CReSIS UAV Critical Design Review: The Meridian
William Donovan
University of Kansas 2335 Irving Hill Road Lawrence, KS 66045-7612 http://cresis.ku.edu
Technical Report CReSIS TR 123
June 25, 2007
This work was supported by a grant from the National Science Foundation (#ANT-0424589). Executive Summary
This report briefly describes the development of the three preliminary configuration
designs proposed for the Meridian UAV. This report details the selection of the
primary configuration and further, more detailed, analysis including Class II weight
and Balance, Class II Stability and Control, Performance Analyses, Systems Design,
Class II landing gear, structural arrangement, a manufacturing breakdown and a cost
analysis.
The design mission for this aircraft is to takeoff from a snow or ice runway, fly to a
designated area, then use low frequency radar to perform measurements of ice sheets
in Greenland and Antarctica. Three designs were developed:
• A Monoplane with Structurally Integrated Antennas
• A Monoplane with Antennas Hanging from the Wing
• A Biplane with Antennas Structurally Integrated Into the Lower Wing
The monoplane with antennas hanging from the wing was selected as the primary
configuration for further development. This report describes the Class II design and
analysis of that vehicle.
i Acknowledgments
This material is based upon work supported by the National Science Foundation under Grant No. AST-0424589. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
ii Table of Contents
Executive Summary...... i Acknowledgments ...... ii Table of Contents ...... iii List of Figures...... v List of Tables ...... vi Nomenclature ...... vii Abbreviations ...... viii 1 Summary of Preliminary Designs...... 9 2 Configuration Selection and Requirement Changes ...... 11 2.1.1 Engine Selection – Turboprop Variant ...... 12 3 Class II Design...... 15 3.1 Class II Weight and Balance...... 15 3.1.1 The Aircraft V-n Diagram ...... 15 3.2 Component Weight Estimations ...... 18 3.3 Class II Stability and Control...... 22 3.3.1 Trim Diagrams...... 22 3.3.2 Open Loop Dynamics ...... 31 3.3.3 Actuator Size and Rate Requirements ...... 37 3.4 Class II Aerodynamics...... 38 3.5 Propulsion ...... 46 3.6 Performance Analysis...... 47 3.6.1 Stall Speed...... 47 3.6.2 Takeoff Distance...... 47 3.6.3 Climb...... 48 3.6.4 Cruise Performance...... 48 3.6.5 Landing Distance...... 50 3.7 Systems ...... 51 3.7.1 Flight Control System...... 52 3.7.2 Electrical System ...... 54 3.7.3 Communications/Telemetry System...... 58 3.7.4 Fuel System...... 59 3.7.5 Anti-Icing System ...... 60 3.8 Class II Landing Gear ...... 61 3.8.1 Tire Selection...... 62 3.8.2 Strut Sizing...... 63 3.8.3 Landing Gear Integration...... 64 3.9 Structural Arrangement...... 66 3.9.1 Wing Structure...... 67 3.9.2 Fuselage Structural Layout ...... 70 3.10 Manufacturing Breakdown...... 74 3.11 Cost Analysis...... 75
iii 3.11.1 Research, Development, Test, and Evaluation Costs ...... 77 3.11.2 Acquisition Cost...... 77 3.11.3 Cost Estimate Summary...... 78 3.11.4 Cost Estimate Justification...... 81 4 Conclusions...... 84 5 References...... 85
iv List of Figures
Figure 1.1: Comparison of Fuel Usage for 3 Fine Scale Missions ...... 10 Figure 1.2: Combined Takeoff Weight Regression Chart ...... 10 Figure 2.1 - Vivaldi Antenna ...... 11 Figure 2.2: CAD Model of Innodyn 165TE...... 13 Figure 3.1 - The Meridian UAV ...... 14 Figure 3.2 - V-n Diagram for the Meridian ...... 17 Figure 3.3 - Center of Gravity Excursion for the Meridian ...... 20 Figure 3.4 – Component C.G. Locations ...... 21 Figure 3.5 - Trim Diagram - Cruise...... 23 Figure 3.6 - Trim Diagram - Takeoff, Gear Down ...... 24 Figure 3.7 - Trim Diagram - Takeoff, Gear Up ...... 25 Figure 3.8 - Trim Diagram – Landing Heavy, Gear Down ...... 26 Figure 3.9 - Trim Diagram - Landing Heavy, Gear Up...... 27 Figure 3.10 - Trim Diagram - Landing Light, Gear Down...... 28 Figure 3.11 - Trim Diagram - Landing Light, Gear Up...... 29 Figure 3.12 - Trim Diagram - OEI...... 30 Figure 3.13 - Drag Polars for the Meridian without Antennas ...... 41 Figure 3.14 - Lift-to-Drag for the Meridian without Antennas ...... 42 Figure 3.15 - Drag Polars for the Meridian with Antennas ...... 43 Figure 3.16 - Lift-to-Drag for the Meridian with Antennas...... 44 Figure 3.17 - Relationship of Parasite Area and Wetted Area for Various Single Engine Aircraft [6]...... 46 Figure 3.18 - Cloud Cap Tech. Piccolo II Autopilot [36]...... 52 Figure 3.19 - Piccolo II Architecture [36] ...... 53 Figure 3.20 - Piccolo Ground Station and Pilot Controller (Operator Interface Not Shown) [36] ...... 53 Figure 3.21 - Electrical Load Profile for the Meridian UAV ...... 56 Figure 3.22 - Electrical System Layout ...... 57 Figure 3.23 - Fuselage Systems Layout...... 58 Figure 3.24 - Fuel Tank Integration...... 60 Figure 3.25 - Lancair Legacy Landing Gear Strut [37] ...... 64 Figure 3.26 - Lancair Legacy Landing Gear Installation [38]...... 65 Figure 3.27 - Matco Tailwheel Assembly [39]...... 65 Figure 3.28 - Wing Structural Layout...... 69 Figure 3.29 - Fuselage Structural Layout ...... 71 Figure 3.30 - Wing-Fuselage Attachment...... 72 Figure 3.31 - Standard 20 Foot Shipping Container Door [9] ...... 73 Figure 3.32 - Typical Engine Mount for the Innodyn 165TE...... 74 Figure 3.33 - Manufacturing Breakdown...... 75 Figure 3.34 - Cost Breakdown by Overall Category ...... 80 Figure 3.35 - UAV Cost in Terms of Payload Weight ...... 82
v Figure 3.36 - UAV Cost Based on System Cost Versus Payload Weight ...... 83
List of Tables
Table 1.1: Summary of Preliminary Design Concepts ...... 9 Table 3.1 - V-n Diagram Parameters ...... 16 Table 3.2 - Design Speeds and Load Factors for the Meridian ...... 16 Table 3.3 - Class II Weight and Balance for the Meridian ...... 19 Table 3.4 - Weight and Balance Summary for the Meridian...... 19 Table 3.5 - Meridian Flight Conditions ...... 22 Table 3.6 - Dynamic Analysis Flight Conditions ...... 31 Table 3.7 - Stability Deriviatives for the Meridian...... 33 Table 3.8 - Control Derivatives for the Meridian ...... 34 Table 3.9 - Longitudinal Transfer Functions for Cruise...... 35 Table 3.10 - Lateral Transfer Functions for Cruise ...... 35 Table 3.11 - Directional Transfer Functions for Cruise...... 36 Table 3.12 - Meridian Dynamic Stability Parameters ...... 36 Table 3.13 - Roll Control Requirements - Time to Achieve Bank Angle (Seconds) . 37 Table 3.14 - Roll Control Results ...... 37 Table 3.15 Wing Geometry for Drag Calculations...... 38 Table 3.16 - V-Tail Geometry for Drag Calculations...... 38 Table 3.17 - Fuselage Geometry for Drag Calculations ...... 39 Table 3.18 - Flap and Landing Gear Geometry for Landing Gear Calculations ...... 39 Table 3.19 - Drag Analysis Results ...... 40 Table 3.20 - Resultant Oswald's Efficiency and Parasite Area for the Meridian ...... 45 Table 3.21 - Stall Speed Summary ...... 47 Table 3.22 - Landing Gear Strut Sizing...... 64 Table 3.23 - Engineering and Manufacturing Rate Estimation ...... 76 Table 3.24 - RDT&E and Acquisition Cost Summary ...... 79 Table 3.25 - Cost Breakdown by Overall Category...... 79 Table 3.26 - Cost Breakdown by RDT&E and Production Categories ...... 80 Table 3.27 - Current UAV Procurement Cost [40]...... 81
vi Nomenclature
Symbol Description Units AR Aspect Ratio ~ b Wing Span ft, in c Wing chord ft, in CD Drag Coefficient ~ CD0 Zero-Lift Drag Coefficient ~ CL Lift Coefficient ~ -1 CLα Lift-Curve Slope Rad cp Specific Fuel Consumption Lbs/hp-hr D Drag Lbs Dp Propeller Diameter Ft e Oswald’s Efficiency ~ f Equivalent Parasite Area Ft2 L Lift Lbs M Munk’s Span Factor ~ np Number of Propeller Blades ~ P Engine Power hp 2 Pbl Blade Power Loading hp/ft R Range Nm S Wing Area Ft2 2 SWet Wetted Area Ft WE Empty Weight Lbs WF Fuel Weight Lbs Wpay Payload Weight Lbs WTO Takeoff Weight Lbs Γ Dihedral Angle Deg α Angle of Attack Deg ε Wing Twist Deg η Wing Station ~ ηp Propeller Efficiency ~ σ Biplane Interference Factor ~
vii Abbreviations
Abbreviation Description CReSIS Center for Remote Sensing of Ice Sheets FAR Federal Aviation Regulations NSF National Science Foundation UAV Uninhabited Air Vehicle
viii 1 Summary of Preliminary Designs
Three preliminary designs have previously been presented to satisfy the
requirements for the CReSIS polar research mission:
• A monoplane with flush-mounted antennas
• A monoplane with hanging antennas
• A biplane with flush mounted antennas
The three designs are summarized in Table 1.1. The fuel required to complete 3
fine-scale mission is shown in Figure 1.1 for the three designs as well as the
Lockheed P-3 and the De Havilland Twin Otter. Figure 1.2 shows the aircraft plotted on the takeoff weight regression chart.
