supported at the tips while the aircraft is in flight and support the weight of the aircraft assumed to act at the middle of the rod, a 3 point load arrangement emerges as shown by Figure 107 below. Acceptable Unacceptable
- 182. Design and Development of a Hybrid UAV Brett M. 162 | P a g e ME5308 – Major Group Project Figure 107 Three Point Physical Stress Test Results Graph During the test, the rod was rested on the forks of a pallet truck, the forks of which were spread by a distance of 621mm. With no weight on the rod, a measurement was taken between a straight edge laid across the pallet truck’s forks and the centreline of the rod. Following definition of this deflection datum, a hanger was secured using a clamp at the middle of the rod, midway between the end supports (the pallet truck forks) and incremental weights were added to the hangar. With each addition of weight the rod deflected a small amount and a measurement between the straight edge and the rod’s centreline was taken. The difference between this measurement and the datum was the total deflection as a direct result of the weight being applied. The calculated deflection was then plotted against the applied load to create a deflection curve for carbon rods under a 3 point bending action and is shown in Figure 108.
- 183. Design and Development of a Hybrid UAV Brett M. 163 | P a g e ME5308 – Major Group Project Figure 108 Three Point Physical Stress Test Results Graph The curves shown by Figure 108 describe a near linear relationship between applied load and deflection particularly for the 8mm rod while the 10mm rod showed a small deviation from the straight line projection, probably as the result of measurement errors. Using this graph, the team selected a combination of the 20mm and 10mm rods for the front and rear spars respectively as together they would provide the greatest resistance to bending in flight while still being light enough for the purpose. Acceptable Unacceptable Approx. weight of the aircraft
- 184. Design and Development of a Hybrid UAV Arturs D. 164 | P a g e ME5308 – Major Group Project Second rod test After the first carbon fibre reinforced rod test analysis, some systematic errors were detected which seemed to have a significant effect on the test results. It was detected that the mount used for cantilever test has some systematic sources of error. The metal clamp which was used to clamp the rods was too big therefore, flat balsawood rings were especially made for each of the rod being tested. The inner diameter of the rings was the same as of the rods but the outer was large enough to be clamped by the metal clamp. One end of the rod was inserted through 5 balsawood rings approximately 1cm apart which were clamped by the metal clamp. The other end of the rod was gradually loaded with the weights. Main source of error came from the balsawood rings when the other end of the rod was subjected to load. The properties of balsa wood are such that it is more elastic than other types of woods and deforms a lot easier. The purpose of the test was to measure the deflection of the rods under a load however, when the rod was subjected to a load the balsawood rings were subjected to various stresses especially compression, bending and torsion which also caused the rings to deform. This deformation caused increased deflection of the rods. The main concern was about the 20mm diameter rods which were used for main wing spar and the tail boom. A new test was designed to test the 20mm diameter rod. A flat wooden piece was used to which one end of the rod was clamped using 3 plastic semi-circular fittings as clamps. The fitting radius was a little smaller than the diameter of the rod. The rod was fixed on to the wooden flat piece using 3 clamps which were drilled onto the wooden place securing the rod. 3 clamps were used to minimise any deflection caused by the clamping mechanism. The wooden piece was securely fixed onto the edge of a bench and at the other end of the bench a ruler was set upright perpendicularly to the rod against which the deflection ridings were taken. The datum was taken at the initial position of the rod after which the end of the rod was systematically loaded with weights and the deflection readings were taken. The weight of the hanger and the individual weights were measured prior to the experiment using a scale to minimise the errors and uncertainties due to equipment used. The rod was loaded up to 54N which is approximately 2g of the whole aircraft. The rod used for the test was 1m long which is longer than the spars used in the wings of the aircraft. Also the lift forced which the wing spar would be subjected
- 185. Design and Development of a Hybrid UAV Arturs D. 165 | P a g e ME5308 – Major Group Project during the flight are distributed along the whole wing span while the rod was subjected to point loading at the end which generated larger moment hence, larger deflection. The rod was subjected to higher forces than it would be subject during the flight and the test has proven it to be suitable for the application. The results of the test are demonstrated below on the Figure 109 graph. Figure 109 demonstration of carbon fiber rod deflection with cantilever point loading Figure 109 demonstrates the result of the second test on 20mm diameter, 1m long carbon fibre reinforced rod. The graph demonstrates linear deformation of the rod in a form of deflection from its original datum when it is being subjected to a load. The graph follows similar to stress-strain linear deformation graph which indicates that under the subjected loads the material is still within the linear, elastic region and that there is no permanent deformation being done. Almost perfect linearity of this graph also indicates the improvement of the test technique and minimal error. 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 Deflection(mm) Load (N) 20 mm rod deflection
- 186. Design and Development of a Hybrid UAV Abbinaya T.J. 166 | P a g e ME5308 – Major Group Project 6.1.2. Connections Test description: As mentioned previously, the connections that were manufactured out of 3D printed P400 ABS are very crucial to the aircraft. Even though basic information of the material itself was studied through research, the parts manufactured had to be tested in order to ensure safety. Since all of the connections had similar dimensions in terms of thickness only one the samples were tested. The tested component was one of the connections between the wing spars and the fuselage rods. The component was tested in order to check if it could sustain the loads during flight. During manoeuvres such as a turn, the aircraft would be experiencing more than 1g load and therefore it was required that the component is capable of sustaining a load of at least 5g. A 5g load would mean 5 times the weight of the aircraft all up weight (MTOW) which is a load of approximately 300 N. The test was conducted in the structures and materials laboratory in Brunel University. The test conducted was a destruction stress test using the Instron tensile testing equipment. The aim of this particular component test was to determine how much load the component could sustain before failure. Figure 110(a) demonstrates the experimental setup of the test and Figure 110(b) shows the geometry of the test component. The room temperature was 21° at the time of the test and the humidity level in the atmosphere was 32%.
- 187. Design and Development of a Hybrid UAV Abbinaya T.J. 167 | P a g e ME5308 – Major Group Project (a) (b) Figure 110: Experimental setup of the test conducted (left) and a drawing of the component (right) The holes for both the spars were tested to see how much load it could sustain before failure. The hole for the main spar has a diameter of 20mm and the hole for the secondary spar has a diameter of 10mm. Two separate tests were conducted in order to determine the strength of each of the holes. The test rig that can be seen in Figure 110 was designed using steel. The purpose of the rig was to hold the component in place during the test. A metal (steel) rod was threaded through the hole for the spar and the plates attached to the Instron tensile tester so that a tensile stress would be applied in the inner wall of the hole. Results The test was completed successfully. Table 32 shows the final results obtained from testing each of the two holes. The results obtained showed that the component is more than capable of fulfilling the required tasks. The hole with the diameter of 10mm could withstand 1876 N of force before failure and the hole with the 20mm diameter could withstand up to 1664 N of force before failure. Results: 10 mm diameter hole 20 mm diameter hole Maximum load (N) 1875.875 1663.775
- 188. Design and Development of a Hybrid UAV Abbinaya T.J. 168 | P a g e ME5308 – Major Group Project Maximum Tensile extension (mm) 3.22 2.2 Tensile stress at maximum load (MPa) 32.02 28.4 Tensile strain at maximum extension (mm/mm) 0.555 0.379 Table 32: Results obtained from the stress test conducted on the 3D printed component. Given below in Figure 111 and 112 are the graphs of the load vs. the tensile extension for each of the two holes. Figure 111: Load vs Tensile extension for the 10mm diameter hole
- 189. Design and Development of a Hybrid UAV Abbinaya T.J. 169 | P a g e ME5308 – Major Group Project Figure 112: Load vs Tensile extension for the 20mm diameter hole. From the results obtained, it was concluded that the component was strong enough and safe for use on the aircraft.
- 190. Design and Development of a Hybrid UAV Bennie M. 170 | P a g e ME5308 – Major Group Project 6.2. Motor Characterisation It is important to characterise the rotors and EDF that are used on the UAV. This is done due to the fact that linear behaviour of these components can vary due to many factors such as production defects or electromagnetic interference (EMI). The motors used in both cases are brushless and utilise electronic commutation [94]. The rotation of the motors is created the permanent magnetic rotors chasing a revolving magnetic field which is induced by the current in stator windings inside the motor [94]. Using PWM signals that are either on or off the magnetic field varies causing the drum to rotate. The motor will thus accelerate or decelerate depending on the current density changes caused by the changing PWM state [94]. Because of this use of magnetic fields EMI can cause behaviour and performance of the motor to be non-linear at certain PWM duty cycles and it is important to know what range of PWM signal allows for linear behaviour and thus performance. Once this region is known the amount of thrust produced within the linear region is analysed in order to ensure the required thrust does not lie within the nonlinear region which is normally at very low and very high PWM duty cycle. By doing a characterization test we are able to check and verify that the motor and propeller combinations assumed in section 4.2.9 are viable. 6.2.1. Set-Up In order to perform motor characterisation the thrust bench in the aerospace engineering was used. It consisted of pivoted arm with a pulley that allowed for the loading of weights. The motor to be tested is mounted on the top of the arm such that it creates a moment that activates the force sensor. The bottom of the arm connects to a force sensor which will send data to a computer. On the computer a LabView code is used which gives us numeric data that is then used to determine the thrust and characterise the motor. The apparatus is shown in Figure 113.
- 191. Design and Development of a Hybrid UAV Bennie M. 171 | P a g e ME5308 – Major Group Project Figure 113 Thrust bench and NI High USB carrier used for motor characterisation. The components that were used to test the motor and EDF were a 4000 mAh 6S LiPo battery (Figure 119), a customised bench mount for the motor and EDF (Figure 114 & Figure 115), an 80A Superbrain ESC (Figure 116), an APM 2.6 flight controller, a Turnigy RPM/Current reader (Figure 117), National Instruments High USB Carrier (Figure 118) and the thrust bench (Figure 113). Figure 114 EDF Mount for thrust bench. Figure 115 Motor Mount for thrust bench.
- 192. Design and Development of a Hybrid UAV Bennie M. 172 | P a g e ME5308 – Major Group Project Figure 116 80A ESC Turnigy Superbrain. Figure 117 Turnigy KV-RPM Meter. Figure 118 National Instruments Hi-Speed USB Carrier. Figure 119 Turnigy 4000 mAh LiPO Battery (6s). Motor Testing Code To run the motors the microcontroller that was to be used for flight control on the UAV was used (APM 2.6). Since it is based on the Arduino AT2560 Mega board a custom code had to be written to allow for selectable PWM signals to be sent to the motor being tested. The custom code called “Manual_Input_MotorTest” is included in the report Appendix A. The code used the Arduino servo library which actually uses PPM signals which are then converted to PWM since the width of the pulse determines the position or angle the motor will rotate through [95] as shown in Figure 120.
- 193. Design and Development of a Hybrid UAV Bennie M. 173 | P a g e ME5308 – Major Group Project Figure 120 PWM changing the angle of a dc motor [95]. Since an ESC was being used the motor would have to be armed. To accomplish this a function was made called void_arm that would send a high signal to the ESC while the battery’s positive terminal was not connected. Once the program displayed “Connect battery” the positive terminal would be connected to the ESC and the motor would be armed. It should be noted that when performing the EDF test the ESC it was using during tests was the 80A ESC rather than the 100A ESC it was going to be coupled with on the UAV. This was due to the 100A ESC not being able to be armed using this method. Once the motor is armed a loop statement is run that allows the tester to input a PWM signal that would run the motor. Due to the risk and hazard associated with this type of testing a fail-safe was implemented in the code that allowed for the motor to be stopped at any point in case of an emergency. Another fail-safe was implemented to prevent PWM values that were higher than 100 as to prevent damage to the motors or mistakes such as mistyped values being reCGnised. Code was also implemented to slowly step down the running speed of the motor this would allow for a controlled stop at the end of every test run.
