DESIGN AND ANALYSIS OF A RESPONSIVE AERIAL FIRE

FIGHTING VEHICLE

by

Saad Bin Masood

Affaf Ahmad

Muhammad Salman Sajid

Abdul Rahman Akbar

Supervisor

Mr. Izhar Hussain Kazmi

Department of Aeronautics and Astronautics

Institute of Space Technology, Islamabad

2022
 Institute of Space Technology

Department of Aeronautics and Astronautics

Design and Analysis of a Responsive Aerial Fire Fighting Vehicle

by

Saad Bin Masood

Affaf Ahmad

Muhammad Salman Sajid

Abdul Rahman Akbar

Approval Page

APPROVAL BY BOARD OF EXAMINERS

_____________________
Mr. Izhar Hussain Kazmi

ii
 Certificate

This is to certify that the research work described in this thesis is the original work

of author(s) and has been carried out under my direct supervision. I have personally gone

through all the data/results/materials reported in the manuscript and certify their

correctness/authenticity. I certify that the material included in this thesis is not plagiarized

and has not been used in part or full in whole script already submitted or in the process of

submission in partial/complete fulfillment of the award of any other degree from my

institution. I also certify that the thesis has been prepared under my supervision according

to the prescribed format and I endorse its evaluation for the award of Bachelor of Science

in Aerospace Engineering degree through the official procedures of the Institute.

_____________________
Mr. Izhar Hussain Kazmi

iii
Copyright © 2022

This document is jointly copyrighted by the authors and the Institute of Space Technology

(IST). Both authors and IST can use, publish or reproduce this document in any form.

Under the copyright law, no part of this document can be reproduced by anyone, except

copyright holders, without the permission of authors.

iv
 Sustainable Development Goals

1. No Poverty

2. Zero Hunger

3. Good Health and Well-Being ✓

4. Quality Education

5. Gender Equality

6. Clean Water and Sanitation ✓

7. Affordable and Clean Energy

8. Decent Work and Economic Growth

9. Industry, Innovation and infrastructure

10. Reduced Inequality

11. Sustainable Cities and Communities

12. Responsible Consumption and Production

13. Climate Action ✓

14. Life Below Water ✓

15. Life On Land ✓

16. Peace, Justice and Strong Institution

17. Partnerships for the Goal

v
 Dedication

This work is dedicated to our parents who were always patient, supportive and understanding

throughout the time we sat alone in our ivory towers.

vi
 Abstract

The AIAA RFP requests the design of a dedicated firefighter aircraft design since the

existing fleet of firefighter aircrafts consist of converted commercial and transport aircraft

which makes for unsafe operation as their structurally not designed to handle retardant

loads. The main goal of this project is to present a design that is optimized for the role of

firefighting as a Large Air Tanker (LAT). The design effort followed an inverted pyramid

approach with regards to fidelity. Historical methods coupled with iterative sizing

algorithms were used to constrain the design space provided by the RFP with regards to

performance and capacity. A feasibility study was done to determine whether autonomous

or remote flight capability should be included. Class 1 design phase 1 was aimed at

applying low fidelity design approaches to multiple configurations in order to select a best

configuration based on performance and logistical advantages. Class 1 design phase 2 used

medium fidelity and low fidelity methods in iterative algorithms to arrive at an optimized

external layout of the selected configuration such that an iterative approach could be

avoided for Class 2 methods since those were extensive and needed hundreds of

engineering man-hours to finish a single run. Class 2 design was aimed at covering every

facet of the design in greater detail than previous phases rather than to optimize the design.

Class 2 design was then followed by CFD and FEA analyses to verify the aerodynamic

approximations of Lift, drag and Moment; As well as structural approximations of Load

bearing capabilities made in Class 2 design. To design groups which may aim at continue

the design refinement of the AFV-21, the team recommends using the results of CFD and

FEA analyses in this thesis along with the suggested improvement strategies as a starting

point in future design iterations.

vii
 PHOENIX AVIATION Presents:
AFV-21 ICESPIKE

AIAA Undergraduate Team Aircraft Design

Competition 2021-2022

viii
 General Characteristics AFV-21
Compliance Matrix
Empty Weight [lb] 70669
Description Requirement Compliance Page
Max Takeoff Weight [lb] 153284 Performance
Fuel Weight [lb] 28215 Retardant Capacity 4000 to 8000 gal 6000 gal 27
Power Plant CFM LEAP-1A Retardant reload ≥ 500 gal/min 1000 gal/min 94
Drop Speed 125 to 150 kts 131 kts (clean) 41,126
Installed Thrust [lb] 49902
Drop Altitude ≤ 300 ft AGL 300 ft AGL 16
Absolute Ceiling [ft] 43068 Design Radius 200 to 400 nm 370 nm 124
Service Ceiling [ft] 42000 Ferry Range 2000 to 3000 nm 2740 nm 126
Dash Speed 300 to 400 kts 355.5 kts 79
Wing Loading [lb/ft2] 78.16
Takeoff Field Length 5000 to 8000 ft 5413 ft (Dry) 123
T/W 0.3255 Landing Field Length 5000 to 8000 ft 2855 ft (Dry) 123
Max Mach (@39000 ft) 0.83 Proposal
Max Mach (@MSL) 0.7 Performance Analysis - - 122
Flight envelope - - 124
Cruise Mach (@34000 ft) 0.624
V-n diagram - - 107
Max R/C (@MSL) [fpm] 2242.8 Materials Selection - - 118
Radius (@ Full Payload) [nm] 370 Structural Design - - 107
Ferry Range [nm] 2740 Geometric Description - - x
Weight Statement - - 96
Stall Speed (full Flaps) [kts] 94.5 Propulsion System - - 46
Takeoff BFL (Dry Concrete) [ft] 5413 Stability and Control - - 99
Unit Cost [M$] 93.7 Cost Analysis - - 129
Concept Selection - - 29
Operating Cost [$/hour] 11284

ix
 AFV-21 ICESPIKE
Wing Area 1961.1 ft2
HT Area 600 ft2
VT Area 250 ft2
Wing Twist -3 deg
Wing Incidence 2.3 deg
HT Incidence 0.0 deg
Nose Gear Height 5.15 ft2
Main Gear Height 5.5 ft2
Neutral Point (X) 25.24 ft
MTOW (Forward) CG 24.8 ft
OEW (Shown) CG 25.08 ft
EW (Aft) CG 25.17 ft
All Fig dimensions are in [ft]

x
 Table of Contents

Approval Page ..................................................................................................................... ii

Certificate ........................................................................................................................... iii
Copyright ........................................................................................................................... iv
Sustainable Development Goals ......................................................................................... v
Dedication .......................................................................................................................... vi
Abstract ............................................................................................................................. vii
Table of Contents ............................................................................................................... xi
List of Tables .................................................................................................................... xx
List of Figures ................................................................................................................ xxiii
List of Abbreviations and Symbols................................................................................ xxxi
Acknowledgements ...................................................................................................... xxxiii
EXECUTIVE SUMMARY ......................................................................................... xxxiv
1. REQUIREMENTS ...................................................................................................... 1
1.1 Explicit Requirements .......................................................................................... 1

1.2 Implicit Requirements .......................................................................................... 2

2. MARKET ANALYSIS ............................................................................................... 3

2.1 Selection of Major Market for Design ................................................................. 3

2.2 Ferry Range Target Value .................................................................................... 4

2.3 Payload Target Value ........................................................................................... 5

2.4 Estimating Market Potential ................................................................................. 5

2.5 Production Rate .................................................................................................... 8

3. FEASIBILITY STUDY FOR INTEGRATION OF AUTONOMOUS

CAPABILITIES ................................................................................................................ 10
3.1 System Architecture ........................................................................................... 10

3.2 Required Technologies ....................................................................................... 11

3.3 Weights Trends for RPAs vs. Manned Aircraft ................................................. 12

xi
 3.4 Conclusion of Study ........................................................................................... 13

4. MISSION DESIGN AND ANALYSIS .................................................................... 14

4.1 Flexible Operation .............................................................................................. 14

4.2 Mission Profiles.................................................................................................. 14

5. DESIGN SPACE EXPLORATION AND WEIGHT SIZING ................................. 18

5.1 The Weight Sizing Algorithm ............................................................................ 18

5.1.1 WE and WTO Statistical Prediction w.r.t Payload........................................ 19

5.1.2 Empty Weight Fraction ............................................................................... 20

5.1.3 Engine Mathematical Model ....................................................................... 21

5.1.4 Drag Polars.................................................................................................. 21

5.1.5 Remaining Equations .................................................................................. 22

5.1.6 Sizing Algorithm Steps ............................................................................... 22

5.2 Design Space Exploration for Design Radius .................................................... 23

5.2.1 Ferry Range Constraint ............................................................................... 24

5.2.2 Operating Cost Constraint Line .................................................................. 24

5.2.3 Carpet Plot Generation Algorithm .............................................................. 26

5.2.4 Design Space Exploration Using Carpet Plot ............................................. 27

6. CONFIGURATION DOWNSELECT...................................................................... 28
6.1 Concept Generation by Morph Matrix ............................................................... 28

6.2 Concept Selection ............................................................................................... 29

6.2.1 Rejected Configurations.............................................................................. 30

xii
 6.2.2 Summary of Rejected Configurations ......................................................... 32

6.2.3 Accepted Configurations ............................................................................ 33

6.2.4 Summary of Accepted Configurations........................................................ 34

7. AIRFOIL SELECTION ............................................................................................ 37

8. PERFORMANCE SIZING ....................................................................................... 41
9. GEOMETRIC SIZING ............................................................................................. 43
9.1 Wing ................................................................................................................... 43

9.2 Fuselage .............................................................................................................. 44

9.3 Empennage ......................................................................................................... 44

10. PROPULSION SYSTEM ......................................................................................... 46

10.1 Thrust Required .............................................................................................. 46

10.2 Engine Selection ............................................................................................. 46

10.3 Engine Characteristics .................................................................................... 47

10.4 Inlet and Diffuser Geometry ........................................................................... 49

10.5 Capture Area Calculation ............................................................................... 49

10.5.1 Engine Mass Flow....................................................................................... 49

10.5.2 Statistical Estimate of Capture Area ........................................................... 50

10.5.3 Analytical Estimate of Capture Area .......................................................... 50

10.6 Nozzle Design................................................................................................. 51

10.7 Summary of Important Results ....................................................................... 51

11. AERODYNAMICS .................................................................................................. 53

11.1 Lift Calculations ............................................................................................. 53

xiii
 11.1.1 Maximum Lift – Clean................................................................................ 53

11.2 Maximum Lift – High Lift Devices ................................................................ 53

11.2.1 HLD System Type Trade Study.................................................................. 53

11.2.2 HLD System Arrangement ......................................................................... 56

11.2.3 HLD System Design ................................................................................... 57

11.2.4 Lift Curves .................................................................................................. 58

11.3 Drag Calculations ........................................................................................... 59

12. STABILITY ANALYSIS ......................................................................................... 61

13. PERFORMANCE ..................................................................................................... 62
13.1 Criteria 1 – Aerodynamic Efficiency.............................................................. 62

13.2 Criteria 2 – Fuel Efficiency ............................................................................ 63

13.3 Criteria 3 – Ferry Range ................................................................................. 63

13.4 Criteria 4 – Takeoff Performance ................................................................... 63

13.5 Criteria 5 – Landing Performance .................................................................. 64

14. FINAL CONFIGURATION SELECTION .............................................................. 65

14.1 FOD Protection Concept Testing ................................................................... 65

14.2 Conclusion ...................................................................................................... 66

15. WING OPTIMIZATION .......................................................................................... 68

15.1 Aspect Ratio ................................................................................................... 68

15.2 Taper Ratio, Incidence and Twist ................................................................... 69

16. WEIGHTS AND BALANCE ANALYSIS .............................................................. 70

17. LANDING GEAR .................................................................................................... 71

xiv
18. HORIZONTAL AND VERTICAL TAIL SIZING .................................................. 75
18.1 Horizontal Tail ................................................................................................ 75

18.2 Vertical Tail .................................................................................................... 77

19. CONDITIONS FOR OPTIMAL PERFORMANCE ................................................ 78

19.1 Cruise Altitude ................................................................................................ 78

19.2 Speeds of Major Segments ............................................................................. 79

19.2.1 𝑽𝒄𝒓𝒖𝒊𝒔𝒆 𝟏 and 𝑽𝒄𝒓𝒖𝒊𝒔𝒆 𝟐 Optimization .................................................. 80

19.2.2 𝑽𝒍𝒐𝒊𝒕𝒆𝒓 𝟏 and 𝑽𝒍𝒐𝒊𝒕𝒆𝒓 𝟐 Optimization .................................................... 80

20. SUBSYSTEMS ......................................................................................................... 82

20.1 Fuel System .................................................................................................... 82

20.2 Flight Control Systems ................................................................................... 83

20.3 Engine Controls .............................................................................................. 85

20.4 Hydraulic System ........................................................................................... 85

20.5 Electric Systems ............................................................................................. 86

20.6 Pneumatic System........................................................................................... 87

20.7 Environmental Control System (ECS) ........................................................... 87

20.8 Avionics .......................................................................................................... 88

20.9 Complete Systems Layout .............................................................................. 90

21. RETARDANT EVACUATION SYSTEM AND TANK DESIGN ......................... 92

21.1 Retardant Tank Design ................................................................................... 92

21.2 Retardant Evacuation Mechanism .................................................................. 93

21.3 Maintenance and Reloading ........................................................................... 94

xv
22. WEIGHT AND BALANCE ..................................................................................... 95
23. STABILITY AND CONTROL ANALYSIS............................................................ 99
23.1 Stabilizer Sizing.............................................................................................. 99

23.2 Control Surface Sizing.................................................................................... 99

23.3 Trim Analysis ............................................................................................... 100

23.4 Longitudinal Stability Characteristics .......................................................... 102

23.5 Lateral-Directional Stability Characteristics ................................................ 104

24. STRUCTURAL DESIGN AND MATERIAL SELECTION................................. 107

24.1 Manoeuvre and Gust Loads .......................................................................... 107

24.2 Wing Design ................................................................................................. 108

24.3 Empennage Design ....................................................................................... 113

24.4 Fuselage Design ............................................................................................ 114

24.5 Miscellaneous Material Selection ................................................................. 117

24.6 Final Structural Layout Assembly ................................................................ 119

25. DRAG POLARS ..................................................................................................... 120

26. PERFORMANCE ANALYSIS .............................................................................. 122
26.1 Takeoff and Landing Performance ............................................................... 122

26.2 Payload Radius Diagram .............................................................................. 123

26.3 Operational Envelope ................................................................................... 124

26.4 Specific Excess Power Diagram ................................................................... 125

26.5 Other Performance Parameters ..................................................................... 126

26.6 Climb Gradient ............................................................................................. 127

xvi
27. COST ANALYSIS.................................................................................................. 129
27.1 Research, Development, Test and Evaluation (RDTE) Costs ...................... 129

27.2 Production (Flyaway) Costs ......................................................................... 130

27.3 Unit Sales Price ............................................................................................ 131

27.4 Break-Even Analysis .................................................................................... 131

27.5 Direct Operating Cost ................................................................................... 132

28. MANUFACTURING PLAN .................................................................................. 134

29. COMPUTATIONAL FLUID DYNAMICS ........................................................... 135
29.1 Geometry ...................................................................................................... 136

29.2 Meshing ........................................................................................................ 140

29.2.1 Overall Procedure ..................................................................................... 140

29.2.2 Global Mesh Setup .................................................................................... 140

29.2.3 Part Mesh Setup ........................................................................................ 141

29.2.4 Density Boxes ........................................................................................... 142

29.2.5 Prism Layers and Y+ ................................................................................ 144

29.2.6 Mesh Output with Boundary Conditions .................................................. 146

29.3 Problem Setup............................................................................................... 146

29.3.1 General ...................................................................................................... 146

29.3.2 Models....................................................................................................... 147

29.3.3 Materials ................................................................................................... 147

29.3.4 Boundary Conditions ................................................................................ 147

xvii
 29.3.5 Reference Values ...................................................................................... 148

29.3.6 Report Definitions ..................................................................................... 148

29.4 Results and Discussion ................................................................................. 148

29.4.1 Drag (CD) .................................................................................................. 152

29.4.2 Lift (CL)..................................................................................................... 155

29.4.3 Pitching Moment (Cm) .............................................................................. 157

29.5 Post Processing and Flow Visualization ....................................................... 159

29.5.1 Contours and Vector Diagrams for Flow Visualization ........................... 159

29.5.2 Visualizing Different Flight Phenomenons .............................................. 166

30. FINITE ELEMENT ANALYSIS (FEA) ................................................................ 185

30.1 Geometry ...................................................................................................... 185

30.2 Material ......................................................................................................... 186

30.3 Meshing ........................................................................................................ 186

30.4 Boundary Conditions and Loading ............................................................... 187

30.5 Solution and Results ..................................................................................... 189

30.5.1 Total Deformation ..................................................................................... 190

30.5.2 Equivalent Stress ....................................................................................... 190

30.5.3 Safety Factor ............................................................................................. 192

31. SUSTAINABLE DEVELOPMENT GOALS ........................................................ 193

31.1 Good Health and Well-Being ....................................................................... 193

31.2 Clean Water and Sanitation .......................................................................... 193

xviii
 31.3 Climate Actions ............................................................................................ 194

31.4 Life Below Water ......................................................................................... 194

31.5 Life on Land ................................................................................................. 195

32. CONCLUSION ....................................................................................................... 196

REFERENCES ............................................................................................................... 197
APPENDICES ................................................................................................................ 201
Appendix – A (Softwares) .......................................................................................... 201

xix
 List of Tables

Table 1.1 Summary of Explicit Requirements .................................................................... 2

Table 2.1 Acres burned and LAT requirement estimates for different countries ............... 6

Table 2.2 Capability targets finalized after market analysis ............................................... 9

Table 3.1 Summary of roles of several technologies in RPAs, and their TRLs [18]........ 11

Table 6.1 Initial Morph Matrix with 367290 Possible Configurations ............................. 28

Table 7.1 Candidate Airfoil Characteristics...................................................................... 39

Table 7.2 Airfoil selection by scoring method.................................................................. 40

Table 8.1 Performance Sizing Results .............................................................................. 41

Table 9.1 Wing Geometric Parameters ............................................................................. 43

Table 9.2 Fuselage Geometric Parameters........................................................................ 44

Table 9.3 Empennage Geometric Parameters ................................................................... 44

Table 10.1 Candidate Engine Parameters ......................................................................... 46

Table 10.2 Important results from engine selection and nacelle design ........................... 51

Table 11.1 Existing HLD arrangements for a CLmax of 3.0............................................... 56

Table 11.2 Zero lift drag coefficient of all components of the four configurations ......... 59

Table 11.3 Drag coefficients of all configurations in clean, takeoff, and landing settings

........................................................................................................................................... 60

Table 12.1 Fundamental static stability derivatives for competing configurations (per

radian) ............................................................................................................................... 61

Table 13.1 Aerodynamic Efficiencies of Different Mission Segments of four

Configurations................................................................................................................... 63

Table 13.2 Fuel Weight of Four Configurations. .............................................................. 63

xx
Table 13.3 Range of Four Configurations. ....................................................................... 63

Table 13.4 Balanced Field Length of Four Configurations. ............................................. 64

Table 13.5 Landing Field Length of Four Configurations. ............................................... 64

Table 17.1 Landing gear longitudinal characteristics ....................................................... 71

Table 17.2 Landing gear loads .......................................................................................... 72

Table 17.3 Strut design parameters ................................................................................... 72

Table 17.4 Tire selections and specifications. .................................................................. 72

Table 17.5 Landing gear lateral stability. ......................................................................... 74

Table 20.1 Avionics Systems List .................................................................................... 89

Table 20.2 Complete subsystem key ................................................................................ 90

Table 22.1 Empty Weight Breakdown w.r.t Nose ............................................................ 96

Table 22.2 Operational Items CG Locations w.r.t Nose ................................................... 97

Table 23.1 Control Surface Sizing .................................................................................. 100

Table 23.2 Aircraft Drag Coefficients in cruise for incidence and elevator trim scenarios

......................................................................................................................................... 101

Table 23.3 Longitudinal Stability Derivatives (per radian) ............................................ 102

Table 23.4 Static Margin Check ..................................................................................... 103

Table 23.5 Longitudinal Dynamic Stability Characteristics ........................................... 104

Table 23.6 Lateral-Directional Stability Derivatives (per radian) .................................. 104

Table 23.7 Lateral-Directional Dynamic Stability Characteristics ................................. 105

Table 24.1 Shear flow analysis wall properties .............................................................. 112

Table 24.2 Wing Substructure Key Parameters .............................................................. 113

Table 24.3 HT Substructure Key Parameters ................................................................. 114

xxi
Table 24.4 VT Substructure Key Parameters ................................................................. 114

Table 24.5 Fuselage Substructure Key Parameters......................................................... 117

Table 24.6 Material Assignment and Properties ............................................................. 118

Table 25.1 Wetted Area and Parasite Drag Breakdown for the AFV-21 ....................... 120

Table 26.1 Summary of Takeoff and Landing performance parameters ........................ 123

Table 26.2 Performance Coefficients ............................................................................. 126

Table 27.1 Operating costs Per Flight Hour Per Aircraft (Top Rows) and Per Year Per

Aircraft (Bottom Rows) .................................................................................................. 133

Table 29.1 Boundary conditions. .................................................................................... 147

Table 29.2 Tabulation of CFD results............................................................................. 151

Table 29.3 Pitching Moment Coefficient comparison .................................................... 158

xxii
 List of Figures

Fig 2.1 No. of wildfire disasters in different countries from 1900 to 2013 [2] .................. 3

Fig 2.2 Farthest regions in the Continental US with High Burn Count [3] ........................ 4

Fig 2.3 Payload selection niche plot ................................................................................... 5

Fig 2.4 Acres Burned by Wild Fires in the US (1990 - 2020) [5] ...................................... 6

Fig 2.5 Comparison of EIS and Retirement Years of Firefighting Aircraft ....................... 8

Fig 3.1 RPA System Interfaces [17] ................................................................................. 11

Fig 3.2 Mass Domains of Manned and Un-manned Aircraft [19] .................................... 13

Fig 4.1 Ferry Mission ........................................................................................................ 15

Fig 4.2 Fire mission with Payload Drop ........................................................................... 16

Fig 4.3 Fire mission without Payload Drop ...................................................................... 16

Fig 5.1 Weight sizing algorithm for fire mission .............................................................. 19

Fig 5.2 WE and WTO statistical prediction curves ............................................................. 19

Fig 5.3 Empty Weight fraction trend ................................................................................ 20

Fig 5.4 Ferry Range and Cost Constraint Data Builder Algorithms ................................. 25

Fig 5.5 Carpet plot generation algorithm for design radius selection ............................... 26

Fig 5.6 Weight Sizing Carpet Plot with Ferry (red) and Cost (blue) Constraints ............. 27

Fig 6.1 Configurations Considered for Class 1 Design .................................................... 29

Fig 7.1 Drag polar comparison (XFLR-5) ........................................................................ 38

Fig 7.2 Lift slope comparison (XFLR-5) .......................................................................... 38

Fig 7.3 Lift to drag ratio comparison (XFLR-5) ............................................................... 38

Fig 7.4 Moment coefficient comparison (XFLR-5) .......................................................... 38

Fig 7.5 Drag polar comparison (OpenVSP) ...................................................................... 39

xxiii
Fig 7.6 Lift slope comparison (OpenVSP) ....................................................................... 39

Fig 7.7 Lift to drag ratio comparison (OpenVSP) ............................................................ 39

Fig 7.8 Moment coefficient comparison (OpenVSP) ....................................................... 39

Fig 8.1 Constraint diagram for performance sizing .......................................................... 42

Fig 10.1 Engine Performance Graphs ............................................................................... 48

Fig 11.1 Airfoil and HLDs in landing configuration ........................................................ 58

Fig 11.2 Low Speed Lift Curves ....................................................................................... 59

Fig 14.1 Vector Diagram to test the FOD protection hypothesis ..................................... 65

Fig 15.1 Finalized Wing Geometry .................................................................................. 68

Fig 15.2 Aspect Ratio Optimization ................................................................................. 69

Fig 15.3 Semi Span Lift Distribution................................................................................ 69

Fig 17.1 Longitudinal Landing Gear Geometric Parameters & Loads ............................. 71

Fig 17.2 Landing Gear Lateral Clearance Requirement. .................................................. 71

Fig 17.3 Lateral tip-over criteria. ...................................................................................... 74

Fig 18.1 Longitudinal X-Plot ............................................................................................ 75

Fig 18.2 Directional X-Plot............................................................................................... 75

Fig 18.3 Finalized HT (Left) and VT (Right) geometries ................................................ 75

Fig 19.1 Fuel optimization ................................................................................................ 79

Fig 19.2 Time optimization............................................................................................... 79

Fig 19.3 Cruise Speeds optimization ................................................................................ 80

Fig 19.4 Loiter Speeds optimization ................................................................................. 80

Fig 20.1 Layout of Basic Aircraft Systems ....................................................................... 91

Fig 21.1 Retardant Tank Cross-Section ............................................................................ 93

xxiv
Fig 21.2 Retardant Tank Dimensioned Isometric View ................................................... 93

Fig 22.1 Empty Weight CG Locations ............................................................................. 95

Fig 22.2 Operational items CG and aircraft CG locations ................................................ 96

Fig 22.3 CG Excursion Diagram ...................................................................................... 98

Fig 23.1 DATCOM Model of the AFV-21 ....................................................................... 99

Fig 23.2 Trim Diagrams for different scenarios ............................................................. 101

Fig 23.3 Lateral SAS Block Diagram ............................................................................. 106

Fig 24.1 V-n Diagram with Gust Envelope .................................................................... 107

Fig 24.2 Wing Lift Distribution Approximation............................................................. 109

Fig 24.3 Total Wing Load Distribution .......................................................................... 109

Fig 24.4 Spanwise Wing Shear Loading......................................................................... 109

Fig 24.5 Spanwise Wing Bending Moment .................................................................... 109

Fig 24.6 Idealized Wing Rib at Root for Bending Analysis ........................................... 110

Fig 24.7 Idealized Wing Rib at Root for Shear Flow Analysis ...................................... 111

Fig 24.8 Wing Internal Structure .................................................................................... 113

Fig 24.9 HT Internal Structure ........................................................................................ 114

Fig 24.10 VT Internal Structure ...................................................................................... 114

Fig 24.11 Idealized Fuselage Section at Wing Rear Spar ............................................... 115

Fig 24.12 Fuselage Internal Structure ............................................................................. 117

Fig 24.13 Aircraft Final Structure ................................................................................... 119

Fig 25.1 Multiple Flight Configurations Drag Polars ..................................................... 120

Fig 25.2 Drag Polar at Various Cruise Mach No. ........................................................... 120

Fig 26.1 Takeoff BFL Analysis – (Dry Concrete) .......................................................... 122

xxv
Fig 26.2 Payload-Radius Diagram at different fuel loads............................................... 124

Fig 26.3 Operational Envelope at MTOW ...................................................................... 124

Fig 26.4 Specific Excess Power Diagram ....................................................................... 125

Fig 26.5 Performance Coefficients ................................................................................. 127

Fig 26.6 Second segment Climb Gradient VS Gross Weight ......................................... 127

Fig 27.1 RDTE costs breakdown in different cost models ............................................. 130