Table 1.1: Summary of Preliminary Design Concepts Parameter Units Red Design White Design Blue Design Geometry Wing Area ft2 49 82 88 Wing Span ft 15.33 25.6 17.2 Length Overall ft 16 17.5 16.5 Height Overall ft 5.5 5.6 5.6 Weights Takeoff Weight lbs 760 1,270 950 Empty Weight lbs 450 720 550 Payload Weight lbs 121 121 121 Fuel Weight lbs 185 425 270 Performance Range nm 1,750 1,750 1,750
L/DCr ~ 12.5 8.0 10.0 Powerplant Engine ~ Rotax 912-A Rotax 914-F Rotax 914-F Power hp 81 115 115
9 1,200
8,000
119
188
82 Red DesignRed Design White Design Blue Orion P-3 Otter Twin
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 Fuel Required for 3 Fine Missions, gallons
Figure 1.1: Comparison of Fuel Usage for 3 Fine Scale Missions
10,000
Predator B
E-Hunter
Shadow 600 1,000 Predator Shadow 200 I-Gnat
Dakota Empty Weight, lbs
100
log10(WTO) = A + B*log10(WE) A = -0.0183 B = 1.0930
10 100 1,000 10,000 Takeoff Weight, lbs
Figure 1.2: Combined Takeoff Weight Regression Chart
10 2 Configuration Selection and Requirement Changes
The three preliminary designs presented were compared based on several criteria including weight, aerodynamic performance, and antenna integration. The selection of the configuration for further study was made primarily based on antenna integration issues. While the Red and Blue designs were more efficient than the
White design aerodynamically, the antenna integration methods used for these designs would limit the possible bandwidth of the radar, thereby decreasing the total system performance. The success of the CReSIS project is heavily dependent on the level of synergy between the radar and aircraft systems. Design decisions must be made based on high level, systematic concerns.
L = 0.5 m H = 0.5 m T = 0.125 m
Figure 2.1 - Vivaldi Antenna
11 The White design was therefore selected as the configuration for further
development. However, the antenna type must be changed to accommodate higher
operating bandwidths. The primary antenna will be a Vivaldi antenna as shown in
Figure 2.1. These are essentially flat plates that hang from the wing of the aircraft.
While these have been selected as the current antenna for the radar system, it is clear
that the antenna type may change again. Therefore, the White design was modified
slightly such that eight mounting points will be integrated into the wing structure
from which several types of antennas could be mounted. The aircraft will be
designed for the Vivaldi antennas, but if another type of antenna of equal or smaller
size proves to be more effective, then it may be mounted on the wing without any
structural modifications. Essentially, the new design philosophy is that the aircraft is
relatively insensitive to the type and size of antennas used within reason. This
solution will yield a highly adaptable, high performance vehicle capable of multiple
missions.
2.1.1 Engine Selection – Turboprop Variant
The engines selections shown for the three preliminary designs were driven by power and specific fuel consumption requirements. The reliability of these engines in a cold-weather environment is questionable. Also, from a logistics standpoint, the
Rotax engines are suboptimal as the primary fuel used in Antarctica is Jet-A, not aviation gas. For these reasons, the Innodyn 165TE (Figure 2.2) has been selected for further investigation. The Innodyn is a fairly new, small turbopropeller engine that
12 has yet to be fully tested. Therefore, the specific fuel consumption of the engine is
somewhat unknown.
Figure 2.2: CAD Model of Innodyn 165TE The current specific fuel consumption estimate for the Innodyn engine is 0.70 lbs/hp-hr, which is much higher than the Rotax 912 and 914 value of 0.56 lbs/hp-hr.
Nonetheless, the reliability and maintainability issues make this engine very appealing for this mission. The Innodyn has been selected as the primary engine for the Meridian pending testing that will be performed in the Fall of 2006. The Rotax
914 will be considered as a backup to the Innodyn in the case of poor test results.
13
Figure 2.3 - The Meridian UAV
14 3 Class II Design
The purpose of this document is to expand upon the chosen aircraft configuration
(Monoplane with antennas hanging from the wing) through Class II design. The new
Meridian design shown in Figure 2.3 is the result of several design iterations focused
on manufacturability and operational constraints.
3.1 Class II Weight and Balance
The purpose of this section is to describe the Class II weight and balance performed
for the Meridian UAV. This consisted of first calculating and plotting a V-n diagram
to determine the limit and ultimate loading for the Meridian. The results of the V-n
diagram were then used to create weight estimates for each vehicle component.
Finally, a weight and balance analysis is presented to show the aircraft center of gravity travel.
3.1.1 The Aircraft V-n Diagram
A V-n diagram was constructed for the Meridian UAV to help determine the
maximum load factors and design speeds that will be used for structural sizing. The
V-n diagram was created based on FAR 23 requirements for Normal class aircraft as
there are currently no certification requirements for UAVs. The inputs to the V-n
diagram creation are shown in Table 3.1.
15 Table 3.1 - V-n Diagram Parameters Parameter Value Units Altitude 0 ft
Wgross 1,125 lbs S 69.6 ft2 W/S 16.2 psf m.g.c. 2.64 ft -1 Cla 3.98 rad
CLmax (+) 1.3 ~
CLmax (-) -0.97 ~
CD @ CLmax (+) 0.085 ~ C @ C (-) D Lmax 0.064 ~
The V-n diagram for the Meridian is shown in Figure 3.1. The design speeds and
limit load factors are shown in Table 3.2. The positive load factor was set to 3.8 based on FAR 23 requirements [35]. The negative load factor was set to 40 percent of the positive load factor according to FAR 23 requirements [35].
Table 3.2 - Design Speeds and Load Factors for the Meridian Parameter Value Units
Vs 61 kts
VC 133 kts
VD 186 kts
VA 118 kts
VS,neg 66 kts
nlimit (+) 3.8 ~ n (-) limit -1.5 ~
16 4
Positive g Limit = 3.8
3
+VC Gust Line
+VD Gust Line 2
1
VC VD 0 Load Factor, n Factor, Load VS VA
-1 Negative g Limit = -1.5
-VD Gust Line
-2
-VC Gust Line
-3 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 Speed, KEAS
Figure 3.1 - V-n Diagram for the Meridian
17 3.2 Component Weight Estimations
Four methods were used for estimating aircraft component weights: the Cessna,
Torenbeek, General Dynamics, and USAF methods. These methods are integrated
into the AAA software, which was used for the weight estimations [7].
These methods are designed for conventional, inhabited aircraft, which the
Meridian is not. For this reason a certain amount of designer intuition was employed to select the most applicable methods for each component. For example, the Cessna method produced a wing weight of approximately 300 lbs, while the USAF and
Torenbeek methods resulted in weights of approximately 100 lbs. The latter results were deemed to be reasonable, therefore the Cessna method was not used for the wing weight estimation. Table 3.3 shows the component weights as well as the methods used. The data shown in Table 3.3 are the result of several iterations.