- 194. Design and Development of a Hybrid UAV Carlos C.M. 174 | P a g e ME5308 – Major Group Project 6.2.2. Calibration of Equipment The equipment was calibrated using a set of weights of which the exact mass is known and the software intended for the experiment use, Lab-View. The software attains the mean value of a numeric reading force differences at the sensor where the force is being applied. In the experiment the force is applied by the motors thrust and in the calibration the weights apply an exact force to the sensor. The calibration process consists of gradually hanging weights from the sensor to obtain the corresponding numeric reading for specific forces, or the weights. The software provides a reading or mean numeric that can be plotted against the force of the weights to obtain a calibration curve. Figure 121 shows a sample calibration curve for one of the motors tested. A trend line is fitted to the curve and then the equation is displayed, where y is the force and x is the numeric value. To obtain the forces from the motors in the experiment the numeric obtained at different motor RPMs may be inserted in the equation to determine the force being produced. Figure 121 Sample calibration curve for the test bench. 6.2.3. Procedure for Testing The test procedure used during the motor characterisation test is show in the below table (Table 33). Step Procedure Result y = 147.56x - 0.7495 0 5 10 15 20 25 30 35 0 0.05 0.1 0.15 0.2 0.25 CalibrationWeight(N) Mean Numeric Value Calibration Curve
- 195. Design and Development of a Hybrid UAV Carlos C.M. 175 | P a g e ME5308 – Major Group Project 1 Perform Pre-checks - Ensure props are clear - Ensure battery is not completely connected. - Ensure all connections other than battery are fastened and secure. - Ensure test bench and motor are facing away from test personnel and by standers. - Ensure the amount of weight on the Thrust bench pulley is the correct amount. It must be 30N in the case of our UAV.’ - Perform Loading of weights and note the numeric for each weight until 30N have been loaded onto the thrust bench. - Once 30N is loaded create a trend line and display the equation (linear y=mx+c) Once all Pre-checks are performed continue to step 2. 2 Motor run code in relative IDE on Computer. Once codes are running proceed to step 3 3 Load Motor test code APM and then open serial command window and await arming instructions. Once instructed to ‘Connect Battery’ connect positive terminal. Proceed to step 4 4 Once motor is armed a test low signal will run the motor for a short period in the serial command window input 0 to stop the motor from running. Once motor is stopped proceed to step 5. 5 Take initial numeric reading and then input a PWM of 4 and take the following readings:- - Acting numeric - Current draw - RPM - PWM input Repeat this step until readings are acquired for an increasing number of PWM values (lower than 100). Once enough readings have been taken input 0 to stop the motor running. 6 Post-test checks:- - Make sure prop is clear - Disconnect battery positive Once battery is disconnected close codes and change out motor and/or propellers if necessary. 7 Using gathered data and M and C from the trend line in step 1 we can determine the thrust for each different PWM using the equation: Plot results and check that within linear region thrust being produced is adequate for required propulsion needs.
- 196. Design and Development of a Hybrid UAV Carlos C.M. 176 | P a g e ME5308 – Major Group Project Table 33 Testing Procedure for Motor Test. In step 5 & 7 of the procedure we gather the relative information we need in order to characterise the motor being tested. These results are presented in sections 6.2.1 and 6.2.2.
- 197. Design and Development of a Hybrid UAV Bennie M. 177 | P a g e ME5308 – Major Group Project 6.2.4. EDF Results As mentioned before the EDF was not run with the ESC that it would be using on the aircraft (K-Force 100A) instead it used the Superbrain 80A ESC that allowed us to arm it for the test. The EDF normally has to be run on a high timing which is set in the ESC. At the time of experimentation these setting could not be programmed due to the fact that the team did not have any programming boxes. The ESC have a default of medium and this is what was used while testing. For the EDF test the numeric loading data used (post calibration) to determine the trend line and M and C values that are used to calculate the thrust of the EDF were M = 147.56 and C= - 0.7495 (see Figure 122). Figure 122 Numeric Loading for EDF and trend line. Once the numeric loading had been performed the test procedure proceed onto inputting the various values of PWM to run the EDF. These results are presented in Figure 123. y = 147.56x - 0.7495 -5 0 5 10 15 20 25 30 35 0 0.05 0.1 0.15 0.2 0.25 Numerics Loading EDF
- 198. Design and Development of a Hybrid UAV Bennie M. 178 | P a g e ME5308 – Major Group Project Figure 123 Thrust Results for the EDF. As can be seen in Figure 34 the EDF has a fairly linear behaviour even at very low PWM. The results also show that within the PWM range of 4 to 50 the EDF is more than adequate to produce the amount of thrust needed for forward flight. 0 5 10 15 20 25 30 35 0 10 20 30 40 50 60 Thrust(N) PWM Thrust vs. PWM input (EDF)
- 199. Design and Development of a Hybrid UAV Carlos C.M. 179 | P a g e ME5308 – Major Group Project 6.2.5. Motor Results The final propellers considered for the VTOL motor test were the two bladed APC 6x6 and the three bladed E-series slow fly 9x7.5 These are two and three bladed respectively. As shown in Figure 124 the three bladed propeller should be more efficient for thrust load coefficient over 0.5, but the current thrust load coefficient is almost 1.5. According to the experimental results of J. Dang and H. Laheij [96] the two bladed propeller should provide more thrust for the same amount for the same input signal. Figure 124 Thrust efficiency of two and three bladed propellers [96]. Due to the nature of VTOL and the yawing moments generated when the propellers spin, the tricopter should have two motor spinning in one direction and the other one in the opposite. Since it was very difficult to find the appropriate counter rotating propellers, then pusher and tractor props where used. They are the same as counter rotating propellers, but have a different name. All three motors where used in the testing phase with the three different ESCs. The corresponding propellers tested are: 3 Bladed Pusher Propeller Pusher Propeller Tractor Propeller
- 200. Design and Development of a Hybrid UAV Carlos C.M. 180 | P a g e ME5308 – Major Group Project Figure 125 VTOL Motor test with different propellers. The results of the test are shown in Figure 125. It is observed that the two bladed tractor propellers provide more thrust than the two bladed pusher propeller for a given PWM value, just as expected. The results for the two bladed propellers are very similar, they have the same gradient, or variation of thrust with signal, and they vary almost linearly. This is very important since it allows the program/controller to know what signal input to the motor to achieve specific thrust output. The slight variation in the thrust form the pusher to the puller propeller could had arisen from the set-up of the experiment. To hold the motor in place a large plate was needed, this restricts the incoming mass flow of air for the pusher prop. It also creates a ground effect for the tractor propeller increasing the actual static thrust. This may not be the only reason behind that difference, but it’s the most likely even though different motors and ESCs were used in the three experiments. The amount of thrust provided is more than enough to sustain the aircraft with a mass of in a steady hover and climb. The motor test were stopped once the thrust reached a value close to because the calibration of the equipment was only done up to of force, any results above that value would be inaccurate. 0 5 10 15 20 25 30 35 0 5 10 15 20 25 30 35 StaticThrust(N) PWM Siganl VTOL Motor Test with Different Propellers 3 Bladed Pusher 2 Bladed Tractor 2 Bladed Pusher Hover Thrust Required
- 201. Design and Development of a Hybrid UAV Carlos C.M. 181 | P a g e ME5308 – Major Group Project The motors have potential to produce a lot more force. This is deduced from the fact that the maximum current drawn in the motor test was no more than . The maximum current draw according to the manufacturer of the motors is up to . Looking at Figure 126, if the current varies almost linearly with the thrust then the maximum potential thrust could be around . Figure 126 Current Draw of the motor for any given thrust. For hover of the aircraft the current draw should be about per motor (taking the highest value of the two), meaning that the motors draw for a sustained hover flight. This is substantially smaller than the initial used to calculate the initial endurance and battery size. It is worthwhile reminding the fact that the current drawn was taken as a maximum since no information was available on the current draw of the motor for specific thrusts. With this essential information the final endurance calculation may be made. Another solution would be having a smaller battery, and therefore reduce the aircraft weight, to achieve the same mission profile. 0 5 10 15 20 25 30 0 5 10 15 20 25 30 CurrentDrawn(A) Thrust (N) Current Drawn Vs. Thrust 2 Bladed Tractor 2 Bladed Pusher Hover Thrust Required
- 202. Design and Development of a Hybrid UAV Camilo V. 182 | P a g e ME5308 – Major Group Project 7. Build & Manufacturing Methods & Materials in Chronological Order 7.1. Logistics In any project involving a construction of a vehicle it is essential to plan ahead purchases of potentially critical components. These were done through the order Forms provided to the group by the university. A copy of all orders purchased online is listed in the Appendix B, As such the organization of orders was split into phases of purchase early on in the design process in order to buy it all in steps and have the components arrive as they were required. Below is a list of the suppliers used for out sourced components. Supplier Time taken for delivery Comments Foamwings.co.uk 4 weeks Bad craftsmanship, inaccurate dimensions, poor finish on foam wings Billkits.co.uk 1 week Excellent service, great finish on wings and empennage Hobby king UK Warehouse 2 weeks Orders received as described Hobby King International Warehouse 12 weeks Items took a long time to arrive BRC Hobbies/ Robot Birds 2 days Item arrived very quickly, in good condition Slough rc retail shop N/A Excellent customer service B-Composites 1 ½ weeks Excellent customer service, Quick turnaround time and shipping, woven carbon rods. Carboncopy.co.uk 1 week Rapid manufacture and custom work on rear landing strut
- 203. Design and Development of a Hybrid UAV Camilo V. 183 | P a g e ME5308 – Major Group Project Espirit Model.com 2 weeks Items in good condition as advertised Build your own drone.co.uk 2 weeks Items in good condition as advertised 3DRobotics 4 weeks Overseas item arrived in good condition Max amps.com 4 weeks Overseas item arrived fully charged and in working order DIY drones.com 4 weeks Overseas item arrived fully charged and in working order Foam-board.co.uk 2 days Great customer service, items in pristine condition Unmanned Tech shop.co.uk 3 weeks Reason for long turnaround due to back- order Easy Composites 1 week Quick service Sussex Model center 3 weeks Item arrived in good order, delivery was a bit slow Steve Webb Models 2 weeks Items in good condition as advertised Table 34 List of Suppliers and any comments surrounding orders and components delivered 7.1.2.