Fig 27.2 Production costs breakdown in different cost models ...................................... 130

Fig 27.3 Payload Capacity VS Sales Price ..................................................................... 131

Fig 27.4 Break-Even Analysis (Nicolai) ......................................................................... 132

Fig 27.5 Break-Even Analysis (Raymer) ........................................................................ 132

Fig 27.6 Operating costs breakdown in different cost models........................................ 133

Fig 29.1 OpenVSP model after ....................................................................................... 137

Fig 29.2 OpenVSP model after ....................................................................................... 137

Fig 29.3 CATIA model of AFV-21 ................................................................................ 138

Fig 29.4 SPACECLAIM model of AFV-21 ................................................................... 139

Fig 29.5 Halved fluid domain with the aircraft geometry removed ............................... 139

Fig 29.6 Part Mesh Setup in ICEM CFD ........................................................................ 142

Fig 29.7 Surface meshes to show contrast in detail between aircraft and domain mesh detail

......................................................................................................................................... 142

Fig 29.8 Density boxes, one behind the outer wing and the other behind the empennage

......................................................................................................................................... 143

Fig 29.9 Top view of a cut plane to demonstrate the effect of density boxes on mesh detail

behind aircraft ................................................................................................................. 144

xxvi
Fig 29.10 Different zoomed-in views of prism mesh to demonstrate detail required in BL

detail capturing................................................................................................................ 146

Fig 29.11 The Mesh imported in FLUENT (Green – inlet / Red - outlet) ...................... 148

Fig 29.12 Simulations results for AOA = 00 ................................................................... 149

Fig 29.13 Simulation Results for AOA = 40 ................................................................... 150

Fig 29.14 Simulation Results for AOA = 80 ................................................................... 150

Fig 29.15 Simulation Results for AOA = 120 ................................................................. 151

Fig 29.16 Comparison of Drag Coefficient Estimates from CFD and Design Phase ..... 152

Fig 29.17 Comparison of Lift Coefficient Estimates from CFD and Design Phase ....... 155

Fig 29.18 Comparison of Moment Coefficient Estimates from CFD and Design Phase 158

Fig 29.19 Velocity Contours (Mach No.) for AOA = 00 Fig 29.20 Velocity

Contours (Mach No.) for AOA = 40 ............................................................................... 161

Fig 29.21 Velocity Contours (Mach No.) for AOA = 80 Fig 29.22 Velocity Contours

(Mach No.) for AOA = 120 ............................................................................................. 161

Fig 29.23 Pressure Coefficient for AOA 00 Fig 29.24 Pressure Coefficient for AOA 40

162

Fig 29.25 Pressure Coefficient for AOA 80 Fig 29.26 Pressure Coefficient for AOA

120 162

Fig 29.27 Total Pressure for AOA = 00 Fig 29.28 Total Pressure for AOA = 40

163

Fig 29.29 Total Pressure for AOA = 80 Fig 29.30 Total Pressure for AOA = 120

163

Fig 29.31 Static Pressure for AOA = 00.......................................................................... 164

xxvii
Fig 29.32 Static Pressure for AOA = 40.......................................................................... 164

Fig 29.33 Static Pressure for AOA = 80.......................................................................... 164

Fig 29.34 Static Pressure for AOA = 120........................................................................ 164

Fig 29.35 Flow Past Aircraft at AOA = 00 Fig 29.36 Flow Past Aircraft at AOA = 40

165

Fig 29.37 Flow Past Aircraft at AOA = 80 Fig 29.38 Flow Past Aircraft at AOA =

120 165

Fig 29.39 Velocity Contours at AOA = 00 ...................................................................... 168

Fig 29.40 Velocity Contours at AOA = 120 .................................................................... 168

Fig 29.41 Inboard Airfoil Velocity Contours at AOA = 00 ............................................ 168

Fig 29.42 Inboard Airfoil Velocity Contours at AOA = 120 .......................................... 169

Fig 29.43 Static Pressure Contours at AOA = 00 ............................................................ 170

Fig 29.44 Static Pressure Contours at AOA = 120 .......................................................... 170

Fig 29.45 Inboard Airfoil Static Pressure Contours at AOA = 00 .................................. 170

Fig 29.46 Inboard Airfoil Static Pressure Contours at AOA = 120 ................................ 171

Fig 29.47 Static Temperature Contours at AOA = 00 ..................................................... 171

Fig 29.48 Static Temperature Contours at AOA = 120 ................................................... 172

Fig 29.49 Inboard Airfoil Static Temperature Contours at AOA = 00 ........................... 172

Fig 29.50 Inboard Airfoil Static Temperature Contours at AOA = 120 ......................... 172

Fig 29.51 Total Pressure at AOA = 00 ............................................................................ 173

Fig 29.52 Total Pressure at AOA = 120 .......................................................................... 173

Fig 29.53 Inboard Airfoil Total Pressure at AOA = 00 ................................................... 174

Fig 29.54 Inboard Airfoil Total Pressure at AOA = 120 ................................................. 174

xxviii
Fig 29.55 Wall Y+ (Same for all cases) ........................................................................... 176

Fig 29.56 Close-up Boundary Layer Visualization at AOA = 120 ................................. 176

Fig 29.57 Boundary layer Visualisation for AOA = 00 .................................................. 177

Fig 29.58 Boundary layer Visualisation for AOA = 120 ................................................ 177

Fig 29.59 Close-up of Boundary Layer Velocity............................................................ 177

Fig 29.60 Close-up of Boundary Layer Velocity............................................................ 177

Fig 29.61 Velocity Vectors around wings at AOA = 00 ................................................. 178

Fig 29.62 Velocity Vectors around wings at AOA = 120 ............................................... 178

Fig 29.63 Close-up of Boundary Layer Velocity............................................................ 178

Fig 29.64 Close-up of Boundary Layer Velocity............................................................ 178

Fig 29.65 Flow at wing location Y = 8 ft ........................................................................ 180

Fig 29.66 Flow at wing location Y = 30 ft...................................................................... 180

Fig 29.67 Flow at wing location Y = 56 ft ...................................................................... 180

Fig 29.68 Comparison of chord-wise Cp distributions at different spanwise locations .. 181

Fig 29.69 Downwash region and vortex behind wing at AOA = 40 ............................... 182

Fig 29.70 Downwash region and vortex behind wing at AOA = 120 ............................. 183

Fig 29.71 Downwash region and vortex behind empennage at AOA = 40 ..................... 183

Fig 29.72 Downwash region and vortex behind empennage at AOA = 120 ................... 183

Fig 29.73 Downwash region and vortex at the end of density box at AOA = 40 ........... 184

Fig 29.74 Downwash region and vortex at the end of density box at AOA = 120 ......... 184

Fig 30.1 Wing Structure Geometry in ANSYS Mechanical ........................................... 186

Fig 30.2 Wing Structure Mesh ........................................................................................ 187

Fig 30.3 Wing Structure Mesh in Close-up .................................................................... 187

xxix
Fig 30.4 Pressure Loading for Front Spar ....................................................................... 188

Fig 30.5 Pressure Loading for Rear Spar ........................................................................ 189

Fig 30.6 Boundary and Loading Conditions for Wing FEA ........................................... 189

Fig 30.7 Total Deformation of Wing Structure .............................................................. 190

Fig 30.8 Equivalent Stress on Wing Structure ................................................................ 191

Fig 30.9 Close-up showing Maximum Equivalent Stress near Root .............................. 191

Fig 30.10 Safety Factor for Wing Structure.................................................................... 192

xxx
 List of Abbreviations and Symbols

AAA = Advanced Aircraft Analysis

ADP = Air Driven Pump
AFV = Aerial Firefighting Vehicle
AGL = Above Ground Level
AMF = Aircraft Manufacturing Factory
APU = Auxiliary Power Unit
AR = Aspect Ratio
ATC = Air Traffic Control
AUP = All Up Mass
BTB = Bus Tie Breaker
CG = Center of Gravity
EBF = Externally Blown Flap
ECS = Environmental Control System
EHA = Electro-Hydrostatic Actuator
EIS = Entry Into Service
EMA = Electro-Mechanical Actuator
EW = Empty Weight
FAA = Federal Aviation Administration
FADEC = Full Authority Digital Engine Control
FAR = Federal Aviation Regulations
FBW = Fly By Wire
FCU = Fuel Control Unit
FMS = Flight Management System
FOD = Foreign Object Damage
FRDS = Fire Retardant Dispersal System
HLD = High Lift Device
HT = Horizontal Tail
IAB = Interagency Airtanker Board
IAP = Integrated Actuator Package
IDG = Integrated Drive Generator
IFR = Instrument Flight Rules

xxxi
IMA = Integrated Modular Avionics
LAT = Large Air Tanker
MSL = Mean Sea Level
MTBS = Monitoring Trends in Burn Severity
MTOW = Maximum Takeoff Weight
NASA = National Aeronautics & Space Administration
OBIGGS = On Board Inert Gas Generation System
OEW = Operational Empty Weight
OTH = Over The Horizon
PAC = Pakistan Aeronautical Complex
PMG = Permanent Magnet Generator
PRSOV = Pressure Regulator Shut Off Valve
RAT = Ram Air Turbine
RDTE = Research Development Testing & Evaluation
RFP = Request For Proposal
RPAS = Remotely Piloted Aircraft System
RPS = Remote Pilot Station
SAS = Stability Augmentation System
SFC = Specific Fuel Consumption
SSLW = Second Segment limiting Weight
TRL = Technology Readiness Level
TRU = Transformer Rectifier Unit
USAF = United States Air Force
USB = Upper Surface Blown
USFS = United States Forest Service
VFR = Visual Flight Rules
VT = Vertical Tail
WOF = Wash Out Filter

xxxii
 Acknowledgements

Many academic and industry professionals were consulted at different points in the design.
The team would like to express their gratitude towards those who lent their time and
expertise.

• Institute of Space Technology

Prof Dr. Syed Hossein Raza Hamdani

Prof Dr. Jamshed Riaz

Mr. Syed Abbas Rameez Gilani

Mr. Muhammad Umar

• Holy Cows Inc.

Mr. Bill Galbraith

• Pakistan Aeronautical Complex

Mr. Saad Kamal

xxxiii
 EXECUTIVE SUMMARY

RFP calls for a purpose-built Large Air Tanker (LAT) due to compromises, inefficiencies,

safety issues, and falling numbers of airworthy LATs in the existing fleets of converted

aircraft. Studies also indicate a substantial increase in both frequency and sizes of wildfires

due to global change in meteorological patterns. We at Phoenix Aviation firmly believe

that we have understood the importance and challenge of introducing a capable, cost-

efficient, and reliable airtanker in the aerial firefighting arsenal worldwide. In this spirit,

we proudly present AFV-21 ICESPIKE as a purpose-built LAT suitable for the

requirements put forth in the RFP and other requirements specified by the team.

A high level summary of the design process employed by Phoenix Aviation is that the

development of design from low fidelity (Class 1) methods to higher fidelity (Class 2)

methods, as presented in “Airplane Design” by Dr. Jan Roskam, was adopted but with a

twist, i.e., Class 1 design was preceded by market analysis, design space exploration and

was supplemented by higher fidelity methods from numerous other design texts and

softwares to ensure an optimal solution was developed before the beginning of Class 2

design. This allowed significant reduction in Class 2 design iterations while retaining the

detailed design definition achieved as a result of utilizing Class 2 methods.

Comprehensive risk assessments were conducted to determine the feasibility of producing

the AFV-21 as a Remotely Piloted Aircraft. Unfortunately, the investigations concluded

that despite having the requisite technologies, the introduction of such a version is limited

to improvement of Technology Readiness Levels (TRLs) and to when remote aircraft in

the size range of AFV-21 are flight proven.

xxxiv
Market analysis was conducted to finalize production rate, ferry range, and payload

capacity. The highly iterative nature of the design process was facilitated by the

development of a sophisticated sizing code that enabled the calculation of fuel weight by

determining the drag as the AFV-21 flew through the mission, calculating the thrust

required and the resulting fuel flow. The sizing code acted as a foundation of complex

algorithms for design space exploration, sizing, mission design, and performance analysis

throughout the design effort. Sixteen potential configurations were assessed, and four were

carried forward to class 1 design - phase 1, where they were developed and analyzed using

low fidelity methods. A subsidiary to the main project was also started to investigate the

feasibility and benefit of including powered HLDs in the design but was abandoned due to

safety concerns in the One Engine Out (OEI) scenario. Conventional configuration, being

the best performer, was finalized and optimized using higher fidelity methods in class 1

design - phase 2. Class 2 design utilized high fidelity methods and softwares geared

towards addressing details of the design, such as control surface sizing, subsystems

selection and layout, landing gear design, trim analysis, stability augmentation, retardant

tank design, detailed drag breakdown, and performance analysis. Detailed structural

design, material selection, and analysis for major aircraft substructures, i.e., Wing, HT, VT,

and fuselage, were also carried out as a part of class 2 design after finalizing aircraft

geometry and subsystem layout. A unit price was finalized using several methods in the

cost analysis for a 15% profit, and a manufacturing plan proposed for low volume

production of the 200 aircraft market share estimated in the initial stages of the design.

The AFV-21 features a conventional design. It has a ferry range of 2740 nm and can carry

a retardant payload of 6000 gal (54000 lb) while executing two sorties of its design mission

xxxv
(380 nm radius) on a single fuel load; capabilities uncommon in contemporary aircraft. The

aircraft has a fuselage length of 64.35 ft, a wing span of 122.08 ft and is powered by two

CFM LEAP-1A high-bypass turbofan engines. The MTOW and EW are 153284 lb and

70669 lb, respectively. The unit price is expected to be $ 93.7 million, and the operating

cost is estimated at $ 11,815 per hour. The AFV-21 is capable of meeting all requirements

mentioned in the RFP.

xxxvi
 1. REQUIREMENTS

The RFP [1] does not provide a specific mission profile nor give exact values against

capacity or performance capabilities; instead, it provides ranges for the latter two. The

designer is left to generate a concept of operations or mission profiles based on existing

literature and the operational history of existing firefighter aircraft. Requirements may be

divided into two general categories, explicit and implicit. Explicit requirements are directly

provided by the customer, which in this case is the AIAA RFP. Implicit requirements are

those not stated by the customer but are derived to meet the explicit requirements.

1.1 Explicit Requirements

Table 1.1 summarizes the requirements stated in the Request For Proposal (RFP) [1]. It is

worthy to note that all of the requirements were ranges that generated a large design space,

the extremes of which produced completely different designs, so they had to be anchored

to singular values.

1
 Table 1.1 Summary of Explicit Requirements

[R] = 4000 gal; [O] = 8000 gal

Fire Retardant Capacity
Density = 9 lb/gal
Drop Capability [R] ≥ 2000 gal/drop
Retardant reload [R] ≥ 500 gal/min
Drop Speed [R] ≤ 150 kts; [O] ≤ 125 kts
Drop Altitude [R] ≤ 300 ft AGL
Design Radius [R] = 200 nm; [O] = 400 nm
Ferry Range [R] = 2000 nm; [O] = 3000 nm
Dash Speed [R] = 300 kts; [O] = 400 kts
[R] = BFL ≤ 8000 ft, @5000 ft & 350F
Field Requirements
[O] = BFL ≤ 5000 ft, @5000 ft & 350F
- VFR and IFR flight with an
autopilot
Certifications - Flight in known icing conditions
- Certification via FAA 14 CFR Part
25
[O] = Objective, [R] = Requirements
1.2 Implicit Requirements

Following implicit requirements were needed to meet the explicit requirements. The

aircraft required a high lift system to meet the takeoff and landing requirements. An

Axillary Power Unit (APU) was necessary since the availability of equipment for engine

start in all airports near fire-prone areas was uncertain.

2
 2. MARKET ANALYSIS

Not all design decisions were technical in nature; many were driven by customer

satisfaction and market trends. Before beginning the design, the team decided to conduct a

market and stakeholder study in hopes of discovering what the operators and buyers of

firefighter aircraft desired; what the trends for this part of aviation were, and what they

were pointing toward; but most importantly, what level of capability should the design be

anchored at.

2.1 Selection of Major Market for Design

Although the demand for firefighter aircraft is rising worldwide with rising temperatures,

the team optimized the design by focusing on the largest market.

Fig 2.1 No. of wildfire disasters in different countries from 1900 to 2013 [2]

Fig 2.1 shows that the United States has the largest number of wildfire disasters which

included fatalities, millions in damages or a state of emergency (criteria for event inclusion

specified in [2]), effectively making it the largest market for Large Air Tankers (LATs).

The design was therefore optimized to serve in the United States.

3
2.2 Ferry Range Target Value

A target value for the ferry range of the AFV-21 had to be selected between 2000 and 3000

nm. a reasonable selection criterion was set to be the distance between the two farthest fire-

prone areas in the Continental United States. This information was obtained using the

interactive tool available on the Monitoring Trends in Burn Severity (MTBS) website [3].

The two farthest regions in the Continental United States with high burn severity and burn

count from 2000 to 2019 were shortlisted, and the distance between them was measured.

Fig 2.2 Farthest regions in the Continental US with High Burn Count [3]

From Fig 2.2, it was deduced that the two farthest regions with high burn severity and burn

counts are the regions near Seattle and Miami, which translates to a distance of almost 2669

miles. Compensation for some loiter and diversion distance was added, and a minimum

ferry range requirement of 2700 nm was set.

4
2.3 Payload Target Value

Fig 2.3 Payload selection niche plot

The California Department of Forestry and Fire Protection has a recognition guide [4] that

contains specifications of every rotor or fixed wing aircraft being used for firefighting in

the United States, making it a good source for determining which market segment was

vacant with regards to payload capacity.

It was clear from Fig 2.3 that all of the fixed wing aircraft currently have a payload capacity

below 4000 gal or above 10000 gal; therefore, a 6000 gal payload capacity would allow

the AFV-21 to sit in an untapped niche. The Fig did not provide enough valuable

information to decide on top speed. Thus determining a top speed was reserved as an

optimization problem.

2.4 Estimating Market Potential

One of the design objectives in RFP [1] emphasized that manufacturing methods and

materials selections be appropriate for the annual production rate. The estimation of market

potential for the AFV-21 became necessary since the annual production rate depended on

market size and, of course, the market share that was expected to be acquired.

5
 Fig 2.4 Acres Burned by Wild Fires in the US (1990 - 2020) [5]

Establishing a method to determine the number of firefighting aircraft demanded by the

world was made very difficult by the fact that there had never been a Large Air Tanker

built specifically for firefighting and that the United States Forest Service (USFS) relies on

private contractors whose aircraft data was not readily available. The team’s method used

the available United States (US) data and extrapolated it to the world.

Fig 2.4 shows an average of 6 million acres burned each year between 1990 and 2020 in

the US with a LAT fleet size of 24, which suggests that 4.1 LATs covered one million

acres. Table 2.1 uses this metric to calculate the number of LATs needed based on 2020

acres burned data for eight major countries.

Table 2.1 Acres burned and LAT requirement estimates for different countries

Country Acres Burned (Millions) Estimated LATs needed

United States [5] 10.1 41.41
Australia [6] 46 188.08
Russia [7] 45 183.98
Canada [8] 10.33 42.24
Spain [9] 0.0182 0.075

6
 Greece [10] 0.4046 1.65
Indonesia [11] 9.9 40.48
Portugal [12] 0.1111 0.45
Total LATs needed in worldwide firefighting 498.37

Looking at the History of Aircraft Manufacturers in the US, for specialized aircraft such as

bombers and fighters used by government agencies, if a contract is awarded to a

manufacturer, that basically means that the winning company will capture at least 50% of

the market since there are very few customers (Army, Navy, Air force) in that market

segment. This was applicable here since the only ultimate customer for the AFV-21 is a

country’s government, be it directly through its forest service or indirectly through

contractors for the forest service.

A reasonable estimate for the AFV-21 market share was 40% of the LAT market size or

200 aircraft. This estimate was supported by the fact that there were no aircraft currently

being used by the USFS in the selected payload range and also by the prediction that aircraft

currently being used in firefighting would mostly have been retired by the time the AFV-

21 approaches its planned Entry Into Service (EIS) year, i.e., 2030. The prediction is

supported by Fig 2.5, which plots firefighting aircraft by their EIS year and retirement year

based on the assumption that the average aircraft lifespan is 30 years [13].

7
 Fig 2.5 Comparison of EIS and Retirement Years of Firefighting Aircraft

2.5 Production Rate

The selected production rate of the AFV-21 had to match the demand for LATs. An

estimate for demand was made by dividing the number of LATs in fleets of major US aerial

firefighting contractors such as Neptune Aviation by the number of years those contractors

have been operating LATs, giving an average acquisition rate for each contractor.

According to information on their websites, e.g. [14], an average of eight LATs were being

acquired per year by all USFS contractors combined. Forty percent of the US market meant

almost three AFV-21s could be sold in the US after EIS, combined with a conservative

estimate of four sales per year to the rest of the world meant an annual production rate of

seven aircraft.

Production of the JF-17 Thunder (4th generation fighter) by Pakistan Aeronautical Complex

(PAC) was found relevant to the AFV-21, as PAC has produced almost 100 of these aircraft

in 12 years [15] since the launch of production in 2009. This translates to an average annual

production rate of almost eight aircraft. The feasibility of low volume production was

confirmed after observing the similarity in projections for the AFV-21 and the history of

8
the successful low volume production of JF-17, which allowed the team to finalize the

annual production rate of seven aircraft and emulate the manufacturing concepts used in

the production of JF-17 as explained in section 28.

2.6 Conclusion of Market Analysis

Market analysis allowed the finalization of the ferry range and payload capacity targets as

well as the annual production rate for the AFV-21.

Table 2.2 Capability targets finalized after market analysis

Attribute Value
Range 2700 nm
Retardant Payload 6000 gal
Production Rate 7 Aircraft / Year

9
3. FEASIBILITY STUDY FOR INTEGRATION OF AUTONOMOUS

CAPABILITIES

Aviation related incidents are one of the leading causes of death among wildland

firefighters. From 2000 to 2013, 78 firefighters were fatally injured while performing

wildland fire duties involving aircraft [16].

The team attempted to minimize these events by improving and optimizing the design,

structure, and equipment. However, the danger to human lives could only be entirely

avoided by using autonomous or remotely piloted aircraft. This section contains the

feasibility assessment of using RPA for firefighting.

3.1 System Architecture

Several systems and subsystems are integrated into the Remotely Piloted Aircraft System

(RPAS) to make safe flight possible while completing all the required missions smoothly.

Following are the main components of RPAS and a pictorial representation of their

interdependencies;

• Remotely Piloted Aircraft (RPA)

• Remote Pilot Station (RPS)

• Command and Control Link (C2 Link)

• Devices to Interoperate RPA with ATC (Air Traffic Control)

10
 Fig 3.1 RPA System Interfaces [17]

3.2 Required Technologies

Technologies required to operate an RPAS were found to be complex, advanced, and more

in number than those required for a manned aircraft. Some of these technologies were

already available, and some were still in the development phase. Future technologies would

take some time to be available in the market. Based on the progress scale developed by

NASA, i.e., the Technology Readiness Level (TRL) scale [18], critical technological

requirements, their descriptions, and current development levels were determined and

summarized in Table 3.1.

Table 3.1 Summary of roles of several technologies in RPAs, and their TRLs [18]

Technology Need and Development TRL

A high autonomy level of an RPA allows the user to get the
Autonomous maximum benefit as it leads to less human interaction,
Mission reducing the O&M cost for the RPAS operations. The
Management 4
mission management system also allows on-ground
contingency management. Verification and validation of
these systems will be a challenge.

11
 The on-board contingency management system reacts to
unforeseen events and failures. One of the primary
contingencies to be planned for is the loss of link between a
Sophisticated
RPA and its operator. Contingency management for RPAs
Contingency 4
requires sophistication that currently does not exist.
Management
Relatively little development of this capability has occurred
to date, although several promising concepts have been
proposed.
To fly with few restrictions, RPAs require a collision
avoidance system. The intent is to have an “equivalent level
Collision of safety” compared to piloted aircraft. Some work has
2
Avoidance already been done in this area, such as the USAF project for
the evaluation of an avoidance algorithm coupled with an
automatic evasion maneuver.
This technology allows contingency management based on
Intelligent the failed state of the vehicle and is essential for access to air
System Health space by RPAs. Limited health monitoring systems have 5
Monitoring been around for some time, but comprehensive and generic
systems have not been developed due to a lack of funding.
Regarding RPAs, FAA reports indicate that they expect
reliability comparable to a piloted aircraft. One approach is
simply to increase the redundancy of flight systems which
Reliable Flight comes with an initial cost and weight penalty. Another
6
Systems approach would be to add onboard intelligence to recognize
and remedy a failure. Simulations of adaptive flight control
systems have shown promise for many years, and several
adaptive control methods have been tested.
A key technology is the ability to transmit data over the
horizon (OTH). The OTH capability is a “web-based”
network approach to communication for a given mission.
Over-the- Satellite communication links have been used in flight
Horizon operations to provide over-the-horizon communication. 3
Communication However, the concept described here significantly expands
beyond what has been flown to date. Adjustable bandwidths
and a ‘web-based’ use as needed approach are concepts that
still require significant technology developments.

3.3 Weights Trends for RPAs vs. Manned Aircraft

Unmanned aircraft, in general, have been of lower mass than manned aircraft, as indicated

in Fig 3.2. In terms of all-up mass (AUM), manned aircraft range from the smallest single-

12
seater jet, the Bede BD-5J of about 299 kg, to the 640,000kg of the Antonov An-225. RPAs

are on a lower scale, from about 6 kg for the Raphael Skylight, up to the 12,000 kg of the

Northrop-Grumman Global Hawk. Hence the largest fixed-wing RPA is an order of

magnitude smaller, in terms of mass, than its manned counterpart, as per Fig 3.2.

Fig 3.2 Mass Domains of Manned and Un-manned Aircraft [19]

3.4 Conclusion of Study

The significant variance of the AFV-21 from the RPA historical weight trend in Fig 3.2;

coupled with the fact that the TRLs of several important technologies mentioned in Table

3.1 had only matured up to the development and demonstration stage (TRL6) at best;

swayed the team’s decision against the adoption of RPA technology for the current design

and initial productions of the AFV-21. The possibility of inculcating these technologies in

the productions of later years in the program remains open as their TRLs are sure to

improve in the coming years.

13
 4. MISSION DESIGN AND ANALYSIS

In accordance with the requirements outlined in the RFP [1], the aircraft capabilities were

translated into three mission profiles. A common feature was that the aircraft would take

off from and land at an altitude of 5,000 ft MSL elevation on a +35°F hot day in all

missions. The altitudes and speeds in the following profiles were determined in the class 1

design and optimization studies.

4.1 Flexible Operation

The RFP [1] required a design radius with ‘Full Payload’ between 200 to 400 nm. This

statement did not clarify whether the aircraft ‘must’ drop its retardant or not if it is to reach

a design radius in this range. Both meanings could have led to differences in the weight

and, therefore, several other design characteristics depending upon which mission was kept

as the design mission. The team decided to design for the latter possibility with a fuel

capacity that would allow the aircraft to make two sorties with full payload without

dropping it for a design radius of 380 nm, which enabled the AFV-21 to service both

scenarios, i.e., allowed for more operational flexibility.

4.2 Mission Profiles

The ferry mission was designed to allow the relocation of the aircraft across the Continental

United States and was to be executed without any payload and a full fuel tank.