18 Table 3.3 - Class II Weight and Balance for the Meridian
Method Class II Weight XCG YCG ZCG lbs Structure Wing USAF, Torenbeek 100.0 101.6 0.0 40.0 Empennage Cessna, USAF 22.5 224.0 0.0 50.0 Fuselage Cessna, USAF 59.3 121.2 0.0 50.0 Landing Gear Main Gear Cessna 51.5 96.0 0.0 38.0 Tail Wheel Cessna 9.0 220.0 0.0 48.0 Main Gear - Retracted Cessna 51.5 101.0 0.0 42.0 Tail Wheel - Retracted Cessna 9.0 225.0 0.0 48.0 Propulsion Propeller Torenbeek/GD 34.8 50 0 50 Engine Manufacturer 188.0 64.0 -0.8 48.0 Fuel System USAF, Torenbeek 31.7 102.0 0.0 40.0 Engine Systems Torenbeek/GD 43.6 70.0 0.0 0.0 Fixed Equipment Flight Control System Cessna, Torenbeek 22.4 120.0 0.0 50.0 Avionics/Electronics Class I/Manufacturer 11.2 120.0 0.0 50.0 Electrical System Cessna, Torenbeek 24.4 118.0 0.0 50.0 Icing System USAF, Torenbeek 15.6 120.0 0.0 50.0 Paint Torenbeek 3.6 120.0 0.0 50.0 Fuel and Payload Mission Fuel 239.9 107.0 0.0 50.0 Fuel Reserves 54.0 107.0 0.0 50.0 Trapped Fuel and Oil 5.4 107.0 0.0 50.0 Payload 165.0 104.0 0.0 50.0 Totals Structure - Gear Extended 242.3 120.97 0.00 43.25 Structure - Gear Retracted 242.3 122.22 0.00 44.10 Powerplant 298.1 67.28 -0.50 40.36 Fixed Equipment 77.2 119.37 0.00 50.00 Empty Weight 617.6 94.86 -0.13 42.34 Useful Load 464.3 105.93 0.00 50.00 Total - Gear Extended 1081.9 97.88 -0.13 44.56 Total - Gear Retracted 1081.9 98.06 -0.13 44.69
The c.g. locations of each component are shown in Figure 3.3. The c.g. travel due to fuel and payload loading is shown in Table 3.4 and Figure 3.2.
Table 3.4 - Weight and Balance Summary for the Meridian Parameter Inches % mgc Most Forward c.g. 94.86 0.18 Most Aft c.g. 99.89 0.34 Total Excursion 5.03 0.16 Fuel Excursion 2.76 0.09
19 Wing Chord, %mgc 0.16 0.17 0.18 0.19 0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28 0.29 0.3 0.31 0.32 0.33 0.34 0.35 1200.0
Retract Gear
1100.0
WTO
1000.0
+ Fuel
900.0 - Fuel
Weight, lbs Extend Gear
800.0 Trapped Fuel and Oil
+/- Payload 700.0
WOE
WE 600.0 94 95 96 97 98 99 100 Fuselage Station, in
Figure 3.2 - Center of Gravity Excursion for the Meridian
20
Figure 3.3 – Component C.G. Locations
21 3.3 Class II Stability and Control
The purpose of this section is to describe the Class II stability and control analyses performed for the Meridian UAV. These include:
• Trim Diagram (Power on and power off)
• Roll Performance
• Crosswind Control During Final Approach and While on the Runway
• Open Loop Dynamic Handling
• Actuator Size and Rate Requirements
3.3.1 Trim Diagrams
Trim diagrams were created for the flight conditions listed in Table 3.5. The trim diagrams shown in Figure 3.4 through Figure 3.11 were created using the AAA software [7]. The V-tail incidence was adjusted to ivee = -3.5 deg so that the aircraft could be trimmed at the stall speed with the most forward center of gravity.
Table 3.5 - Meridian Flight Conditions Flight Condition Altitude Speed Weight Flaps Gear ft kts lbs deg ~ Clean 5,000 120 963 0 Up Takeoff Gear Down 0 60 1,082 0 Down Takeoff Gear Up 0 60 1,082 0 Up Landing Heavy, Gear Down 0 65 1,082 30 Down Landing Heavy, Gear Up 0 65 1,082 30 Up Landing Light, Gear Down 0 65 843 30 Down Landing Light, Gear Up 0 65 843 30 Up OEI 5,000 80 963 0 Up
22
Figure 3.4 - Trim Diagram - Cruise
23
Figure 3.5 - Trim Diagram - Takeoff, Gear Down
24
Figure 3.6 - Trim Diagram - Takeoff, Gear Up
25
Figure 3.7 - Trim Diagram – Landing Heavy, Gear Down
26
Figure 3.8 - Trim Diagram - Landing Heavy, Gear Up
27
Figure 3.9 - Trim Diagram - Landing Light, Gear Down
28
Figure 3.10 - Trim Diagram - Landing Light, Gear Up
29
Figure 3.11 - Trim Diagram - OEI
30 The trim diagrams shown in Figure 3.4 through Figure 3.11 show that the aircraft can be trimmed throughout the entire flight envelope requiring no more than 20 degrees of control surface deflection.
3.3.2 Open Loop Dynamics
The open loop dynamics were calculated for the Meridian using the AAA program
[7]. The longitudinal and lateral-directional dynamics and flying qualities were calculated for the takeoff, cruise, and approach flight conditions as specified in Table
3.6. These were compared to the flying quality requirements specified in MIL-F-
8785C [6] and MIL-STD-1797A [6] for a Class I aircraft. While it is not necessary for a UAV to meet the military specifications, it is common design practice to use the military flying quality requirements as a basis for the dynamic analysis.
Table 3.6 - Dynamic Analysis Flight Conditions Parameter Units Flight Condition Takeoff Cruise Approach Altitude ft 0.00 5000.00 0.00 ΔT deg F 0.00 -40.00 0.00
U1 kts 60.00 120.00 65.00 W lbs 1083.4 962.9 843.2 α deg 13.28 0.63 6.82 CL1 ~ 1.15 0.30 0.84 n g 1.00 1.00 1.00
δF deg 0.00 0.00 40.00
Xcg in 99.97 99.21 98.25
Zcg in 45.66 45.12 44.42
εvee deg 0.96 0.41 1.28
ηvee, p. off ~ 1.00 1.00 1.00 η vee ~ 1.96 1.15 1.00
31 The stability and control derivatives for Meridian are shown in Table 3.7 and Table
3.8, respectively. These values represent the result of several iterations of control surface sizing.
32 Table 3.7 - Stability Deriviatives for the Meridian
Parameter Units Flight Condition Takeoff Cruise Approach
CTx ~ 0.54 0.08 0.05
CMT ~ -0.072 -0.011 -0.007
CDu ~00 0
CLu ~ 0.010 0.012 0.008
CMu ~ 0.002 0.003 0.002
CTXu ~ -1.61 -0.23 -0.135027
CMTu ~ 0.22 0.03 0.0237792 -1 CDα rad 0.39 0.10 0.29 -1 CLα rad 4.01 4.06 4.01 -1 CMα rad -0.42 -0.52 -0.64 -1 CMTα rad -0.46 -0.31 -0.07 -1 CDα rad 00 0 -1 CLα rad 0.53 0.55 0.54 -1 CMα rad -2.09 -2.17 -2.16 -1 CDq rad 00 0 -1 CLq rad 3.58 3.82 4.03 -1 CMq rad -9.63 -9.85 -9.94 -1 CYβ rad -0.42 -0.42 -0.42 -1 Clβ rad -0.13 -0.09 -0.12 -1 Cnβ rad 0.11 0.11 0.11 -1 CnTβ rad -0.001 -0.001 -0.001 -1 CYβ rad 0.0125 0.0003 0.0107 -1 Clβ rad -0.0005 0.0000 0.0000 -1 Cnβ rad 0.0051 0.0001 0.0044 -1 CYP rad -0.06 -0.12 -0.09 -1 ClP rad -0.46 -0.46 -0.46 -1 CnP rad -0.16 -0.04 -0.10 -1 CYr rad 0.27 0.27 0.27 -1 Clr rad 0.31 0.10 0.20 -1 Cnr rad -0.13 -0.11 -0.12 -1 CDivee rad 0.01 0.01 0.01 -1 CLivee rad 0.30 0.30 0.30 -1 C rad mivee -1.17 -1.19 -1.19
33 Table 3.8 - Control Derivatives for the Meridian
Parameter Units Flight Condition Takeoff Cruise Approach -1 CDδrv rad 0.003 0.004 0.006 -1 CLδrv0 rad 0.14 0.14 0.14 -1 CLδrv rad 0.05 0.14 0.14 -1 CMδrv0 rad -0.56 -0.56 -0.56 -1 CMδrv rad -0.18 -0.56 -0.56 -1 Chβrv rad -0.03 -0.07 -0.03 -1 Chδrv rad -0.28 -0.35 -0.29 -1 CYδa rad 00 0 -1 Clδa rad 0.12 0.13 0.12 -1 Cnδa rad -0.03 -0.01 -0.02 -1 Chαa rad 0.21 0.18 0.21 -1 Chδa rad 0.04 -0.03 0.03
CL0 ~ 0.26 0.26 0.26
CL0 ~ 0.26 0.26 0.39
CM0 ~ 0.002 -0.003 -0.021
C wf m0 ~ -0.05 -0.05 -0.08
The stability and control derivatives shown in Table 3.7 and Table 3.8 were used to
calculate the open loop transfer functions for the Meridian. This was done with the
AAA software [7]. The transfer functions for the Cruise condition are shown in
Table 3.9 through Table 3.11. The dynamic stability parameters related to the aircraft
flying qualities are shown in Table 3.12. The Meridian met Level I flying quality requirements for all flight conditions.