- 204. Design and Development of a Hybrid UAV Abbinaya T.J. 184 | P a g e ME5308 – Major Group Project Key materials used Foam Expanded Polystyrene Styrofoam (EPS) was selected for wing and empennage cores. EPS Styrofoam has a strong honeycomb like structure. This material selected because it was found to be comparatively stronger, cheaper and also more precise than the typically used balsa wood. Commercial balsa used in model airplanes weigh between 96 and 288 kg/m3 whereas, EPS foam weighs only about 16 kg/m3 . Manufacturing the wing and tail with EPS cores would also give it a more accurate profile than if balsa was used. EPS foam has high compression strength of 4882.43 kg/m2 .Since balsa wood weighs higher, the wing is usually made in a skeleton fashion resulting in loss of strength and uneven strength along the wing. 3D printed ABS Due to the complexity in design for some custom parts required, 3D printed manufacturing method was used. The parts that were 3D printed were i. Rod connections between the wing and the fuselage. ii. Connections between wing spars and the boom rods. iii. Connections between the boom rods and the vertical tails. iv. Connections between the horizontal and vertical tails. v. Connections between the boom rods and the rods connecting the VTOL motors. The connections to be made from this material are very critical to the UAV. Therefore, the material was tested before confirming whether it was safe to use on the aircraft. Balsa wood Balsa wood was used in the aircraft for components that do not have to sustain loads. The main components made out of balsa wood are the mount for the camera and the mount for the ultrasonic range finder. Balsa wood was selected as it is light
- 205. Design and Development of a Hybrid UAV Abbinaya T.J.& Brett M. 185 | P a g e ME5308 – Major Group Project weight and easily available. Another reason for the use of balsa wood is that it is very easy to machine and hence very suitable for making custom parts. Metal(s) Metal had to be used in some parts of the aircraft which were required to sustain high loads. In some cases it was selected due to its high tensile strength and heat resistance capacities. The metal used in the aircraft was mainly aluminium and some of aluminium alloys. Carbon Rod(s) Carbon rods are made of a carbon and other fibre mix (such as glass, Kevlar or aramid) solidified in a resin epoxy matrix. The rods are as the name suggests are formed as a rod (solid piece) or tube (hollow) made by either pultrusion (through a die at high pressure for rod forms) or wrapped using a mandrel and pressure bandage technique (for tubes). Carbon rods have been used throughout the aircraft, as fuselage stringers, tail booms and wing spars because of their lightweight but very stiff and high strength characteristics. Carbon Fibre Landing Gear(s) Carbon fibre as a landing gear material offers the high strength and stiffness required to cope with the dynamic loads endured by the landing gear during heavy landings or when rolling over uneven ground. Carbon fibre is highly flexible in the way it can be shaped, allowing the gear to be streamlined and thus producing as little drag as possible while the aircraft is airbourne. Additionally the lightweight character of the carbon fibre means further savings in terms of aircraft performance due to a reduction in extra carried weight and therefore lift induced drag. Obeche Wing veneer Obeche timber comes from African Triplochiton Scleroxylon tree. Most of aircraft foam wings are veneered with a thin layer of obeche since they add rigidity to the wing however, it is easy to work with since it is not too rigid and can be shaped around the wing. It has very low density almost like balsa wood but better properties for wing veneering. It works very well with glues and can easily be glued on to the foam. Smooth surface finish is easier to achieve that with balsa wood. To further
- 206. Design and Development of a Hybrid UAV Arturs D. 186 | P a g e ME5308 – Major Group Project improve surface finish the wing was covered with plastic film using heat shrinking technique which improves the properties further and protects from moisture and other external influences. Sourced pieces Main sourced parts that were obtained for the project are carbon fibre reinforced resin rod. Rods mainly connected major sections of the aircraft using rod connections. Rods were used as the main load bearing structure which reinforced the wings, the tail and held the whole fuselage together. For wing reinforcement and the boom tail rods were all 20mm outer diameter with 1mm wall thickness. Since the larger, 20mm diameter rods are main load bearing and will be exposed to higher forced, the braided type was selected for improved bending performance properties. The wings and the tail were also sourced externally by a wing manufacturer. The design of both the tail (two vertical stabilisers and one horizontal stabiliser) and wings were done by the team and the technical drawings were supplied to the manufacturer with material descriptions. The wing manufacturer used CNC hot wire technique to cut out the wing shape foam core. Then the foam core was manually veneered with thin layer of obeche, the control surfaces were cut out and both leading edge and the trailing edge were reinforced with balsa wood. Some procedure were carried out for the both vertical stabilisers and the horizontal stabilizer.
- 207. Design and Development of a Hybrid UAV Camilo V. 187 | P a g e ME5308 – Major Group Project 7.2. Fuselage As-well as the design aspect of the project, the build of the aircraft presented its own unique challenges. This involved the manufacture of all connections and components on the aircraft, in addition to this, some aspects of the aircraft underwent alterations in the process which took theory and design, into practical construction, time & budget restrictions. The first component of the aircraft to be put together was the fuselage. This consisted of 3 Bulkheads of varying thicknesses of Plywood held firmly by a triangular skeletal support of 3 carbon fiber rods with a 10mm diameter and 1 m length. “Laser Ply” was used as the structural component of the bulkheads due to the strength and light weight nature of the material. The abundance of it in varying thicknesses in the lab allowed for fast and cheap turnaround time throughout development of crucial components, which in turn allowed for rapid prototyping and testing. A laser cutter was implemented to cut the material. This was able to cut through up to 10 mm of any organic material, although tolerances meant that final cut pieces had slightly different dimensions due to material melting or burning off. Figure 127 Laser cutting the aft EDF bulkhead Of the 3 Bulkheads, the forward two were comprised of 4 mm Plywood, whilst the last one would be where the load from the rear landing gear would be displaced to the rods. As a result the aft fuselage bulkhead was of a much thicker and stronger 6 mm plywood construction. Further to the rear of the carbon rods, the EDF mounting system was placed, this comprised of a couple of 4 mm plywood bulkheads
- 208. Design and Development of a Hybrid UAV Camilo V. 188 | P a g e ME5308 – Major Group Project embedded on the outside of the duct, split through the vertical and horizontal axis respectively to hold the EDF Firmly in place, along with a third bulkhead further down the duct to prevent the motor from tilting around its main two bulkhead mounts. all the bulkheads were permanently fixed to the rods using epoxy resin, in a process which began front the nose of the aircraft, and traversed to the rear as more of the fuselage detail was completed and added on. The first of the bulkheads was the first component fixed to the bottom two rods, as well as the first to be designed in detail due to the complexity of the system it encases. Supported on the bulkhead itself are both the tilting propeller mechanism and the moveable forward landing gear. The tilting mechanism was designed such that the maximum displacement of the servo translated to a maximum rotation of +/- 25 degrees of the propeller above it. This was achieved by mechanically limiting the effect of the rotation of the servo even at the maximum rotation value onto the propeller arm by altering the length of the arm connections between the two of them. By extending the top propeller rod arm and shortening the servo arm any movement on the servo caused a proportionally smaller translation of motion to the propeller. Below is a picture of the aforementioned component. Figure 128 Fuselage during initial Epoxy resin stage of construction (left), tilting Propeller mount (Right) As can be seen above to make construction simpler the fuselage was suspended upside down with the top two carbon rods 50 mm above the surface of the table in the lab. The paper seen below is a scale drawing of the fuselage that was used as a
- 209. Design and Development of a Hybrid UAV Camilo V. 189 | P a g e ME5308 – Major Group Project reference throughout construction. It reflected dimensions and locations of components of the structure and avionics from both the CAD model and the excel sheet which was created for the purposes of CG estimation. The nose gear was also mounted on the front Bulkhead, spaced out around the tilting propeller servo a micro servo was installed alongside a mount and strut for the nose landing gear. This can be seen in the figure below. Figure 129 rear view of the front Bulkhead displaying the nose gear mechanism The nose gear mechanism itself consisted of the servo mounted facing down screwed into two blocks of wood which had been glued onto the bulkhead. This servo was linked to the rod of the nose gear strut via a linkage and arm attached onto the top of the plastic mount. The servo was tested to ensure the translation of motion between the different arms of the mechanism allowed for sufficient range for good ground roll directional control. Moving backwards from the front both the plywood battery tray along the top and the foam-board secondary deck were installed. The reason for the two trays was to separate the fuselage into 3 sections laterally to help in spreading out the components that would be housed within. The plywood plate was then additionally used as the spacer to mark out where the next bulkhead would be epoxied into place. In order to help support the loads the tray would be experiencing it was zip tied to the carbon rods along 4 points throughout its length on either side, leading to 8 hard-points to the carbon rods. This minimized the bending that the plywood battery tray would experience. The system chosen to keep all the avionics in place
- 210. Design and Development of a Hybrid UAV Camilo V. 190 | P a g e ME5308 – Major Group Project during flight implemented the use of velcro strips, with opposite ends on the battery tray and on the components mounting onto it. Due to the weight of the battery however an extra strap which looped all the way around the battery through to the top of the battery tray was developed to assist the Velcro on the underside of the plate as the battery is essentially hanging in place underneath the tray. Next was the middle fuselage bulkhead, and from here the wing -fuselage connections were fixed in place, as well as the APM and GPS module tray. This tray was also zip tied to the carbon rods; however this time to allow enough space for the APM to be placed on the CG the tray was mounted on a different level than that of the Battery tray. This then allowed for the placement of the third and final fuselage bulkhead, the one which would absorb the loads from the rear landing gear. For this reason it was the thickest of the bulkheads of the fuselage, consisting of 6mm Plywood. Wrapping the skeletal structure was a rigid skin in the form of 3 mm foam- board. It was chosen due to the ease in shaping the material, and the turnaround time to produce components from it. The foam-board skin was then encased in pro- film to protect the material from any moisture encased in the take-off surface in-case of an undercarriage collapse as-well as provide a smooth surface finish to which slide gently to a stop. on the skin three access points were cut to provide access to key wires and components housed in the underside and top section of the battery compartment as-well as the compartment underneath the wings and the APM boards. In RC aircraft, it is commonplace to use metal, fiberglass or carbon fiber landing struts for the rear landing gear due to the loads experienced on the component during operation. Ultimately the material chosen for the project aircraft was a carbon fiber landing strut. This was due to the lightweight, load bearing capacity of the material over conventional heavy metal struts, and the relative weakness of available fiberglass alternatives for the size required. There was also an example in the lab of an aircraft with a similar weight to the project aircraft, and it used a carbon fiber strut for its main landing gear. The back-up landing gear initially arrived as a single carbon fiber component within a few days of ordering. This had to have holes drilled for the mounting screws to the fuselage as-well as for the axle which would hold the wheels in place. Originally from
- 211. Design and Development of a Hybrid UAV Camilo V. 191 | P a g e ME5308 – Major Group Project manufacturers specification the landing gear was only rated to 2 kg, however the manufacturers agreed to stiffen the strut by adding a couple more fiber layers to ensure the strut was rated for the 6 kg fully loaded vehicle weight that was being estimated. As a result the strut which arrived was incredibly stiff which led to an issue in drilling holes for the axles. Normally the landing struts flex outwards a small amount to upon being loaded, which is why the straight plate at the bottom of the strut itself is offset inwards. Due to the rigidity of this particular strut however, there was no flex, so the holes for the axles had to be drilled in such a way as to ensure that the wheel would be perpendicular to the ground. The figure below shows the rig that was made to accomplish this task. Figure 130 drilling axle holes on the non-vertical mounting plate of the carbon fiber Landing gear Once the axle holes were drilled, balsa blocks tapered on one side were inserted onto the axles to act as spacers to keep the wheels from hitting the angled carbon fiber strut. This set up and the test rig devised to see the strength the back-up mount can be seen in the figure below.
- 212. Design and Development of a Hybrid UAV Camilo V. 192 | P a g e ME5308 – Major Group Project Figure 131 Rear landing gear assembly Figure 132 Fuselage structure with back-up rear undercarriage (left), Nose gear (right) The figure above displays the structure of the fuselage along with the undercarriage and EDF mount prior to build finish. This was the configuration for the first flight test.
- 213. Design and Development of a Hybrid UAV Abbinaya T.J. 193 | P a g e ME5308 – Major Group Project 7.3. Connections Different types of connections were designed in order to hold the main airframe structure together. As already mentioned, rod connections between the wing and the booms, the wing and the fuselage, the vertical tails and the rods, the connections between the horizontal and vertical stabilisers and rod connections between the rods holding the VTOL motors and the boom were all manufactured by 3D printing P400 ABS plastic. The 3D printed parts were manufactured in campus with the 3D printing facility available in Brunel University. The required parts were first designed on Solidworks (CAD software) and then uploaded as ‘stl’ files on the computer connected to the printer. Once the parts were printed off, they had to be cured for around 18 hours in UV light. The surface quality obtained from the printer was not very smooth. Even though a better quality printer was available, it was not used in order to reduce costs. Also the surface finish is not very critical for the connections. The printed parts had to be sanded down slightly in order to make up for the manufacturing tolerances. Motor mounts that connect the VTOL motors to the carbon fibre rods running horizontally behind the wings had to be designed and custom made. A connection made of aluminium plates was designed using CAD SoildWorks. Given below in Figure 133 is the assembly of the connections designed. These parts were manufactured in the metal workshop in Brunel University. Figure 133: Schematic of assembly of the aluminum VTOL motor mounts. This connection was designed such that the three aluminium plates were supporting the rods and the motors were held together using 3mm screws and nuts. The type of
- 214. Design and Development of a Hybrid UAV Abbinaya T.J. 194 | P a g e ME5308 – Major Group Project nuts used were “Nyloc nuts” which are equipped nylon collar inserts. This type was used in order to prevent any unscrewing due to vibrations on the aircraft. 7.4. Wings & Tail The wing and tail cores were manufactured externally. The EPS core used was CNC cut. Slots were cut for the servos and the 3-D printed fuselage and boom connections. The holes for the spars were also cut using hot-wire. The core was then veneered with balsa wood. After receiving the outsourced components, channels for wires were drilled in the wing and tail sections. The wing, tail and the respective control surfaces were received, they were laminated with profilm to give it a smoother surface finish. After laminating with profilm, plastic hinges were glued in to slots to attach the ailerons to the wing, elevator to the horizontal tail and the rudders to the vertical tails. The servos were fitted in their slots and glued in place with epoxy resin. The control horns for the servo arms were then screwed into each of the control surfaces The 3-D printed rod connections that adjoin the wing to the fuselage were fitted in. Due to manufacturing tolerances, some small sections of the slots and the 3-D printed parts and had to be sanded down. The connections were connections were then adhered on to the slots in the wing. This was done with Z-epoxy resin. Two set squares were used during this process in order to ensure perpendicular alignment. The attachment was also measured over an outlined 2-D drawing to further check the alignment. The wing to boom connections were also sanded down for a smoother surface finish and to account for manufacturing tolerances. They were then bolted on to the boom rods once slotted in. The other ends of the boom rods attach to the vertical tail. The connections designed for this were slotted into the vertical tails, and then secured with bolts. Finally, the spars were inserted in to the wing and threaded through the connections to allow the wing to be connected to the fuselage.