14
 Fig 4.1 Ferry Mission

The Fire Attack mission was a retardant/water payload delivery to a fire 380 nm away from

the airport; all the operations in the mission zone, including the drop maneuver, were

simplified to an overall 20 minutes of loiter for reconnaissance, loiter and drop. After

conducting some research simultaneously with the design’s development, the team found

through regulations from the Interagency Airtanker Board (IAB) [20] that the drop time

itself would last only a few seconds depending upon the coverage level decided by the

operators. However, the mission segment over the fire was set as a 20 minute loiter since

the extended loiter time will allow the AFV-21 to have observation capabilities, i.e., it

could loiter before and after each drop to examine the accuracy of the current drop and plan

for the next drop while also communicating with ground units with regards to planning the

drop. This removed the need for a lead plane that currently performs some of the

aforementioned tasks if a payload of such magnitude is involved.

There are two variations to the ‘Fire Mission’, one with payload drop (Fig 4.2) and one

without (Fig 4.3). The no-drop mission was identical to the mission with payload drop,

except for the drops, as the LAT would return to base with its full payload in each sortie.

15
This made the profile in Fig 4.3 the design mission for the AFV-21 as it was the most

demanding mission.

The AFV-21 would complete two design mission sorties in a single fuel tank. The

possibility of doing four sorties (complete fire line) in a single tank was also explored;

however, it was found that the time saved in refueling would be nothing compared to what

it will cost to operate such a large aircraft.

Fig 4.2 Fire mission with Payload Drop

Fig 4.3 Fire mission without Payload Drop

16
The target design radius of 380 nm was finalized after a design space exploration of

payload and radius constrained by cost and ferry range targets mentioned in Table 2.2. This

is elaborated upon further in the following section.

17
 5. DESIGN SPACE EXPLORATION AND WEIGHT SIZING

The sizing techniques suggested in several design texts were highly statistical in nature,

where weight fractions were determined for different mission segments using the Brequet

range and endurance equations with statistical estimates of almost all the values to be used

in them. The team was skeptical of using these existing approaches since none of the data

applies to firefighter aircraft, as no LAT has yet been purposely built for a firefighting

mission.

It was decided that a high fidelity sizing code, with minimal reliance on statistics, would

be developed where the fuel weight would be calculated by determining the actual drag as

the AFV-21 flies through its mission, and from that, the required thrust level and resulting

fuel flow could be determined for the brequet equations. According to Raymer [21], a

similar approach is followed by highly detailed sizing codes of major airframe companies.

The team found this sizing code to be quite helpful despite the time it took to develop since

it was used throughout the design to quantify the effect of changing power plants and

geometry during trade studies. The code also allowed the team to collect data, perform

sanity checks on the aircraft’s behavior and investigate any abnormalities using values of

performance coefficients (CL/CD) and angle of attack required to generate lift for any

particular point in the mission.

5.1 The Weight Sizing Algorithm

The weight sizing algorithm used several equation sets or components, which have been

visualized in Fig 5.1 and explained in the following subsections:

18
 Fig 5.1 Weight sizing algorithm for fire mission

5.1.1 WE and WTO Statistical Prediction w.r.t Payload

Data for existing airtankers and cargo aircraft was used to find a linear fit equation that

allowed a rough weights estimation for any payload between 4000 and 8000 gal. This

allowed the sizing code to predict how the empty and takeoff weight of the aircraft will

change with the payload requirement and thereby use a more accurate starting value for the

iterative algorithm.

Fig 5.2 WE and WTO statistical prediction curves

19
 𝑊𝑇𝑂𝑠𝑡𝑎𝑡𝑖𝑠𝑡𝑖𝑐𝑎𝑙 = (1.385𝑊𝑃𝐿 ) 𝑊𝐸𝑠𝑡𝑎𝑡𝑖𝑠𝑡𝑖𝑐𝑎𝑙 = (1.133𝑊𝑃𝐿 ) + (9571)
+ (5.772 × 104 )

5.1.2 Empty Weight Fraction

Initially, the empty weight fraction equation convergence method suggested by Raymer

[21] was used to estimate the takeoff and empty weights using fuel and statistical takeoff

weights. Later in the design, component weights methods, such as those in chapter 15 of

Raymer [21] and Part V of Roskam [22], became applicable.

Fig 5.3 Empty Weight fraction trend

Fire mission Ferry mission

𝑊𝑒 𝑊𝑒
= 0.58566 𝑊𝑜−0.011 = 2.9829 𝑊𝑜−0.124
𝑊𝑜 𝑊𝑜
Two sizing codes were developed, one for the fire mission and the other for the ferry

mission, since one had payload and the other did not; this led to different takeoff weights

and the need for separate sizing codes. The empty weight fraction trend lines and equations

are shown in Fig 5.3, where the ferry mission takeoff weight was roughly estimated by

subtracting the payload weight of an aircraft from its maximum takeoff weight.

20
5.1.3 Engine Mathematical Model

The mathematical model for the engine was taken from Howe [23]. General equations of

the engine model are given as follows:

𝑐 = 𝑐 ′ (1 − 0.15𝑅 0.65 )[1 + 0.28(1 + 0.063𝑅 2 )𝑀𝑁 ]𝜎 0.08

𝑇 𝑇
𝑐𝑂𝐷 = 𝑐 [1 + 0.01 ( − 1)] ; 𝑓𝑜𝑟 < 10
𝑇𝑂𝐷 𝑇𝑂𝐷

𝑇 = 𝜏𝑇0

𝜏 = 𝐹𝜏 [𝐾1𝜏 + 𝐾2𝜏 𝑅 + (𝐾3𝜏 + 𝐾4𝜏 𝑅)𝑀𝑁 ]𝜎 𝑠

Where ‘R’ is the bypass ratio, ‘T0’ is the datum sea level static thrust, and ‘c’ is the SFC

calculated for full throttle. ‘cOD & TOD’ are the off-design SFC and thrust which were most

relevant since corresponding to every flight condition, there was usually an off-design

thrust value equal to the drag in that condition, and corresponding to that was an off-design

SFC value which ended up being used in the brequet equations. The values of the constants

in the above equations kept changing with different ranges of altitude and Mach number

and also when the hypothetical engine was replaced with the final engine selection.

5.1.4 Drag Polars

The drag calculation and, thereby, the thrust required and fuel flow for each engine in a

mission segment required the drag polar for that segment. For initial iterations of the

design, where the geometry was unknown, the statistical drag polar estimation methods

given in Roskam [22] were used to determine drag polars for sizing. These were later

replaced with a function that would use the drag buildup approach to give drag polars

specific to the design’s geometry and segment flight condition. The statistical equations

from Roskam [22] are as follows:

21
 𝐶𝑙𝑒𝑎𝑛 ⟹ 𝐶𝐷 = 0.0186 + 0.0415 𝐶𝐿2
𝑇𝑎𝑘𝑒𝑜𝑓𝑓 ⟹ 𝐶𝐷 = 0.0536 + 0.0458 𝐶𝐿2
𝐿𝑎𝑛𝑑𝑖𝑛𝑔 ⟹ 𝐶𝐷 = 0.1036 + 0.0483 𝐶𝐿2
5.1.5 Remaining Equations

Following are a few other basic equations employed in determining the weight fraction and

fuel weight for each mission segment:

𝑉 𝐿 𝑊𝑜 1 𝐿 𝑊𝑜
𝑅= 𝑙𝑛 ; 𝐸= ln
𝑐 𝐷 𝑊1 𝑐 𝐷 𝑊1
1 1
𝐿 = 𝑊 = 𝜌𝑉 2 𝑆𝐶𝐿 ; 𝑇 = 𝐷 = 𝜌𝑉 2 𝑆𝐶𝐷
2 2
5.1.6 Sizing Algorithm Steps

The weight fraction and fuel weight for a particular mission segment were determined

through the following steps:

1. The Engine Mathematical Model function would run in the very beginning to

supply the sizing function with a 70x30000x35 matrix for the off-design or partial

throttle SFC and a 70x35 thrust matrix. The size of the matrices was like that

because they contained cOD and TOD values for all Mach numbers (0:0.01:0.7) and

all altitudes (0:1000:35000).

2. The statistical weight estimate corresponding to the payload supplied to the sizing

function was multiplied with all segment weight fractions prior to the current

mission segment.

3. Once the weight for the current segment was found, the lift coefficient was

determined, and then using the appropriate drag polar, the drag coefficient and

hence the drag for that segment was determined.

22
 4. Since the AFV-21 has two engines, the drag was divided by two. The SFC matrix

was already supplied, so the off-design SFC corresponding to a thrust equal to half

of the drag of the entire aircraft was selected from the available matrix.

5. The SFC of only one engine was used in each segment of the sizing code as SFC

normalizes over the thrust. The L/D and cOD values obtained were used in the range

and endurance equations to get the weight fraction for the current segment.

In this fashion, the weight fractions for all cruise, loiter, and climb segments were

determined, and historical data was used for the remaining smaller segments. Finally, the

statistical takeoff weight was used in conjunction with all weight fractions to get the fuel

expended in each segment, as was suggested by Raymer [21].

Individual segment fuel weights were added up to get the total fuel weight, which was then

used with the statistical takeoff weight in the convergence method based on the empty

weight equations mentioned in section 5.1.2, to get the takeoff and empty weight.

5.2 Design Space Exploration for Design Radius

In order to construct the design space, a 5x5 matrix containing the combinations of design

radii and payloads from the ranges prescribed in the RFP, i.e., payloads set

(4000:1000:8000) and design radius set (200:50:400), was supplied to the sizing code to

determine the corresponding mission fuel, takeoff, and empty weights. Since the payload

was already anchored at 6000 gal, this exploration aimed to anchor the design radius at a

particular value. The design space needed to be constrained to be useful, so the two

constraints used were a ferry range constraint and a cost constraint. Making these

constraints was more data-intensive than making the design space and is explained in the

following subsections.

23
5.2.1 Ferry Range Constraint

The ferry range constraint was that the WF selected for the fire mission must be more than

the fuel required to cover the minimum ferry range of 2700 nm.

The working of the constraint data builder is simple, against each payload (4000:1000:8000

gal), the ferry sizing code is called once to determine the WF, which would correspond to

the ferry mission if the aircraft was being built for that particular payload capacity. Then

at that payload, the fire sizing code is called for the entire design radius range (200:1:400

nm), and the WF for each value is stored. Out of these 200 WF values, the index of the one

with the least difference from the ferry constraint WF is found, and the corresponding

design radius value is recorded. This makes one point of payload and design radius of the

ferry constraint in the entire design space. The process is repeated for each of the five

payloads (4000:1000:8000 gal), and it gives a set of five payload-radius combinations to

plot the red ferry constraint as the lower threshold in the design space.

5.2.2 Operating Cost Constraint Line

According to the Wildland Fire Management Aerial Application Study (2005) [24], based

upon the collective results of analysis of fixed wing airtankers in service, in the desired

specifications of a future airtanker, a positive economic cost benefit was mentioned for an

operating cost rate of $6000 per hour or less (2005 dollars).

Based on inflation and fuel price trends over the years, a conservative estimate of a fuel

constraint of WF = 28000 lb was found. This result was found using the operating cost

module of the software AAA which is based upon Roskam [22]. The five cost constraint

points were determined in the same manner as the ferry constraint to plot the blue cost

24
constraint line as the upper threshold in the design space. Ferry range and cost constraint

data builder algorithms are shown in Fig 5.4.

Ferry Range Constraint Operating Cost (Fuel) Constraint

Fig 5.4 Ferry Range and Cost Constraint Data Builder Algorithms

25
5.2.3 Carpet Plot Generation Algorithm

Fig 5.5 Carpet plot generation algorithm for design radius selection

26
5.2.4 Design Space Exploration Using Carpet Plot

Fig 5.6 is the final carpet plot, made using the algorithm in Fig 5.5, which used the design

space and constraint line generation algorithms of Fig 5.4 and showed the fuel weight (WF)

against each radius and payload combination of the design mission. The reason the Y-axis

was set to WF and not WTO is that the WE and WTO values were obtained using different

empty weight equations in the convergence component for the ferry and fire missions due

to their difference in payload, as discussed in section 5.1.2, which meant that the

consistency of WTO and WE between the ferry line constraint and the fire mission data

points became questionable.

Fig 5.6 Weight Sizing Carpet Plot with Ferry (red) and Cost (blue) Constraints

Since the payload was fixed as 6000 gal in section 2.3, the maximum design radius, which

did not violate the operating cost constraint against this payload, i.e., 380 nm was selected

as the target value.

27
 6. CONFIGURATION DOWNSELECT

Before beginning the design process, it was necessary to select viable configurations for

Class 1 design.

6.1 Concept Generation by Morph Matrix

Table 6.1 Initial Morph Matrix with 367290 Possible Configurations

Syst
Subsystem Option 1 Option 2 Option 3 Option 4
em

Wing location High Wing Low Wing Mid Wing

Wing type Rectangular Tapered Delta Swept Back
Wing

High lift device Slotted Flap Split flap Kruger flap Leading Edge Slot
Strut braced Strut braced
Structural config. Cantilever
(faired) (un-faired)
Tail

Tail Type Conventional Cruciform H-tail T-tail

Engine type Turboprop Turbofan Turbojet Hybrid

Propulsion
System

No. of Engines Single-Engine Twin-Engine Tri-Engine Multi-Engine

Inside vertical Side of fuselage at
Engine Location Top of Wing Under Wing
tail Aft-section
Landing gear Fixed
Fixed (faired) Retractable Partially Retractable
Landing

mechanism (unfaired)
Gear

Landing gear Tricycle (or

Multi-wheel Tandem
type nose gear)
Flight Deck Hybrid
Analog Gauges Glass Cockpit
Fuselage

Layout Glass/Analog
Pressurized Unpressurized
Pressure system
Cabin Cabin
Power
Sub-Systems

Mechanical Hydraulic Fly-by-Optic Fly-by-Wire

transmission
Fuel tank Inside Fuselage Inside Wings External Tank
Payload Store Inside Fuselage External Tank

There are 367290 possible configurations from the above morph matrix; however, most of

them were found to be infeasible from the very start. For example, any combinations with

28
strut-braced structural configuration are impractical due to the high drag penalty at M0.6

and upward.

With the help of competition analysis and reasoning, configurations with obvious problems

and shortcomings with respect to their compatibility with the design mission were

eliminated. The team selected 16 configurations that could deliver on the mission

requirements.

Fig 6.1 Configurations Considered for Class 1 Design

6.2 Concept Selection

From the 16 configurations in Fig 6.1, four were to be selected for class 1 design and

analysis. At this stage, however, exact geometric or performance parameters were not

known for these configurations; therefore, the team relied on literature review and research

29
to list all the Pros and Cons of these configurations based upon which the four

configurations were chosen.

6.2.1 Rejected Configurations

Configuration 1
+ Good stability and − Downwash of 1st
control wing on 2nd
− Increased control
+ High L/D
surfaces complexity
− Certification cost and
time
− Risk of engine
retardant ingestion

Configuration 2
− More expensive due
+ Tail Area Reduction to mid-wing
intersection
− Drag reduction
+ Less interference due
outweighed by
to mid-wing
complexity
− Certification cost and
time

Configuration 3
− Certification cost and
+ High L/D
time
− Less engine
+ Low trim drags
accessibility
− Novel design, many
unknowns

Configuration 4
− Less engine
+ High L/D
accessibility
+ Engines protected − Certification cost
from FOD and time
+ Low skin friction
− Novel design, many
and interference
unknowns
drag

30
 Configuration 6
− Canard/Engine
+ Slightly better L/D
Interference
− Certification cost and
+ Low trim drags
time

Configuration 7
− Overturn risk in
+ Low Trim Drag
main-plane stall
+ Co-Locate Wing
− Landing gear
Spar and Rear
integration
Bulkhead
− CG excursion

Configuration 8
+ Easier ground − Landing gear
logistics integration
+ T-tail out of
downwash and wake − Heavier empennage
regions
− Deep stall risk due to
+ T-tail end plate effect
T-tail

Configuration 9
+ Landing gear
− CG excursion
integration in wing
+ Tail out of
− Certification cost and
downwash and wake
time
regions
+ Combines
advantages of
− Low speed
conventional and T-
performance
tail, i.e., lightweight
and endplate
− Risk of engine
retardant ingestion

Configuration 10
+ Good engine
− Heavy airframe
accessibility
− Complicated Control
+ Low Trim Drag
System
− Roll Yaw coupling

31
 Configuration 12
+ Good stability and − Downwash of 1st
control wing on 2nd
− Increased control
+ High L/D
surfaces complexity
− Certification cost and
+ High AR
time
− Landing gear
integration
− Engine placement

Configuration 14
+ Landing gear
− CG excursion
integration in wing
+ Tail out of
downwash and wake − Engine maintenance
regions
− Deep stall risk due to
+ T-tail end plate effect
T-tail
− Risk of engine
retardant ingestion

Configuration 15
+ Easier ground
− Engine maintenance
logistics
− Unreliable OEI
+ Tail out of
conditions as it relies
downwash and wake
on powered HLDs at
regions
low speeds
− Ducting to counter
+ T-Tail end plate adverse OEI effects
effect is complicated and
expensive
+ Powered HLDs lead
to smaller wing and − Certification cost and
HT, i.e., efficient time
cruise
6.2.2 Summary of Rejected Configurations

For many of the rejected configurations shown in section 6.2.1, a major issue would have

been the FAA certification since many of them involved novel or unconventional aspects

32
to the configuration, which may have been difficult to certify in time, given the EIS and

the dangers involved.

Several configurations such as the tandem wing, joined wing and the canard designs

showed promise for higher aerodynamic efficiency; however, the novelty of these

concepts, i.e., not enough literature being available for them, would have made their design,

development, and certification costly and their response to adverse conditions

unpredictable.

Other reasons for rejecting the designs included; CG excursion, which can become a major

problem after dropping a large payload and would require a larger horizontal tail to

maintain longitudinal stability, i.e., result in more trim drag; Risk of retardant ingestion,

configurations with the longitudinal location of the engines behind the retardant release

hatches carried the risk of retardant ingestion and engine shutdown; Risk of Foreign Object

Damage (FOD) to the engines was a major concern due to the debris and ash over forest

fires, configurations with engines more exposed to these conditions were therefore

penalized.

6.2.3 Accepted Configurations

Configuration 5
+ Engines protected − Wing Flutter
from FOD
+ Light airframe − Engine accessibility
+ Tail protected from
downwash and jet
wash
+ Better lower than
horizon visibility
+ Short and light
fuselage

33
 Configuration 11
+ Weight reduction by − Landing gear
twin tail integration
+ Cheaper and quicker − Heavier than
certification Conventional tail
+ Height reduction of
VT, more hangers
can accommodate
the aircraft
+ Vertical tails in
undisturbed air and
smaller HT through
end plate effect

Configuration 13
+ Engines protected − Engine maintenance
from FOD
+ Tail out of − Deep stall risk due to
downwash and wake T-tail
regions
+ T-tail end plate effect
+ Low CG excursion
+ Co-locate landing
gear and engine
pylon structure

Configuration 16
+ Maintenance and − Higher risk of FOD
accessibility
+ Cheaper and quicker − Landing gear
certification integration
+ Abundant literature
+ available
+ Easier ground
logistics
+ Low CG excursion

6.2.4 Summary of Accepted Configurations

The selected configurations had more pros than cons and were relatively conservative

choices, such that there have been many aircraft of these configurations, and there is plenty

34
of literature available on each of them, making for a shorter development time and cost,

along with a potentially easier time getting FAA certification. These configurations had

better CG excursion characteristics as compared to the novel or unconventional

configurations, which was a major consideration in their selection due to the possible

instability issues that could arise after retardant release if such a large payload was placed

too far from the empty weight CG. Another common contributing factor was shielding the

engines from retardant ingestion which was ensured by the disposition of engines in the

selected configurations.

35
 CLASS 1 DESIGN – PHASE 1
The purpose of class 1 design - phase 1 was to provide an increased resolution or a more

quantitative outlook on the competing configurations by determining the impact of

configuration on performance, to guide one of the most important choices in the entire

design process, i.e., the final configuration selection. Phase 1 contained low fidelity design

methods from Raymer [21] and preliminary design sequence 1 of Roskam [22].

The team was divided such that each of the four members were given a configuration to

take through design phase 1, with the team leader responsible for performance evaluation,

effectively ensuring the removal of any biases.

36
 7. AIRFOIL SELECTION

Airfoil analysis was performed using the vortex-lattice software XFLR-5 and VSPAERO,

the CFD module of OpenVSP. XFLR-5 was used for analyzing low speed characteristics,

and VSPAERO for high-speed characteristics since XFLR-5 cannot predict drag rise due

to local shocks, and OpenVSP cannot predict stall.

Ideal and maximum lift coefficient (𝐶𝑙𝑖 and 𝐶𝐿𝑚𝑎𝑥 ) requirements were determined, where

ideal lift coefficient took into account the cruise condition while maximum lift coefficient

took into account take-off and landing conditions for the aircraft. Required values of 𝐶𝑙 𝑖

and 𝐶𝐿 𝑚𝑎𝑥 were 0.618 and 1.798, respectively. Based on these values, 10 airfoils were

selected from a NASA database [25]. Analysis of each airfoil in XFLR-5 and VSPAERO

allowed construction of the following selection matrices (Table 7.1 and Table 7.2).

37
 Fig 7.1 Drag polar comparison (XFLR-5) Fig 7.2 Lift slope comparison (XFLR-5)

Fig 7.3 Lift to drag ratio comparison (XFLR- Fig 7.4 Moment coefficient comparison
5) (XFLR-5)

38
 Fig 7.5 Drag polar comparison (OpenVSP) Fig 7.6 Lift slope comparison (OpenVSP)

Fig 7.7 Lift to drag ratio comparison Fig 7.8 Moment coefficient comparison
(OpenVSP) (OpenVSP)
Table 7.1 Candidate Airfoil Characteristics
S.No Airfoil Cli CLmax Cdmin Cm0 αstall α0 (deg) (L/D)max Clα Stall

1 653-618 0.8 1.824 0.004 -0.128 23.5 -4.6 192 0.0649 Docile

2 641-212 1.0388 1.843 0.004 -0.0424 19 -1.6235 139.841 0.0894 Moderate

3 642-415 0.62 1.874 0.004 -0.085 20 -3 135.745 0.0815 Docile

4 652-415 0.563 1.809 0.004 -0.062 21 -3 156.657 0.0754 Docile

5 651-412 0.55 1.622 0.005 -0.0809 17.5 -3 136.685 0.0791 Docile

6 2412 0.85 1.997 0.005 -0.054 19.5 -2.0721 152.742 0.0926 Moderate

7 4412 1.05 2.064 0.005 -0.108 18.5 -4.1287 163.451 0.0912 Moderate

8 23012 1 1.958 0.005 -0.01104 19 -1.1486 171.101 0.0972 Sharp

9 23015 0.9 1.944 0.005 -0.011 19 -1.1208 173.592 0.0966 Moderate

10 631-412 0.9 1.943 0.004 -0.0828 18 -3.1171 143.694 0.0920 Docile

39
Once all the parameters to be compared were determined, scoring allowed the selection of

NACA 653-618 as the best airfoil. Although it suffers from a high moment coefficient, it

is also the airfoil with the best low speed and lift characteristics which are far more critical

in firefighting than reducing trim drag.

Table 7.2 Airfoil selection by scoring method

Design Weig 653- 641- 642- 652- 651- 241 441 2301 2301 631-
Objective ht 618 212 415 415 412 2 2 2 5 412
Cdmin 20% 18.0 18.0 18.0 18.0 16.0 16. 16. 16.0 16.0 18.0

Cm0 15% 1.5 10.5 6.0 9.0 6.0 9.0

0 3.0
0 15.0 15.0 6.0

αstall 15% 15.0 4.5 7.5 9.0 1.5 6.0 4.5 4.5 4.5 3.0

α0 10% 10.0 2.0 6.0 6.0 6.0 3.0 9.0 1.0 1.0 6.0

(Cl/Cd)max 10% 10.0 2.0 1.0 4.0 1.0 4.0 5.0 7.0 7.0 2.0

Clα 5% 0.5 4.0 3.0 2.0 2.5 4.5 4.0 5.0 5.0 4.5

Stall 25% 25.0 12.5 25.0 25.0 25.0 12. 12. 2.5 12.5 25.0

Sum
Quality 100% 80.0 53.5 66.5 73.0 58.0 55.
5 54.
5 51.0 61.0 64.5

0 0

40
 8. PERFORMANCE SIZING

The matching plot technique as outlined in Roskam [22] was used to select the following

key parameters:

Table 8.1 Performance Sizing Results

Parameter Value Parameter Value

Thrust to Weight (T/W) 0.3165 Landing Lift Coefficient 3.0
(𝐶𝐿max )
𝐿

Wing Loading (W/S) 80 Max Clean Lift Coefficient 1.7

lb/ft2 (𝐶𝐿max )
Takeoff Lift Coefficient 2.2
(𝐶𝐿max )
𝑇𝑂

FAR – 25 does not have a stall speed requirement, but to enhance the safety attributes of

the aircraft, the team decided to set an aircraft stall speed requirement at 96 kts for landing

configuration, i.e., full flaps deployed. Keeping in view the requirement of a drop speed

between 125 to 150 kts, the team assigned a clean configuration stall speed requirement of

130 kts for safer and easier operation over the fire.

In order to determine the thrust required and wing area, the takeoff weight estimate was

required. The sizing code explained in section 5.1 was used in conjunction with the

statistical drag polars from section 5.1.4 since aircraft geometric sizing was still

downstream in the design.

41
Fig 8.1 Constraint diagram for performance sizing

42
 9. GEOMETRIC SIZING

9.1 Wing

The wings and high lift devices were designed according to the procedures in [21] and [26],

respectively. Although a higher fidelity approach could have been taken at this point to

optimize the lift distribution of the wings by altering the aspect ratio, taper ratio, and twist,

however, the competition was between configurations and not team members; therefore,

trends in [21] were relied upon for all wings in this design phase.

Table 9.1 Wing Geometric Parameters

Conventional H-Tail Low Wing Twin Boom

Wing Area, S 1994.2 ft2
Aspect Ratio,
9.0
AR
Span, b 134 ft
LE Sweep, 𝚲𝑳𝑬 100
Taper Ratio, 𝝀 0.35
Twist, 𝜽𝒕𝒘 -30 -30 -20 -30
Dihedral, 𝚪 -20 -20 50 -20
Vertical
High High Low High
Location
Root Chord, 𝑪𝒓 20.475 ft
Mean Chord,
14.889 ft
MAC
Airfoil NACA 653-618

43
9.2 Fuselage

The fuselage geometry only differs for the twin boom configuration since twin boom

designs have shorter fuselages with lower fineness ratios, while booms take care of the

empennage moment arms.

Table 9.2 Fuselage Geometric Parameters

Conventional H-Tail Low Wing Twin Boom

Length, 𝒍𝒇𝒖𝒔 64.35 ft 47.67 ft
Diameter, 𝒅𝒇𝒖𝒔 8.0 ft 9.0 ft
Fineness Ratio 8.04 5.3
Tail-cone
16.0 ft 11.295
length
Tail Cone
8.530 2.660
Angle
9.3 Empennage

For an initial approximation, the tail volume coefficient method, coupled with the trends

in [21], was used to size the empennage. In cases with two vertical tails, dimensions for

one of them are given. Common factors in tail parameters were that the horizontal tails

were given smaller aspect ratios and larger sweeps than the wing for better stall

characteristics, and the common tail airfoil was thinner than the wing airfoil for a higher

critical Mach number than the wing.