34 Table 3.9 - Longitudinal Transfer Functions for Cruise
Table 3.10 - Lateral Transfer Functions for Cruise
35 Table 3.11 - Directional Transfer Functions for Cruise
Table 3.12 - Meridian Dynamic Stability Parameters Flight Condition Parameter Units Takeoff Cruise Approach Longitudinal
ωsp rad/s 2.9 4.63 2.31
ζsp ~ 0.57 0.45 0.59
ωp rad/s 0.34 0.25 0.2
ζp ~ 0.23 0.1 0.09 n/α g/rad 3.2 12.1 5.8 2 CAP 1/gsec 2.63 1.77 0.92 Lateral-Directional
TCSpiral sec -25.6 106.5 -650.7
TCroll sec 0.3 0.16 0.27
ωD rad/s 2.2 2.9 2.1
ζD ~ 0.12 0.1 0.09 Note: Level I flying qualities met for all flight conditions.
Roll Control Effectiveness The roll control effectiveness is a vital parameter for aircraft controllability, especially during approach and landing. The roll control requirements for a Class I
36 aircraft are specified in Table 3.13. The requirement states that the aircraft must be able to achieve the specified bank angle in the specified amount of time.
Table 3.13 - Roll Control Requirements - Time to Achieve Bank Angle (Seconds)
Cat A Cat B Cat C
Level φt = 60 deg φt = 60 deg φt = 30 deg I 1.3 1.7 1.3 II 1.7 2.5 1.8 III 2.6 3.4 2.6
The roll control analysis results are shown in Table 3.14. These values were calculated using AAA [7]. As can be seen, the Meridian meets Level I flying qualities for all flight conditions.
Table 3.14 - Roll Control Results
Takeoff Cruise Approach Cat. C B C
φt 30 60 30 t r 1.25 1.1 1.2
3.3.3 Actuator Size and Rate Requirements
The actuators were sized using the AAA program to calculate the hinge moment derivatives. The actuators were then sized for a factor of safety of 2.0 at maximum deflection. The most critical actuator size was determined to be the flaps, which required a servo with a maximum torque of at least 100 in-lbs. The servo selected is the Model 820 manufactured by Moog Components Group (www.polysci.com). This servo has a peak torque of 150 in-lbs and accepts a PWM signal, which is compatible with the selected autopilot.
37 3.4 Class II Aerodynamics
The Class II drag analysis was performed for the Meridian with the AAA software
[7]. The geometries used in the drag calculations are shown in Table 3.15 through
Table 3.18. The drag analysis was performed for the takeoff, cruise, approach, and
OEI flight conditions with and without antennas. These flight conditions are
described in Table 3.5.
Table 3.15 Wing Geometry for Drag Calculations Parameter Units Value 2 S ft 69.6 AR ~ 10 λ ~1
Λc/4 deg 0
(t/c)r %18
(t/c)t %18 LER/c % 1.5 L' ~ 2 xlam/c %10 -1 Clα rad 4.3
fgap ~ 0.97 -3 k 10 ft sand 0.00167
Table 3.16 - V-Tail Geometry for Drag Calculations Parameter Units Value S ft2 6.1 AR ~ 4 λ ~0.5
Λc/4 deg 26.3
(t/c)r %12
(t/c)t %12 LER/c % 1.58 L' ~ 2
xlam/c %10 -1 Clα rad 6.25
fgap ~ 0.96 -3 k 10 ft sand 0.00167
38 Table 3.17 - Fuselage Geometry for Drag Calculations Parameter Units Value 2 Sb ft 0.01 2 Swet ft 75.1 Lft14.8 2 Sfrontal ft 3
xlam/L %0 2 Swet-lam ft 0 2 Splf ft 23.9
Df max ft 24 -3 k 10 ft sand 0.00167
Table 3.18 - Flap and Landing Gear Geometry for Landing Gear Calculations Parameter Units Value Flaps Flap Type ~ Plain
ηi f %27
ηo f %57
cf/cw %25 Main Gear 2 Sft ft 0.25
Lstrut ft 2.5 Tailwheel
CDref ~0.5 2 Sref ft 0.02 F D ~0
39 Table 3.19 - Drag Analysis Results Component Takeoff Cruise Approach OEI Zero-Lift Drag Wing 0.0124 0.0107 0.0122 0.0117 Vee Tail 0.0021 0.0018 0.0021 0.002 Fuselage 0.0018 0.0034 0.0018 0.0041 Flap 0 0 0.0188 0 Retract 0.0108 0 0.0108 0 Fixed Gear 0.0001 0.0001 0.0001 0.0001 Trim 0.0018 0.0005 0.0015 0.0004 Propeller 0 0 0 0.0267 Inlet 0.0001 0.0001 0.0001 0.0001 Nozzle 0.0015 0.0015 0.0021 0.0015 Power 0.0064 0.0004 0.0013 0 Gear Pod 0.0014 0.0012 0.0014 0.0013 Total Zero-Lift 0.0384 0.0197 0.0522 0.0479 Drag Due to Lift Wing 0.0548 0.0034 0.0254 0.0142 Vee Tail 0.0009 0 0 0 Fuselage 0.0051 0 0.0006 0 Total Due to Lift 0.0608 0.0034 0.026 0.0142 Total Drag 0.0992 0.0231 0.0782 0.0621
Antenna Drag (6) 0.0042 0.0037 0.0042 0.0042 Total Drag w/ Ant. 0.1034 0.0268 0.0824 0.0663
The drag results are shown in Table 3.19 and Figure 3.12 through Figure 3.15. The mid-cruise lift-to-drag ratio was found to be 14.5 as indicated on Figure 3.13. This is slightly less than the value of 16.0 estimated in the Class I design. However, the performance analysis results show that the range requirement is still met in the current configuration so no further iteration is necessary.
40 2 Takeoff Cruise 1.8 Approach OEI
1.6
1.4 Reference Data: S = 68.6 ft2
L 1.2 AR = 10.0 Takeoff: W = 1,083 lbs 1 e = 0.84 f = 2.84 ft2 Cruise: W = 963 lbs
Lift Coefficient, C 0.8 e = 0.83 f = 1.4 ft2 0.6 Approach: W = 843 e = 0.75 2 0.4 f = 3.7 ft OEI: W = 963 e = 0.81 0.2 f = 3.8 ft2
0 0 0.05 0.1 0.15 0.2 0.25
Drag Coefficient, CD
Figure 3.12 - Drag Polars for the Meridian without Antennas
41 2 Takeoff Cruise 1.8 Approach OEI Reference Data: 1.6 S = 68.6 ft2 AR = 10.0 Takeoff: 1.4 W = 1,083 lbs e = 0.84 f = 2.84 ft2
L 1.2 Cruise: W = 963 lbs e = 0.83 1 f = 1.4 ft2 Approach: W = 843 e = 0.75
Lift Coefficient, C Coefficient, Lift 0.8 f = 3.7 ft2 OEI: 0.6 W = 963 e = 0.81 f = 3.8 ft2 0.4
0.2
0 02468101214161820
Drag Coefficient, CD
Figure 3.13 - Lift-to-Drag for the Meridian without Antennas
42 2 Takeoff Cruise 1.8 Approach OEI
1.6
1.4 Reference Data: S = 68.6 ft2 L 1.2 AR = 10.0 Takeoff: W = 1,083 lbs 1 e = 0.84 f = 3.13 ft2 Cruise:
Lift Coefficient, C Coefficient, Lift 0.8 W = 963 lbs e = 0.83 f = 1.65 ft2 0.6 Approach: W = 843 e = 0.75 0.4 f = 3.9 ft2 OEI: W = 963 0.2 e = 0.81 f = 4.13 ft2
0 0 0.05 0.1 0.15 0.2 0.25
Drag Coefficient, CD
Figure 3.14 - Drag Polars for the Meridian with Antennas
43 2 Takeoff Cruise 1.8 Approach OEI Reference Data: 1.6 S = 68.6 ft2 AR = 10.0 Takeoff: 1.4 W = 1,083 lbs e = 0.84 f = 3.13 ft2
L 1.2 Cruise: W = 963 lbs e = 0.83 1 f = 1.65 ft2 Approach: W = 843 e = 0.75
Lift Coefficient, C 0.8 f = 3.9 ft2 OEI: 0.6 W = 963 e = 0.81 f = 4.13 ft2 0.4
0.2
0 0 5 10 15 20 25
Drag Coefficient, CD
Figure 3.15 - Lift-to-Drag for the Meridian with Antennas
44 To verify the validity of the drag analysis the Oswald’s efficiency and parasite area were calculated for each flight condition using Equation 3.1 and Equation 3.2 respectively. These values are shown on the drag polar plots in Figure 3.12 through
Figure 3.15 and Table 3.20. The cruise values of the parasite area with and without the antennas were plotted against known aircraft in Figure 3.16 from [6] for further verification. As can be seen, Meridian falls somewhere between the lines for overall skin friction coefficient values of Cf = 0.005 to 0.006 depending on whether the antennas are installed or not. This is a good indication that the Class II drag results are reasonable.