- 215. Design and Development of a Hybrid UAV Abbinaya T.J. 195 | P a g e ME5308 – Major Group Project 7.5. Propulsion The mounts for all motors both for horizontal and vertical flight were designed to withstand the forces and thrust outputs expected during operation. For example the EDF and tri-copter propeller motors used in the aircraft are all capable of a significant amount of thrust; around 3 kg static thrust each. Due to this the bulkheads and motor mounts connecting the propulsion to the fuselage were required to be suitably rigid and well designed. 7.5.1. VTOL Propeller motors Mount The Turnigy 3648 6S motor has a shaft with an adapter with which to mount a propeller. The base of this mount was 3 mm from the bottom of the shaft, so this distance was the design thickness limitation for the propeller mounting plate. Three solutions were crafted from machined Aluminium, 3d Printed ABS type plastic, and 3 mm plywood acquired from the lab. The original idea however was to use parts that were made out of metal. This part had to be sourced alternatively because of time constraints in manufacturing the parts made of metal. Hence, the first alternative solution was to manufacture the motor mounts by 3D printing the P400 ABS plastic. Due to uncertainty with the material properties of the part itself, a simple load test was conducted. The component started to deflect and crack when a load of 20 N was applied to it. This made the component incompetent for use on the aircraft. Also, during operation, the temperature of the motors is likely to go up. This would further make the plastic material less favourable for the part because of its poor heat resistance. Due to this, other materials were considered for the part. The second material considered for the part was 3mm plywood. The mount made of the 3mm plywood was manufactured using a laser cutting machine. On testing the plywood motor mount, it was found that the component was capable of sustaining up to 30 N of load. Even though these motor mounts were strong enough, they were only used as a temporary solution and not for the actual flight. This was because of the material being flammable in case of an accident. The actual motor mounts used were the ones preferred initially made out of aluminium. This was possible because the parts ordered were ready just a couple of
- 216. Design and Development of a Hybrid UAV Abbinaya T.J. 196 | P a g e ME5308 – Major Group Project weeks prior the first test flight. The aluminium metal motor mounts used are light, strong and also durable. Figure 134 displays the pictures taken during the load tests of the components. The deflection on the mount made of P400 ABS plastic is visible in the image. Figure 134: Load tests conducted on the P400 ABS plastic (left) and the 3mm (right) plywood motor mounts.
- 217. Design and Development of a Hybrid UAV Camilo V. 197 | P a g e ME5308 – Major Group Project 7.5.2. Lander 90 mm EDF Mount The Ducted fan was nestled by three 3mm plywood bulkheads. A side and top view of the assembly is shown in the figure below. Figure 135 EDF Mount to the fuselage, Side view (left), top view (right) The Mount was designed to take advantage of the outline of the EDF due to the variable geometry around the outside of the duct as can be seen on the figures above. There was a 6 mm wide indent where the manufacturer’s motor mount would normally be placed, however due to the way in which the motor was being implemented on the UAV; the space was taken up by two 3mm bulkheads which had been split in half on the vertical and horizontal axis respectively. Due to the high torque of the motor causing it to disconnect itself, the outside shell was super glued to the internal sleeve of the bulkheads containing it to prevent any slip of the EDF.
- 218. Design and Development of a Hybrid UAV Bennie M. 198 | P a g e ME5308 – Major Group Project 7.6. Avionics Installation of the avionics system had to be done in such a way that all electronic systems would operate optimally. Some avionic systems had to be specially positioned in order to avoid electromagnetic interference from rendering them faulty. Phenomena such as electromagnetic and acoustic/mechanical noise were key factors in the positioning of many components. Servo Installation The servos that were to be used had to be installed on their relevant control surfaces. Servos were used for the actuation of the ailerons, rudder, elevator, nose landing gear, tilt mechanism for the front rotor. Due to the positions of the radio receiver and APM it was necessary to run several servo extension cables as the cables on the servos would not be able to reach. For this purpose the wings were modified to have an additional cut out channel that would allow for the cables to be fed through the wing and into the fuselage section of the UAV. For the servos placed on the tail the cables were run through the boom tail beams and then fed into the wing from the boom-wing junction. Servos on the wing and tail control surfaces were then epoxied into their relative cut-out sections. For the tilt and nose gear servos, cut-out sections were made on the front fuselage bulk head that allowed for them to be epoxied into place. Extension cables were then run to the relative inputs on the radio receiver and APM. The servos where then checked to make sure they were securely in place and a servo tester was used to check that they actuated in the right direction and were responsive. Camera, Live stream & OSD Installation The camera was installed so that it would be easily accessible from the outside of the UAV. It was placed on the bottom mid-section of the fuselage. A special mount was made that would secure the mount the camera came with to the bottom rod of the fuselage. Between the two mounts a layer of Depron foam was used to add damping of mechanical vibration that the rod may experience in flight this would provide a steady stable picture feed. The Live stream and OSD chip were then connected to the APM and camera and placed in on the upper loading tray of fuselage and held in place with Velcro straps.
- 219. Design and Development of a Hybrid UAV Bennie M. 199 | P a g e ME5308 – Major Group Project The live stream board had to position so that it would not cause any radio frequency interference (RFI) or EMI with either the radio receiver or the GPS/Compass uBlock which was mounted on the wing. GPS & Compass installation The GPS/Compass uBlock was mounted on the wing. This was due to it needing to be exposed to the environment to get a GPS lock and maintain said lock. This also was done to prevent EMI caused by the EMF produced by DC current sources such as the PDH. The uBlock module is held in place with pvc tape. Sonar Installation The ultrasonic range finder (sonar) was also placed in the bottom mid-section of the fuselage this allowed for the sonar to be directed directly to the ground. Due to EMI it was necessary to modify the sonar to reduce this effect by using a 100uF Capacitor, a 10Ohm resistor, and shielded cabling. This solution was recommend by the manufacturer of the ultrasonic range finder MaxBotix [97]. At this position the sonar would also be fair away from any air turbulence created by the rotors which could cause acoustic or mechanical noise that could result in inaccurate readings. Flight Control System & Air speed sensor Installation The flight control system (APM) was installed in the rear section of the fuselage under the wings. A mounting tray was fabricated for it and a male and female Velcro strip was used to secure the APM into its position. The tray was placed in an elevated position in the fuselage to avoid as much as possible the EMI from the PDH cables running at the bottom of the fuselage. The Airspeed sensor came in three parts, the Pitot tube, silicon tubes, and airspeed sensing chip. The Pitot tube was connected to one and of the silicon tubes and a custom mounting brace was made to keep the Pitot tube outside of the boundary layer when the UAV was in flight. The tubes were then connected to the airspeed chip which was placed inside the fuselage on the top mounting tray. This was then connected to the APM via a servo cable. ESC Installation and calibration ESCs are known to produce high levels of heat and EMF. For this reason the ESCs had to be positioned; where possible, far away from sensitive systems. Two
- 220. Design and Development of a Hybrid UAV Bennie M. 200 | P a g e ME5308 – Major Group Project 80A ESCs were installed on the adjoining rear motor beam. The 100A K-Force ESC was placed between the two holding bulkheads of the EDF. By positioning these ESCs in the open it would allow them to dissipate their heat much easier to the surroundings and reduce the risk of overheating in an enclosed area like the fuselage. However this was not possible for the front rotors ESC as it had to be placed inside the fuselage due to the position required for the front rotor from the CG. Battery and Power Distribution Harness (PDH) Installation The batteries that were to be used to power the avionics and propulsion were placed in positions that were calculated to allow for the CG of the UAV to be in a required position. The batteries could be moved if the CG needed to be altered for flight performance issues. The power distribution harness was positioned at the bottom of the aircraft. This would prevent its EMF from interfering with other key systems in the fuselage. The sonar however was placed underneath the PDH as well as the camera, when the first flight test was conducted only the camera was on board and functional and there was no sign of EMI. Further testing with the sonar on board would be needed to determine the strength of the EMF of the PDH and if it inducing high levels of inaccuracy in the sonar. Telemetry (MAVlink) Installation Installation of the telemetry system was similar to that of the Livestream board. However to prevent any form of RFI or EMI the telemetry was positioned on the middle fuselage bulkead on the top mounting tray. The telemetry antenna was placed stream wise on the side of the fuselage to reduce as much as possible any drag on the UAV.
- 221. Design and Development of a Hybrid UAV Abbinaya T.J. 201 | P a g e ME5308 – Major Group Project 8. After Build Testing 8.1. Flight Tests 8.1.1. Horizontal Flight Test 1 The first flight test was an attempt to do the conventional flight to make sure that the UAV is capable of doing STOL. From the results obtained it can be concluded that the first flight test was a partial success. Although the flight itself was not successful, a lot of lessons were learned from this experience. The test did not go as expected and was not successful due to a number of reasons. The first factor that contributed to the failure was the surface of the runway. The runway conditions for the take-off were unanticipated. It was a grass runway which was wet, mushy and very uneven. Even though the runway conditions were poor, the aircraft should have still been able to take-off. The aircraft also did not reach the take-off velocity at the point of rotation. During the take-off run, the aircraft was not able to reach the desired take-off velocity with the allocated distance due to the poor surface conditions. The aircraft lifted off the ground and then again pitched back to the ground since it did not reach the desired velocity at rotation. The low velocity at the point of rotation did not allow the aircraft to generate sufficient lift and caused the aircraft to stall and lose height. When the aircraft pitched down to the ground upon impact the aircraft’s rear landing gear sheared off from the fuselage. The EDF continued downwards and dragged along the ground due to the lack of rear landing gear. This led to ingestion of debris and other objects which then caused a sequential failure of 4 of the 5 fan blades. The test pilot tried to recover the aircraft by pitching up. Aided by the bounce due to the contact with the ground the aircraft climbed a few feet and stalled again as there was no thrust provided by the EDF. At this point the aircraft stalled, collided with the ground and continued its forward track along the ground due to the momentum it had built up for take-off on the grass. It came to a stop a few metres from the position of the final bounce.
- 222. Design and Development of a Hybrid UAV Abbinaya T.J. 202 | P a g e ME5308 – Major Group Project Another reason for the aircraft not being able to lift of the runway was the unusually high take-off weight on the day of the test flight. The aircraft had an additional weight of 600 grams of ballasts and weighed a total of 6.2 kg. The ballasts were added to the nose of the aircraft in order to move its CG forward and closer to the aerodynamic centre of the wing. This was done because moving the CG forward would help increase its static margin. A higher static margin was preferred for the first flight as this would help increase its pitch stability and provide a safer flight. Outcomes While inspecting the aircraft after the test flight, it was found that the main structure of aircraft sustained impacts quite well and remained undamaged. The components that had been damaged were the EDF blades and the under carriage. The balsa formers that were meant to prevent the rear landing gear from pivoting around the bulkhead sheared. The metal strut of the nose landing gear had bent. All 5 of the EDF blades were also damaged. The outcomes of the test show that a longer take-off ground run could have been implemented. This would have helped mitigate the consequences of the poor runway conditions and the additional aircraft weight. Design modifications One of the components which failed in the flight test was the mount for the rear landing strut to the fuselage. In order to improve the resistance against the shear forces acting on the strut in the rough take-off surface the modifications shown below were added.