Table 9.3 Empennage Geometric Parameters

Conventional H-Tail Low Wing Twin Boom

HT VT HT VT HT VT HT VT
Aspect Ratio 5.0 1.7 4.0 2.0 5.0 1.0 5.0 1.37
Area (ft2) 922.8 302 876.68 290.69 876.68 286.91 665.03 150.0
Span (ft) 67.93 22.66 59.22 24.11 66.207 16.94 57.66 14.34

44
Root Chord
16.98 17.77 16.92 16.07 13.55 18.82 11.53 10.46
(ft)
LE Sweep (0) 140 250 140 250 150 200 00 250
Taper Ratio 0.6 0.5 0.75 0.5 0.7 0.8 1.0 1.0
Volume
1.0 0.04 0.95 0.07 0.95 0.038 1.0 0.04
Coefficient
Airfoil NACA 0012

45
 10. PROPULSION SYSTEM

It was clear that having a new engine designed and manufactured specifically for the AFV-

21 was not an option since a relatively small number of aircraft were expected to sell

compared to commercial aircraft, even with the most optimistic projections. A comparison

was therein made between existing engines that fulfilled the thrust requirement; to select

the most suitable engine. This was followed by inlet design, capture area calculation, and

nozzle design.

10.1 Thrust Required

The following thrust requirement was found using the results from weight and performance

sizing:

𝑇
= 0.3165 ; 𝑊𝑇𝑂 = 159536 𝑙𝑏 ; 𝑇𝑚𝑎𝑥−𝑖𝑛𝑠𝑡𝑎𝑙𝑙𝑒𝑑 = 50493 𝑙𝑏
𝑊

Since typical thrust losses due to installation are 4% for low bypass and 8% for high bypass

ratio engines [27], a 10% increment was added to the uninstalled thrust requirement to

account for installation losses of a high bypass engine, and any unforeseen growth in the

weight or shortfall in lift capabilities as the design progressed.

𝑇𝑚𝑎𝑥−𝑢𝑛𝑖𝑛𝑠𝑡𝑎𝑙𝑙𝑒𝑑 = 55542 𝑙𝑏 ; 𝑇𝑚𝑎𝑥−𝑝𝑒𝑟 𝑒𝑛𝑔𝑖𝑛𝑒 = 27771 𝑙𝑏

10.2 Engine Selection

Table 10.1 Candidate Engine Parameters

Rated Dry
Engine 𝐒𝐅𝐂𝐜𝐫𝐮𝐢𝐬𝐞 Unit
Manufacturer Thrust Weights Reliability Application
Identification (lb/lb/hr) Cost
(lbf) (lb)
A320/A340/B-
737/ KC-135R /
CFM $10
CFM-56-7B 27300 0.545 5359 9 E-6 Mercury / E-
International million
3D Sentry / DC-8
Super 70

46
 CFM $16 A320neo
LEAP-1A 27120 0.51 6451 8
International million
Pratt & PW1127G1- $14.8
27076 0.482 6300 5 ACJ319/ACJ320
Whitney JM million
A320ceo family
$10.5
IAE V2528-D5 28000 0.574 5721 7 (except A318)/
million
KC-390 / MD-90

Although the CFM-56 came out on top in a weighted scoring table due to lower cost and

better reliability due to its maturity, the team decided to go with the LEAP-1A instead.

Although its per-unit cost would increase the aircraft acquisition cost, it was a one-time

investment that would attract more operators in terms of fuel saved, owing to a lower SFC

and higher Bypass Ratio. The PW1127G1 had the lowest SFC; however, it had more

reports of reliability issues as compared to LEAP-1A, which had minimal reliability issues

since its EIS and had found more widespread acceptance amongst airframe companies [28].

10.3 Engine Characteristics

The mathematical model used to model the LEAP-1A was the one for a high bypass

turbofan engine in Howe [23]. The validity and relevance of this model were corroborated

by the fact that the team was able to recreate the engine performance graphs of the

hypothetical high bypass turbofan engine given in Appendix E of Raymer [21] with

reasonable accuracy, despite the fact that the hypothetical engine performance graphs in

[21] were based upon the engine design text by Mattingly [29], meaning that they were

more physics based. The general equations of the engine model have been discussed in

section 5.1.3; coefficients of these equations were found using data of the LEAP-1A to

give the performance graphs in Fig 10.1.

47
Fig 10.1 Engine Performance Graphs

48
The graphs show a decrease in full throttle SFC with increasing altitude and an increase as

the Mach number increases since less thrust is produced with the same fuel flow rate. The

partial throttle SFC shows little variance for a thrust above 10000 lb; however, at lower

thrust levels, there is a sharp increase in the SFC as the operating condition strays further

from the design condition of the engine. The Partial Throttle SFC increases with increasing

Mach number for a fixed altitude. For a fixed Mach Number, however, the Partial Throttle

SFC decreases with altitude, explaining the high cruising altitude obtained for the AFV-21

in section 19.1.

10.4 Inlet and Diffuser Geometry

Pitot type inlet was the obvious choice since it has virtually 100% pressure recovery and

works extremely well in subsonic and transonic regimes [21]. The nacelle houses accessory

gearbox, power management systems, and fuel delivery systems, and its cross-section was

shaped like an airfoil to minimize drag.

The diffuser, which is the interior portion of the inlet where the flow is slowed down

further, was made as short as possible without exceeding a 10-degree internal angle.

𝐼𝑛𝑙𝑒𝑡 & 𝐷𝑖𝑓𝑓𝑢𝑠𝑒𝑟 𝐿𝑒𝑛𝑔𝑡ℎ = 5.92 𝑓𝑡

10.5 Capture Area Calculation

10.5.1 Engine Mass Flow

A fundamental quantity required to find the capture area is the mass flow of the engine,

which was unavailable for the LEAP-1A; however, the empirical method in Raymer [21]

allowed its estimation. The method's accuracy was corroborated by its close approximation

of the CFM-56 engine’s mass flow which was known.

49
 𝑚𝑒 = 26 ∗ (𝐸𝑛𝑔𝑖𝑛𝑒 𝐹𝑟𝑜𝑛𝑡 𝐷𝑖𝑎)2 = 26 ∗ (6.5)2 = 1100 𝑙𝑏/𝑠

10.5.2 Statistical Estimate of Capture Area

Using a Fig in Raymer [21] relating capture area/mass flow to design Mach number, with

a design Mach of M0.624 at 34000 ft, the capture area was found as:

𝐶𝑎𝑝𝑡𝑢𝑟𝑒 𝐴𝑟𝑒𝑎 𝑓𝑡 2
= 0.0250
𝑚𝑎𝑠𝑠 𝑓𝑙𝑜𝑤 𝑙𝑏/𝑠

𝐶𝑎𝑝𝑡𝑢𝑟𝑒 𝐴𝑟𝑒𝑎 = 0.0250 ∗ 1100 = 27.5 𝑓𝑡 2

10.5.3 Analytical Estimate of Capture Area

For turbofans, the air must slow down to M0.4 before it reaches the fan. So a good

approximation of the Mach number at the inlet was halfway between M0.4 and the cruise

Mach (M0.624), i.e., M0.512. The following equation was used at the throat and engine

front:

3
𝐴 1 1 + 0.2𝑀2
= ( )
𝐴∗ 𝑀 1.2

The throat: The engine front:

3 3
𝐴 1 1 + 0.2(0.512)2 𝐴 1 1 + 0.2(0.4)2
( ∗) = ( ) ( ∗) = ( )
𝐴 𝑡ℎ𝑟𝑜𝑎𝑡 0.512 1.2 𝐴 𝑒𝑛𝑔𝑖𝑛𝑒 0.4 1.2

= 1.3175 = 1.59014

The following ratio was obtained using above values:

𝐴
𝐴𝑡ℎ𝑟𝑜𝑎𝑡 (𝐴∗ )𝑡ℎ𝑟𝑜𝑎𝑡
= = 0.8285
𝐴𝑒𝑛𝑔𝑖𝑛𝑒 𝐴
(𝐴∗ )
𝑒𝑛𝑔𝑖𝑛𝑒

50
Since the engine frontal area was known from its diameter, the capture area was found as:

𝜋𝑑2 𝜋 ∗ (6.5)2
𝐴𝑒𝑛𝑔𝑖𝑛𝑒 = = = 33.18 𝑓𝑡 2
4 4

𝐴𝑐𝑎𝑝𝑡𝑢𝑟𝑒 = 0.8285 ∗ 33.18 = 27.49 𝑓𝑡 2

The capture area estimate in section 10.5.2 supports this result.

10.6 Nozzle Design

The fixed convergent nozzle was used since it is almost universally used for subsonic

turbojet and turbofan engines. The nozzle exit area was set for cruise efficiency, resulting

in a slight loss of performance at lower speeds. However, the simplicity and weight

reduction of the fixed nozzle more than make up for the performance loss [21]. According

to the trends in Raymer [21], the exit area was estimated to be:

𝐸𝑥𝑖𝑡 𝐴𝑟𝑒𝑎 = 0.7 ∗ 𝐼𝑛𝑙𝑒𝑡 𝐴𝑟𝑒𝑎 = 0.7 ∗ 27.5 = 19.244 𝑓𝑡 2

The nozzle length was estimated as 10% of the engine length [21].

𝑁𝑜𝑧𝑧𝑙𝑒 𝐿𝑒𝑛𝑔𝑡ℎ = 0.1 ∗ 10.92 = 1.092 𝑓𝑡

10.7 Summary of Important Results

Table 10.2 Important results from engine selection and nacelle design

Parameter Value
Selected Engine LEAP 1A
Inlet and Diffuser
5.92 ft
Length
Engine Mass Flow 1100 lb/s
Capture Area
27.5 ft2
(Statistical)

51
Capture Area
27.49 ft2
(Analytical)
Exit Area 19.244 ft2
Nozzle Length 1.092 ft

52
 11. AERODYNAMICS

11.1 Lift Calculations

Due to the constraints specified at the beginning of the design phase, the wing design and,

therefore, the lift characteristics were the same for all configurations. The maximum lift

coefficient prediction methods by Raymer [21] and [26] were used to size the high lift

devices (HLDs), and the low speed lift curves were estimated using the methods in

Schaufele [30] which corroborated these predictions.

11.1.1 Maximum Lift – Clean

Two methods were used to estimate the maximum clean lift coefficient, and a conservative

approach was taken by selecting the lower value for performance analyses.

11.1.1.1 Blunt Leading Edge Method[21]

For a high aspect ratio wing with a large airfoil leading-edge radius, the maximum lift

depends mostly upon the airfoil characteristics. Finite wing and sweep effects on the

maximum clean lift were estimated using the following equation;

𝐶𝐿𝑚𝑎𝑥 = 0.9 𝐶𝑙𝑚𝑎𝑥 𝑐𝑜𝑠Λ 𝑐 = 1.719

4

11.1.1.2 DATCOM 1978 Method [26]

The wing CLmax-clean was calculated using the following equation from [26]:

𝐶𝐿𝑚𝑎𝑥
𝐶𝐿𝑚𝑎𝑥 𝑐𝑙𝑒𝑎𝑛 = ( ) . 𝐶𝑙𝑚𝑎𝑥 𝑐𝑙𝑒𝑎𝑛 + ∆𝐶𝐿𝑚𝑎𝑥 = 1.7226
𝐶𝑙𝑚𝑎𝑥

11.2 Maximum Lift – High Lift Devices

11.2.1 HLD System Type Trade Study

There are two types of High lift devices used in modern aircraft:

53
 1. Mechanical High Lift Devices – these are the high lift devices that do not require

any equipment other than the drive system for extending and retracting the flaps.

2. Powered High Lift Devices – these devices rely on the manipulation of the jet

exhaust in different ways to generate high lift coefficients

Initially, substantial work and research were done by two members of the team who were

tasked to explore the feasibility of inculcating powered high lift systems and developing a

separate design that sports an Upper Surface Blowing (USB) or an Externally Blown Flap

(EBF) powered high lift system using the methods in [31] and [32].

The team had high hopes for this branch of the design effort since the conventional

approach, i.e., Mechanical HLDs, led to a large wing compared to the fuselage. Initial

estimates of the powered high lift approach led to a smaller wing and more efficient cruise

due to a higher L/D ratio; however, as the development of this approach progressed and

knowledge grew, it was realized that there were several development issues with this

approach and potential for future operational issues as well, which could lead to the

proposal’s rejection.

It was found that if the wing was downsized to satisfy cruise lift only and a large percentage

of lift for low speed operations was expected from the powered high lift system; the aircraft

would have a wing loading of almost 150 lb/ft2 and a maximum lift coefficient of 5.0 would

be needed to satisfy the stall speed requirement. This was not ideal since a higher wing

loading could only be possible by the use of the HLD system in cruise configuration, which

was not desirable for safety reasons. Furthermore, according to the information in Nicolai

[31], it was found that even with a maximum lift coefficient of 5.0, full advantage of the

capabilities of powered high lift systems could not be taken since EBF and USB systems

54
are capable of 7.5 and 8.0 maximum lift coefficients respectively. Nonetheless, all the

complexities in the development and maintenance of these systems would have to be

accepted with the application of such a system, meaning the outcome would not justify the

effort.

The problems in the above discussion could have been reframed as an optimization

problem had the decision been made to inculcate powered HLDs in the design regardless

of the arguments against it. An optimization study could have led to a good compromise

with a higher maximum lift coefficient and a higher wing loading; however, there was one

major issue that could not be overlooked or the potential issues due to it be overruled, i.e.,

the OEI condition and its implications on safety. Since the aircraft would rely on its high

lift system for a major part of its lift in low speed operations, a question is what happens

when an engine fails? Especially for a firefighting aircraft, OEI is a very real danger over

the fire and ash. The answer to this problem lies in a complicated ducting mechanism [32]

which, in case of an engine failure, supplies part of the thrust from the remaining engine(s)

to the side with the failed engine; apart from this, during the design phase, the control

surfaces and flaps need to be exaggerated to counter this very problem to a large extent

since ducting alone cannot make up for the lost lift and the moments which are generated

therein.

A successful aircraft that employs the blown flap technology is the Boeing C-17

Globemaster which uses EBF and has a high wing loading of 158 psf, but it also has four

engines which make it easier to counter the OEI condition. A two-engine blown flap

aircraft, the YC-14 Clipper was Boeing’s entry into the United States Air Force's Advanced

Medium STOL Transport (AMST) competition in the 1970s; it was a twinjet aircraft and

55
used USB technology. Although the YC-14 met or exceeded AMST specifications in most

cases, it was still not put into production. The team deduced that the OEI condition could

have been at least a factor in its rejection since the competing four-engine EBF design entry

by McDonnel Douglas into the AMST program went on to be enlarged and upgraded into

the C-17 Globemaster under the C-X program. The YC-14’s history served as a warning

to the team and was considered a precedent to potential certification problems and even

rejection.

All of the issues outlined in the above discussion, especially the OEI condition, became the

determining factor in the abandoning of further development of this branch of the design

effort, and a mechanical high lift system was designed for the AFV-21.

11.2.2 HLD System Arrangement

According to the data provided in [31], there are three transport aircraft whose CLmax values

are around 3.0. These aircraft and their HLD arrangements are given in Table 11.1:

Table 11.1 Existing HLD arrangements for a CLmax of 3.0

Aircraft AR CLmax Leading Edge Trailing Edge

Krueger IB, Slats
737-200 8.83 3.05 Triple Slotted Fowler
OB
A321- Double Slotted Fowler + Drooped
9.5 3.2 Full Span Slats
200 Ailerons
DC-9 8.5 2.96 Full Span Slats Full Span Double Slotted Flap
[IB] = inboard, [OB] = outboard

Full span double slotted flap of the DC-9 was rejected to avoid complexities of combined

flap-aileron analysis and the potential problems that full span flaps would bring by creating

a massive lift disparity among the two wings as they are differentially deflected for banking

56
at low speeds. The most complex triple fowler system used in the 737-200 was not selected

due to cost concerns. The arrangement of the A321-200 was chosen for the AFV-21, and

since the CLmax requirement was 3.0, the drooped ailerons were excluded. The double

slotted Fowler flaps gave two advantages, extending the flap increased the lift by increasing

the projected area, shifting the lift curve to the left; the slots increased lift by allowing high

energy airflow from the bottom surface to the top, extending the same lift curve. The flap

alone would cause the wing to stall much quicker; this was countered by including a full

span leading edge slat to delay the stall and, at the same time, increase the lift due to the

increase in the projected area, extending the lift curve and shifting it to the right.

11.2.3 HLD System Design

The high lift device design guide [26] was followed, and the procedure was two-fold. The

Clmax increment caused by the selected high lift devices for the airfoil was determined and

was translated into the CLmax increase for the wing after fine-tuning the area ratios for the

different leading edge and trailing edge HLDs being used.

• Increase in airfoil Clmax by trailing edge HLDs:

Δ𝐶𝑙𝑚𝑎𝑥𝑓 = 𝑘1 𝑘2 𝑘3 (Δ𝐶𝑙𝑚𝑎𝑥 )𝑏𝑎𝑠𝑒 = 1.62

• Increase in airfoil Clmax by leading edge HLDs;

𝑐′
Δ𝐶𝑙𝑚𝑎𝑥𝑠 = 𝐶𝑙𝛿𝑚𝑎𝑥 𝜂𝑚𝑎𝑥 𝜂𝛿 𝛿𝑓 = 0.41382
𝑐
• Increase in wing CLmax by trailing edge HLDs:
𝑆𝑊𝑓
Δ𝐶𝐿𝑚𝑎𝑥 𝑓 = ∆𝐶𝑙𝑚𝑎𝑥 𝑓 𝐾 = 0.92823
𝑆𝑊 Λ
• Increase in wing CLmax by leading edge HLDs:
𝑆𝑤𝑠
Δ𝐶𝐿𝑚𝑎𝑥 𝑠 = Δ𝐶𝑙𝑚𝑎𝑥 𝑠 𝑐𝑜𝑠𝜙𝐻𝐿 = 0.3872
𝑆

57
The result of iterative HLD design is as follows and verifies that the chosen arrangement

of the HLDs is adequate to meet the lift requirements set in performance sizing, chapter 8.

𝐶𝐿𝑚𝑎𝑥𝐻𝐿𝐷 = 𝐶𝐿𝑚𝑎𝑥𝐶𝑙𝑒𝑎𝑛 + Δ𝐶𝐿𝑚𝑎𝑥 𝑓 + Δ𝐶𝐿𝑚𝑎𝑥 𝑠

𝐶𝐿𝑚𝑎𝑥𝐻𝐿𝐷 = 3.034

Fig 11.1 Airfoil and HLDs in landing configuration

11.2.4 Lift Curves

The method outlined in Schaufele [30] was used to estimate the low speed lift curves of

the AFV-21 for the cruise, takeoff, and landing settings. The method took into account

airfoil lift characteristics, wing geometry, and flap geometry and used an amalgamation of

graphical techniques, historical data for jet transport aircraft, and physics to estimate the

lift curves shown in Fig 11.2 for the aforementioned settings.

58
 Fig 11.2 Low Speed Lift Curves

11.3 Drag Calculations

The parasite drag coefficients were determined using the component buildup method of

Raymer [21]. The compressibility drag was accounted for above M0.6 using the data given

in Roskam [22] for jet transports, and the fuselage upsweep drag was estimated from [33].

A MATLAB function was developed so that the drag buildup method could be executed at

any flight condition; this function replaced the statistical drag polars in the sizing code from

section 5.1.4 and was used for the performance analysis in chapter 8 as well as phase 2

optimization studies where aircraft weight sizing was repeated. Below is the parasite drag

buildup of all configurations in the cruise condition.

Table 11.2 Zero lift drag coefficient of all components of the four configurations

H-Tail Conventional Low Wing Twin Boom

Compon Compon Compon Compon
CD0 CD0 CD0 CD0
ent ent ent ent
Wing 0.0072 Wing 0.0070 Wing 0.0080 Wing 0.00721
HT 0.0030 HT 0.0031 HT 0.0030 HT 0.00273

59
 VT 1 0.0008 VT 0.0011 9.5580e VT1 0.00056
VT
-4
VT 2 0.0008 Fuselage 0.0014 VT2 0.00056
Fuselage 0.0013
Fuselage 0.0013 Nacelle 0.00064 Booms 0.0008
1 14 Nacelle 8.3449e
Nacelle Fuselage 0.0015
0.0006 1 -4
1 Nacelle 0.00064
Nacelles 0.0013
2 14 Nacelle 8.3449e
Nacelle
0.0006 2 -4 Upswee
2 Upswee 0.0015
0.0007 p
p Upswee
Upswee 0.0007
0.0007 p 0.00012
p 0.00012 Pylons
Pylons 035
035 0.00012
0.00012 Pylons
Pylons 035 L&P
035 L&P 5%
5% Drag
Drag L&P
L&P 5%
5% Drag CD0 0.0171
Drag CD0 0.0154
CD0 0.0165
CD0 0.0159

Using takeoff and landing conditions in the RFP [1] and the speeds decided earlier in

performance sizing, parasite drag values for these settings were found using the methods

in Raymer [21]. The induced drag for the cruise setting was estimated using the Oswald

span efficiency method, while the impact of HLDs on the induced drag was predicted using

estimates in Part 1 of Roskam [22] to account for the wing’s departure from elliptical lift

distribution caused by flap deployment.

Table 11.3 Drag coefficients of all configurations in clean, takeoff, and landing settings

𝑪𝑫𝟎𝑻𝑶 (𝛿𝑓𝑙𝑎𝑝 𝑪𝑫𝟎𝑳𝑮 (𝛿𝑓𝑙𝑎𝑝

Configuration 𝑪𝑫𝟎𝒄𝒍𝒆𝒂𝒏 𝑲𝒄𝒍𝒆𝒂𝒏 𝑲𝑻𝑶 𝑲𝑳𝑫
= 200 ) = 550 )
Conventional 0.0141 0.03989 0.0688
H-Tail 0.0146 0.04039 0.0693
0.0512 0.0555 0.0580
Low Wing 0.0152 0.04099 0.0699
Twin Boom 0.0151 0.04089 0.0698

60
 12. STABILITY ANALYSIS

A perspective on the stability characteristics of each configuration was also instrumental

in finalizing the configuration for class 2 design. The pitch stiffness, weathercock stability,

and the dihedral effect stability derivatives for the four configurations were obtained by

running a VSPAERO-CFD analysis. The center of gravity for each configuration was

estimated using the weights module in Raymer Design Software (RDS). At this stage in

the design, only the three derivatives mentioned above were checked since they

characterize an aircraft’s inherent static stability;

𝐶𝑚𝑎 < 0; 𝐶𝑛𝛽 > 0; 𝐶𝑙𝛽 < 0

Once the center of gravity for each configuration was specified in OpenVSP, a CFD

analysis was run in VSPAERO, which used the vortex lattice method. The software then

output a data file with longitudinal, directional, and lateral moment coefficients for the

aircraft recorded against the range of angle of attack and sideslip angles specified prior to

the simulation. This data was then used to generate stability graphs for all configurations

in MATLAB, and the slopes of these graphs gave the values for 𝐶𝑚𝑎 , 𝐶𝑛𝛽 and 𝐶𝑙𝛽 .

Table 12.1 Fundamental static stability derivatives for competing configurations (per

radian)

Configuration Conventional H-Tail Low Wing Twin Boom

Cmα -2.698 -2.407 -1.163 -4.085
Cnꞵ 0.049 0.019 0.476 0.006
Clꞵ -0.038 -0.097 0.189 -0.024

61
 13. PERFORMANCE

As the class 1 design - phase 1 came to its end, the sizing code was further refined such

that the engine mathematical model, which used a scaled version of the hypothetical engine

from Raymer [21], was switched to a mathematical model tailored to the LEAP-1A and

the statistical drag polars from Roskam [22] were replaced with a function which used the

drag buildup approach of section 11.3 to supply a drag coefficient for any mission segment

in the sizing code specific to the geometry of the aircraft being analyzed and the flight

conditions of that segment.

For performance analysis of all the configurations, a traditional performance approach such

as the one used in Anderson [34], albeit an excellent introduction to performance, was not

used owing to assumptions such as constant weight throughout the mission, constant drag

polar values during climb and many other simplifications, as the results could have led to

the team making wrong design decisions downstream. Therefore, it was decided to adopt

a physics-based approach in the performance analysis that would utilize the updated sizing,

drag buildup, and engine codes coupled with some numerical methods and iterative

algorithms.

13.1 Criteria 1 – Aerodynamic Efficiency

There is one loiter segment over the fire, and two cruise segments that are long enough to

have a substantial impact on aircraft takeoff and fuel weight, the L/D ratios for these

segments are shown as follows:

62
 Table 13.1 Aerodynamic Efficiencies of Different Mission Segments of four

Configurations.

Configuration (L/D)cruise-1 (L/D)loiter-2 (L/D)cruise-2

Conventional 17.8056 16.5056 17.8017
H-Tail 17.5184 16.3433 17.5103
Low Wing 17.1857 16.1525 17.1724
Twin Boom 17.3505 16.2474 17.3398

13.2 Criteria 2 – Fuel Efficiency

All configurations are taken through the same design mission, and the fuel weight required

by each configuration to complete the design mission is given below:

Table 13.2 Fuel Weight of Four Configurations.

Configuration Conventional H-Tail Low Wing Twin Boom

Fuel Weight (lb) 28852 29933 29160 30397

13.3 Criteria 3 – Ferry Range

The aircraft in a ferry mission is going to fly without its payload as per RFP [1], and

assuming each configuration flies with the respective fuel weight obtained for it in section

13.2, the ferry ranges are:

Table 13.3 Range of Four Configurations.

Configuration Conventional H-Tail Low Wing Twin Boom

Range (nm) 2950 2730 2820 2690

13.4 Criteria 4 – Takeoff Performance

The Takeoff Balanced Field Lengths were calculated using the method in Schaufele [30]

for all configurations and are compared as follows:

63
 Table 13.4 Balanced Field Length of Four Configurations.

Configuration Conventional H-Tail Low Wing Twin Boom

BFL (ft) 5431 5651 5584 5783

13.5 Criteria 5 – Landing Performance

The Landing Field Lengths of the configurations were also calculated using the method in

Schaufele [30] and are compared as follows:

Table 13.5 Landing Field Length of Four Configurations.

Configuration Conventional H-Tail Low Wing Twin Boom

LFL (ft) 2293 2369 2333 2390

64
 14. FINAL CONFIGURATION SELECTION

14.1 FOD Protection Concept Testing

In section 6.2 the configurations with their engines above the wings or fuselage were given

a positive point for protection against FOD. This may certainly hold true in the case of

runways where such configurations may afford the AFV-21 more flexibility with where it

operates out of, as engines placed above the wing will not ingest dirt or debris. However,

this was countered by the fact that the range capabilities for which the AFV-21 was being

designed made it unlikely that it would ever need to operate from unimproved runways.

The team hypothesized before phase 1 that placing engines above the wings would also

protect the engines from the harsh environment over the fire, where strong updrafts

generated by the heat can make tree branches up to 18 inches long airborne, as well as

protect them from the heavy airborne ash which follows the updrafts. This hypothesis was

evaluated in greater detail at this stage.