1 Equation 3.1 e = δCD 2 eA δCL
f = C S Equation 3.2 D0 w
Table 3.20 - Resultant Oswald's Efficiency and Parasite Area for the Meridian
No Antennas With Antennas 2 2 Swet = 240 ft Swet = 275 ft Flight Condition e f f 2 2 ~ ft ft Takeoff 0.84 2.84 3.13 Cruise 0.82 1.39 1.65 Approach 0.75 3.68 3.97 OEI 0.81 3.84 4.13
45
Figure 3.16 - Relationship of Parasite Area and Wetted Area for Various Single Engine Aircraft [6]
3.5 Propulsion
The installed thrust of the Innodyn engine was calculated using the AAA program
[7]. The Innodyn is rated at 165 SHP. The extracted power is estimated at 5 hp based on the electrical power generation required. The total installed power is 125 hp, which is above what is required for takeoff and climb performance.
The AAA program was also used to calculate an estimate for the inlet area. This resulted in an inlet with an area of 0.2 ft2.
46 3.6 Performance Analysis
The purpose of this section is to describe the performance requirements imposed on this aircraft design and to verify that these requirements have been met by the
Meridian. This includes:
• Stall speed
• Takeoff Distance
• Cruise Performance
• Landing Distance
3.6.1 Stall Speed
The stall speed of the Meridian was calculated for heavy and light flight conditions with zero and full flaps as shown in Table 3.21. The maximum trimmed lift coefficients determined in Section 3.3.1 were used for the clean and full flap configurations. Power effects were ignored for the stall speed calculations.
Table 3.21 - Stall Speed Summary Clean Full Flaps Parameter Units Light Heavy Light Heavy Weight lbs 843 1,082 843 1,082 Flaps deg 0 0 30 30
CLmax ~ 1.42 1.42 1.70 1.70 Altitude ft 0 0 0 0 Vs ft/s 50 57 46 52
3.6.2 Takeoff Distance
The takeoff distance for the Meridian was calculated for conventional tires as well as ski operations using the AAA program [7]. This process uses methods found in [6]
47 to calculate the takeoff distance to clear an obstacle of a specified height. The following assumptions were used in the takeoff distance calculations:
• Ground Friction Coefficient (μG = 0. 02 for Tires, μG = 0. 15 for skis)
• The obstacle height is 50 ft
• Weight = 1,082 lbs
• Standard sea level conditions
• Drag is based on Class II drag analysis for Takeoff condition:
CD 0 = 0.041
• V3/VTO = 1.3 (Based on FAR 23)
The takeoff analysis resulted in the following takeoff distances:
• Standard Tires on Asphalt: STO = 415 ft
• Skis on Snow: STO = 635 ft
The takeoff distances with and without skis both exceed the required distance of
1,500 ft by a large amount. This is due to the fact that the selected engine has more power than required by performance matching.
3.6.3 Climb
3.6.4 Cruise Performance
The cruise performance calculations for the Meridian were performed with the
AAA program [7]. This consisted of estimating the range and endurance assuming constant speed cruise.
48 Range
The range of the Meridian was calculated using the constant speed range equation found in [6]. The lift-to-drag value was calculated using the mid-cruise Class II drag polar (Section 3.4). The following assumptions were used for the range calculation:
• Wbegin = 1,083 lbs
• Wfuel = 240 lbs, Wfuel, res = 55
• ηp = 0.80
• cp = 0.90 lbs/hp-hr
The range of the Meridian was determined to be:
• Without Antennas
940 nm (1,735 km) without reserves
1,200 nm (2,220 km) with fuel reserves
• With 6 Antennas
830 nm (1,530 km) without fuel reserves
1,030 nm (1,900 km) with fuel reserves
The range for the Meridian without antennas is acceptably close to the required range of 1,750 km without antennas and the range with antennas exceeds the required
1,500 km.
Endurance
The endurance of the Meridian was calculated using Equation 3.3. The following assumptions were made for the endurance calculation:
• Constant speed cruise
49 • Wbegin = 1,083 lbs
• Wfuel = 240 lbs, Wfuel, res = 55 lbs
• ηp = 0.80
• cp = 1.2 lbs/hp-hr (From manufacturer data)
• U1 = 80 kts
• Drag based on mid cruise Class II drag polar
⎡ ⎤ Equation 3.3 ⎛ 550 ⎞ η p ⎛ L ⎞ ⎛ Wbegin ⎞ E = 60⎜ ⎟⎢ ⎜ ⎟ln⎜ ⎟⎥ 1.688 c U D ⎜W −W ⎟ ⎝ ⎠⎣⎢ p 1 ⎝ ⎠ ⎝ begin fuel ⎠⎦⎥
The loiter speed was set at 80 kts as this is the speed for maximum L/D. The results from the endurance calculations are:
• Without Antennas
12.7 hours without reserves
16.1 hours with fuel reserves
• With 6 Antennas
11.5 hours without fuel reserves
14.8 hours with fuel reserves
The Meridian exceeds the specified endurance requirements with and without the antennas and fuel reserves.
3.6.5 Landing Distance
The landing distance for the Meridian was calculated with the AAA program [7].
The landing distance includes the distance from a 50 ft obstacle to the ground and the
50 distance from touchdown to a full stop. The following assumptions were used for the landing gear calculation:
• Weight = 1,082 lbs
• CLmax = 1.7 (Based on maximum trimmed lift coefficient)
• Drag based on Class II drag polar for the Approach flight condition
• Average ground acceleration = 0.45 g
• Δn = 0.10 (Correction factor due to pilot technique)
The results of the landing distance calculation are:
• Sair = 1,110 ft (Distance in air from obstacle to ground)
• SLG = 430 ft (Ground run distance)
• SL = 1,540 ft (Total distance)
The total landing distance is acceptably close to the required landing distance of
1,500 ft. This distance is based on conventional tires, which was determined to be the critical requirement as skis actually have a higher coefficient of friction than wheels on asphalt.
3.7 Systems
The purpose of this section is to describe the systems both on and off the Meridian that are required for operation. These include:
• Flight Controls
• Electrical System
• Communications
51 • Fuel
• Anti-Icing
3.7.1 Flight Control System
The Meridian will utilize a fly-by-wire control system based around the Piccolo autopilot, which is produced by Cloud Cap Technologies [36]. The Piccolo requires dynamic and static pressure inputs and electrical power. The Piccolo interfaces with the servo actuators using a Pulse Width Modulated (PWM) signal, which is standard for remote control aircraft. The architecture for the Piccolo is shown in Figure 3.18.
Figure 3.17 - Cloud Cap Tech. Piccolo II Autopilot [36]
52
Figure 3.18 - Piccolo II Architecture [36] The ground equipment associated with the Piccolo autopilot consists of a ground station, operator interface (PC), and a pilot control unit (Futaba Controller) as shown in Figure 3.19.
Figure 3.19 - Piccolo Ground Station and Pilot Controller (Operator Interface Not Shown) [36]
53 3.7.2 Electrical System
This section describes the electrical system layout for the Meridian including an electrical load profile for a typical mission. The Meridian will require both 12 and
24VDC power busses. The power system consists of:
• Electrical Generator
• Battery
• Electrical Bus
• Electrical Wiring
The first step in developing the electrical system layout was to generate an electrical load profile for the Meridian. This was done by listing all necessary systems required during each phase of a given flight. The results of the load profile are shown in Figure 3.20. The total load was estimated assuming the radar system is turned on at takeoff, while the essential load assumes the radar system only requires power during the on-station flight phase. The most critical flight phases are the takeoff and landing segments as the landing gear and flap actuators will be operated in addition to the other systems. The emergency flight phase is representative of an engine flame-out situation. The battery was sized such that all necessary systems could remain operating will the aircraft descends and attempts tot restart the engine.