- 223. Design and Development of a Hybrid UAV Abbinaya T.J & Brett M. 203 | P a g e ME5308 – Major Group Project Figure 136 Reinforced rear landing gear mount First off, the formers around the bulkhead were re-made from 6 mm plywood layers instead of the balsa to reinforce the area local to the landing gear bulkhead. Thereafter the landing gear plate to which the carbon strut was screwed into was extended towards the nose. This allowed for additional connections to tie the landing plate to the fuselage. These connections were the 3D printed 10 mm carbon rod connections that had been initially made for the VTOL mounts. These then acted as extended arms to counteract and anchor the plate to the fuselage and resist the pivoting motion that had sheared off the previous plate from the bulkhead. The EDF blades were replaced with new ones. In order to gain access to the EDF, the surrounding bulk heads had to be sawed off as they were adhered with epoxy. The EDF was then dismantled to replace the blades and clean it. New bulkheads were laser-cut and put back in position after refurbishing the EDF. After an initial attempt at flying the aircraft in the conventional flight mode from a rough take off field, the front and rear landing gear arrangements failed to withstand the dynamic loading placed on them and the front landing gear in particular bent very early into the take-off ground roll, making the rest of the take-off ground roll very challenging for the aircraft. The front landing gear fixture was subsequently redesigned using new parts and a more rigid and capable mounting structure of plywood and aluminium plate as shown in Figure 137. This heavier duty
- 224. Design and Development of a Hybrid UAV Abbinaya T.J & Brett M. 204 | P a g e ME5308 – Major Group Project arrangement was expected to prevent the landing gear from bending again on the aircraft’s second attempt at take-off. Figure 137 Strengthened Retro-fit Nose Landing Gear
- 225. Design and Development of a Hybrid UAV Arturs D. 205 | P a g e ME5308 – Major Group Project Horizontal Flight Test 2 The second test flight was carried out after taking consideration and improving upon the weaknesses detected during the first test flight. As mentioned before one of the main weaknesses of the first test was front landing gear which was too weak for the weight of the aircraft. So the landing gear was replaced with a higher weight rating gear and the mount mechanism was significantly improved by fixing an additional plate to the core structure of the fuselage in the form of aluminium plat to which the landing gear shaft was clamped using a U-shape aluminium clamp. This prevents the landing gear bending backwards as the aircraft accelerates which happened during the first flight test. Since it was as a horizontal flight test the excess avionics required for VTOL configuration was removed from the aircraft in order to decrease the weight and additional dead weight was introduced to the nose of the aircraft to increase the longitudinal stability of the aircraft for safety reasons. During the first test the aircraft had few attempts to take of but stalled due to that it was decided to allow aircraft to gain velocity before taking off for longer. Another reason for not having high enough velocity for take-off was the surface of the runway of the field the test took place at. In order to improve on it, after the wind direction was determine, the field was inspected to find which part of the field has the best surface along the take-off path. The purpose of the test was to demonstrate the horizontal flight capability of the experimental aircraft and its ability to fulfil the basic mission profile segments such as take-off, climb, cruise with basic manoeuvres such as turn, descent and landing. Figure 138 Second flight test ground roll demonstration
- 226. Design and Development of a Hybrid UAV Arturs D. 206 | P a g e ME5308 – Major Group Project Prior to the test the wing direction was determined in order to decide on the aircraft take-off direction. After the direction decided, a good location to set up the aircraft for take-off was chosen which did not have any obstacles on the on the runway line and had enough distance for full ground roll with the safety margin. Figure 3 is a screenshot taken from the test video recorded on the day. On the screenshot the take-off direction and approximate ground roll length is indicated by the black line with the arrow. The aircraft started the take-off procedure where the line begins and took-off from the ground where the tip of the arrow is. Instead of the previously calculated during the design phase ground roll of 42m the aircraft took-off after approximately 7-8m. The initial calculation of the ground roll is the maximum with the safety margin also, some of the avionics and the VTOL motors have been removed from the aircraft for this test which decreased the weight making the take-off ground roll requirements smaller. Figure 139 Second flight test tip stall demonstration Straight after the aircraft took-off it experienced wing tip stall of the right wing and rolled onto the right side almost touching the ground with the right wing tip which is demonstrated on figure 4, top screenshot.it can clearly be seen on the screenshot that the aircraft is pitching down. At that moment the ailerons were deployed in such way to roll the aircraft to the left in order to stabilise the aircraft. On the bottom screenshot one can see the aircraft tilting left however, by that time the aircraft is generating enough lift and can be seen that it is pitching up. One of the reasons the
- 227. Design and Development of a Hybrid UAV Arturs D. 207 | P a g e ME5308 – Major Group Project aircraft experienced stall straight after it took-off could be because of not gaining enough speed since the ground roll was only 8m. One of the solutions to that would be to let it run for longer in order to gain speed. Also, the ailerons were programmed into another channel as elevons to aid take-off and landing. Slight deflection downwards would increase the lift however, during the take-off the elevon configuration was not used because for the first successful flight it is important to find out the capabilities and performance of the aircraft without using flap in order to determine the operational safety and contingency. After the aircraft stabilised, it climbed very rapidly to altitude approximately of 30m. After 3 laps flown around the field the throttle was reduced until powered off. The aircraft steadily glided down and just before the landing stalled hitting the left wing first on the ground which can be seen on figure 5, below. The aircraft experienced harsh landing which damaged the front landing gear by slightly bending it backwards together with the reinforcement aluminium plate. Since the EDF was completely powered off for landing the speed dropped down to stall speed which could have triggered the stall at landing. Using ailerons as elevons to increase lift at landing and maintaining a little thrust in the EDF could easily solve the problem. Figure 140 Second flight test landing stall demonstration Overall the second flight test was successful apart from few minor problems which mainly arise due to aircraft operational aspects and can be tackled easily by slight changes in the test procedures. In order to prove and demonstrate that all of the major problems have been resolved there is a need to carry out a third flight test. The third flight test could be also carried out with original static margin to which the
- 228. Design and Development of a Hybrid UAV Camilo V & Carlos C.M. 208 | P a g e ME5308 – Major Group Project initial aircraft design was carried to. For future test the aircraft front landing gear need further reinforcement and increased damping in case of a harsh landing as well as introduce flap deflection at take-off and landing to increase the lift generated by the wing and avoid stalling. 8.1.2. Vertical Flight Tests The first flight attempt was unsuccessful. This was due to the powering down and non-response of two VTOL motors on idle speed. This meant that the aircraft was not able to lift off the ground under its own power. The Body propeller worked as expected with the tilting servo mechanism operating as it should. After extensive troubleshooting and discarding sources of error, It was concluded that The ESC’s had to be programmed to a higher timing set-up due to the number of poles on the motors. Secondly part of the solder on the connectors had come loose causing a faulty response from the front right tri-copter motor (aft left motor in relation to the Aircraft configuration). Upon fixing the issues, the tri-copter was run again to attempt VTOL. During the test, the same motor which failed in the previous attempt, failed again, however with the aid of one student preventing the tri-copter from drifting due to the loss of thrust from the motor, it continued to climb and stabilized under its own power. Once it landed again, it was established that the source of the failure was one of two things. Either it was similar to that of the first test involving faulty soldered connections within the power distribution cables, or the motor itself was faulty. The power distribution cables were checked and re soldered appropriately. A third attempt would ultimately prove where the failure originated from. The third attempt ended with a successful run of all three VTOL motors. The tri- copter was able to support itself, with no external input from the pilot, handled by members of the group for safety reasons to avoid any injury to bystanders or damage to the vehicle as-well as any drifting due to the gusts. Below is a screenshot from the video taken showing the aircraft in hover under its own power.
- 229. Design and Development of a Hybrid UAV Camilo V & Carlos C.M. 209 | P a g e ME5308 – Major Group Project Figure 141 UAV in Tri-copter mode From the flight test, the roll, pitch and yaw were functioning as expected. The Arducopter APM and relay sent the signals correctly through to the motors and servo. There were some problems which require fine tuning to improve the tri-copter handling characteristics. These include modifying the gains and settings on the PID inputs within the firmware in the APM to allow the tri-copter to better correct itself from gusts and other external perturbations.
- 230. Design and Development of a Hybrid UAV Arturs D. 210 | P a g e ME5308 – Major Group Project 9. V-n Diagram The V-n diagram in figure 6 demonstrates the variation of load factor with airspeed for manoeuvres. In other words it is the aircraft’s operational envelop. The operational envelope of an aircraft is mainly determined by its structural capabilities, the stresses and forces the structure of it can withstand. Some of the parameters in the diagram are estimates and the gust diagram is there only for demonstration, showing the gust which falls within he envelope. Also the diagram only represents the horizontal flight envelope without taking into consideration the tri-copter configuration. All of the aircraft velocities demonstrated on the diagram are in terms of actual airspeed rather than equivalent because, the aircraft is designed to fly relatively close to the ground and the difference between actual speeds and equivalent can be assumed negligible. The Lower limit load factor was calculated using the following formula: The maximum lift coefficient is 1.08, wing area is 0.45m^2 and the weight of the aircraft is 6kg which were used to calculate the lower limit load factor which is 2.05 however, due to margin of safety the design of the aircraft structure was aimed to be able to withstand the load factor of minimum 3 which is used on the diagram. Since it has been designed to limit load factor of at least 3 the gust diagram goes up to load factor of 3. The minimum speed indicated on the diagram is the stall speed of 14.021m/s below which the aircraft is not allowed to fly because the wing will stall. The cruise speed of 22.272m/s is used to calculate the dive speed, which is 1.25 of the cruise speed or 27.84m/s. Dive speed is also maximum allowable speed at which the aircraft is permitted to fly at. The take-off curves on the left hand side of the graph are estimations of clean configuration. If the flaps are deployed during the take-off the curves would be steeper and the wings would generate more lift at take-off. Negative side of the graph was just a rough estimate with negative limit load factor of -1 which includes
- 231. Design and Development of a Hybrid UAV Arturs D. 211 | P a g e ME5308 – Major Group Project the margin of safety. As mentioned previously, many of the parameters indicated on the V-n diagram are rough estimations and their accuracy have to been improved and proven through aircraft testing. Figure 142 V-n Diagram and Gust Loading graph VS VDVC -2 -1 0 1 2 3 0 5 10 15 20 25 LoadFactor(n) Air Speed (m/s) V-n Diagram with Gust Loading
- 232. Design and Development of a Hybrid UAV Camilo V. 212 | P a g e ME5308 – Major Group Project 10. Budget Where ever possible to keep costs to a minimum, components have been recycled from spares as the University has some of the necessary construction materials as well as electronic components for the fabrication of the project aircraft onsite. In these cases the unit price has been left at £0.00. In order to better display use of budgeting to minimize the cost of the project, the Final budget as well as the budget as it was at the end of term 1 are shown Below on Tables 35 and 36. Table 35 Mid-Project Budget
- 233. Design and Development of a Hybrid UAV Camilo V. 213 | P a g e ME5308 – Major Group Project By the end of the project extra costs were invoked upon completing the detailed design phase. This was due to the wings and several components of the fuselage including various connections of the aircraft being finalized and sent to manufacture as well as having materials ordered in. Table 36 Final-Project Budget
- 234. Design and Development of a Hybrid UAV Abbinaya T.J. 214 | P a g e ME5308 – Major Group Project 11. Conclusion The project proposal itself was an ambitious idea but realistic goals were set. The project was limited in several ways. However, these obstacles were overcome and the requirements of the aims and objectives were met. One of the problems encountered was the delay in the orders of outsourced components which in turn led to problems with the build. Problems such as faulty/unsuitable parts and time required for replacement were not anticipated. Other problems encountered with components were due to the high demand for the technicians in the metal workshop in Brunel University. This led to prolonged delays in ordered parts. The project was also challenging because a balance between the UAV system for conventional flight and tricopter system for the VTOL had to be maintained. The members of the group had limited exposure to designing and building a UAV and this was a first time experience for all members. In spite of this, the group managed to achieve the main aims and objectives set out for the project. All tests performed were successful and problems encountered were solved in time to achieve the goals set out. Overall, the project can be considered a success as the set out aims and objectives were met and the aircraft was able to complete successful flight tests for the STOL and VTOL mission. This project has been very beneficial to the individuals in the group and has helped gain extensive knowledge on the realities involved with engineering an aircraft. This project will also serve as a base/platform for future students who wish to continue working on the UAV.