Fig 14.1 Vector Diagram to test the FOD protection hypothesis

After assuming that the ash and foreign objects over the fire follow the updrafts, the team

found from [35] that the updraft speed spectrum lies from 32 ft/s to 196 ft/s. Fig 14.1 was

made using this data, where 219 ft/s (stall speed in clean configuration) is the loiter velocity

65
over the fire, and it shows that the hypothesized protection will be practically non-existent

since the probability of updraft speeds reaching above 160 ft/s (where the wing shields the

engine), according to [35], is less than 0.01. This led the team to conclude that the earlier

hypothesis was false and precedence such configurations had due to the expected FOD

protection was to be mitigated.

14.2 Conclusion

Conventional dominated all other configurations in performance analysis owing to its

lower wetted area. In stability analysis, other configurations would beat the conventional

in one of the three derivatives but then have undesirable values in the others, whereas

conventional had moderate yet stable values for all derivatives. The twin boom and low

wing had to be rejected based on the concept testing results of section 14.1 since their

perceived advantage of protecting the engines from FOD had vanished, yet their problems

such as engine accessibility (maintenance), flutter, and deep stall remained. The H-tail was

dropped due to subpar performance as compared to the conventional configuration.

Although conventional is considered by many designers as 𝑝𝑎𝑠𝑠𝑒́ , and was not as exciting

as the twin boom configuration, the results of phase 1 were a testament to why conventional

configuration dominates the market in almost every class of aircraft, especially transport,

and so the conventional configuration was finalized.

66
 CLASS 1 DESIGN – PHASE 2

The aim of class 1 design - phase 2 was to complete all aspects required to begin class 2

design, which was based on preliminary design sequence 2 of Roskam [22]. The class 1

methods in Roskam [22] were coupled with some higher fidelity physics based methods of

Sadraey [36] to optimize the class 1 design’s geometry to the extent that the team was sure

that too many adjustments would not be required in class 2 design since those calculations

were much more extensive, time-consuming, and would most likely make it infeasible to

come back too many times in the given time frame.

67
 15. WING OPTIMIZATION

The wing geometry from Table 9.1 was changed and finalized to that shown in Fig 15.1 by

virtue of the optimization studies discussed in this chapter.

Fig 15.1 Finalized Wing Geometry

15.1 Aspect Ratio

The trade-off regarding aspect ratio is simple; a large aspect ratio leads to good

aerodynamic performance; however, it also leads to a heavier reinforcement structure near

the root to support the larger bending moments caused by a larger wingspan. On the other

hand, a smaller aspect ratio is aerodynamically less efficient; however, it also requires a

lighter structure. An optimal point between these two scenarios was found by including the

component weight estimation equation from Raymer [21], which pertained to the wing

geometry, in the sizing algorithms to determine this optimal aspect ratio in an iterative

fashion. Fig 15.2 shows an optimal aspect ratio of 7.6, lower than the 9.0 aspect ratio set at

the beginning of the design.

68
 Fig 15.2 Aspect Ratio Optimization Fig 15.3 Semi Span Lift Distribution

15.2 Taper Ratio, Incidence and Twist

Twist and taper ratio have a significant impact on the lift distribution of the wing. An

elliptical distribution was highly desirable since it provides safety from an outboard stall

and produces the least induced drag and smaller bending moments, simplifying the spar

design. A code based on the lifting line theory and the guidelines in Sadraey [36] was used

to optimize the lift distribution of the wing by changing the aforementioned parameters.

Another important adjustment was made to the incidence angle ‘iw’ to ensure that the wing

produced the required cruise lift coefficient of 0.4677 without needing fuselage angle of

attack. The following values were finalized since they enabled the wing to approximate an

elliptical lift distribution, as shown in Fig 15.3, and achieve a cruise lift coefficient of

0.4724, which slightly exceeded the requirement.

𝜆 = 0.45 ; 𝜃𝑡𝑤 = −30 ; 𝑖𝑤 = 2.30

69
 16. WEIGHTS AND BALANCE ANALYSIS

The class 1 weights analysis was done in order to position the main landing gear behind

the aft most CG position and to aid in sizing the horizontal tail by tracking CG changes

with HT area changes. Empty weights of major component groups were estimated using

the fractions method, where weight fractions for component groups of similar aircraft were

averaged and then multiplied by the gross weight of the AFV-21. CG locations of

components were estimated using the information in Roskam [22] in conjunction with the

mass properties module of OpenVSP. The excursion diagram showed the aft most CG at

26.33 ft, and the landing gear was placed well behind it at almost 30 ft; as is shown in the

following chapter, this larger gap was made necessary due to other constraints to be

satisfied by the Main gear such as the tip over and ground clearance constraints.

70
 17. LANDING GEAR

A pugh matrix was used to select an appropriate landing gear configuration from the

tricycle, tail-dragger, and tandem. The tricycle landing gear configuration was found to be

the best one for the AFV-21 due to good over-nose visibility, steering characteristics, and

stability against ground loops.

Table 17.1 Landing gear longitudinal characteristics

𝒙𝒄𝒈 𝒙𝒏𝒈 𝒙𝒎𝒈,𝟏 𝒙𝒎𝒈,𝟐 𝒍𝒎 𝒍𝒏 𝝓𝒈 𝜽

𝒂𝒇𝒕

AFV-21 25.08 ft. 9 ft. 26.5 ft. 30.09 ft. 3.2135 ft. 16.08 ft. 21.16° 17°

As shown in Fig 17.1 and Fig 17.2, the longitudinal and lateral clearance criteria are

satisfied by angles larger than 150 and 50 respectively, as suggested in Roskam [22]

Fig 17.1 Longitudinal Landing Gear Fig 17.2 Landing Gear Lateral Clearance
Geometric Parameters & Loads Requirement.
Static and dynamic loads acting on the main wheel and nose wheel shown in Table 17.2

were calculated using equations from Roskam [22]. Nose wheel tires were sized to bear the

maximum dynamic load, whereas the main wheel tires were sized for the maximum static

load.

71
Table 17.2 Landing gear loads Table 17.3 Strut design parameters

Loads/Weight Value Main Nose

Gear Gear
𝑾 153284 lb
𝑵𝒈 Landing gear load factor 1.7
𝑷𝒎 𝒔𝒕𝒂𝒕𝒊𝒄 31435 lb
FAR 25 vertical touchdown
𝑷𝒏 𝒔𝒕𝒂𝒕𝒊𝒄 19519 lb 𝒘𝒕 12 ft/sec
rate
𝑷𝒏 𝒅𝒚𝒏𝒂𝒎𝒊𝒄 29457 lb Tire energy absorption
𝜼𝒕 0.47
efficiency
Strut energy absorption
𝜼𝒔 0.80
efficiency
𝑺𝒔 Shock strut stroke 1.5847 0.9305
𝒅𝒔 Shock strut diameter 0.4842 0.6478

Tires were selected from the Goodyear tire book [37]. Out of the type VII and radial tires

that could sustain the loads in Table 17.2, radial tires were chosen; due to their thinner

sidewalls and lower aspect ratios, radial tire usage is increasing in modern aircraft and will

continue to do so, according to Currey [38]. From a performance perspective, radial aircraft

tires behave mostly like other tires in response to variations in vertical load and yaw angle

[38]. Four tires were used for the main gear, i.e., two on each fuselage side, placed

longitudinally one after the other, and each tire with its own strut. Selected tires and their

specifications are shown in Table 17.4.

Table 17.4 Tire selections and specifications.

Outer Max.
Width Pressure Loaded
Tire Diameter load
(in) (psi) Radius (in)
(in) (lbs)
Main
36x11.0R18 36.75 10.92 53,700 305 15.90
Gear
Nose
27.75x8.75R14.5 28.68 9.19 31,175 330 12.30
Gear

72
Oleo shock absorbers were selected due to their high shock absorption efficiency (𝜂𝑠 ) or

the efficiency in transferring loads to airframe structure upon landing. The number of tires

per strut was one for the main gear and two for the nose gear. The main gear strut location

was set under the fuselage. Sizing (of struts) was done according to equations in Roskam

[22] shown below:

𝑊
0.5 ( 𝑔𝐿 ) (𝑤𝑡2 )
( 𝑛𝑃 𝑁 ) − 𝜂𝑡 𝑠𝑡
𝑠 𝑚 𝑔

𝑆𝑠 = → 𝑠𝑡𝑟𝑜𝑘𝑒 𝑜𝑓 𝑠ℎ𝑜𝑐𝑘 𝑎𝑏𝑠𝑜𝑟𝑏𝑒𝑟

𝜂𝑠
𝑑𝑠 = 0.041 + 0.0025√𝑃𝑚 → 𝑠ℎ𝑜𝑐𝑘 𝑠𝑡𝑟𝑢𝑡 𝑑𝑖𝑎𝑚𝑒𝑡𝑒𝑟

The retraction mechanism of the AFV-21 takes after the one used by the C-130, i.e., each

main landing gear leg travels straight up into the fuselage using a gear train that connects

it to a gearbox operated by a hydraulic motor. The main landing gear draws its strength

from landing gear beams, which allow load transfer to the wing’s rear spar. Panels are

closed over the main gears once they are retracted into the landing gear bays. The nose

landing gear retracts in the forward direction, common to several other aircraft, in order to

facilitate deployment by ram air in case of complete hydraulic system failure.

Fig 17.3 defines the lateral tip over criterion, ψ, as being no larger than 55 degrees for the

most forward center of gravity. An appropriate lateral spacing was selected to meet this

requirement.

73
 Table 17.5 Landing gear lateral

stability.

AFV-121 Requirement
𝝍 46° ≤ 55°
𝒀𝒈𝒆𝒂𝒓 ±4.70 ft.
Fig 17.3 Lateral tip-over criteria.

74
 18. HORIZONTAL AND VERTICAL TAIL SIZING

The horizontal and vertical tails were sized using the X-plot method in Roskam [22]. The

VT area was then increased to cater to the OEI condition. Fig 18.3 shows the finalized HT

and VT geometries.

Fig 18.1 Longitudinal X-Plot Fig 18.2 Directional X-Plot

Fig 18.3 Finalized HT (Left) and VT (Right) geometries

18.1 Horizontal Tail

The area of the HT was determined using the X-plot shown in Fig 18.1. It features two

legs; the Center of Gravity leg and the Aerodynamic Center leg:

1. The CG leg represents the rate at which the CG moves aft as a function of HT area.

2. The AC leg represents the rate at which the AC moves aft as a function of HT area.

75
The CG leg was determined using class 1 weights and balance analysis with a twist where

the weights of the majority of components which would not be changed in this analysis

were estimated using the same weight fraction method described earlier, while the weights

of both HT and VT were determined using the following equations obtained from

Torenbeek [39]. This approach allowed tracking CG changes with changes in tail areas to

develop the CG leg while simultaneously avoiding potentially inaccurate weight estimates

for other components.

𝑆𝐻0.2 . 𝑉𝐷
𝑚𝐻 = 𝑘𝐻 𝑆𝐻 (62. − 2.5) ; 𝑚𝑉
1000√𝑐𝑜𝑠𝜑𝐻,50

𝑆𝑉0.2 . 𝑉𝐷
= 𝑘𝑉 𝑆𝑉 (62. − 2.5)
1000√𝑐𝑜𝑠𝜑𝑉,50

The change in aircraft aerodynamic center with change in HT area was tracked using the

following equation from Roskam [22]:

𝑑𝜖ℎ 𝑆ℎ ̅
{𝐶𝐿𝛼 (1 − ) ( 𝑆 ) 𝑋𝑎𝑐ℎ }
ℎ 𝑑𝛼
[𝑋̅𝑎𝑐𝑤𝑏 + ]
𝐶𝐿𝛼
𝑤𝑏

𝑋̅𝑎𝑐𝐴 =
𝐹

𝑑𝜖ℎ 𝑆ℎ
{𝐶𝐿𝛼 (1 − ) ( 𝑆 )}
ℎ 𝑑𝛼
𝐹 = [1 + ]
𝐶𝐿𝛼
𝑤𝑏

The HT area against a static margin of 10% was selected for AFV-21 to ensure inherent

stability, where the static margin is defined as:

𝑑𝐶𝑚
= 𝑋̅𝑎𝑐 − 𝑋̅𝑐𝑔 = −0.10
𝑑𝐶𝐿

76
This area turned out to be 600 ft2, and the HT apex was fixed to be 49.0 ft from the nose

after several iterations.

18.2 Vertical Tail

The VT area was initially determined using the lateral X-Plot shown in Fig 18.2. The 𝐶𝑛𝛽

leg was the only one needed in the plot and was determined using the following equation,

where the coefficients were determined using the methods in [40].

𝑆𝑉 𝑋𝑣
𝐶𝑛𝛽 = 𝐶𝑛𝛽 + 𝐶𝐿𝛼 ( )( )
𝑤𝑏 𝑉 𝑆 𝑏

Assuming the VT has to meet the directional stability criteria 𝐶𝑛𝛽 = 0.0010 per degree,

the VT area required was determined to be 226 ft2. The rudder deflection required to

counter engine out yawing moment (𝑁𝑡𝑐𝑟𝑖𝑡 ) and the windmilling engine drag moment 𝑁𝐷

was given by the following equation, and it was recommended by Roskam [22] that rudder

deflection not exceed 250.

(𝑁𝐷 + 𝑁𝑡𝑐𝑟𝑖𝑡 )
𝛿𝑟 =
𝑞̅ 𝑆𝑏𝐶𝑛𝛿𝑟

Lateral engine location was arbitrarily set in class 1, however, after some iterations using

the above equation, the engine’s lateral location on the wing was reduced from 20 ft to 13

ft from the fuselage center, and this was not reduced any further than necessary in order to

avoid any unforeseen issues for the horizontal tail downstream.

With the earlier selection of 226 ft2, the rudder deflection required to satisfy the OEI criteria

at takeoff conditions was 26.150, so the VT area was increased to 250 ft2 which satisfied

the OEI criteria at 23.40.

77
 19. CONDITIONS FOR OPTIMAL PERFORMANCE

The speeds and altitudes shown in the mission profiles in section 4.2 were obtained using

the optimization studies discussed here. This section only shows the final run of the

optimization algorithms which used the finalized geometry obtained at the end of Class 1

Design Phase 2; however, these codes were run multiple times throughout the design

process whenever the design or mission profile was changed. The aim of these studies was

by no means to constrain the operator to a handful of values that make up a rigid single

mission profile; instead, it was to determine the range of speeds and altitudes the AFV-21

can operate without serious loss in performance and capability while simultaneously

estimating the deterioration in these attributes in case of a large deviation from the optimal

values.

19.1 Cruise Altitude

The optimal altitude was found by running the fire mission sizing code at all possible

altitude and cruise speed combinations and determining the time to fire and fuel spent for

each combination.

78
 Fig 19.1 Fuel optimization Fig 19.2 Time optimization

The optimal point of Fig 19.1 (34000 ft and 610 ft/s) was selected as the cruise condition

as opposed to the optimal condition of Fig 19.2 (15000 ft and 750 ft/s); there were several

reasons for this decision. Firstly, the optimal condition of Fig 19.2 sat at the corner of the

variable ranges considered in the optimization for time which meant that it was not really

an optimum flight condition, i.e., a lower time could be achieved by flying lower and faster

which would be an impractical choice due to high fuel costs, FAA limitation on speed (250

kts) and structural limitation on speed at lower altitudes as shown in section 26.3. Secondly,

the time to fire is relatively insensitive to the altitude compared to fuel weight sensitivity

to altitude, which prompts the selection of cruise altitude for minimum fuel required.

Finally, a comparison of both Figs shows that flying at the optimum time condition of Fig

19.2 would cause the fuel required to increase dramatically; however, flying at the optimum

condition of Fig 19.1 would still give a reasonable time to fire of 65 minutes since the

optimum fuel speed of 610 ft/s is a high speed.

19.2 Speeds of Major Segments

There are four legs of the design mission that are sizable enough to significantly impact

fuel weight and, therefore, merit optimization. Initially, a brute force method was used

79
where all possible combinations of the four speeds were being run simultaneously in a

nested loop to find the combination with the lowest fuel weight; this had an estimated run

time of several days, so the algorithm was divided into optimizers which ran combinations

of two speeds per optimizer. These optimizers were looped to run in succession until the

fuel weights against the optimal speeds matched. This approach reduced the runtime to a

few minutes.

Fig 19.3 Cruise Speeds optimization Fig 19.4 Loiter Speeds optimization

19.2.1 𝑽𝒄𝒓𝒖𝒊𝒔𝒆 𝟏 and 𝑽𝒄𝒓𝒖𝒊𝒔𝒆 𝟐 Optimization

Fig 19.3 shows that the region of 550 to 650 ft/s for both cruise speeds has low fuel

consumption gradients and allows for operational flexibility at lower costs. The optimal

speeds for cruise 1 and cruise 2 are 610 and 600 ft/s, respectively.

19.2.2 𝑽𝒍𝒐𝒊𝒕𝒆𝒓 𝟏 and 𝑽𝒍𝒐𝒊𝒕𝒆𝒓 𝟐 Optimization

Fig 19.4 shows that the entire operational region specified for Vloiter 2 has low consumption,

which means that it has less impact on mission performance; the optimal value is 470 ft/s.

For Vloiter 1, the optimal region is from 280 to 300 ft/s and beyond; however, the constraint

80
of operating under 150 knots or 253 ft/s over the fire as per the RFP [1] refrained the team

from considering the optimal values for this segment.

81
 CLASS 2 DESIGN

20. SUBSYSTEMS

Several combinations and options for each subsystem were evaluated for selection and then

placed appropriately in the airframe. The aim behind the detail in the subsystems selection

and placement was to get an accurate estimate of aircraft empty weight in the Class 2

weight estimation techniques and adjust the CG of the aircraft using the placement of the

subsystems in the aircraft for desirable stability characteristics.

20.1 Fuel System

The fuel system is primarily responsible for storing the fuel and supplying it to the engines

and the APU. It was decided that fuel tanks in the AFV-21 will reside in the wings but not

in its center section. The main reason for this was that major structural members which are

responsible for the strength of the fuselage and wing structures, pass through the upper

fuselage; the landing gear retraction system uses landing gear beams to draw their strength

from this very region; which makes the placement of fuel tanks here hazardous since a hard

landing might lead to fuel tank rupture. This coupled with the routing of several other

systems such as the pneumatic, flight control, and anti-ice systems through the center

section region, makes fuel tank placement there problematic and potentially dangerous;

also, since the wing is relatively large as compared to the fuselage so fuel volume allocation

in the wing is a non-issue.

There are four primary tanks, two inboard tanks which can be described as collector tanks

and two outboard tanks, which contain most of the fuel. These tanks are placed between

the front and rear spars to avoid interference with the flight control, anti-ice, and pneumatic

systems. The fuel required 563 ft3 of volume, which was increased to a volume of 590 ft3

82
among the four tanks to allow for ullage space. Surge tanks were included to cater for a

decrease in fuel density while the tanks were full and to capture any fuel that may enter the

vent lines to avoid spillage. Vent tanks were also included to hold fuel that is about to be

jettisoned in case of an emergency. The surge and vent tanks were sized using historical

data.

Since the aircraft will spend considerable time at high altitudes during cruise, a closed vent

system would prove impractical for an aircraft of this size; therefore, an open vent system

was used to connect ullage space to outside air to avoid the buildup of large pressure

differentials. Recognising the risk of exposure over the fire, the FAA guideline of 9%

oxygen concentration for military aircraft at sea level will be observed by the AFV-21

using an On Board Inert Gas Generation System (OBIGGS), which will supply nitrogen

enriched air to the Flammability Reduction System (FRS) to fill the ullage space. The

collector tanks for both engines will be equipped with two booster pumps each to add

redundancy. A crossfeed valve will be included downstream of the booster pumps to allow

the feeding of both engines using a single collector tank if necessary. The outboard tanks

will be further compartmentalized into two tanks to counter the engine rotor burst scenario.

Fuel jettison will be achieved using jettison masts at the rear of wingtips. Inboard and

outboard transfer valves will be included to reduce the wing bending moment.

20.2 Flight Control Systems

There were several candidate actuator systems whose pros and cons were evaluated to

make the final selection for the primary flight controls, i.e., pitch, roll and yaw. Although

most modern jet transports rely on electro-hydraulic actuators powered by two to three

central hydraulic systems, the AFV-21 was a different design with a much smaller fuselage,

83
which meant that installing such a system would lead to volume scarcity for other

subsystems, linkages, and consequently, difficult maintenance. A way around this issue

was using Integrated Actuator Package (IAP) actuators for primary flight controls. The

benefit of using an IAP is that it is a self-contained unit in which a motor drives a hydraulic

pump which then drives the hydraulic actuator, all within a singular unit. A major issue

with IAPs is that the hydraulic pump in them is always running, even when no control is

needed. The final selection solved this problem, i.e., the Electro-Hydrostatic Actuator

(EHA); it uses the three-phase AC power, to power the drive electronics which in turn drive

a variable speed pump together with a constant displacement hydraulic pump. Power is

only drawn from the aircraft buses while the actuator is moving, resulting in substantial

energy savings. EHAs use microprocessors for signal and controls, allowing digital

interfacing to flight control computers through data buses. Four EHAs were assigned to

ailerons, four to elevators, and three to the rudder to attain redundancy. For Flaps and Slats,

there were two choices, conventional screw jack and Electro-Mechanical Actuator (EMA);

the EMA was selected since it was modern, light, and did not need a central hydraulic

system meaning that several of them will be used for each flap and slat to achieve

redundancy.

A full mode, quadruplicated Fly By Wire (FBW) [42] system, was selected for linkage

with no mechanical fail safes. The FBW system will use a primary flight computer to send

electronic signals relayed from the pilot controls to the control surfaces instead of the

conventional method of using pulleys and long lengths of cable to actuate the control

surfaces. There are several reasons why the FBW system was a better choice than the

conventional flight control systems. On the upside, the benefits outlined by Boeing for

84
making the 777 a FBW system were assured: weight reduction, integration of several

systems, better handling of the aircraft, ease of maintenance, ease of manufacturing, and

greater flexibility with changes and integration of new systems into the aircraft [41]. On

the downside, the cost of the system was significantly increased; however, savings from

reduced weight when the aircraft is in service coupled with the positive traits of the FBW

system just described will outweigh the extra flyaway cost.

20.3 Engine Controls

The AFV-21, like most modern aircraft, will utilize a Full Authority Digital Electronic

Control (FADEC) unit to control the engine and all of its functions. FADEC works on the

basis of various input and output parameters. The input signals provide information from

the aircraft and the engine to be used in control algorithms, while the output signals can

perform a control function [42]. The pilot will define thrust requirement through a

throttling lever; in simple engine control systems, it is connected to the Fuel Control Unit

(FCU) through rods and wires, while FADEC will replace these linkages with electrical

signalling, which will reduce friction and eliminate the possibility of jamming in the

control circuit [42]. Digital integration and automatic control of engine inputs and outputs

will make the AFV-21 safer while removing the possibility of data overload for the pilot.

20.4 Hydraulic System

The hydraulic system was laid out to power the utility systems of the aircraft, namely the

landing gear, nose wheel steering, retardant release hatches, and wheel brakes. Two

identical hydraulic systems were incorporated for redundancy, both of which power the

same utility systems mentioned earlier. One is the main hydraulic system and the other is

the auxiliary system. The auxiliary system will only operate if the main system fails. Both

85
systems will feature separate reservoirs, pumps, and corrosion resistant steel piping. The

systems will be powered by variable displacement piston pumps mounted on the accessory

gearbox on the engine casings. Engine speeds will be geared down in the accessory gearbox

to drive the pumps. The systems will operate under a working pressure of 4000 psi to

minimize components’ mass and volume while utilizing Skydrol 5 as the working fluid. A

DC pump deriving power from the APU will be connected to both reservoirs, and

accumulators will be connected to the wheel brakes and landing gear retraction system to

allow the aircraft to land safely in case of a dual engine and hydraulic system failure.

20.5 Electric Systems

The electric system will power various motors, actuators, avionics systems, sub-system

controllers, cockpit and external lighting. The electric system will comprise two Integrated

Drive Generators (IDGs) driven by their respective engines, i.e., left and right engines.

IDGs will be connected via two Bus Tie Breakers (BTBs) to cater to one generator failure.

If both IDGs fail, the tie breakers will also be connected with the auxiliary power unit

(APU).

To convert AC voltage from both the IDGs to DC, left and right transformer rectifier units

(TRUs) will be used. Both the DC buses will be closed through DC ties to account for

failure. DC power will be stored in the main aircraft battery. An Inverter will be integrated

with the main aircraft battery to convert DC voltage to AC. APU battery will also be added

to ensure that an acceptable level of power is available for emergency scenarios with a

capacity capable of handling electrical loads for 30 minutes following a primary power

loss. As electric loads of the AFV-21 are smaller than other aircraft of the same category,

86
i.e., no passenger cabin and de-icing is being done through pneumatic systems, so Ram Air

Turbine (RAT) or Permanent Magnet Generator (PMG) for backup will not be included.

20.6 Pneumatic System

Bleed air from engines and APU will be supplied to the different systems after regulating

the high pressure air through Pressure Reducing Shut-Off Valves (PRSOV). Air from APU

will be used to start both engines making the aircraft a self-contained system capable of

operating in areas with minimal equipment; cross bleed capability will also be included to

start an engine if only one is running or if one engine flames out mid-flight.

An anti-ice system will be included in the aircraft since the possibility of its operation in

cold environments at high altitudes during missions like relocation cannot not be ignored.

The anti-ice system will supply hot air from the pneumatic system to the inner surfaces of

the leading edges and fan cowl, using telescoping ducts and piccolo tubes.

Other systems to be connected to the pneumatic system are windscreen ice/rain protection

and cockpit environmental control system. Backup hydraulic pressurization pumps will

rely on DC power from the APU instead of the pneumatic Air Driven Pumps (ADP) since

the primary hydraulic pumps already rely on the engines, and the backup pump would

become inoperative in case of dual engine failure since the pneumatic system relies on

bleed air from the engines.

20.7 Environmental Control System (ECS)

Since the aircraft cruises at 34000 ft, there is an inherent need for an ECS to regulate

pressure, temperature, and oxygen levels inside the cockpit. The pilots have to be kept in

comfortable conditions so that they do not lose their functional efficiency. Oxygen masks

87
will be avoided to increase comfort for prolonged operation by pressurizing the entire

cockpit.

A single air conditioning unit will draw air from the pneumatic system, refrigerate it, and

supply it to the pilot cabin. The cockpit exhaust air will be used to cool the avionics and

equipment bays as they can operate in much higher temperatures than can be tolerated by

the pilots. Bleed air acclimatized by the air conditioning system will be used for cockpit

pressurization. To address the oxygen demand, a pair of diluter-demand automatic

pressure-breathing regulators will be used in combination with LOX tanks located behind

the cockpit, with an individual capacity of 25 liters providing up to 96 man-hours of oxygen

to each pilot.

20.8 Avionics

The AFV-21 follows suit with major manufacturers like Airbus and Boeing in the selection

of avionics systems, i.e., all avionics equipment will not be ordered from the same

manufacturer since avionics from a single manufacturer could potentially have the same

problem in different avionics modules. If different modules come from different

manufacturers, they are less likely to fail simultaneously or exhibit the same issues. For

example, Boeing 777 uses Collins Aerospace and Honeywell for Imaging and visual

Systems; and uses BAE Systems, Meggitt Avionics, and Thales Avionics for Indicators

and Instruments [43].