This however, will require the ability to turn some systems off autonomously.
The current configuration of the Innodyn engine is with one 600W, 12 V generator and a separate starter. The current generator is a standard off-the-shelf automotive
54 alternator, and can be replaced with a larger generator for the Meridian. The electrical load profile indicates that a 1,000 W generator would be sufficient.
The electrical system layout is shown in Figure 3.21. The wiring is not shown in
Figure 3.21 for clarity. The aileron and flap servo and landing gear actuator wiring will be located just aft of the aft spar. The antenna wiring will be located just behind the forward spar. A more detailed view of the systems located in the fuselage are shown in Figure 3.22.
55 15 Min 5 Min 5 Min 5 Min 20 Min 105 Min 540 Min 105 Min 15 Min 5 Min 10 Min 1000 Total Load Essential Load 900
800
700
600
500
400 Electrical Power,Electrical W
300
200
100
0
g t xi ff n d y a I T ent nc adin Star keo sc Lan o a Climb se Out uise ge L T i Cr De er Cru On Station Em Flight Segment
Figure 3.20 - Electrical Load Profile for the Meridian UAV
56
Figure 3.21 - Electrical System Layout
57
Figure 3.22 - Fuselage Systems Layout
3.7.3 Communications/Telemetry System
The Meridian will utilize dual line-of-site communication links: the piccolo communications will be used for command and control and a secondary communications link will be used for vehicle health monitoring/telemetry. For beyond line-of-site (BLOS) communications, an Iridium satellite communication link will be utilized. The Piccolo autopilot is configured to transmit and receive data over an Iridium link. This communications link will be used for low-bandwidth health monitoring and limited control.
58 3.7.4 Fuel System
The fuel tank integration was difficult for the Meridian due to the removable wing design. Several options were investigated including a hinged wing joint such that the wing pivots rearward but is not removed. This would theoretically allow for fuel to be placed in the outboard wing section, but this type of fuel system would have an extremely high probability of leaking. Another option is to use quick fuel line connectors at the wing split. Again, this type of integration poses serious leaking problems. The design was iterated such that the fuel could be stored inboard of the wing split. This involved increasing the wing thickness to an 18 percent thick airfoil and adding a tank in the fuselage. The latter decision required the fuselage height to grow.
The required fuel volume is 43.7 gallons or 5.84 ft3. Approximately 45 gallons of fuel fits in the fuel tanks in the inboard wing and fuselage sections as shown in Figure
3.23. Fuel bladders will be utilized for the wing and fuselage tanks. These bladders are commercially available and include all of the pickups, lines, and baffling as required. The fuel tank will be split into 3 separate bladders in the wing (1 center, and two outboard of the inboard rib), and 1 bladder in the fuselage. The center wing bladder will serve as the fuel collection point.
59 Fuel Storage
Figure 3.23 - Fuel Tank Integration
3.7.5 Anti-Icing System
A combination of muffed engine exhaust and electrically heated elements will be utilized for the anti-icing system. The location of the wing is such that the leading edge of the wing is forward of the firewall. Similar engine installations have been performed (www.innodyn.com) utilizing NACA inlets to pressurize the engine cowl volume. This air will be pushed through a valve into the leading edge of the wing.
The temperature of the muffed exhaust air has been measured at 180oF. Much attention will be given to the thermal effects on material properties and stress states in the detail design and analysis phases.
60 3.8 Class II Landing Gear
This section discusses the design of landing gear in terms of stroke length, tire diameter, and strut diameter. The landing gear must be retractable so as not to interfere with the radar. More importantly, the landing gear must be retractable with skis or conventional wheels as the Meridian will be operated from snow and paved runways. The Red, White, and Blue designs all incorporated tricycle type landing gear that retracted on a tilted pivot into the fuselage. While this is feasible for conventional tires, this retraction scheme does not work with skis. The tricycle gear had several other design problems:
• The nose gear had to be mounted far enough from the propeller to
leave room for the nose ski. This required a very wide gear to meet
lateral tipover.
• The Meridian should be able to be shipped in a 20 foot container,
which is approximately 90 inches wide. Lateral tipover requirements
called for a wider gear than this, so the gear would have to be removed
for shipping.
• There were no commercially available landing gear similar to the
previous design.
All of these problems lead to the development of a new landing gear integration scheme. The gear disposition was changed to a tail dragger to solve the nose ski integration and lateral tip-over problems. The landing gear were then moved to pods mounted to the wing. This had two benefits:
61 • The landing gear could be purchased commercially
• The landing gear retract straight aft, which allows for ski retraction
The decision to put the landing gear on the wing calls for the use of either an oleo, pneumatic, or rubber damped type strut. The oleo gear was chosen as this type of gear is commercially available. The landing gear selection will be discussed further in Section 3.8.3.
The following assumptions were made for the landing gear design:
• The main gear shall be able to sustain 100% of the static load. (This is
due to the tail dragger configuration.)
• The gear will be sized for a maximum touchdown rate of 10 ft/s.
• The stroke length will be sized such that a 10 ft/s decent rate imparts
1g on the airframe.
• The strut will have an energy absorption efficiency of 80%.
• The strut will be sized for skis, thereby setting the tire deflection to
zero.
3.8.1 Tire Selection
The main gear tires will be 3.00 x 4 tires. These tires have an outer diameter of 10.0 inches, a width of 3.2 inches, a maximum pressure of 50 psi, and weigh 3.5 lbs each.
The tail wheel tire will be a 6.0 inch diameter solid rubber tire, which weighs 4.75 lbs. As will be discussed in Section 3.8.3, the main and tail gears are commercially available parts currently used on homebuilt aircraft.
62 3.8.2 Strut Sizing
The strut stroke length and diameter sizing were performed using the methods described in [6]. The stroke length is calculated by determining the touchdown energy using Equation 3.4. The gear absorption energy equation is then used to determine the appropriate stroke length using Equation 3.5. The results are shown in
Table 3.22.
2 Equation 3.4 1 wt ET = WL 2 g
ET Equation 3.5 −ηt st ns Pm N g Ss = ηs
Where:
• Ss = Strut stroke length
• Et = Touchdown energy
• ns = Number of struts
• Pm = Max static load per gear
• Ng = Ratio of max load to static load
• ηt = Tire energy absorption efficiency
• st = Tire deflection
• ηs = Strut energy absorption efficiency
63 Table 3.22 - Landing Gear Strut Sizing Parameter Units Value
WL lbs 1082
wt fps 10
ns ~2
Pm lbs 541
Ng g1
ηt ~0
st in 0
ηs ~0.6 Diameter in 0.1 Stroke in 2.6
3.8.3 Landing Gear Integration
The landing gear placement, integration, and sizing were iterated such that a commercially available landing gear could be integrated with the Meridian. This greatly increases the feasibility of manufacturing the Meridian in the time allotted as landing gear development is a fairly complicated process. The landing gear strut produced for the Lancair Legacy homebuilt aircraft [37] will be used for the main gear (Figure 3.24 and Figure 3.25).
Figure 3.24 - Lancair Legacy Landing Gear Strut [37]
64
Figure 3.25 - Lancair Legacy Landing Gear Installation [38] The tail wheel will also be purchased commercially. The tail wheel assembly is manufactured by Matco [39] and is commonly used on homebuilt aircraft.
Figure 3.26 - Matco Tailwheel Assembly [39]
65 3.9 Structural Arrangement
The purpose of this section is to discuss the proposed structural arrangement for the
Meridian UAV. This includes material selection, structural layouts for the wing, v- tail, and fuselage, as well as preliminary structural sizing.
The mission of the Meridian is considered to be extreme in that the locations of operation, Greenland and Antarctica, are known for extremely cold weather. While this must be considered during the structural arrangement and material selection, it must be noted that the temperatures the Meridian will experience are not much different from High Altitude Long Endurance (HALE) UAVs. In fact, the Meridian mission may be less extreme than a HALE UAV because it will not experience large changes in temperature throughout a flight. This is mentioned only to emphasize the fact that the material selections should not be arbitrarily limited due to cold weather operations. Rather, the changes in material properties due to temperature should be acknowledged and accounted for in the design process such that the final product is an optimized solution in terms of weight, manufacturability, and service life.
One of the primary drivers of the material selection and structural layout is the advanced development time requirement. For the Meridian to be a successful project, manufacturability has to play a big role in the structural design process. In addition, many of the structures will be manufactured and assembled by graduate students with limited manufacturing experience. Therefore, the aircraft should be designed in such a way that limits the manufacturing skills and facilities required as is often done with homebuilt aircraft. These two concerns warrant the need for a limited part count as
66 well as a high level of automated processes such as computer numerically controlled machining.