- 235. Design and Development of a Hybrid UAV Carlos C.M. 215 | P a g e ME5308 – Major Group Project 12. Improvements and Further Research After completing the project it was realised that some changes would have to be made in order to improve the overall performance of the aircraft. To compensate the short period of time to research every design choice, rapid prototyping methods were implemented. This resulted in inadequate nose landing gear components, complex avionics circuitry and a very basic camera module. These issues will be highlighted below. Landing Gear Although the motion mechanism and the setup was sound, the gear itself was not optimal for our vehicle weight. It should be improved by damping the shear forces of a rough surface on which to take-off or land. Consideration for larger wheels could be made to increase the shock absorbing capability of the assembly. Due to the weight limitations a retractable landing gear was scraped, but it would be a great improvement in terms of aircraft performance, by reducing drag substantially. STOL Propulsion The horizontal flight propulsion system would have to be reconsidered. The EDF was chosen due its size and thrust capabilities, but it resulted being less efficient than an open propeller, due to the higher current consumption. There are several options to replace it, either with a gas engine or an electric motor. The issue with the gas engine was that it unnecessarily complicated the design and hybridisation of the aircraft, when it sole purpose was for the proof of concept. The fuel consumption would change the position of the centre of gravity of the aircraft in mid-flight, this would negatively affect the stability of vertical take-off and landing. Power Plant Upgrade There is emerging power supply technologies that can double the energy density of a LI-Po battery. This would result in a weight reduction maintaining the same propulsive performance. The drawback of this technology is its costs and its primitive state.
- 236. Design and Development of a Hybrid UAV Carlos C.M. 216 | P a g e ME5308 – Major Group Project Transmitter/Camera Range It would be worthwhile considering investing in a longer range antenna or a signal booster to have longer range of transmission, always taking into account that the national legal limits for first person view and automated flight aircraft. Aircraft Systems Currently there are three microcontrollers that aid the aircraft operation, with an extensive wiring system. This primitive method is suitable for an ongoing design project however it is far too tedious and complex for any practical use. It would be recommended, once all the vehicle and systems are finalised, to condense it all into one tailored circuit board within the aircraft that would encompass all the functions. In addition to the circuitry it would be recommended to improve the first person view functionality by implementing a camera gimbal mount. Material upgrade An extensive use of composite materials is highly recommended throughout the aircraft. This would reduce the weight of the aircraft and increase the strength of the structure, such as the nose landing gear reinforcement. If this vehicle was to be considered to commercial purposes, the weight restriction which severely affected the overall design of the aircraft could be ignored. Transition attempt As already described the system currently exists to attempt a transition manoeuvre from hover to horizontal flight. However due to the time restrictions of the project it was not achieved. Despite not being a main project objective, the manoeuvre will be attempted within the near future. In order to get to the point of running the test, all the functionalities must be tested before linking them together. In particular this refers to the mode switches on the tri copter from stabilize, to altitude hold, to acrobatic. More flight tests would be advisable to further tune the responses from the micro controllers, and experimentally locate the limits of the manoeuvrability of the aircraft. Autonomous Flight capabilities Another functionality which is currently in place but not being implemented is the autopilot aspect of the micro controllers on-board. Further research could be
- 237. Design and Development of a Hybrid UAV Carlos C.M. 217 | P a g e ME5308 – Major Group Project undertaken to take full advantage of this aspect and attempt a fully autonomous flight from take-off, to waypoint flying and finally a landing with Arducopter and Arduplane individually or even both in the case of automated transition. Tilt rotors At the beginning of the project it was widely considered by some team members to have tilting rotors that would provide vertical lift in the VTOL mission and would also provide thrust in horizontal flight. It was not taken forward because of the time restrictions of the project, but now that a functional platform is available one recommendation would be to further research the feasibility of implement such design. This could have great potential benefit in terms of weight optimisation because there would be no redundant motors in VTOL or STOL. Ultimate load testing on wings and fuselage Even though successful stress testing on the connections and spar rods were made, it could be a good idea to replicate the aircraft and asses its ultimate load factor. From this test weak points within the structure could be identified and further improvement could be implemented. It would also provide essential information for extreme manoeuvring closer to the aircrafts performance limits. Stress testing of 3D printed parts Detailed stress testing of 3D printed parts must be undertaken if extensive use of 3D printed manufacturing is to be used in the construction of aircraft and other mechanical system components. Of particular interest is the ability of printed ABS to manage strain energy and its capacity to do so before rupturing, particularly in the perpendicular directions with respect to the deposited ABS layers. The failure of the landing gear mount was a direct result of overstraining the intermediate deposition layers leading to crack formation in the part. Analysis of the relationship between layer-to-load orientations would be of significant importance in the manufacture of printed objects subject to mechanical stress.
- 238. Design and Development of a Hybrid UAV Carlos C.M. 218 | P a g e ME5308 – Major Group Project 13. Bibliography [1] “Flight Global,” 26 August 2011. [Online]. Available: http://www.unmanned.co.uk/unmanned-vehicles-news/unmanned-aerial- vehicles-uav-news/iai-completes-development-work-on-mini-panther-uas/. [Accessed 19 Oct 2013]. [2] H. Stone and G. Clarke, “T-Wing VTOL UAV: Technology Deomnstrator,” The University of Sydney, Sydney, 2002. [3] R. Gallagher, “Coming Soon to a Sky Near You: a Pentagon-Backed "Bat Drone",” 15 March 2013. [Online]. Available: http://www.slate.com/blogs/future_tense/2013/03/15/v_bat_drone_from_mlb_c ompany_can_conduct_urban_surveillance_wildlife_monitoring.html. [4] MLB Company, “Products,” 2013. [Online]. Available: http://spyplanes.com/products-v-bat/. [Accessed 19 Oct 2013]. [5] S. Trimble, “Flight Global,” Reed Business Information, 10 April 2013. [Online]. Available: http://www.flightglobal.com/news/articles/laad-santos-lab-unveils- new-uavs-signs-insitu-pact-384491/. [Accessed 4 October 2013]. [6] T. Skinner, “Shepard Media,” Shepard Press Limited, 11 April 2013. [Online]. Available: http://www.shephardmedia.com/news/uv-online/laad-2013-santos- lab-unveils-vtol-uav/. [Accessed 19 October 2013]. [7] E. Schechler, “Popular Mechanics,” Hearst Communication INc., 15 August 2013. [Online]. Available: http://www.popularmechanics.com/technology/aviation/news/this-unmanned- plane-can-land-on-a-helo-pad-15812116. [Accessed 15 October 2013]. [8] B. Coxworth, “ Gizmag Aircraft,” 20 August 2013. [Online]. Available: http://www.gizmag.com/hybrid-quadrotor-hq-uav/28767/. [Accessed 19 Oct 2013].
- 239. Design and Development of a Hybrid UAV Carlos C.M. 219 | P a g e ME5308 – Major Group Project [9] Advanced VTOL Technologies, “AVT Hammerhead V2,” AVT , 2013. [10] Advanced VTOL Technologies, “Hammerhead VTOL Aircraft,” 2013. [Online]. Available: http://www.avtolt.com/products/hammerhead.php. [Accessed 19 Oct 2013]. [11] Naval Technology, “Projects,” 2013. [Online]. Available: http://www.naval- technology.com/projects/belleagleeyeuav/. [Accessed 21 October 2013]. [12] Federation of American Scientists, “FAS Intelligence Resource Program,” 27 November 1999. [Online]. Available: http://www.fas.org/irp/program/collect/eagle-eye.htm. [Accessed 19 October 2013]. [13] R. Coppinger, “Flight Global UAVs,” 20 March 2008. [Online]. Available: http://www.flightglobal.com/news/articles/fully-autonomous-quad-tilt-wing-uav- makes-maiden-flight-222373/. [Accessed 21 October 2013]. [14] Chiba University Nonami Laboratory, “Unmanned Aerial Vehicle Gallery,” 2013. [Online]. Available: http://mec2.tm.chiba- u.jp/~nonami/english/research/uav/gal.html. [Accessed 2013]. [15] AV AeroVironment, “AV AeroVironment, Raven,” AeroVironment, [Online]. Available: http://www.avinc.com/uas/small_uas/raven/. [16] BAYKAR, “BAYKAR Mini UAS,” BAYKAR, [Online]. Available: http://www.baykarmakina.com/en-EN/category/10/Systems/1/bayraktar-mini- iha.aspx. [17] N. Mckeegan, “BATMAV System with AeroVironment's Wasp III MIcro Air Vehicle achieves full rate production,” Gizmag, 8 January 2008. [Online]. Available: http://www.gizmag.com/aerovironment-wasp-iii-micro-air- vehicle/8605/. [18] INNOCON, “INNOCON Micro Falcon Brochure,” INNOCON Ltd.
- 240. Design and Development of a Hybrid UAV Carlos C.M. 220 | P a g e ME5308 – Major Group Project [19] J. T. VanderMey, “A tilt rotor UAV for long endurance operations in remote environments,” Massachusetts Institute of Technology, 2011. [20] Civil Aviation Authority, “CAP 722- Unmanned Aircraft System Operations in UK Airspace - Guidance,” West Sussex, The Stationery Office (TSO), 2012. [21] Civil Aviation Authority, “Air Navigation: The Order and the Regulations,” London, TSO (The Stationery Offi ce), 2014. [22] D. P. Raymer, Aircraft Design a Conceptual Approach, 5th Revised edition edition ed., American Institute of Aeronautics & Astronautics, 2012. [23] Transonic Aerodynamics Committee, “Aerodynamic principles of winglets - ESDU 98013,” Engineering Sciences Data Unit (ESDU), 1998. [24] San Diego Air & Space Museum, “San Diego Air & Space Museum Archives,” U.S. Army, 1958. [Online]. Available: http://www.flickr.com/photos/sdasmarchives/4564654232/. [Accessed 2013]. [25] S. Nathan, “Flaviir project trials new forms of wing command,” The Engineer, 12 July 2010. [Online]. Available: http://www.theengineer.co.uk/in- depth/analysis/flaviir-project-trials-new-forms-of-wing- command/1003464.article. [Accessed 10 October 2013]. [26] M. James, “Nextcraft.com,” 2012. [Online]. Available: http://nextcraft.com/aircraftdesignconcepts.html. [27] W. H. H. D. T. P. Jack D. Mattingly, Aircraft Engine Design 2nd Edition, Washington D.C: American Institute of Aeronautics and Astronomics Inc., 2003. [28] D. S. M. Niƫă, “Estimating The Oswald Efficiency Factor From Basic Aircraft Geometrical Parameters,” Hamburg University, Hamburg, Germany, 2012. [29] D. P. Raymer, Aircraft Design: A Conceptual Approach, Washington D.C.: American Institute of Aeronautics and Astronautics Inc., 1989.