In order to avoid the design difficulties and maintenance problems associated with

traditional federated architectures, the AFV-21 will use an Integrated Modular Avionics

(IMA) system as its Flight Management System (FMS). IMA uses a centralized processor,

which reduces the communication needed for its systems and, therefore, the weight and

88
complexity of the system [44]. Contrary to traditional federated architectures, the IMA

concept proposes an integrated architecture with application software portable across an

assembly of common hardware modules. Airbus reports that IMA approach cuts the part

numbers of processor units for the new A380 avionics suite in half, while the A380 Super

Jumbo, which uses an IMA system, touts 15 to 20 percent lower operating costs than

previous airliners [45]. The AFV-21 will use Rockwell Collins Pro Line Fusion (IMA

system) since the Pro Line Fusion is suitable for smaller aircraft such as the Airbus A220

and the Bombardier Global 5000, which have similar takeoff weights.

Avionics systems of the A220 were used to create a tentative avionics equipment list. The

navigation package by Honeywell makes the aircraft capable of VFR and IFR flight [46].

It is worthy of note that the avionics systems come in packages, and therefore some overlap

with the same manufacturer between the different systems can be seen in Table 20.1.

Table 20.1 Avionics Systems List

Category Systems Manufacturer

Communication Integrated Cockpit Control Esterline Control &

Panels Communication Systems
Radio Communications (Korry)
Collins Aerospace
Communication
Equipment Communications system
Communication Communication Antennas HR Smith Antennas

Flight and Data Flight Control Computer Collins Aerospace,

Management and FBW system Avionics Division
Flight and Data Cockpit Voice Recorder L3 Aviation Recorders
Management (CVR)
Imaging and Visual MultiScan weather radar Collins Aerospace,
Systems Avionics Division
Indicators and Instruments PilotView aircraft network CMC Electronics Inc.
switch

89
 Indicators and Instruments Head-Up Guidance System Collins Aerospace,
Avionics Division
Indicators and Instruments Air Data Computers Collins Aerospace, Sensors
& Integrated Systems
Indicators and Instruments Electronic Flight L3 Aviation Products
Instrument Systems
Navigation and Guidance Navigation Aids Honeywell Aerospace
Navigation package
Warning Systems Configurable Integrated Collins Aerospace, Sensors
Surveillance System & Integrated Systems

20.9 Complete Systems Layout

Table 20.2 Complete subsystem key

Fuel System 10 Communication Systems ECS and Pneumatic

1 Collector Tank 11 Flight and Data 20 System
Wing Duct
2 Main Tank Management
12 Indicators and 21 Nacelle Duct
3 Surge Tank Instruments
Engine Control System 22 HT Duct
4 OBIGGS 13 FCU 23 VT Duct
Electric System 14 Engine Control unit 24 AC Unit
5 IDGs 15 APU 25 LOX Tank
6 APU Generator Hydraulic System Flight Control System
7 APU Battery 16 Main Reservoir 26 Rudder EHAs
8 Main Battery 17 Auxiliary Reservoir 27 Elevator EHAs
9 Misc. Electric 18 DC Pump 28 Flap EMAs
Equipment
Avionics System 19 Hydraulic Pump 29 Slat EMAs

90
Fig 20.1 Layout of Basic Aircraft Systems

91
 21. RETARDANT EVACUATION SYSTEM AND TANK DESIGN

An integrated system, where the retardant tank and drop system are essentially integrated

into the fuselage structure, was selected for the AFV-21 to minimize wasted volume and

weight.

21.1 Retardant Tank Design

Three major considerations went into the tank design, i.e., it is able to hold 6000 gal (802

ft3) of retardant, is lightweight, and is slosh resistant when partially filled so that the CG of

the aircraft does not shift drastically and in effect compromise the static stability during

flight.

There were two options to mitigate the effects of liquid sloshing in the container during

flight; a pressurized tank or a non-pressurized gravity based evacuation system with

baffles. A pressurized retardant tank would require greater reinforcements on its walls and

a circular cross-section to minimize hoop stresses; it would also require another tank

containing compressed air to keep the main tank pressurized, thus increasing the weight of

the system. The team decided to implement the gravity based retardant evacuation system

with baffles to minimize weight, cost, and volume.

A ‘reuleaux triangle’ was used as a cross-section for the retardant tank shown in Fig 21.1.

This is the optimal tank cross-section that could yield minimum lateral load transfer,

enhance roll stability and lower CG height via a wider base, according to a paper by Dr

Xiaodi Kang [47].

92
 Fig 21.1 Retardant Tank Cross- Fig 21.2 Retardant Tank Dimensioned Isometric View

Section

Extreme CG shifts during maneuvers because of dynamic loads induced due to the liquid

retardant sloshing in the enclosed tank could ensue instability in all three axes. A baffling

system was incorporated to minimize these effects. Initially, the team considered using

only the conventional transverse baffling system, but after further evaluation, it was found

that conventional transverse baffles help suppress the sloshing in the pitch plane but offer

negligible resistance in the roll plane. Therefore, the addition of a longitudinal baffle

became necessary as it would considerably reduce the roll plane sloshing, according to

research conducted by Mr Abhijit Dasgupta [48]. Mr Dasgupta’s suggestion of equally

spacing the transverse baffles, and perforating both the transverse and longitudinal baffles

for minimum CG shifts and optimum roll and pitch stability, was also implemented, as

visible in Fig 21.2.

21.2 Retardant Evacuation Mechanism

Two major considerations went into the evacuation mechanism design, i.e., the aircraft is

multi-drop capable for satisfying the RFP [1] requirements and displays efficient retardant

93
dispersal (cover about 50 ft of fire line per 100 gal for this class of airtanker) according to

requirements set by the Interagency Airtanker Board (IAB) [20].

A hatch release system was incorporated in the AFV-21 where four pairs of hatches are

located underneath the fuselage (one under each tank compartment), and their operation is

controlled by the Fire Retardant Dispersal System (FRDS) Gen III by Trotter Controls [49].

This system had already been successfully implemented in the AirTractor-802F, which

features a single integrated retardant tank with a hatch release payload drop mechanism.

The FRDS can be programmed to perform any number of drops while simultaneously

controlling the individual drop volumes and drop spread, the two elements crucial for

obtaining maximum effectiveness for a single drop. These settings can also be programmed

prior to the drop on a control panel, allowing pilots to focus on the mission.

21.3 Maintenance and Reloading

The internal surface of the tank will be sprayed with HIREC® - Super Hydrophobic Water

Repellent Coating [50] to prevent rusting inside the tank due to the liquid retardant; this

will reduce the need for regular tank maintenance. If the tank requires maintenance, the

maintenance personnel can enter the tank through one of the drop hatches.

The aircraft will be fitted with two 2.9 inch diameter hoses on both sides of the retardant

tank out of the fuselage to accommodate fast retardant reload at rates of about 500 gal/min

to fulfill requirements set by the RFP [1] and the AirTanker Board [20]. The two hoses

coupled with 3 inch Kamloc® couplers, if operated simultaneously, can fill the tank volume

of 6000 gal in about 6 min, which is half the time required by the RFP [1] for a 6000 gal

tank.

94
 22. WEIGHT AND BALANCE

The Class 2 weight and balance analysis was conducted using the methods in Roskam [22].

The weight of each component was estimated using Class 2 General Dynamics, Torenbeek,

and Vought methods for the military bomber category. The CG location of each component

was estimated using methods in Roskam [22] as well. The takeoff and empty weight are

defined as:

𝑊𝑇𝑂 = 𝑊𝐸 + 𝑊𝐹 + 𝑊𝑃𝐿 + 𝑊𝑐𝑟𝑒𝑤 ; 𝑊𝐸 = 𝑊𝑠𝑡𝑟𝑢𝑐𝑡 + 𝑊𝑝𝑤𝑟 + 𝑊𝑓𝑒𝑞

Fig 22.1 Empty Weight CG Locations

95
 Fig 22.2 Operational items CG and aircraft CG locations

Fig 22.1 shows the CG location of each component in the empty aircraft. The placement

of the systems was done such that the CG excursion is minimized after payload drop and

fuel expenditure; and so that the center of gravity stays close to, yet behind the wing

aerodynamic center. This arrangement will require less elevator deflection to trim the

aircraft, and the deflection required will be positive, i.e., the horizontal tail will generate

positive lift for the majority of a flight; thereby increasing aircraft L/D ratio.

Table 22.1 Empty Weight Breakdown w.r.t Nose

Component Weight (lb) XCG (ft) ZCG (ft)

Wing 13356.8 24.69974 2.966
Horizontal Tail 2790.02 56.47161 4.057
Vertical Tail 1013.15 55.51272 11.9769
Pylon 400 15.478 1.15
Fuselage 11313.44 27.027 0
Nacelle 2374.46 17.9618 -1.73
Nose Landing Gear 1288.53 9 -5
Main Landing Gear 4941.22 28.2935 -5.435
Jet Engine 12326.82 16 -1.73

96
 Fuel System 258.77 24.9 3.78
Air Induction System 0 12.174 -1.73
Propulsion System 310.0 16 -1.02
Flight Control System 1262.31 12.18 0.57
Hydraulic and Pneumatic System 1566.22 28.37 3.66
Instruments/Avionics/Electronics 2673.34 5.13 -0.16
Electrical System 1069.94 29.82 1.19
Air Cond./Press./Icing System 250.60 30.2 1.1
Oxygen System 34.97 12.5 -3.57
Retardant storage and release system 8000 24.46 -1.80
Auxiliary Power Unit 1028.58 54.06 2.08
Furnishings 143.33 7.46 -1.32
Operational Items 1626.34 20 5
Other Items 1802.53 23 0
Empty Weight 70669.20 25.17 0.30

Table 22.2 Operational Items CG Locations w.r.t Nose

Component Weight (lb) XCG (ft) ZCG (ft)

Fuel 28215 24.77 3.78
Pilot & Copilot 400 7.87 0.62
Retardant 54000 24.46 -1.80

CG-excursion diagram (Fig 22.3) for the AFV-21 depicts excellent CG excursion range,

which confirms that the placement of all major subsystems was satisfactory from a balance

standpoint. The forward most CG is at 24.8 ft, and aft most CG is at 25.08 ft from the nose.

Since both fuel and retardant are placed ahead of the empty weight CG, as shown in Table

22.2, the CG is bound to travel backward as the fuel reduces and the payload is dropped.

97
Fig 22.3 CG Excursion Diagram

98
 23. STABILITY AND CONTROL ANALYSIS

The Datcom+ Pro package, a set of programs and files which are an improved version of

the USAF Digital Datcom program, was used to conduct the stability and control analysis

of the AFV-21. Datcom allowed the team to quickly determine the stability derivatives of

the aircraft for multiple flight conditions, CG locations, and flap deflections. This

capability enabled the team to conduct the static stability analysis of the aircraft in great

detail and conduct dynamic stability analysis with some post processing of the results to

improve the controllability of the AFV-21 in a reasonable amount of time. The Datcom

model used throughout the analysis is shown in Fig 23.1.

Fig 23.1 DATCOM Model of the AFV-21

23.1 Stabilizer Sizing

The horizontal and vertical stabilizers were sized using the X-plot, and the vertical

stabilizer area was increased to cater to the OEI condition, as mentioned in section 18.2

and shown in Fig 18.3.

23.2 Control Surface Sizing

The AFV-21 utilizes various control surfaces for attitude control, namely an elevator,

rudder, and ailerons. These surfaces were initially sized using historical trends from

99
Raymer [21]. The rudder chord ratio was later on increased to 0.32 along with the reduction

of the lateral distance of the engines from the fuselage center-line to counter the OEI

moment within the 25 degree deflection range suggested in Roskam [22]. As shown in Fig

18.3, rudder and elevator were given complete trailing edges while the ailerons were given

the trailing edge, which remained after sizing flaps.

Table 23.1 Control Surface Sizing

Parameter Elevator Rudder Aileron

Chord Ratio 0.25 0.32 0.25
MAC 2.805 3.852 3.09
Deflection Range ± 25 deg ± 25 deg ± 25 deg
23.3 Trim Analysis

Trim analysis was done using the methods in Nelson [51], while the coefficients were

determined using Datcom. Trim analysis was considered very important and was done with

the highest possible fidelity owing to its relationship with performance and operational

safety. The trim condition is achieved when there is zero moment about the aircraft CG,

and the trim angle of attack depends upon the values of 𝐶𝑚0 , 𝐶𝑚𝛼 and 𝐶𝑚𝛿𝑒 . There were

two means for achieving trim, i.e., using horizontal tail incidence or elevator deflection;

trim analysis for both of these options was conducted for both forward and aft CG locations

at cruise, takeoff, and landing settings in order to determine the feasibility of each with

regards to deflection range and select the best possible option. Datcom also allowed the

inclusion of ground effect into the analysis, and it can be seen in the larger values of 𝐶𝑚𝛼

(aggravated slopes) in Fig 23.2 for the takeoff and landing moment lines while the increase

in moment due to the flap deflection at these conditions can be seen in the higher values of

𝐶𝑚0 .

100
 Forward CG (X = 24.8 ft) @MTOW Aft CG (X = 25.08) @OEW

𝛿𝑒 Trim
𝑖ℎ Trim

Fig 23.2 Trim Diagrams for different scenarios

The cruise condition's drag increments for both of these trim devices determined the most

viable option since a higher drag addition would directly affect the aircraft's performance

capabilities. Table 23.2 shows the aircraft drag coefficients in cruise for both trim devices.

Table 23.2 Aircraft Drag Coefficients in cruise for incidence and elevator trim scenarios

𝒊𝒉 Trim 𝜹𝒆 Trim
CG Location 𝑪𝑫𝒕𝒓𝒊𝒎 𝑪𝑫𝒖𝒏−𝒕𝒓𝒊𝒎 𝚫𝑪𝑫𝒊 𝚫𝑪𝑫𝟎 𝑪𝑫𝒕𝒓𝒊𝒎
Forward @ X=24.8 0.025 0.023 5.18e-04 0.00044 0.0239
Aft @ X = 25.08 0.018 0.016 3.82e-04 0.00067 0.017

Although there is not a stark difference, both CG locations show less drag for elevator trim

which provided grounds for, and finalized the selection of elevator for the trim role,

101
because adding the incidence changing actuator after this result would not only add another

subsystem but also not provide any benefits in performance since the elevator itself can be

used with a pitch hold autopilot to keep the aircraft trimmed at a particular condition with

less drag.

23.4 Longitudinal Stability Characteristics

The longitudinal stability derivatives of the AFV-21 for different conditions are

summarized below; cruise derivatives are for an intermediate CG, takeoff derivatives for

forward-most, and landing derivatives are for aft-most CG location.

Table 23.3 Longitudinal Stability Derivatives (per radian)

Longitudinal Derivatives
Cruise Takeoff Landing Cruise Takeoff Landing
𝑪𝑫𝜶 0.2234 0.1167 0.2023 𝑪𝑳𝒒 6.2390 6.1350 1.1830
𝑪𝑳𝜶 5.5680 4.2200 4.9040 𝑪 𝒎𝒒 -11.100 -11.270 -1.8160
𝑪 𝒎𝜶 -0.3895 -0.8025 -1.0620 𝑪𝑳𝜶̇ 2.7680 1.4300 -1.6886
𝑪𝑫𝒖 -1.17e-6 -3.03e-6 -3.31e-6 𝑪𝒎𝜶̇ -5.2958 -2.6910 -0.7044
𝑪𝑳𝒖 0.2547 0.0173 0.0106 𝑪𝒍𝜹𝒆 -0.6060 -0.5259 -0.5312
𝑪 𝒎𝒖 -0.1677 -0.0539 -0.0426 𝑪𝒎𝜹𝒆 -1.6372 -1.4330 -1.1567

The AFV-21 possesses longitudinal static stability since any change in the angle of attack

is resisted by a counter pitching moment which is characterized by the negative 𝐶𝑚𝛼 values

for all conditions, as shown in Table 23.3 and corroborated by the negative slopes in the

trim plots in Fig 23.2.

Static margin must never fall below 5% for any loading condition, as suggested by Roskam

[22] for this category of aircraft. The static margins for both CG locations are summarized

in Table 23.4. Since the static margin is positive and above 5%, it indicates that the CG is

102
sufficiently ahead of the aircraft aerodynamic center (neural point), further proving that the

AFV-21 is longitudinal statically stable.

Table 23.4 Static Margin Check

Load Case XCG (ft) XAC (ft) Static Margin (%)

MTOW 24.8 25.24 8.9
OEW 25.08 6.6

Using the stability derivatives in Table 23.3, the longitudinal dynamic stability

characteristics were analyzed by populating a state space model of the aircraft in the

longitudinal plane.

Δu̇ 𝑋𝑢 𝑋𝑤 0 −𝑔 Δu
Δẇ 𝑍𝑢 𝑍𝑤 𝑢0 0 Δw
[ ]=[ ][ ]
Δq̇ 𝑀𝑢 + 𝑀𝑤̇ 𝑍𝑢 𝑀𝑤 + 𝑀𝑤̇ 𝑍𝑤 𝑀𝑞 + 𝑀𝑤̇ 𝑢0 0 Δq
Δθ ̇ 0 0 1 0 Δθ

𝑋𝛿𝑒 𝑋𝛿𝑇
𝑍𝛿𝑒 𝑍𝛿𝑇 Δδe
+[ ][ ]
𝑀𝛿𝑒 + 𝑀𝑤̇ 𝑍𝛿𝑒 𝑀𝛿𝑇 + 𝑀𝑤̇ 𝑍𝛿𝑇 ΔδT
0 0

There were few FAR requirements related to stability, control, and handling. On the other

hand, military standards were much more precise and strict. The standards concerning

flying qualities of Class 3 (Large, Heavy) aircraft in MIL-F-8785C were used for

comparison. It was decided that the AFV-21 shall possess Level-1 handling qualities,

defined on the Cooper-Harper scale as "excellent, highly desirable”. Any shortcomings

would be rectified using Stability Augmentation or Feedback Loops. The following

comparison has been made for the cruise condition as it takes priority for being the longest

leg of the design mission.

103
 Table 23.5 Longitudinal Dynamic Stability Characteristics

Mode Short Period Phugoid

𝜻 𝒘𝒏 (𝒓𝒂𝒅/𝒔) 𝜻 𝒘𝒏 (𝒓𝒂𝒅/𝒔)
Level- 1 Requirements 0.3 ≤ 𝜁 ≤ 2.0 0.3 ≤ 𝑤𝑛 ≤ 2.0 𝜁 ≥ 0.04 -
AFV-21 0.626 2.07 1.0 0.0851

As shown in Table 23.5, AFV-21 satisfies most of the level-1 handling criterion except for

the short period frequency; however, the difference was so small that it did not warrant

adding a Stability Augmentation System (SAS).

23.5 Lateral-Directional Stability Characteristics

Table 23.6 shows all lateral-directional stability derivatives for different conditions:

Table 23.6 Lateral-Directional Stability Derivatives (per radian)

Lateral-Directional Derivatives
Cruise Takeoff Landing Cruise Takeoff Landing
𝑪𝒚𝜷 -0.4820 -0.4745 -0.4741 𝑪𝒍𝒓 0.1151 0.3234 0.2642
𝑪𝒍𝜷 -0.0503 -0.0773 -0.0694 𝑪 𝒏𝒓 -0.0318 -0.0524 -0.0318
𝑪 𝒏𝜷 0.0369 0.0376 0.0372 𝑪𝒍𝜹𝒂 0.1898 0.1910 0.1902
𝑪𝒚𝒑 -0.0145 -0.0241 -0.0408 𝑪𝒏𝜹𝒂 -0.0259 -0.1024 -0.0756
𝑪𝒍𝒑 -0.4703 -0.1961 -0.3514 𝑪𝒚𝜹𝒓 0.3821 0.3623 0.3611
𝑪 𝒏𝒑 -0.0362 -0.1360 -0.1040 𝑪𝒍𝜹𝒓 0.0080 0.0144 0.0075
𝑪𝒚𝒓 0.1817 0.1732 0.1709 𝑪𝒏𝜹𝒓 -0.0858 -0.0863 -0.0748

The positive value of weathercock stability derivative 𝐶𝑛𝛽 shows that the AFV-21 is

directionally stable such that it will generate a restoring yaw moment to resist any change

in sideslip angle. Similarly, the negative value of dihedral effect derivative 𝐶𝑙𝛽 shows that

the aircraft will generate a counteracting roll moment for changes in the sideslip angle often

caused by changes in bank angle.

104
Similar to longitudinal dynamic stability, lateral-directional dynamic stability

characteristics were analyzed by populating the following state space model of the aircraft

in the lateral and directional planes using the stability derivatives for cruise condition in

Table 23.6.

Δv̇ 𝑌𝑣 𝑌𝑝 −(𝑢𝑜 − 𝑌𝑟 ) 𝑔 Δv 0 𝑌𝛿𝑟

Δṗ 𝐿 𝐿𝑝 𝐿𝑟 0 Δp 𝐿𝛿𝑎 𝐿𝛿𝑟 Δδ𝑎
[ ]= 𝑣 [ ]+[ ][ ]
Δṙ 𝑁𝑣 𝑁𝑝 𝑁𝑟 0 Δr 𝑁𝛿𝑎 𝑁𝛿𝑟 Δδ𝑟
Δφ̇ [0 1 0 0] Δφ 0 0

Table 23.7 shows the comparison of the inherent dynamic stability characteristics of the

AFV-21 and the requirements for the Level-1 handling qualities for the spiral, roll, and

dutch roll modes:

Table 23.7 Lateral-Directional Dynamic Stability Characteristics

Mode Dutch Roll Roll Spiral

𝜻 𝒘𝒏 (𝒓𝒂𝒅/𝒔) 𝑻𝒓 (𝒔𝒆𝒄) 𝑻𝒔 (𝒔𝒆𝒄)
Level- 1 Requirements 𝜁 ≥ 0.08 𝑤𝑛 ≥ 0.5 𝑇𝑟 ≤ 1.4 𝑇𝑠 ≥ 28.9
AFV-21 (No SAS) 0.0389 0.97 0.491 186
AFV-21 (with SAS) 0.825 0.634 0.491 306

The pole-zero map for the unaugmented system showed that the spiral pole was positive,

which meant that the spiral mode was unstable; however, the 186 second time constant

makes it easy for the pilot to compensate for this, thus making it a non-issue. The dutch

roll mode, though stable, lacked the damping necessary for level-1 handling qualities and

necessitated the addition of a yaw damper, i.e., a yaw rate feedback to the rudder. The yaw

damper also significantly increased the spiral mode stability; however, since the spiral

mode is typically a very slow mode and pilots are used to banking and turning without

constant aileron input, a washout filter was added, which placed a zero at the origin and

105
constrained the spiral pole to remain near the origin. Fig 23.3 shows the block diagram of

the yaw damper developed in Simulink.

Fig 23.3 Lateral SAS Block Diagram

106
 24. STRUCTURAL DESIGN AND MATERIAL SELECTION

24.1 Manoeuvre and Gust Loads

The structural design began with the construction of the V-n diagram using methods in Niu

[52] to comply with the requirements of 14 CFR § 25.333. The limit maneuvering load

factors were determined from 14 CFR § 25.337. A positive limit load factor of 3.5 and a

negative limit load factor of -1 were set, considering the extreme conditions the AFV-21 is

expected to face over the wildfires. The gust lines were calculated using 14 CFR § 25.341

and plotted on the V-n diagram.

Fig 24.1 V-n Diagram with Gust Envelope

The exceptional load-bearing characteristics of the AFV-21 are demonstrated in Fig 24.1

since most of the gust lines lie within the bounds of the V-n diagram.

107
After determining the limit load factors, key aircraft components, i.e., wing, empennage,

and fuselage, were designed. The material selection for the prominent structural members

in these components was based on the analyses conducted on the designed components.

24.2 Wing Design

The wing was designed against the limit load factor of 3.5 with a safety factor of 1.5. The

procedure started with the lift distribution along the wingspan, determined using Schrenk’s

approximation [53], which suggests that the lift distribution along a tapered wing is the

average of elliptic and trapezoidal lift distributions. Using this approximation and the

various loads on the wing, i.e., wing weight, loads of the main and collector fuel tanks, and

the engine weight, a total wing loading diagram was generated. The loads along the

wingspan were then integrated to obtain the wing shear diagram. The shear loads were

again integrated to produce the wing bending moment diagram.

108
 Fig 24.2 Wing Lift Distribution Fig 24.3 Total Wing Load Distribution

Approximation

Fig 24.4 Spanwise Wing Shear Loading Fig 24.5 Spanwise Wing Bending

Moment

A maximum shear load of 190741 lbs and bending moment of 5121750 lbs.ft were obtained

at the wing root from Fig 24.4 and Fig 24.5, respectively, and were used in further analysis.

109
According to the recommendations by Niu [52], the wing ribs were sized to a thickness of

0.18 in. with circular cut-outs in the middle and trailing edge sections to minimize stress

concentration regions. The first ribs outboards from the root were spaced 5 ft away from

the root rib with a central beam running in between, to form the wing carry-through

structure; these ribs and the root rib were assigned a thickness of 0.61 in. to cater to the

high loads near the root. The front and rear spars were then placed at 13.5% and 73.3% of

the chord, respectively, as a large torque box would help distribute the torsional forces

more effectively and reduce the rib thickness. The front spar has an I-beam as it experiences

higher loading than the rear spar, which has a Channel cross-section for ease of inspection

and accommodation of attachment points for ailerons and flaps. Spar placement was

followed by the disposition of

20 stringers with equal cross-

sectional areas in the mid rib

section, where spar caps and

stringers were idealized into

Fig 24.6 Idealized Wing Rib at Root for Bending
direct stress carrying booms.
Analysis
Using the methods in Megson

[54], areas of these booms were calculated at the root and tip of the wing. The areas were

linearly decreased from root to tip to model the stringers. Using the ultimate bending

moment of 5121750 lbs.ft obtained from Fig 24.5, the direct stresses were found at the

wing root rib from the following equation:

𝑀𝑥 (𝐼𝑧𝑧 𝑧 − 𝐼𝑥𝑧 𝑥)
𝜎𝑦 =
𝐼𝑥𝑥 𝐼𝑧𝑧 − 𝐼𝑥𝑧 2

110
It was discovered that a maximum compressive stress of 51454 psi occurs in boom 9.

Therefore, the upper wing stringers will be composed of Al-7075 T3, which has a yield

strength of 73000 psi. Similarly, a maximum tensile stress of 28317 psi occurs in boom 16,

so Al-2024 T3, with a yield strength of 50000 psi, will be used to construct the bottom

wing stringers. The stringers will feature an extruded Z-cross-section due to its high

Farrar’s efficiency factor and easy assembly. The spar caps experienced maximum stress

of 61500 psi; therefore, the spars will be composed of Al 7075 T3.

After the direct stress analysis, shear flow analysis was conducted for the rib at the wing

root as it experiences the highest shear load. For this analysis, the wing skin and spar webs

were idealized into shear

stress carrying ‘panels’ or

‘walls’. The rib was divided

into two cells by the front

spar web, and the trailing Fig 24.7 Idealized Wing Rib at Root for Shear Flow
edge portion was neglected Analysis
based on the assumption that

all of it would be dedicated to flaps and ailerons. Anticlockwise shear flow was taken

positive as the convention. The total torque being applied on the rib was calculated by

multiplying the maximum shear force of 190741 lbs by the length of the torsion box from

the leading edge to the shear center (near the centroid). A set of equations was then

generated, using the following general formula from Megson [54], for the two cells.