3.9.1 Wing Structure
The structural arrangement of the wing for the Meridian was driven by the following concerns and requirements:
• Shipping requirement
• Hard point mounting requirement for antennas
• Fuel system integration
• Cost
• Manufacturability
• Weight
• On site storage facility limitations
The shipping, storage, and hard point requirements were determined to be the most critical and therefore had the biggest effect on the wing structural arrangement. The shipping and storage limitations [5] were such that the wing had to be designed in at least three pieces. In terms of structural optimization, the best solution would be to make the wing in two pieces. This however, does not consider the other system requirements and limitations such as landing gear and fuel tank integration nor does it consider manufacturability.
The structural layout of the wing was determined by integrating the landing gear placement, fuel tank sizing, control surface sizing, shipping requirements, and manufacturing limitations. The final solution is a three-piece wing: the inboard
67 section contains the fuel tanks and landing gear and the outboard sections contain all of the control surfaces. The outboard sections are removable for shipping and storage in small field hangars. In addition the length of the longest part in the wing is less than that of the current composite curing facilities at the University of Kansas (~10 ft).
Several possible arrangements were investigated for the wing structure including:
• Single spar
• Two Spar
• Three Spar
• Tube Spar
The single spar concept was eliminated as the control surfaces will require some sort of closeout mechanism. The three spar concept was eliminated based on preliminary structural sizing analyses. The two spar concept was selected as the primary configuration with the tube spar as a secondary option. The wing is designed with a rectangular forward spar and a c-channel rear spar. The spar of the outboard wing slides into the inboard spar and is held by fasteners on top and bottom. This allows the outer portion of the wing to be removable without adding a great deal of complexity of weight.
68
Figure 3.27 - Wing Structural Layout The material selection for the wing was influenced primary by manufacturability, load types, and thermal considerations. In terms of manufacturability, composite skins allow for a high level of automation in the tooling manufacturing and provide excellent surface finish. In terms of the substructure, there are several locations where loaded fasteners are required such as the landing gear attachment, wing joint, and antenna hard points. This warrants the use of aluminum in several of the structural components such as the forward and aft spar as well as several of the ribs.
The combination of different materials in the wing has implications in terms of thermal expansion. These will be investigated further in the detailed design and analysis of the structure.
69 3.9.2 Fuselage Structural Layout
The structural layout of the fuselage was driven by the following:
• Wing-Fuselage Integration
• Manufacturability
• Weight
• Engine Installation
• Payload Integration
• Accessibility Requirements
The structural layout of the fuselage was integrated with the configuration design in terms of wing and payload placement. The wing placement, which is driven by the aircraft center of gravity was iterated until the main spar of the wing was collocated with the firewall. To produce structurally efficient aircraft designs, this type of synergy must be implemented between the design aspects such that the amount of structural members required is decreased. By locating the landing gear, wing main spar, and fuselage firewall at the same fuselage station, the amount of heavy structural members has been decreased, which provides weight savings and improves the manufacturability of the vehicle.
The primary frames in the fuselage were placed at the locations of the wing spars, payload hatch closure, payload rack mount, fuselage split, and v-tail spars. The remainder of the fuselage frames were spaced according to preliminary buckling calculations. The upper longerons were placed in line with the top engine mounting bracket as well as the payload hatch opening. The lower longerons were located at
70 the upper surface of the wing and coincide with the lower engine mounting brackets.
Two frames were located at the forward and aft v-tail spar locations. These frames were also used to mount the tailwheel assembly.
Figure 3.28 - Fuselage Structural Layout
71
Figure 3.29 - Wing-Fuselage Attachment The aircraft structure was iterated several times such that the fuselage and center wing section would fit in a standard 20 foot container [9]. The goal was to minimize the amount assembly that would have to be performed on-site. This is important for shipping, but also for on-site storage. The projected hangar size is approximately 15 feet wide, which means the wings must be removed after every flight. The aircraft is shown in a 20 foot container in Figure 3.30.
72
Figure 3.30 - Standard 20 Foot Shipping Container Door [9] The engine mount will be procured from the engine manufacturer and will be very similar to the mount shown in Figure 3.31.
73
Figure 3.31 - Typical Engine Mount for the Innodyn 165TE
3.10 Manufacturing Breakdown
The aircraft skin is divided into 6 pieces: 2 for the cowl, left and right sections for the forward fuselage, and top and bottom sections for the aft fuselage sections.
Again, manufacturability was a primary driver in the fuselage design as the leading edge of the v-tail was placed such that it coincides with the mold line of the fuselage.
This allows for the aft fuselage and v-tail skin to be continuous, which improves structural rigidity and reduces the parts count.
74
Figure 3.32 - Manufacturing Breakdown
3.11 Cost Analysis
The cost estimate for the Meridian UAV was created using the AAA software [7] and the methods described in [6]. The vehicle cost is broken down into research, development, test, and evaluation or RDT&E costs; and acquisition cost. The methods of [6] are typically used for production type aircraft that will be sold for some profit. This Meridian is strictly a research aircraft developed for a specific mission. The marketability of the Meridian, while may be exploited at a later date, is not part of this cost estimate. For this reason, the cost estimates of [6] were augmented with quotes from vendors for items such as avionics, tooling, and engines.
75 The first step in the cost estimation is to determine the Aeronautical Manufacturer’s
Planning Report (AMPR) weight. This is defined as the vehicle empty weight less all of the items that will be purchased from vendors such as the engine, actuators, avionics, wheels, etc. The AMPR weight of the Meridian was estimated to be:
• WAMPR = 260 lbs
The next step in the cost estimation was to develop the hourly rates to apply to each cost estimate. This project is different from a typical aircraft manufacturing process as much of the work will be performed by students, whom work at much lower rates than typical industry standards. Average rates for manufacturing and engineering time were developed based on the rates for undergraduate students, graduate students, professors, and industry labor as shown in Table 3.23. The expected breakdown of time is also shown in Table 3.23, which was used to create a time-weighted average.
The industry rate was included in the wage calculations as some of the part manufacturing will be outsourced. The rates shown in Table 3.23 include typical overhead rates.
Table 3.23 - Engineering and Manufacturing Rate Estimation
Engineering Labor Manufacturing Labor % of Total Time Hourly Rate % of Total Time Hourly Rate % $/hr (2006) % $/hr (2006) Undergraduate 15 $16.00 30 $16.00 Graduate 60 $24.00 60 $24.00 Professor 15 $96.00 0 $96.00 Industry 10 $60.00 10 $60.00 Total (Averged) $37.20 $25.20
76 3.11.1 Research, Development, Test, and Evaluation Costs
The total RDT&E cost for an aircraft is defined as the sum of the airframe engineering and design cost; development, support and testing cost; flight test airplane cost; and flight test operations cost. This cost is then adjusted by factors accounting for test facilities, profit, and financing. This was not done for the
Meridian, however, as there will be no profit or financing, and the manufacturing facilities will be paid for by other university funding.
The following assumptions were made in the RDT&E cost estimation:
• The number of aircraft built during the RDT&E phase is 1.
• The workforce is assumed to be relatively skilled in Computer Aided
Design
• The engine cost was estimated at $30,000 per the manufacturer.
• The propeller cost was assumed to be $5,000 per the manufacturer.
• The avionics cost was estimated at $100,000 per the manufacturers.
• No profit or financing were included in the RDT&E phase.
The total cost for the RDT&E phase was determined to be $2.018 million. The cost breakdown is shown in Table 3.24 on page 79.
3.11.2 Acquisition Cost
The acquisition cost of an aircraft is defined as the sum of the manufacturing cost and the profit. As there will be no profit for the Hawkeye, this reduces to simply the manufacturing cost, which is comprised of the airframe engineering and design cost for the production phase; the airplane program production cost; and the production
77 flight test operations cost. The following assumptions were made for the acquisition cost estimation:
• The total number of aircraft produced for the production phase is 1.
• The manufacturing rate is assumed to be 0.1 aircraft.
• The interior costs were set as $0.
• 40 hours of flight testing at $500/hr with an overhead factor of 4.0
were assumed for the production vehicle.
• No profit or financing were included in the production cost estimate.
The total acquisition cost was estimated as $1.009 million. The cost breakdown for the acquisition phase is shown in Table 3.24.
3.11.3 Cost Estimate Summary
The RDT&E and acquisition cost estimates are summarized in Table 3.24. The total costs are broken down into overall categories in Table 3.25 and by RDT&E and production categories in Table 3.26.