- 241. Design and Development of a Hybrid UAV Carlos C.M. 221 | P a g e ME5308 – Major Group Project [30] D. J. Roskam, Airplane Design: Part I Preliminary Sizing of Airplanes, Kansas USA: DARCorp, 2005. [31] S. Gudmundsson, General Aviation Aircraft Design Applied Methods and Procedures, Oxford: Elsevier, 2013. [32] FAA, “Aviation Maintenance Technician Handbook - Airframe,” United States Department of Transportation, Federal Aviation Administration, Airman Testing Standards Branch, Oklahoma, USA, 2012. [33] L. Dimitri, “Numerical Investigation of Blended Winglet Effects on Wing Performances,” Aalborg Univerisity, Aalborg, Denmark, 2008. [34] J. J. a. S. M. S. Guglielmo, “High-Lift Low Reynolds Number Airfoil Design,” Journal of Aircraft, vol. 34, no. 1, January-February 1997. [35] D. J. Roskam, Airplane Design: Part VII Determination of Stability, Control and Performance Characteristics, Kansas USA: DARCorp, 2006. [36] D. J. Roskam, Airplane Design: Part VI Preliminary Calculation of Aerodynamics, Thrust and Power Characteristics, Kansas USA: DARCorp, 2004. [37] P. F. J. J. I. A. Christian Bak, “Wind Tunnel Test of the NACA 63-415 and a Modifie NACA 63-415 Airfoil,” Risø National Laboratory, Roskilde, Denmark, 2000. [38] MIT, “www.mit.edu,” 6 April 2006. [Online]. Available: http://ocw.mit.edu/courses/aeronautics-and-astronautics/16-01-unified- engineering-i-ii-iii-iv-fall-2005-spring-2006/systems-labs-06/spl8.pdf. [Accessed October 2013]. [39] A. K. Kundu, Aircraft Design, vol. 1, Cambridge: Cambridge University Press, 2010, pp. 180-185. [40] M. Sadraey, “Chapter 12: Design of Control Surfaces,” in Aircraft Design: A
- 242. Design and Development of a Hybrid UAV Carlos C.M. 222 | P a g e ME5308 – Major Group Project Systems Engineering Approach, Wiley Publications, 2012. [41] D. M. Jabbal, Advanced Aerodynamics Note Lecture 2 "Vortex Lattice Numerical Method", 2013. [42] D. Y. Liu, “MIT Marine Hydrodynamics Lecture 19,” 2009. [Online]. Available: http://web.mit.edu/2.20/www/lectures/lec19/lecture19.pdf.. [43] R. Hendrickson, D. Roman and D. Rajkovic, “Drag: An Introduction,” 2006. [Online]. Available: http://www.dept.aoe.vt.edu/~mason/Mason_f/ConfigAeroDrag.pdf. [44] D. MacWilkinson, W. Blackerby and J. Paterson, “Correlation of full-scale drag predictions with flight measurements on the C-141A aircraft. Phase II wind tunnel test, analysis and prediction techniques. Volume 1 – Drag Predictions, Wind tunnel data analysis and correlation,” 1974. [45] P. M. Cavcar, “Minimum Drag Speed,” Eskisehir, Turkey, 2004. [46] D. S. Miklosovic, Drag on basic shapes, 2004. [47] S. Gudmundsson, General Aviation Aircraft Design: Applied Methods and Procedures, Elsevier, 2013. [48] ESDU, “SLOPE OF LIFT CURVE FOR TWO-DIMENSIONAL FLOW,” ESDU, 1955. [49] ESDU, “RATE OF CHANGE OF LIFT COEFFICIENT WITH CONTROL DEFLECTION IN INCOMPRESSIBLE TWO-DIMENSIONAL FLOW,” ESDU, 1956. [50] L. J. W. J. B. Laurence K, “The effects of variations in reynolds number between 3 to 25 million upon the aerodyamic characteristics of a number of NACA 6 series airfoil sections.,” Virginia, USA, 1950. [51] D. P. Raymer, Aircraft Design: A Conceptual Approach, AIAA Education Series, 2012.
- 243. Design and Development of a Hybrid UAV Carlos C.M. 223 | P a g e ME5308 – Major Group Project [52] “xflr5-About stability analysis using XFLR5,” A. Deperrois, November 2010. [Online]. Available: http://www.xflr5.com/docs/XFLR5_and_Stability_analysis.pdf. [53] “ Longitudinal stability and control,” [Online]. Available: www.aerostudents.com. [54] Federal Aviation Administration, “Pilot's Handbook of Aeronautical Knowledge,” in Chapter 05: Flight Controls , Washington, Federal Aviation Administration (FAA). [55] National Aeronautics and Space Administration, “Vertical stabilizer - rudder,” March 2014. [Online]. Available: http://www.grc.nasa.gov/WWW/k- 12/airplane/rud.html. [Accessed 13 September 2010]. [56] G. Shpati, “Aircraft CG Envelopes,” 17 September 2011. [Online]. Available: http://www.sawe.ca/download/tech2011/Aircraft%20CG%20Envelopes.pdf. [Accessed 10 February 2014]. [57] D.-W. Yoo, H.-D. Oh, D.-Y. Won and M.-J. Tahk, “Dynamic Modeling and Stablization Techniques for Tri-Rotor Unmanned Aerial Vehicles,” vol. 11(3), no. 167, 2010. [58] B. Mettler, Identification Modeling and Characteristics of Miniature Rotorcraft, Boston: Kluwer Academic Publishers, 2003. [59] D. Moell and J. Nordin, “VTOL UAV - A Concept Study,” Linkopings Universitet, Linkoping, 2008. [60] J. Roskam, Airplane Design Part III: Layout Design of Cockpit, Fuselage, Wing and Empennage: Cutaways and Inboard Profiles, Darcorporation, 1986. [61] F. O. P. P. G. School, “Aircraft Structure - An introduction to major airplane components,” [Online]. Available: http://www.free-online-private-pilot-ground- school.com/aircraft-structure.html. [Accessed 15 February 2014]. [62] F. A. Ltd., “SkyRanger Construction,” Flylight Airsports Ltd., [Online]. Available: http://www.flylight.co.uk/skyranger/construction.htm. [Accessed 1 February
- 244. Design and Development of a Hybrid UAV Carlos C.M. 224 | P a g e ME5308 – Major Group Project 2014]. [63] “Chapter 4: Basic Aircraft Construction,” [Online]. Available: http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0 CCsQFjAA&url=http%3A%2F%2Fwww.ae.metu.edu.tr%2F~ae362%2FAircraft %2520Basic%2520Construction.pdf&ei=XAoiU6OyD- Gc0QXSrIHwCg&usg=AFQjCNHXUWY-7e3WbznllYgtQ5Q399N83g. [64] Flight Global , “Flight Global Cutaways,” [Online]. Available: http://www.flightglobal.com/cutaways/. [Accessed 20 November 2013]. [65] J. Fielding, C. Lawson, R. Pires and G. Monterzino, “Development of the Demon technolody demonstrator UAV,” in 27th International Congress of the Aeronautical Sciences, 2010. [66] BormaTec, “UAV MAJA,” Bormatec Unmanned Vehicles, [Online]. Available: http://www.bormatec.com/index.php?option=com_content&view=article&id=78 &Itemid=59. [Accessed 10 October 2013]. [67] Range Video, “RV Jet Airframe Manual,” Range Video, 2012. [68] A. Daviel, “Building a Popc,” March 2004. [Online]. Available: http://andrew.triumf.ca/andrew/popsicle-bridge/ . [Accessed 10 December 2013]. [69] “Basis of Structural Design: Course 2 Structural Action: cables and arches,” [Online]. Available: http://www.ct.upt.ro/users/AurelStratan/bsd%5Cc02- bsd.pdf. [Accessed 20 October 2013]. [70] D. P. Raymer, Aircraft Design: A Conceptual Approach, Reston, Virginia: American Institute of Aeronautics and Astronautics Inc., 2006. [71] R. S. e. a. S. Brandt, Intorduction to Aeronautics: A Design Perspective, Reston: American Institute of Aeronautics and Astronautics Inc, 2004, pp. 350- 354.
- 245. Design and Development of a Hybrid UAV Carlos C.M. 225 | P a g e ME5308 – Major Group Project [72] canply.org, “CANPLY Plywood Design Fundamentals,” 2012. [Online]. Available: www.tolko.com/documents/plywood_designfund.pdf. [Accessed 20 January 2014]. [73] Federal Aviation Administration, “Aviation Maintenance Technician Handbook, Chapter 1: Aircraft Structures,” Federal Aviation Administration, 2013. [74] Civil Aviation Safety Authority, “DESIGN STANDARDS: UNMANNED AERIAL VEHICLES - AEROPLANES,” Civil Aviation Safety Authority, 2000. [75] Federal Aviation Administration, “Chapter 02: Aircraft Structure,” in Pilot's Handbook of Aeronautical Knowledge, Federal Aviation Administration, 2013. [76] Nitroplanes.com, “Nitroplanes.com's,” Nitroplanes.com's , [Online]. Available: http://www.nitroplanes.com/transall160.html. [Accessed 18 03 2014]. [77] G. Landolfo, “AERODYNAMIC AND STRUCTURAL DESIGN OF A SMALL NONPLANAR WING UAV,” UNIVERSITY OF DAYTON, Dayton, Ohio, 2008. [78] BillKits, “billkits.com/,” BillKits, 2012. [Online]. Available: http://www.billkits.com/. [Accessed 18 03 2014]. [79] D. W. Song, “Composite materials lecture notes part 3 Manufacturing Processing,” Brunel University, London, 2013. [80] Federal Aviation, “Chapter 3: Aircraft Fabric Covering,” in Aviation Maintenance Technician Handbook - Airframe, Federal Aviation, 2013. [81] RangeVideo, “RangeVideo.com,” RangeVideo, 2013. [Online]. Available: http://www.rangevideo.com/en/18-rvjet. [Accessed 18 03 2014]. [82] J. Woodward, “Matching Engine and Propeller,” The University of Michigan, 1973. [83] MIT, “DC Motor/ Propeller Matching,” MIT, 2005. [84] “Aircraft Datasheet,” 2013. [Online]. Available: http://aircraft- world.com/prod_datasheets/hp/emeter/hp-propconstants.htm. [Accessed 2 Nov
- 246. Design and Development of a Hybrid UAV Carlos C.M. 226 | P a g e ME5308 – Major Group Project 2013]. [85] D. Raymer, Aircraft Design: A Conceptual Approach, AIAA, 2012. [86] MaxAmps, “MaxAmps,” MaxAmps, [Online]. Available: www.maxamps.com. [Accessed 10 November 2013]. [87] E. Cetinsoy, S. Dikyar, C. Hancer, E. Oner, E. Sirimoglu, M. Unel and M. Aksit, “Design and contstruction of a novel quad tilt-wing UAV,” vol. 26 Nov 2012, 2012. [88] G. Owen, “How to build your own quadcopter flight controller,” 2013. [Online]. Available: http://ghowen.me/build-your-own-quadcopter-autopilot/. [Accessed 10 October 2013]. [89] J. Gauterin, “StackOverlow,” 9 December 2009. [Online]. Available: http://stackoverflow.com/questions/1840314/when-do-i-need-to-use- quaternions. [Accessed 10 October 2013]. [90] B. Horn, Some Notes on Unit Quaternions and Rotation, MIT, 2001. [91] Mathworks, “Tuning Multi-Loop control systems,” 2014. [Online]. Available: http://www.mathworks.co.uk/help/robust/examples/tuning-multi-loop-control- systems.html. [Accessed 2014]. [92] D. Heath-Caldwell, “Relay code,” 2014. [93] AdruCopter, “Adrucopter- Instructions,” [Online]. Available: http://copter.ardupilot.com. [Accessed 2 March 2014]. [94] G. Hunt, “How to select a DC Motor,” Micro Motion Solutions, 2013. [95] Aniss1001, “Adruino Hardware Interfereence,” 2009. [Online]. Available: http://forum.arduino.cc/index.php/topic,14146.0.html. [Accessed Dec 2013]. [96] J. Dang and H. Laheij, “Hydrodynamic Aspects of Steerable Thrusters: Session Thrusters/Propulsion,” Wartsila propulsion Netherlands BV, Drunen,
- 247. Design and Development of a Hybrid UAV Carlos C.M. 227 | P a g e ME5308 – Major Group Project Netherlands, 2004. [97] MaxBotix, “MaxSonar Operation on a multicopter,” MaxBotix, Dec 2013. [Online]. Available: http://www.maxbotix.com/articles/067.htm. [Accessed Jan 2014]. [98] S. Shrouder, “DEMON UAV - FLYING WITHOUT FLAPS,” BAE Systems, [Online]. Available: http://www.baesystems.com/product/BAES_051903/demon-uav---flying- without- flaps.?_afrLoop=162667540574000&_afrWindowMode=0&_afrWindowId=null& baeSessionId=1pRTTbnQpDRkwNHLrVhCfXQvbyJJ5npWKxYp3Cxpylh1sG1n 5ktS!-909532306#%40%3F_afrWindowId%3Dnull%26baeSession. [Accessed 15 October 2013]. [99] D. J. Roskam, Airplane Design: Part III Layout of Cockpit, Fuselage, Wing and Emepennage, Kansas USA: DARCorp, 2002. [100] D. J. Roskam, Airplane Design: Part II Preliminary Configuration Design and Integration of Propulsion Systems, Kansas USA: DarCorp, 2004. [101] D. J. Roskam, Airplane Design: Part IV Layout of Landing Gear and Systems, Kansas USA: DARCorp, 2004. [102] D. J. Roskam, Airplane Design: Part V Component Weight Estimation, Kansas USA: DARCorp, 2003. [103] D. J. Roskam, Airplane Design: Component Weight Estimation, Kansas USA: DRACorp, 2003. [104] DARCorporation, Advanced Aerodynamics Analysis Help, Kansas USA: DARCorp, 2013.