𝑑𝜃 1
= [−𝑞𝑅−1 𝛿𝑅−1,𝑅 + 𝑞𝑅 𝛿𝑅 − 𝑞𝑅+1 𝛿𝑅+1,𝑅 ]
𝑑𝑦 2𝐴𝑅 𝐺𝑅𝐸𝐹

111
In the above equation, ‘ẟ’ is a correction factor to account for the difference in the materials

used to construct each skin panel and spar web, as is the case in real-life applications.

‘GREF’ is the shear modulus of a single wall selected as reference, which in this case was

wall 24. ẟ is calculated as the integral of the reciprocal of corrected thicknesses (t*) over

the lengths of the walls.

Table 24.1 Shear flow analysis wall properties

Wall Length t (in) t* (in) Material G (ksi) Cell area

12outer (in)
86.7 0.059 0.057 Al 2024 4060 A1 =
12inner 30.7 0.23 0.22 T3 (in2)in2
715.9
34 24.1 0.23 0.22 Al 7075 A2 = 5815
3900
13 157 0.059 0.059 T6
24 151 0.059 0.059 in2

A third equation for the overall torque in the cells was generated based on:

𝑇 = ∑ 2𝐴𝑅 𝑞𝑅
𝑅=1

A MATLAB code was developed to calculate the areas of the cells; the shear flows in each

wall were calculated by simultaneously solving the equations whose development is

described above, and the shear stresses were found by dividing these shear flow values

with the wall thicknesses. It was found that the maximum shear stress of 42887 psi occurs

in the center wing skin panels, while the outer wing skin panels experienced that of about

22093 psi. Al 7075 T6 with a shear strength of 48000 psi and Al 2024 T3 with a shear

strength of 41000 psi were finalized as the materials for center wing skin panels and outer

wing skin panels respectively, since the shear strengths of these materials exceed the

maximum shear stress faced by their respective panels.

112
 Table 24.2 Wing Substructure Key

Parameters

Parameter Value Parameter Value

Stringer
Rib 0.180 2.93
area (at
thickness in in2
root)
Stringer
Rib 25.0 0.385
area (at
spacing in in2
tip)
Spar web 0.230 Skin 0.059
thickness in thickness in
Front spar Rear spar
Fig 24.8 Wing Internal Structure 4.397 3.10
cap area cap area
in2 in2
(at root) (at root)
24.3 Empennage Design

Empennage design was initiated with an analysis of the horizontal tail, where the loads

were calculated through methods in a paper by James L. Decker [55]. The maximum

bending moment and shear loads calculated through these methods were then used to

conduct analyses similar to the wing, whereby the critical tail substructure parameters were

determined. In case of the vertical tail, the maximum sideslip force was calculated by the

following formula from Nelson [51]:

𝑌 = 𝐶𝑦𝛽 𝑞̅ 𝑆𝑣 𝛽

This force was then used to calculate spanwise bending moments and shear loads to carry

out analyses similar to the wing and HT to determine critical VT substructure parameters.

Both tails will utilize Z-section stringers composed of Al 2024 T3 due to lower strength

requirements as compared to the ribs and spars, which will be constructed of Al 7075 T6

113
due to the high strength requirements for these structural members. The front and rear spars

in HT and VT will have similar cross-sections to the wing spars.

Table 24.3 HT Substructure Key Parameters

Parameter Value Parameter Value

Stringer
Rib 0.180 1.77
area (at
thickness in in2
root)
Rib Stringer 0.350
19.2 in
spacing area (at tip) in2
Spar web 0.230 Skin 0.047
thickness in thickness in
Fig 24.9 HT Internal Structure

Table 24.4 VT Substructure Key Parameters

Parameter Value Parameter Value

Stringer
Rib 0.180 2.01
area (at
thickness in in2
root)
Rib Stringer 0.350
18.5 in
spacing area (at tip) in2
Spar web 0.230 Skin 0.047
thickness in thickness in
Fig 24.10 VT Internal Structure

24.4 Fuselage Design

The fuselage features a semi-monocoque structure with a combination of longerons

running throughout its length and stringers placed between sections that experience high

shear loads and bending moments such as those near the wing. Two pressure bulkheads are

placed to seal the cockpit in the front part of the fuselage, whereas two hollow bulkheads

114
are placed below the middle and rear spar for structural support and to provide wing

attachment points. Frames are placed along the fuselage length with close spacing in the

middle section as it is expected to experience high loads from the retardant and wing.

Frames in the cockpit are placed with great care so as to not interfere with pilot viewing

angles. The frames in the aft fuselage are placed with equal spacing.

The design and analysis process began with modeling the fuselage as a cylindrical

cantilever beam supported at the wing by the middle and rear spars. The fuselage structural

weight, subsystem weights, and their c.g locations along the fuselage length were obtained

from the weight statement in Table 22.1. The fuselage structural weight distribution was

approximated as a trapezoidal distribution, whereas the subsystems were considered as

point loads. In a similar fashion to the wing analysis, Bruhn’s [56] method was used with

this load distribution to obtain the shear loads and bending moments for the fuselage

stations along its length. It was observed

that the critical fuselage section, which

experiences the highest load, lies at the rear

wing spar location. Ten I-section longerons

and eighteen Z-section stringers were laid

out along this critical section’s perimeter.

This section was idealized such that the

longerons and stringers were approximated

as direct stress carrying booms. The boom Fig 24.11 Idealized Fuselage Section

areas were determined using methods from at Wing Rear Spar

115
Megson [54], and direct stress analysis was conducted using the following equation:

𝑀𝑦 (𝐼𝑧𝑧 𝑧 − 𝐼𝑦𝑧 𝑦)
𝜎𝑥 =
𝐼𝑦𝑦 𝐼𝑧𝑧 − 𝐼𝑦𝑧 2

For a maximum bending moment of 9539 kips.ft, the maximum stresses of 62222 psi and

59770 psi occurred in the longerons and upper section stringers, respectively, while the

bottom section stringers faced a maximum stress of 47712 psi. Al 7075 T6, which has a

yield strength of about 73000 psi, was finalized as the material for the longerons and upper

section stringers, while Al 2024 T3, which exhibits a yield strength of 50000 psi, was

finalized for the bottom section stringers; since all of their yield strengths exceed the

maximum stresses faced by their respective components.

A torsional moment analysis was also conducted on the critical section, whereby the

thickness of the skin, composed of Al 2024 T3, was finalized as 0.091 in. As the AFV-21

is a manned aircraft that operates above 10000 ft, cockpit pressurization is essential for

safe operations. The cockpit bulkheads allow pressurization up to a pressure altitude of

15000 ft; a choice made in compliance with 14 CFR § 25.841, as the aircraft can reach an

absolute ceiling of 43068 ft and is expected to operate at a cruise altitude of 34000 ft. Using

the most demanding scenario where the aircraft operates with the maximum internal

pressure of 8.29 psia (15000 ft pressure altitude), at an altitude of 43068 ft ASL, the hoop

stress in the fuselage was calculated using the following equation:

𝑃0 − 𝑃𝑖
𝑃𝑖 𝑟𝑖2 − 𝑃0 𝑟02 − 𝑟𝑖2 𝑟02 ( )
𝜎ℎ = 𝑟2
𝑟02 − 𝑟𝑖2

Where Pi and ri are the internal and Po and ro are the external pressure and radius of the

idealized cylindrical cockpit. Skin panels for the cockpit, composed of Al 2024 T3, with a

116
thickness of 0.101 in. proved to be satisfactory to bear the shear stresses and a hoop stress

of 4234 psi inside the cockpit, a consequence of extreme pressurization conditions.

Fig 24.12 Fuselage Internal Structure

Table 24.5 Fuselage Substructure Key Parameters

Parameter Value Parameter Value

Frame Thickness 1.30 Longeron Area 5.32 in2

in
Frame Spacing 12.3 Stringer area 0.42 in2
(fore fuselage) in
Frame Spacing 23.3 Skin thickness 0.091
(mid fuselage) in in
Frame Spacing 0.23 Skin thickness 0.11 in
(aft fuselage) in (Cockpit)
Pressure Bulkhead 2.00 Hollow 1.70 in
Thickness in Bulkhead
Thickness

24.5 Miscellaneous Material Selection

Material selection for the various substructures has already been discussed in the preceding

sections of this chapter. Al 7178 T6 was specifically selected to construct the bulkheads

117
because of their extreme load-bearing nature. The landing gear will be constructed using

AISI 4340 steel alloy due to its high fatigue strength, yield strength, and elastic modulus,

all desirable characteristics to bear the extreme cyclic static and dynamic loads from

takeoff/landing.

Table 24.6 Material Assignment and Properties

Yield Elastic Shear Shear

Density
Component Material Strength Modulus Strength Modulus
(lb/in3)
(ksi) (ksi) (ksi) (ksi)
wing bottom
stringers, wing
outer skin panels, Al 2024-
50.0 10600 41.0 4060 0.100
HT/VT skin T3
panels, HT/VT
stringers
wing top stringers,
wing center upper
and bottom skin
panels,
wing/HT/VT
spars,
Al 7075-
wing/HT/VT ribs, 73.0 10400 48.0 3900 0.102
T6
fuselage frames,
longerons,
fuselage top
stringers, fuselage
bottom stringers,
rivets, fittings.
fuselage pressure
Al 7178-
and hollow 78.0 10400 52.2 3920 0.102
T6
bulkheads
AISI
landing gear 125 29000 - 11300 0.284
4340

Composites were not considered in the design despite their appeal of empty weight saving

because of their high costs and difficulty in maintenance and repair in case of damage.

118
24.6 Final Structural Layout Assembly

Fig 24.13 Aircraft Final Structure

119
 25. DRAG POLARS

Following the finalization of all geometric parameters of the AFV-21, the drag polars were

recalculated using the methods in Schaufele [30]. Fig 25.1 represents the drag polars for

different flight configurations, and Fig 25.2 depicts the relationship between cruise speed

and aircraft drag.

Fig 25.1 Multiple Flight Configurations Drag Fig 25.2 Drag Polar at Various Cruise Mach

Polars No.

The component-wise wetted area and parasite drag contributions for cruise condition are

given in Table 25.1.

Table 25.1 Wetted Area and Parasite Drag Breakdown for the AFV-21

L
Fuse Win Pyl Nac Tr Compres Tot
Component HT VT &
lage g ons elles im sibility al
P
3820 172 123 467 67. 610. 792
Swet (ft2) _ _ _
.10 7.90 6.00 .30 03 45 8.78
Swet
48.1 21.7 15.5 5.8 0.8 100.
Contribution 7.70 _ _ _
8 9 9 9 5 00
(%)

120
 ΔCD0 (1
25.4 76.2 24.4 9.3 1.3 12.7 4.4 10. 165.
Count = 1.17
5 6 4 1 1 1 0 55 60
0.0001)
ΔCD0
15.3 46.0 14.7 5.6 0.7 2.6 6.3 100.
Contribution 7.68 0.71
7 5 6 2 9 6 7 00
(%)

121
 26. PERFORMANCE ANALYSIS

This section aims to verify that the AFV-21’s performance meets the requirements set forth

by the RFP [1]

26.1 Takeoff and Landing Performance

The takeoff Balance Field Length (BFL) was determined at an elevation of 5000 ft and ISA

+350C (as per the RFP [1]) using the methods outlined in Schaufele [30]. The takeoff

performance code calculates the distances for both ‘accelerate continue’ and ‘accelerate

stop’ scenarios for several OEI recognition speeds (V1) until they become equal, which

gives the V1 for the Balanced Field Length.

Fig 26.1 Takeoff BFL Analysis – (Dry Concrete)

The Landing Field Length (LFL) was calculated through the methods given in

Gudmundson [57] as well as Anderson [34], which are based on integrating the equations

of motion along the length of the runway while accounting for the changes in the forces

acting on the aircraft with each passing segment of the runway, i.e., reduction in lift and

drag.

122
Above mentioned methods were used to evaluate takeoff and landing performance at

different runway conditions the AFV-21 may face; the typical values for rolling and

braking coefficients were taken from Anderson [34]. The BFL and LFL for all conditions

were summarized in the Table 26.1:

Table 26.1 Summary of Takeoff and Landing performance parameters

Condition BFL@MTOW LFL@0.9*MTOW LFL@0.9*MTOW

[30] [57] [34]
Dry 5413 ft 2674 ft 2855 ft
Concrete
Wet 7018 ft 3894 ft 4100 ft
Concrete
Hard turf 5806 ft 2910 ft 3103 ft

From Table 26.1, the AFV-21 approaches the 5000 ft objective BFL requirement in the dry

conditions and satisfies the 8000 ft requirement in wet conditions. The aircraft satisfies the

5000 ft LFL objective in all conditions without the use of thrust reversers or spoilers, owing

to the wing’s design for good low speed performance. Such results meant that these devices

need not be installed, saving cost.

26.2 Payload Radius Diagram

From Fig 26.2, the maximum radius capability of the AFV-21 for the design mission

reduced from 380 nm to 370 nm at full payload, which could be attributed to the increase

in drag after Class 2 drag calculations. The payload-radius diagram demonstrates the

changes in achievable radius in the design mission with changes in fuel and payload, where

the outermost line represents the achievable design radius at MTOW and full fuel (28215

lb) while the inner lines show the performance at lower fuel levels.

123
 Fig 26.2 Payload-Radius Diagram at different fuel loads

26.3 Operational Envelope

The operational envelope of the AFV-21 is shown in Fig 26.3; the service ceiling at MTOW

was 42000 ft. At low altitudes, the speed was limited by structural strength, while at higher

altitudes, it was limited by engine thrust. The maximum Mach at sea level was roughly

M0.7 (785 ft/s).

Fig 26.3 Operational Envelope at MTOW

124
26.4 Specific Excess Power Diagram

Specific excess power contours were plotted to determine the power available to reach

certain altitudes and speeds and the limits of the AFV-21’s capabilities. The dents in the

contours occur due to replacements in the different coefficients of the engine mathematical

model, as the bounds of their definitions are reached at different altitudes and Mach

numbers.

A maximum Mach of M0.83 was achievable at 39000 ft, while the absolute ceiling was

estimated to be 43068 ft which confirms the validity of the operational envelope (Fig 26.3)

and the service ceiling. The red line denotes the maximum specific excess power at certain

altitudes, which allows the best rate of climb, and all of these rates of the climb are

achievable as they lie within the contours. The maximum rate of climb at MSL was 37.38

ft/s (2242.8 fpm).

Fig 26.4 Specific Excess Power Diagram

125
26.5 Other Performance Parameters

The ferry range of the AFV-21 was determined to be 2740 nm which, for the reasons

discussed in section 26.2, has slightly degraded from earlier performance estimates made

in section 13.3. Stall performance was also investigated for the finalized weight, wing

loading, and maximum lift coefficients (section 11.1) for the clean and landing

configurations. The clean and landing stall speeds were 130.75 kts and 94.5 kts,

respectively; both satisfy the low speed performance requirements set by the team in

chapter 8.

Performance coefficients were calculated for MTOW at the cruise altitude of 34000 ft and

the range of Mach numbers afforded by the aircraft's flight envelope. The values of these

performance coefficients represent the aircraft's performance in different segments of the

mission. Contrary to the predictions made using Anderson [34], Mach numbers against the

maximum values of the performance coefficients did not match with the optimal cruise and

loiter segment velocities obtained in the altitude and speed optimization studies of chapter

19. This discrepancy was caused by the use of an engine model and the concepts of partial

throttle SFC in the optimization studies instead of the assumption of constant values of

thrust and SFC in lower fidelity methods.

Table 26.2 Performance Coefficients

Coefficient Value Mach Number

(𝑪𝑳 ⁄𝑪𝑫 )𝒎𝒂𝒙 17.31 0.62

𝟑/𝟐
(𝑪𝑳 ⁄𝑪𝑫 )𝒎𝒂𝒙 14.79 0.47
𝟏/𝟐
(𝑪𝑳 ⁄𝑪𝑫 )𝒎𝒂𝒙 25.52 0.74

126
 Fig 26.5 Performance Coefficients

26.6 Climb Gradient

The climb gradient curves were drawn for the second take-off climb segment in which the

speed was taken as 1.2 times the stall speed at take-off as per the method in Schaufele [30].

AFV-21 has a climb gradient of 2.9 at MTOW and an altitude of 5000 ft, which is greater

than the FAR requirement of 2.4, i.e., well above the limit and can climb with one engine

at take-off thrust.

Fig 26.6 Second segment Climb Gradient VS Gross Weight

127
The second segment climb gradient at a particular altitude decreases with increasing gross

weight until it reaches the minimum point below which it cannot climb with one engine

inoperative; such a gross weight is called Second Segment Limiting Weight (SSLW). For

the AFV-21, SSLW at 5000 ft is 158500 lbs, greater than its MTOW of 153284.5, as shown

in Fig 26.6. The Fig also depicts that on increasing the take-off altitude, the SSLW value

decreases since thrust decreases with altitude and reduces the gradient, making takeoff at

altitudes above 6000 ft unsafe for takeoff at MTOW in the OEI condition.

128
 27. COST ANALYSIS

Estimations for RDTE, Production and Operating costs were conducted using three

methods. The first two methods are different variations of the DAPCA IV cost model

provided in Nicolai [31] and Raymer [21], followed by the third method laid out in Roskam

[22], which is similar to the DAPCA IV in principle; however, it uses a larger database of

aircraft and is higher fidelity since it requires more aircraft parameters as inputs and factors

in details such as the cost increment due to tapered wings.

Since the RDTE efforts will primarily take place before the EIS date, the associated costs

were given an inflation factor for 2025, whereas the costs associated with the production

phase were given an inflation factor for the year 2030 since most of the production efforts

would be just before and after the EIS. Operating and maintenance costs were given an

inflation factor for 2030 as well.

27.1 Research, Development, Test and Evaluation (RDTE) Costs

After consulting the case studies from [58], it was decided that three prototypes would be

sufficient for the development phase, and the calculations were done as such. The RDTE

cost for the AFV-21 program was estimated to be $2.1 Billion, $1.77 Billion, and $1.4

Billion from Nicolai [59], Raymer [21], and Roskam [22] methods, respectively. The

breakdown of RDTE costs from different methods is given in Fig 27.1:

129
 Fig 27.1 RDTE costs breakdown in different cost models

27.2 Production (Flyaway) Costs

An estimate of 200 aircraft sales was made through the market analysis in section 2.4 and

used to determine the production costs. Production costs of the AFV-21 program were

estimated to be $15.6 Billion, $16.1 Billion, and $13.4 Billion from Nicolai [59], Raymer

[21], and Roskam [22] methods, respectively. The breakdown of production costs from

different methods is given in Fig 27.2. Note that the engine cost is dominant due to the lack

of internal furnishings.

Fig 27.2 Production costs breakdown in different cost models

130
27.3 Unit Sales Price

In order to stay competitive with the current market of firefighting aircraft, the price was

to be set such that no major part of the market should be turned off by it. This was done by

comparing the retardant payload capability of the firefighter aircraft currently in service to

their unit prices and determining where the price point of the AFV-21 should lie. The unit

costs estimates from Nicolai [59], Raymer [21], and Roskam [22] were adjusted for a 15%

profit and included in the plot as well; it is clear from the location of these estimates in Fig

27.3 that a specialized firefighter is cheaper since there are substantial weight and material

savings owing to a smaller fuselage and no retrofitting costs. For a conservative approach,

the Roskam estimate of $93.7 million was finalized as the unit price.

Fig 27.3 Payload Capacity VS Sales Price

27.4 Break-Even Analysis

The break-even analysis was conducted to determine how many aircraft would need to be

sold to break even and begin profit. The fixed and variable cost estimates from Nicolai [59]

and Raymer [21] were used for the break-even analysis. According to the estimates from

131
Nicolai [59], break-even occurs at 65 aircraft, while estimates using Raymer [21] predict

break-even at 76 aircraft.

Fig 27.4 Break-Even Analysis (Nicolai) Fig 27.5 Break-Even Analysis (Raymer)

27.5 Direct Operating Cost

Direct operating and maintenance cost per flight hour and per year was calculated using

methods given in Nicolai[31], Raymer [21], and Roskam [22], assuming 1200 flight hours

per year as per the RFP [1]. Total operation and maintenance cost per flight hour per aircraft

was calculated to be $11,815 and $ 12,201 from Nicolai, and Raymer, respectively. A direct

operating cost of $ 11,284 was also calculated using AAA, which is based on the methods

in Roskam [22]. A target operating cost of $ 10,354 per hour (inflation for 2030 applied)

was mentioned in section 5.2.2, which according to [24], will give a positive economic cost

benefit; the AFV-21 outperforms the LATs currently in service since its hourly operating

cost estimates closely match this value. Pie charts with the breakdown of the total operating

costs for each cost model are also shown for comparison.

132
Table 27.1 Operating costs Per Flight Hour Per Aircraft (Top Rows) and Per Year Per

Aircraft (Bottom Rows)

Nicolai Raymer Roskam

Fuel Cost $ 5,684 $ 7,984 $ 7,376
$ 6,820,704 $ 9,580,929 $ 8,850,800
Oil Cost $ 18 $ 26 $ 37
$ 22,027 $ 31,368 $ 44,254
Crew Cost $ 3,085 $ 2,619 $ 2,420
$ 3,701,750 $ 3,143,040 $ 2,903,520
Labour Cost $ 1,512 $ 1,095 $ 1,011
$ 1,813,857 $ 1,313,465 $ 1,213,370
Materials Cost $ 1,516 $ 477 $ 440
$ 1,819,872 $ 572,214 $ 528,608
Total OPM Cost (Per FH Per AC) $ 11,815 $ 12,201 $ 11,284
Total OPM Cost (Per Year Per AC) $ 14,178,210 $ 14,641,016 $ 13,540,553

Fig 27.6 Operating costs breakdown in different cost models

133
 28. MANUFACTURING PLAN

In line with the low volume production model for the JF-17, where PAC holds 58% work

share of JF-17 airframe co-production [60] and the rest is held by China; Airframe

manufacturing of the AFV-21 will be distributed amongst two facilities, a final assembly

plant in the United States and a component manufacturing facility in Mexico. This

distribution is based on [61], which states a saving of 30% on labor for Bombardier despite

added logistical costs of importing parts from Mexico. Using this distribution, labor

intensive work of manufacturing parts such as ribs, lugs, and bolts, will be outsourced to

the external facility; however, the intricate work of systems integration, wiring, assembly

and testing will be done by a smaller yet better trained and skilled workforce in the United

States.

Due to the low annual production rate of seven aircraft, developing a production/assembly

line was deemed unnecessary as it would be impossible to recover the cost of making a

setup like Boeing with large scale automation, which reduces development times to days.

Similar to the approach followed by PAC for the JF-17 [62], small yet sophisticated

facilities with conventional and CNC machining, and a small group of well trained

technicians capable of manufacturing multiple types of components while operating within

the bounds of a certified Quality Management System would be well suited to the low

volume production of the AFV-21. Such an approach would reduce the negative cash

bucket associated with infrastructure development while facilitating better quality control

as it would become feasible to perform inspections at every stage in the manufacturing

process with fewer and smaller facilities.

134
 29. COMPUTATIONAL FLUID DYNAMICS

By the time the team had reached the CFD portion, less than a month had remained to

complete the project. The main objective was to design an aircraft that fulfilled the

requirements outlined in the AIAA RFP [1]; which, as mentioned in Chapter 26, has been

achieved since the AFV-21 meets its performance requirements quite well, despite the use

of conservative approaches and high fidelity methods in the performance estimation; this,

however, prompts the question, why perform a CFD analysis? The answer lies in the fact

that the validity of the assessments of lift/drag characteristics and stability derivatives of

the AFV-21, which formed the foundation of its performance analyses and stability

analysis, is still not set in stone. The reason behind the preceding statement is that despite

using high fidelity methods from texts like Schaufele [30], Gudmundson [57] and the Class

2 design sequence of Roskam [22], a thorough read leads to the understanding that even

these high fidelity estimation methods rely (to some extent) on historical data and empirical

techniques which means that tools like CFD must verify their results before any serious

consideration is given to prototyping and production.

An effort was made in the earlier chapters to minimize the amount of writing and only

focus on the results. This was done for two reasons. The first reason was that this is how

scientific literature is written, i.e., results are focused on, and the reader is not bothered

with reading about how the calculations were done; instead, the code is provided in a disk

or an online directory. The second reason was that the AIAA RFP [1] required the report

to be no more than 100 pages which meant only major results and algorithms could be

discussed. Since the CFD portion (Chapter 29) onwards has been added to the project report

post submission, the same approach of expression has not been followed in the following

135
chapters; instead, the explanation follows the same structure followed by the ‘Fluid Flow’

analysis system in ANSYS Workbench. Due to a shortage of time and computational

resources, only the longitudinal pitch moment (𝐶𝑚 ), drag coefficient (𝐶𝐷 ) and lift

coefficient (𝐶𝐿 ) were investigated at different angles of attack to determine the pitch

stiffness, drag polar and lift curve, respectively.

29.1 Geometry

The geometry of the AFV-21 was developed on multiple platforms and kept changing as

the project evolved due to its iterative nature and the use of higher fidelity methods as time

went on. For the initial phases of the design in class 1, OpenVSP was used to generate the

geometry since it provided the capability of rapid changes in the configuration, and such

was the requirement of the design effort right up until the final phases of Class 2 design.

In addition, the major internal components were being iteratively adjusted to satisfy weight

and balance requirements, which again was facilitated by the rapid adjustment capabilities

of OpenVSP as demonstrated by the intricacy of systems layout shown in Fig 29.2.

136
 Fig 29.1 OpenVSP model after Fig 29.2 OpenVSP model after

Class 1 design was finalized Class 2 design was finalized

After the Class 2 design was finalized, the geometry and systems layout was completed.

The team felt the need for a detailed geometry representation for a variety of reasons, such

as enhancing the aesthetics of the report using detailed renders of the aircraft model, such

as the one on page ix, a visual representation of the aircraft’s complete internal structure

which would require a model of its exterior for cross-checking, utilizing the exterior model

for CFD analysis and structures model for FEM. For CFD, a model of the AFV-21’s

exterior was developed using CATIA since it was known to be the preferred choice of

engineers in the aerospace industry for CFD. Individual parts of the aircraft, such as the

wings, HT and VT, were designed in the part design module; the fuselage was imported

directly from the OpenVSP model due to difficulty in replicating the intricacy of its curves.

All the parts were then assembled in the assembly design module to give the final CATIA

model shown in:

137
 Fig 29.3 CATIA model of AFV-21

Upon completing the CAD modeling of the exterior surface, the geometry was then

exported to SPACECLAIM, the geometry engine of ANSYS Workbench. SPACECLAIM

is not so much known for geometry creation as it is for geometry preparation, cleaning, and

repair, which is exactly what the team used it for. Fig 29.3 shows several small faces on

the fuselage in the CATIA model; this was a drawback of importing the fuselage directly

from OpenVSP rather than designing it in CATIA. The problem was rectified to some

extent in SPACECLAIM, which allowed the removal of several small faces on the

fuselage.

SPACECLAIM was also used for fixing holes, sharp edges, and overlapping faces/curves

in the original geometry, which could have led to problems in meshing. Finally, once the

geometry had been prepared and cleaned, a cylindrical domain was created around the

aircraft, and a Boolean operation was performed to subtract the aircraft geometry from the

domain geometry, which is standard practice for CFD. The fluid domain was also halved

since only longitudinal parameters were of interest, as explained in the introduction. The

138
prepared aircraft geometry is shown in Fig 29.4 and is followed by the final fluid domain

shown in Fig 29.5 with the aircraft geometry subtracted from it.