78 Table 3.24 - RDT&E and Acquisition Cost Summary
Item Cost 10$ RDTE Engineering and Design 0.232 Development, Support, and Testing 0.087 RDTE Labor Costs 0.872 Material Costs 0.379 Avionics Equipment 0.1 Tooling 0.15 Quality Control 0.113 Engine 0.035 Flight Test Operations 0.05 2.018 Production Cost Airframe Engineering and Design 0.031 Labor 0.386 Production Materials 0.277 Production Avionics 0.15 Manufacturing Tooling 0 Manufacturing Quality Control 0.05 Engines 0.035 Flight Test Operations 0.08 1.009
Acquisition Cost (2 Vehicles) 3.027
Table 3.25 - Cost Breakdown by Overall Category
Item Cost 10$ Labor 1.608 Materials 0.656 Avionics 0.25 Tooling 0.15 Quality Control 0.163 Engines 0.07 Flight Test Operations 0.13 Total 3.027
79 Table 3.26 - Cost Breakdown by RDT&E and Production Categories Item Cost 10$ RDTE Engineering and Design 0.232 Development, Support, and Testing 0.087 Test Aircraft 1.649 Flight Test Operations 0.05 2.018 Production Cost Airframe Engineering and Design 0.031 Production Manufacturing 0.898 Flight Test Operations 0.08 1.009 Acquisition Cost (2 Vehicles) 3.027
Flight Test Operations 4% Engines 2% Quality Control 5%
Tooling 5%
Avionics 8%
Labor 54%
Materials 22%
Figure 3.33 - Cost Breakdown by Overall Category
80 3.11.4 Cost Estimate Justification
The cost estimate performed is based on conventional aircraft production costs.
The viability of using these methods for a UAV is questionable. Therefore, the estimated cost was tabulated against several current UAV systems for comparison as shown in Table 3.27. It is important to note that the cost of these systems is listed in terms of the vehicle costs and the system costs, which include ground support equipment and a certain number of vehicles. It is difficult to estimate the cost of one vehicle with ground support equipment, therefore the system cost was divided by the number of aircraft per system. This gives a more reasonable estimate as to the actual cost of functional UAV. The aircraft cost and the cost per aircraft based on system cost were plotted versus payload weight in Figure 3.34 and Figure 3.35 respectively.
Figure 3.34 shows that the estimate vehicle cost of the Meridian is almost exactly on the linear regression. This indicates that the vehicle cost estimate is reasonable. The cost per aircraft based on system cost plotted in Figure 3.35 shows how the Meridian is a more cost-effective system because the ground support equipment (ground station, charging system, etc) is already included in the vehicle cost estimate. (The ground station costs are included in the avionics cost estimates.)
Table 3.27 - Current UAV Procurement Cost [40]
System Aircraft Weight Payload Aircraft Cost System Cost Number Cost Per Aircraft lbs lbs FY06$ mil FY06$ mil Acft/System In System FY06$ mil Dragon Eye 4 1 0.03 0.14 3 0.05 RQ-7A Shadow 216 60 0.41 13.24 4 3.31 RQ-2B Pioneer 307 75 0.68 17.93 5 3.59 RQ-8B Fire Scout 1,765 600 4.27 22.83 4 5.71 RQ-5A Hunter 1,170 200 1.25 27.62 8 3.45 MQ-1B Predator 1,680 450 2.81 25.75 4 6.44 MQ-9A Predator 3,050 750 5.42 47.01 4 11.75 RQ-4 (Block 10) Global Hawk 9,200 1,950 19.81 60.15 1 60.15 RQ-4 (Block 20) Global Hawk 15,400 3,000 27.62 64.84 1 64.84 Meridian 1,082 165 1.51 3.03 2 1.51
81 100.00
Global Hawk Global Hawk
10.00 Meridian Predator Fire Scout Predator
Hunter 1.00 Cost(FY06$ 10^6) = (0.0089)WPL Pioneer Shadow
Cost (FY06$ 10^6) (FY06$ Cost 0.10
Dragon Eye
0.01
0.00 1 10 100 1,000 10,000 Payload Weight, lbs
Figure 3.34 - UAV Cost in Terms of Payload Weight
82 100.00 Global Hawk Global Hawk
Predator 10.00 Predator
Fire Scout Shadow Pioneer Hunter
Cost (FY06$ 10^6) = (0.0225)WPL 1.00
Meridian Cost (FY06$ 10^6) (FY06$ Cost
0.10 Dragon Eye
0.01 1 10 100 1,000 10,000 Payload Weight, lbs
Figure 3.35 - UAV Cost Based on System Cost Versus Payload Weight
83 4 Conclusions
This document summarizes the redesign of the Meridian UAV based on the response from the Preliminary Design Review. The antenna type has been changed from a bow-tie to a Vivaldi or exponential antenna. In addition, the aircraft has been modified to incorporate several commercially available off-the-shelf parts. The goal with the redesign of the Meridian is to produce a design that is not only novel, but is feasible considering the extremely short development time. This meant integrating manufacturability, performance, and operational constraints into the design process.
The product is a vehicle that can be shipped in a standard 20 foot container and quickly assembled and disassembled with minimal tools. The Meridian is the smallest turboprop powered UAV in the world. It is also one of the only UAVs with retractable ski landing gear. The purpose of this continued development of the
Meridian is to completely flush out the ‘best’ new UAV design based on the mission specification.
84 5 References
1 www.worldaerodata.com/countries/Antarctica.php 2 www.is.northropgrumman.com 3 www.uavforum.com 4 www.uav.com 5 “Mission Concepts for Uninhabited Aerial Vehicles in Cryospheric Science Applications”. University of Kansas Remote Sensing Laboratory. KS, 2004. 6 Roskam, Jan. Airplane Design: Parts I-VIII. DARCorporation. Lawrence, KS. 1997. 7 Advanced Aircraft Analysis Software. DARCorporation. Lawrence, KS. 2005. 8 Simons, Martin. Model Aircraft Aerodynamics, 4th Edition. Nexus Special Interests, 1999. 9 www.oceanairlogistics.com 10 www.piaggioamerica.com/ 11 Raymer, Daniel P. Enhancing Aircraft Conceptual Design using Multidisciplinary Optimization. Ph. D. Thesis, Swedish Royal Institute of Technology, Stockholm, Sweden, 2002. 12 Worldwide UAV Roundup. www.aiaa.org/images/PDF/WilsonChart.pdf. 2003. 13 Munson, Kenneth. Jane’s Unmanned Aerial Vehicles and Targets Issue 11. 1999. 14 Lambert, Mark ed. Jane’s All the World’s Aircraft 1993-94. Jane’s Information Group. Alexandria, VA. 1993. 15 Sobieszczanski-Sobieski, J., “Multidisciplinary Design Optimization: An Emerging New Engineering Discipline,” Advances in Structural Optimization (483-496), Kluwer Academic Publishers, the Netherlands, 1995. 16 www.airliners.net. April 25, 2006. 17 www.aviataircraft.com. April 25, 2006. 18 www.staggerwing.com. April 25, 2006. 19 www.comnap.com. May 19, 2005 20 http://www.scoop.co.nz/stories/PO0601/S00042.htm. May 19, 2006. 21 www.aaicorp.com. May 19, 2006. 22 www.genaero.com. May 19, 2006. 23 www.diamondair.com. May 19, 2006. 24 http://www.diamond-air.at/en/press/pressarchive/40820.htm. May 19, 2006. 25 Donovan, William. “CReSiS Airborne Platform Summary”. The University of Kansas. 2006. 26 www.rotax-aircraft-engines.com. May 19, 2006. 27 Hoerner, Sighard. Fluid-Dynamic Drag. Published by Author. Great Britain, 1958. 28 Von Mises, Richard. Theory of Flight. Dover Publications. New York, 1959. 29 Barrett, Ron. “Discussion Regarding Biplane Design”. The University of Kansas. May 10, 2006.
85 30 Allen, Christopher. “Discussion Regarding Antenna Design.” The University of Kansas. April 2006. 31 Palais, July, et al. “Meeting with NSF Representatives.” February, 2006. 32 Munk, Max M. “General Biplane Theory.” NACA Report No. 151. 1923. 33 www.nsf.gov. May 19, 2006. 34 “Science Requirements for Field Work in CReSIS. The University of Kansas. September 15, 2005. 35 www.faa.gov. August 11, 2006. 36 www.cloudcaptech.com. August 11, 2006. 37 www.lancair.com. August 11, 2006. 38 http://www.lancairlegacy.com/links.html. August 11, 2006. 39 www.aircraftspruce.com. August 11, 2006. 40 Cambone, Stephen, et al. “Unmanned Aircraft Systems Roadmap 2005-2030.” Office of the Secretary of Defense. August 2005.
86
1