- 248. Design and Development of a Hybrid UAV Carlos C.M. 228 | P a g e ME5308 – Major Group Project 14. Appendices 14.1. Appendix A – Technical Details Roskam Constraint Analysis It shows, compared to Mattingly that the conceptual design approach is less critical than the Mattingly et All, that does not predict any values, instead it allows the user to input all of them to tailor the constraint analysis to the specification needed, rather than commercial aviation. Figure 143 Roskam Constraint Analysis Design Space 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 50 100 150 Trust-to-Weightratio(T/W) Wing Loading (W/S) (N/m^2) Roskam Constarint Analysis Due to Load Factor At Cruise At Take Off Landing Final Configuration Cruise Stall
- 249. Design and Development of a Hybrid UAV Carlos C.M. 229 | P a g e ME5308 – Major Group Project Wing Profile Analysis: Additional Aerofoils Additional Aerofoils Aerofoil *Aerofoil NACA 63(2)-615 NACA 63- 412 NACA 65- 210 NACA 65(2)-415 S1223 E387 NACA 9412 NACA 4412 NACA 4312 S8055 (12%) FX 63-137 (13.7%) *Alpha Stall (α-max) 9 9 7 8 11 10 12 12 12 12 10 *Alpha cruise (α-cruise) 2 3 2 3 0 2 2 2 2 5.5 0 Alpha Margin (deg.) 7 6 5 5 11 8 10 10 10 6.5 10 *Alpha Rotation (deg.) 5 6 5 6 3 5 5 5 5 8.5 3 Coefficients *Oswald Co-eff ( e) 0.85 0.85 0.85 0.85 0.85 0.85 0.85 0.85 0.85 0.85 0.85 *Cl Cruise 0.557 0.510 0.300 0.513 1.105 0.463 0.943 0.517 0.491 0.628 0.733 *Cl Max 1.139 1.011 0.724 0.929 1.969 1.128 1.740 1.337 1.131 1.160 1.547 *Cd Cruise (parasite) 0.014 0.012 0.012 0.013 0.021 0.010 0.016 0.010 0.010 0.010 0.014 *Cl Climb 0.811 0.763 0.556 0.764 1.354 0.717 1.193 0.771 0.746 0.878 0.986 *Cd Climb (parasite) 0.014 0.011 0.014 0.012 0.019 0.011 0.016 0.011 0.012 0.014 0.014 *Cm cruise -0.262 -0.208 -0.116 -0.210 -0.635 -0.199 -0.475 -0.235 -0.213 -0.198 -0.418 *Cm max -0.405 -0.330 -0.220 -0.312 -0.855 -0.364 -0.651 -0.438 -0.417 -0.259 -0.621 Lift Lift TO (N) 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 Lift Cruise (N) 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 49.050 Lift max from Cl max (N) 70.632 70.632 70.632 70.632 70.632 70.632 70.632 70.632 70.632 70.632 70.632 (L/D)max wing 39.948 43.776 25.842 40.268 49.834 47.274 55.308 53.032 49.885 58.684 52.148 (L/D)max aircraft 7.450 6.853 4.157 6.884 14.498 6.383 12.488 7.079 6.750 8.504 9.845 Lift Climb 86.318 88.786 109.858 88.459 72.728 91.965 75.073 88.514 90.075 82.997 79.866 Table 37 Wing profile: Additional Analysis
- 250. Design and Development of a Hybrid UAV Carlos C.M. 230 | P a g e ME5308 – Major Group Project Comparison between the two analysis methods in XFLR5 Results for Vortex Lattice Method (VLM) and the Panel Method with a percentage comparison. Angle of Attack VLM Panel Method Comparison (%) CL CD Cm CL CD Cm CL CD Cm -2 - 0.080416 0.009465 -0.004775 -0.07759 -0.00965 0.010199 1.036369 -0.98042 -0.46818 -1 0.004671 0.009307 -0.016064 0.012234 0.009518 -0.01268 0.381805 0.977831 1.266977 0 0.089772 0.009678 -0.036938 0.102066 0.009958 -0.0356 0.879549 0.971882 1.037672 1 0.174809 0.010429 -0.05782 0.191844 0.010852 -0.05853 0.911204 0.961021 0.98792 2 0.259702 0.011728 -0.078686 0.281512 0.012349 -0.08144 0.922526 0.949713 0.966172 3 0.344372 0.013561 -0.099509 0.371012 0.014428 -0.10431 0.928196 0.939909 0.953965 4 0.428741 0.015966 -0.120265 0.460289 0.017169 -0.12711 0.93146 0.929932 0.946149 5 0.512732 0.018932 -0.140928 0.549286 0.020551 -0.14981 0.933452 0.92122 0.940718 6 0.596266 0.022612 -0.161473 0.637948 0.024813 -0.17238 0.934662 0.911296 0.936722 7 0.67927 0.027166 -0.181875 0.726219 0.02998 -0.1948 0.935351 0.906137 0.933655 8 0.761667 0.032336 -0.202109 0.814045 0.035705 -0.21703 0.935657 0.905643 0.931232 9 0.843384 0.037861 -0.22215 0.901373 0.041893 -0.23906 0.935666 0.903755 0.929261 10 0.924349 0.044145 -0.241975 0.988149 0.049161 -0.26085 0.935435 0.897968 0.927633 11 1.004491 0.051249 -0.261558 1.074322 0.057517 -0.28238 0.935 0.891024 0.926262 12 1.083743 0.060276 -0.280876 - - - - - - Table 38 VLM and Panel Method Result comparison from XFLR5
- 251. Design and Development of a Hybrid UAV Carlos C.M. 231 | P a g e ME5308 – Major Group Project Figure 144 To obtain for Step 4 in Table 17 [49] Figure 145 To obtain for Step 5 in Table 17 [49].
- 252. Design and Development of a Hybrid UAV Carlos C.M. 232 | P a g e ME5308 – Major Group Project Final VTOL and STOL endurance details. STOL Phase Time (s) Current Draw (A) Battery Needed (mAh) Take-off 10 50 138.89 Climb 15 30 125.00 Cruise 450 17 2125.00 Loiter 30 8 66.67 Descent 20 8 44.44 Landing 20 10 55.56 Total Time in Seconds 545 Battery Capacity 2556 in Mins. 9.083333333 Safety battery capacity 3194 Table 39 STOL Mission profile and current specifications VTOL Phase Time (s) Current Draw/motor (A) Total Current Draw (A) Battery Needed (mAh) 1st Climb 30 32 96 800 1st Hover 45 30 90 1125 2nd Climb 30 32 96 800 2nd Hover 45 30 90 1125 Descent 30 30 90 750 Total Time in Seconds 180 Battery Capacity 4600 in Mins. 3 Safety battery capacity 5750 Table 40 VTOL Mission profile and current specifications
- 253. Design and Development of a Hybrid UAV Bennie M. 233 | P a g e ME5308 – Major Group Project
- 254. Design and Development of a Hybrid UAV Bennie M. 234 | P a g e ME5308 – Major Group Project
- 255. Design and Development of a Hybrid UAV Bennie M. & Carlos C.M. 235 | P a g e ME5308 – Major Group Project Motor Test Code // Sweep // by BARRAGAN <http://barraganstudio.com> // This example code is in the public domain. // Authors: Carlos Calles & Bennie Mwiinga #include <Servo.h> // library in use Servo myservo; // create servo object to control a servo // a maximum of eight servo objects can be created int i; // declating i as an integer char junk; // clear buffer holder int myspeed = i; //needed for slow down sequence void arm() // arming sequence procedure { Serial.println("Sending In High=10"); myservo.write(10); delay(1000); Serial.println("Connect Battery"); delay(4000); Serial.println("Sending In Low=0"); myservo.write(0); delay(1000); Serial.println("Sending In Test Signal=4"); myservo.write(4); } void setup() // seting up the system { Serial.begin(9600); myservo.attach(12); // attaches the servo on pin 12 to the servo object arm(); // call arm() function } void emergencyStop() // emergency stop function { Serial.println("Emergency Stop"); for(myspeed = 50; myspeed >= 10; myspeed -= 5) { myservo.write(myspeed); Serial.println(myspeed); delay(50); } for(myspeed = 10; myspeed >= 1; myspeed -= 1) { myservo.write(myspeed); Serial.println(myspeed); delay(200); } myservo.write(0); delay(1000); Serial.println("Stopped Safe to Disconnect +ve wire."); } void loop() { while(Serial.available() == 0); //check value in serial command window
- 256. Design and Development of a Hybrid UAV Bennie M. & Carlos C.M. 236 | P a g e ME5308 – Major Group Project { i = Serial.parseInt(); Serial.print("Current Selected="); Serial.println(i, DEC); if(i > 1 && i <= 100) //FAILSAFE: upper limit servo/motor speed { myservo.write(i); //send power value to motor/servo } else if (i = 1 || i > 100) // FAILSAFE emergency stop initiated at either I is input as 0 or > 100 eg 101. { emergencyStop(); } else // FAILSAFE emergency stop initiated { emergencyStop(); } // ------------ CLEAR BUFFER ----------- // while(Serial.available() > 0) //clear buffer { junk = Serial.read(); //---------------------------------------// } } }
- 257. Design and Development of a Hybrid UAV Carlos C.M. 237 | P a g e ME5308 – Major Group Project 14.2. Appendix B – Project Plan and Management 14.2.1. Gantt Chart The Gantt chart below shows the progress of the group relative to the project deadlines. Up to the preliminary report (the green vertical line) the blue bars represent the actual timing and progress of each phase. After the green line the blue bars were the estimates of future progress, the arrows superimpose on the diagram depict the actual progress of the specific project phase.
- 258. Design and Development of a Hybrid UAV Carlos C.M. 238 | P a g e ME5308 – Major Group Project
- 259. Design and Development of a Hybrid UAV Camilo V. 239 | P a g e ME5308 – Major Group Project 14.2.2. Logistics
- 260. Design and Development of a Hybrid UAV Camilo V. 240 | P a g e ME5308 – Major Group Project
- 261. Design and Development of a Hybrid UAV Camilo V. 241 | P a g e ME5308 – Major Group Project
- 262. Design and Development of a Hybrid UAV Camilo V. 242 | P a g e ME5308 – Major Group Project
- 263. Design and Development of a Hybrid UAV Camilo V. 243 | P a g e ME5308 – Major Group Project
- 264. Design and Development of a Hybrid UAV Camilo V. 244 | P a g e ME5308 – Major Group Project
- 265. Design and Development of a Hybrid UAV Camilo V. 245 | P a g e ME5308 – Major Group Project
- 266. Design and Development of a Hybrid UAV Camilo V. 246 | P a g e ME5308 – Major Group Project
- 267. Design and Development of a Hybrid UAV Camilo V. 247 | P a g e ME5308 – Major Group Project
- 268. Design and Development of a Hybrid UAV Camilo V. 248 | P a g e ME5308 – Major Group Project
- 269. Design and Development of a Hybrid UAV Camilo V. 249 | P a g e ME5308 – Major Group Project
- 270. Design and Development of a Hybrid UAV Camilo V. 250 | P a g e ME5308 – Major Group Project