Fig 29.4 SPACECLAIM model of AFV-21

Fig 29.5 Halved fluid domain with the aircraft geometry removed

139
29.2 Meshing

Explaining the entire meshing process is somewhat tricky since the mesh was gradually

improved over time as problems in convergence and results of the CFD were recognized

and rectified. Following are the significant features of the mesh used in the final case:

29.2.1 Overall Procedure

Although there were several small and large steps involved in the creation of the mesh, the

following are major steps in the final mesh creation conducted in the meshing module

ICEM CFD:

1. Set sizes on different parts and define size function if needed.

2. Create Density boxes to capture wake regions.

3. Create tetra mesh using the octree method.

4. Delete Octree volume mesh and smooth surface mesh.

5. Create volume mesh using the Delaunay method and smooth if needed. Delaunay

mesh was created because it produces smoother mesh and converges faster in

fluent.

6. Create prism layers on aircraft, fix the Y+ using the first layer to set it between 0

and 200, and then split the rest of the layers using a ratio or exponential function

into six or more layers. All of this allowed effective Boundary Layer capturing.

7. Set boundary conditions and export mesh.

29.2.2 Global Mesh Setup

The global element seed size was not set in the global settings since it would be set for

individual parts later. The scale factor was set to 1.0 since it helps visualize mesh size and

avoids a skewed perception due to it being constantly multiplied to maximum and

140
minimum mesh sizes. Curvature and proximity refinement were enabled to refine mesh

near curves and smaller areas; minimum mesh size was set at 0.1 ft since the smallest edge

the team found in the model was of this order; doing this would prevent the mesh from

leaking or developing errors like holes. A refinement of 10 was set, meaning ten edges

would cover a complete circular arc. ‘Elements in gap’ was set to 4 as more of an arbitrary

choice since the minimum mesh size was already too small to be bothered.

The global prism settings were also set here with an exponential growth law, height ratio

of 1.3, and 3 layers. One might inquire why the number of layers was kept so low; the

answer is that a high number of prism layers set in the initial meshing leads to an

exponential increase in meshing time, as was experienced by the team firsthand. Therefore,

creating a small number of layers and splitting them later on, is a much more efficient way

of handling this problem from an iterative standpoint.

29.2.3 Part Mesh Setup

Part mesh setup was relatively straightforward; detailed meshing was required only on the

aircraft surfaces since they are mostly curves and have small edges, which explains the

maximum size of 1.5 ft. The rest of the surfaces, i.e., the Cylindrical wall, symmetry wall,

inlet, and outlet, do not require high detail since they are far removed from the aircraft and

are large and simple shapes compared to the aircraft surfaces; thus, the large size of 40 ft.

Fig 29.6 shows the part mesh setup. Note that only the aircraft surfaces have prism setting

enabled since only the mesh over the aircraft surfaces needs to have boundary layer

capturing capability.

141
 Fig 29.6 Part Mesh Setup in ICEM CFD

Fig 29.7 helps appreciate the impact of the part mesh setup and the differences it imposes

on the detail for the mesh near aircraft surface and the mesh near domain walls.

Fig 29.7 Surface meshes to show contrast in detail between aircraft and domain mesh

detail

29.2.4 Density Boxes

Aircraft generate lift, and therefore they generate wake. Wake capturing is often neglected

in the CFD analysis of such cases, and although extra high detail near the aircraft surface

will lead to effective capture of parasite drag, however, capturing induced drag caused by

vortices generated at the outer edges of the wing while generating lift, requires a fine mesh

behind the regions susceptible to generating these vortices. The team reasoned that

142
prominent regions for vortex generation would be the tips of the wing and the tips of the

horizontal and vertical tails; to capture the vortices, two density boxes with element sizes

of 3.5 ft (smallest possible size with available computational constraints) were generated

behind the wing and the empennage (HT and VT). Fig 29.8 shows that one density box is

placed behind the outer half span of the wing to capture the vortices generated near the tip,

and the other is placed behind the empennage and is large enough to encompass any

possible vortex generation by the HT as well as the VT in case a lateral/directional CFD

case is pursued in the future.

Fig 29.8 Density boxes, one behind the outer wing and the other behind the empennage

Fig 29.9 demonstrates the effect of density boxes on the volume mesh elements using a cut

plane that intersects the aircraft and both density boxes.

143
 Fig 29.9 Top view of a cut plane to demonstrate the effect of density boxes on mesh

detail behind aircraft

This added detail will lead to quicker convergence of the energy equation and a better

estimate of the aircraft’s induced drag.

29.2.5 Prism Layers and Y+

Prism layers are generally preferred in the unstructured tetra mesh to capture or resolve the

strong adverse gradients of the solution within the boundary layer. They can allow for

suitably flow-aligned high aspect ratio cells (without any excessive stream-wise

resolution), thus trying to capture the boundary layer with less numerical diffusion.

Boundary layer flow details can be captured with tetra mesh, but that requires highly

refined tetra cells in the boundary layer region, which can substantially increase the mesh

count, thereby increasing computational effort and hence not preferred. Since prism mesh

144
allows the capture of the boundary layer with a relatively less number of cells, it is preferred

for inflation layers at the boundary [63].

For the prism layer mesh, the total number of layers, growth rate, and first layer thickness

is defined; this is where Y+ comes in since it is a non-dimensional measurement of first

cell height from a wall. The Y+ is usually kept below 200 as a rule of thumb. A CFD course

[64] spreadsheet for Y+ calculation was used to set the first cell height at 0.001312 ft against

a Y+ of 63.10 for the final mesh of the AFV-21.

As explained earlier in section 29.2.2, initially, only three prism layers meshed with default

settings. Then the three layers were split into two layers, giving six prism layers. Following

this, the first layer was adjusted to a height of 0.001312 ft for reasons already discussed.

Finally, to increase fidelity even further, all five layers above the first one were split into

two layers each, giving ten layers. So a total of 11 prism layers were generated to capture

boundary layer flow details. Fig 29.10 shows some zoomed-in photos which demonstrate

the detail of the prism mesh over different areas.

145
Fig 29.10 Different zoomed-in views of prism mesh to demonstrate detail required in BL

detail capturing

29.2.6 Mesh Output with Boundary Conditions

Named selections developed in SPACECLAIM, i.e., Inlet, Outlet, Cylindrical wall,

symmetry, and Aircraft, were instrumental in creating the part mesh setup seen in Fig 29.6.

After the meshing was completed, the mesh file had to be exported to FLUENT. Before

exporting the mesh, the same named selections or parts were used to assign boundary

conditions to different surfaces in ICEM CFD before the mesh file was exported. This

allowed FLUENT to recognize and auto-assign several boundary conditions, reducing

workload in the problem setup phase.

29.3 Problem Setup

This section details how the problem was set up in FLUENT. Essential settings of each

section of the solver tree have been mentioned. The models were selected according to the

recommendations in the CFD course [64].

29.3.1 General

The mesh was checked, upon which FLUENT gave error messages but also suggested

using the inbuilt mesh repair command, which was implemented, and no further issues in

146
the mesh were detected. The cruise conditions reach above M0.6, making the flow

compressible; therefore, the solver type was switched to density-based. The units were

switched to imperial, and the mesh was scaled accordingly.

29.3.2 Models

Since the flow was compressible, the Energy equation was turned on. For the viscous

model, the ‘Spalart-Almaras’ model was selected since it is well suited to external flow

CFD cases.

29.3.3 Materials

Air was selected as the fluid material; density was switched to ‘ideal gas,’ and viscosity

was switched to ‘Sutherland.’ Finally, aluminum was selected as the solid material.

29.3.4 Boundary Conditions

The following table specifies the boundary conditions applied to all named selections:

Table 29.1 Boundary conditions.

Named Condition / Group

S.No Boundary Condition
(Part)
Pressure Far-Field, Mach M0.624 (cruise
1 Inlet Mach), Gauge Pressure 523.48 lb/ft2
(@34000 ft).
Pressure Outlet, Gauge Pressure 523.48
2 Outlet
lb/ft2 (@34000 ft – cruise Altitude).
3 Symmetry Symmetry.
Wall, Specified Shear 0 lb/ft2 in all (X,Y,Z)
4 Cylindrical Wall
components.
5 Aircraft Wall, No Slip Condition.

147
29.3.5 Reference Values

Dimensions of the AFV-21 from Class 2 design were defined here, followed by the cruise

condition values as flow parameters.

29.3.6 Report Definitions

Force reports applicable to a half domain that is symmetric in the aircraft's vertical (XZ

plane) were defined in each case set up for the angles of attack 0, 4, 8, 12. The reports

defined drag, lift, and pitching moment values as well as coefficients. The values obtained

after a reasonably converged solution would be used for further analysis.

Fig 29.11 The Mesh imported in FLUENT (Green – inlet / Red - outlet)

29.4 Results and Discussion

As was mentioned in the introductory section of this chapter, four individual simulation

cases were run for the angles of attack 0, 4, 8, and 12 at cruise conditions, i.e., M0.624 and

34000 ft. Major results which were drawn from these simulations were the Drag, Lift, and

Pitching Moment forces for a symmetric half of the AFV-21; forces and moments thus

obtained were doubled and converted into coefficients. The report plots of the simulations

148
are shown in the following Figs, and all of them depict sufficient convergence for use in

further analysis and discussion.

Fig 29.12 Simulations results for AOA = 00

149
Fig 29.13 Simulation Results for AOA = 40

Fig 29.14 Simulation Results for AOA = 80

150
 Fig 29.15 Simulation Results for AOA = 120

Table 29.2 summarizes the final or converged values of CFD analyses at the angles

mentioned above and doubles them to account for the entire aircraft.

Table 29.2 Tabulation of CFD results

AOA Drag CD Lift CL Moment Cm

0 11723.68 0.041889 73496.6 0.262608 310715.7 0.069112
4 14430.9 0.051563 136101.7 0.4863 45529.16 0.010127
8 24399.28 0.08718 206967.5 0.739509 -10469.8 -0.00233
12 38201.18 0.136495 231877.5 0.828514 -416505 -0.09264

The upcoming sections follow a comparison of the CFD results, the prediction of these

values in Class 2 design, and a discussion of why they may or may not complement each

other.

151
29.4.1 Drag (CD)

Fig 29.16 shows a comparison of the drag coefficients obtained from CFD analysis, the

drag polar obtained from Class 2 methods, and the drag polar estimated by solving for CD0

and K values using drag coefficients obtained at lower angles of attack (00 and 40).

Fig 29.16 Comparison of Drag Coefficient Estimates from CFD and Design Phase

29.4.1.1 Major Observations

Major observations from the above graph are as follows:

1. The drag polar predicted from Low AOAs, i.e., 00 and 40 matches the shape of the

drag polar predicted in chapter 0, indicating that the production of induced drag, at

least at low angles of attack, was accurately estimated. The drag polar derived from

CFD data for low AOAs given in the following equation:

𝐶𝐷0 = 0.0379 + 0.0577 𝐶𝐿2

2. The upward offset of the predicted drag polar from the Class 2 estimate shows that

the parasite drag was underestimated, and the parasite drag characteristics of the

AFV-21 have to be improved, or a significant performance loss will have to be

152
 accepted. The Class 2 drag polar is rewritten in the following equation for

convenience; note the lower value of parasite drag estimate, i.e., CD0:

𝐶𝐷0 = 0.0166 + 0.0503 𝐶𝐿2

3. The curve fit line generated to match CFD drag estimates at high AOAs indicates

that the drag polar format of representation cannot be used to represent drag

behavior at such high AOAs at high speeds (610 ft/s at 34000 ft). Such AOAs are

uncommon at such flight conditions, and extra drag components are generated,

which are not modeled by the drag polar since these angles are unlikely to occur.

The best estimate at this point is that the induced drag factor ‘K’ accurately models

the induced drag generated by vortex generation at low AOAs but fails to accurately

model the large amounts of pressure drag generated at high AOAs.

29.4.1.2 Explanations

The following are reasons behind the higher value of parasite drag at lower AOAs:

1. There is local shock formation above the wing-fuselage junction of the AFV-21 at

all angles of attack in the cruise condition.

2. High-pressure zones form in front of the nose of the aircraft and the closed-off

engines, which leads to extra drag generation. Both are contributors of drag as in

the CFD case; however, the geometry had to be adjusted in this manner to ease

meshing so it is safe to say that the parasite drag value CD0 would not be as high as

predicted by CFD in actual flight.

3. High-pressure zones form at the wing-fuselage junction, and the tips of the wing,

HT, and VT causing excessive pressure drag, which is not accounted for by the

153
 methods used for drag estimation in the design phase since those methods rely on

historical data from aircraft that have served and are aerodynamically refined.

The reason for Sudden drag rise at high AOAs:

1. Flow separation begins soon at high speeds and causes a large increase in pressure

drag not modeled by drag polar.

2. Stronger shock formation above wing-fuselage junction since the Cp is lower above

the wing at higher AOA, as is explained in section 29.5.2.1, which leads to higher

velocities in that region and ultimately causes stronger shock formation, i.e., more

wave drag.

29.4.1.3 Recommended Design Changes

Although the CFD results predict a much higher drag at high AOAs than expected by the

design drag polar, flight in cruise conditions at such high AOAs is unlikely, and these

effects will not have a significant impact on performance as such high AOAs occur at

conditions like take-off and landing, which have short durations and low speeds.

There is a significant offset between the Class 2 drag polar and the low AOA CFD drag

polar, which means that the predicted parasite drag coefficient (CD0) of 0.0166 was much

less than the one estimated using CFD, i.e., 0.0379. The validity of the CFD result may be

debatable; however, the values obtained are not out of the realm of possibility and therefore

merit consideration. The following improvements are recommended for the aircraft

exterior to rectify the increase in drag.

1. Improved wing, HT, and VT tips to smooth out airflow at the tips.

2. Tail cones to streamline the rear extremity of a fuselage by eliminating any base

area, which is the source of base drag.

154
 3. Fillets to smooth out the airflow at the junction between two components like the

fuselage and wing.

29.4.2 Lift (CL)

Fig 29.17 compares the class 2 lift curve estimate for cruise conditions developed using

the techniques in Scahufele [30], which has already been shown in Fig 11.2, and the lift

curve obtained from CFD performed for the AOAs mentioned above at cruise conditions.

Fig 29.17 Comparison of Lift Coefficient Estimates from CFD and Design Phase

29.4.2.1 Major Observations

Major observations from the above graph are as follows:

1. The trend followed by lift coefficient in CFD results matches the estimated

behaviour relatively better than the drag estimates; however, as before, the

deviations increase at high AOAs.

2. The CFD lift curve is lower than the Class 2 estimate; however it is quite close at

lower angles of attack, which matter more than high AOAs due to being prevalent

155
 in the cruise condition, this deficiency may be remedied by an increase in wing

incidence angle or some similar technique to improve aircraft lift generation.

3. Although the linear nature of the Class 2 lift estimate leaves much to be desired in

terms of realism, it seems to be effective at estimating the onset of stall (𝛼𝑠𝑡𝑎𝑙𝑙 =

110 ). This is corroborated by the behaviour of the CFD trend line, which seems to

be approaching stall as well at 120.

29.4.2.2 Explanation

The fact that the CFD lift estimate is lower than the Scahufele [30] estimate is no surprise.

The methods used to make estimates of such characteristics as lift and drag coefficients

rely on historical data and empirical methods since it seems almost impossible to develop

methods so complex that they may fathom the variability and complexity of each design

conceived in the conceptual phase; the fact that both curves are close confirms that the

Class 2 estimate was, to a large extent, accurate. While the validity of the CFD analysis is

also debatable, every precaution was taken in meshing and case set up to ensure high

fidelity and accuracy, so we assume it is accurate. The Class 2 lift estimate at angles higher

than 80 seems to deviate from reality as the CFD trend shows a significant departure from

linear behaviour. This presents the conclusion that for the next iteration of the design, the

Scahufele [30] method is not to be trusted at AOAs higher than 80 and that the HLDs should

be overdesigned to compensate for the expected loss of lift at higher AOAs.

29.4.2.3 Recommended Design Changes

In light of the above discussion, the following changes are recommended for the AFV-21

in the next iteration of its design:

156
 1. CFD trend shows slightly lower lift coefficients at low AOAs; this is problematic

as the pilot would need to cruise at higher AOA, which means more drag and less

performance. A simple remedy would be to increase the wing incidence angle (2.30)

by 30 (judging from Fig 29.17); this, however, could lead to an increase in induced

drag due to more lift generation by the wing.

2. The wing area (S) could be increased instead of the wing incidence to avoid

excessive induced drag while simultaneously achieving the required lift increase.

However, increasing wing area would lead to increased parasite drag due to a larger

wetted area.

3. An optimization algorithm could be designed to reach the best compromise between

wing area increase and incidence increase.

4. Winglets could also be considered to increase lift and reduce the drag due to vortex

generation. Winglets, however, are only useful for a single flight condition and can

be detrimental to others, which might impact the aircraft's versatility in its mission

capabilities.

29.4.3 Pitching Moment (Cm)

Fig 29.18 compares the Class 2 estimates of pitching moment for the cruise condition

discussed in section 23.3 and the moment estimates from CFD at the cruise condition for

the AOAs mentioned above. Again, the curves seem to be in reasonable agreement, and

there seem to be no noticeable deficiencies in the CFD trend, which may prompt a design

improvement in a later iteration.

157
 Fig 29.18 Comparison of Moment Coefficient Estimates from CFD and Design Phase

The values of Class 2 analysis and CFD analysis are compared in Table 29.3, where both

curves show a positive value of the pitching moment at zero AOA (𝐶𝑚0 > 0) and show a

negative slope of pitching moment with increasing AOA i.e. (𝐶𝑚𝛼 < 0). This indicates

that the AFV-21 is inherently stable in the longitudinal axis and does not require any

improvements. Changes in the Dynamic stability characteristics caused due to this change

in these values could reveal problems and the need for design changes. Still, such an

analysis would only be legitimate if all longitudinal stability coefficients were determined

using CFD analysis which is out of this project’s scope.

Table 29.3 Pitching Moment Coefficient comparison

Class 2 (DATCOM) Analysis CFD Analysis

𝑪 𝒎𝟎 0.0823 0.0707
𝑪 𝒎𝜶 -0.0065 - 0.0124

158
29.5 Post Processing and Flow Visualization

The major thrust of the entire CFD analysis was to determine the values of ‘lift,’ ‘drag’,

and ‘pitching moment’ coefficients at different AOAs and compare them to the coefficients

which were predicted using textbook methods during the design effort; if the CFD analysis

was done accurately (which it was, as suggested by the reasonable ranges so accrued) it

would allow the team to:

1. Determine the extent of inaccuracy in lift, drag and moment estimations, i.e.,

streamlining the successive design iterations by shortlisting accurate estimation

methods and discarding/replacing inaccurate methods.

2. Decide which aspects of the design need to be improved and to what extent. The

suggestions made in sections 15429.4.1.3 and 29.4.2.3 are an example.

3. Find out exactly how the airflow interacts with the aircraft exterior and which areas

need the most improvement, such as the nose and wing-fuselage junctions which

seem to generate high values of pressure drag and could be improved by some

streamlining / filleting.

4. Uncover any peculiar or intriguing phenomenon which can only be explained or

compensated for if ‘found’ through flow visualization.

The purpose of section 29.5 is to cater to points 3 and 4 since points 1 and 2 have been

addressed in section 29.4.

29.5.1 Contours and Vector Diagrams for Flow Visualization

The discussion and comparisons in section 29.4 suggest that the CFD conducted has given

quite accurate results; the team thereby recognised this as a perfect opportunity to conduct

extensive post-processing of the available data to understand how the flow interacts with

159
AFV-21’s exterior in cruise and how AOA changes affect that interaction. The Figs in this

section represent the team’s attempt to recognise how different flow properties change

across the aircraft’s exterior at different AOAs and to diagnose problem areas like high

static pressure zones, which can be corrected in later design iterations.

160
29.5.1.1 Mach No. Contours

Fig 29.19 Velocity Contours (Mach No.) for AOA = 00 Fig 29.20 Velocity Contours (Mach No.) for AOA = 40

Fig 29.21 Velocity Contours (Mach No.) for AOA = 80 Fig 29.22 Velocity Contours (Mach No.) for AOA = 120

161
29.5.1.2 Pressure Coefficient Contours

Fig 29.23 Pressure Coefficient for AOA 00 Fig 29.24 Pressure Coefficient for AOA 40

Fig 29.25 Pressure Coefficient for AOA 80 Fig 29.26 Pressure Coefficient for AOA 120

162
29.5.1.3 Total Pressure Contours

Fig 29.27 Total Pressure for AOA = 00 Fig 29.28 Total Pressure for AOA = 40

Fig 29.29 Total Pressure for AOA = 80 Fig 29.30 Total Pressure for AOA = 120

163
29.5.1.4 Static Pressure Contours

Fig 29.31 Static Pressure for AOA = 00 Fig 29.32 Static Pressure for AOA = 40

Fig 29.33 Static Pressure for AOA = 80 Fig 29.34 Static Pressure for AOA = 120

164
29.5.1.5 Velocity Vectors

Fig 29.35 Flow Past Aircraft at AOA = 00 Fig 29.36 Flow Past Aircraft at AOA = 40

Fig 29.37 Flow Past Aircraft at AOA = 80 Fig 29.38 Flow Past Aircraft at AOA = 120

165
29.5.2 Visualizing Different Flight Phenomena

29.5.2.1 Investigating Shock Formation and Drag Rise at High AOAs

A high Parasite drag was observed in comparison to the estimated parasite drag, as has

been observed in section 29.4.1.1; some of it could be attributed to the fact that the engine

mouth was closed, and the nose was not rounded. Both of these adjustments were made to

ease the meshing process, and both generate a lot of pressure build-up in the front of the

aircraft, as seen in Fig 29.23 to Fig 29.25; this causes extra pressure drag which would not

exist in actual flight. These factors, however, do not explain the entire disparity among the

design and CFD drag values, especially at higher AOAs; the possibility of local shock

formation had to be investigated since the AFV-21 cruises at M0.624 which is considered

high subsonic and is close to the transonic bracket specified by most authors. Changes in

flow properties across a shockwave are:

1. Flow Velocity decreases

2. Static pressure increases

3. Static temperature decreases

4. Total pressure decreases

The following Figs are contours of the above-mentioned properties and have been observed

for the lowest and highest AOAs, i.e., 00 and 120, respectively. This allows the observer to

investigate shock formation at the aforementioned AOAs and their role in the associated

drag behaviour. Furthermore, the contours have been developed at two locations, i.e.

fuselage symmetry line and mid-wing, since fuselage and wing are likely regions for shock

development.

166
29.5.2.1.1 Velocity Decrease Check

The top of the centre fuselage shows a local shock formation in both Fig 29.39 and Fig

29.40. The velocity decrease across the shock wave is more pronounced in AOA 12 0 case

(923 ft/s to 527 ft/s) than AOA 00 case (931 ft/s to 665 ft/s); this explains why the AFV-21

has a higher CD0 value than predicted in the design phase and why the drag rises so

dramatically at high speed and deviates from both design and CFD low AOA drag polars,

i.e., the wave drag need to be catered for in the drag estimations. The wave drag can be

eliminated by area-ruling at this wing-fuselage region since the addition of a wing here

causes a sharp rise in cross-sectional area at this fuselage station which is a known

contributor to shock formation and wave drag; this phenomenon has been explained well

in chapter 12 of Raymer [21].

Airfoils show normal behaviour as there is no sharp decline in velocity values at both

AOAs 00 and 120. There is a region of low velocity above the trailing edge in the AOA 120

case; however, it is not indicative of shock formation as flow tends to slow down there due

to area increase after the streamline passes the airfoil’s top region. There is reverse flow in

that region which indicates the beginning of the stall phenomenon due to boundary layer

separation caused by adverse pressure gradients, better visualized in Fig 29.62.

167
 Fig 29.39 Velocity Contours at AOA = 00

Fig 29.40 Velocity Contours at AOA = 120

Fig 29.41 Inboard Airfoil Velocity Contours at AOA = 00

168
 Fig 29.42 Inboard Airfoil Velocity Contours at AOA = 120

29.5.2.1.2 Static Pressure Increase Check

Apart from the shock region already pointed out in the preceding section, i.e., above the

wing-fuselage centre, several other points of interest are visible in Fig 29.43 and Fig 29.44.

For example, there are points of high static pressure in front of the nose and the closed-off

engines, which, as was mentioned earlier in section 29.5.2.1, are likely contributors to the

drag rise discussed in section 29.4.1. Another interesting point is the difference in low-

pressure region and size above the wings between the AOA 00 and 120 cases, where the

low-pressure region in Fig 29.44 is larger since more lift is generated at high AOA.

Consequently, the region has travelled more towards the LE, i.e. the classic forward motion

of the Centre of Pressure at a higher AOA can be observed, and the airfoil contour plots

corroborate this contrast between the AOAs for the same respective cases in Fig 29.45 and

Fig 29.46.

169
 Fig 29.43 Static Pressure Contours at AOA = 00

Fig 29.44 Static Pressure Contours at AOA = 120

Fig 29.45 Inboard Airfoil Static Pressure Contours at AOA = 00

170
 Fig 29.46 Inboard Airfoil Static Pressure Contours at AOA = 120

29.5.2.1.3 Static Temperature Increase Check

Same observations can be made from the static temperature plots as were made in the above

sections, i.e., shock occurs at both AOAs and is more aggravated for the AOA 120 case as

is made prominent in Fig 29.48, by a larger shock wave and a larger temperature increase

across said shock wave (480 0R to 544 0R) as compared to the AOA 00 case with a smaller

increase (499 0R to 521 0R).

Fig 29.47 Static Temperature Contours at AOA = 00

171
 Fig 29.48 Static Temperature Contours at AOA = 120

Fig 29.49 Inboard Airfoil Static Temperature Contours at AOA = 00

Fig 29.50 Inboard Airfoil Static Temperature Contours at AOA = 120

172
29.5.2.1.4 Total Pressure Decrease Check

Same observations can be made with the total pressure contours that have been made in

sections 29.5.2.1.1, 29.5.2.1.2 and 29.5.2.1.3. In addition to the obvious shock above the

wing-fuselage junction, sharp decreases in total pressure are also visible at the tips of Wing

and VT, which are suggestive of local shock formations. Therefore, smoothing of these

tips or additions of winglets along with appropriate area ruling of the fuselage is

necessitated for the next design iteration of the AFV-21 to achieve the drag characteristics

required to meet the performance requirements of the RFP [1].

Fig 29.51 Total Pressure at AOA = 00

Fig 29.52 Total Pressure at AOA = 120

173
 Fig 29.53 Inboard Airfoil Total Pressure at AOA = 00

Fig 29.54 Inboard Airfoil Total Pressure at AOA = 120

29.5.2.2 Boundary Layer, Viscous Flow and Flow Separation

As explained in section 29.2.5, during the meshing process, prism layers were made to

capture or resolve the strong transverse gradients of the solution within the boundary layer.

Y+ was also discussed in section 29.2.5 as a non-dimensional description of fist layer

height. Fig 29.55 shows the Y+ values for the entire aircraft surface, and almost all Y+

values lie below 30, which is an indicator of a good prism mesh per the guidelines of the

174
CFD course [64], as Y+ is recommended to be kept below 200 for accurate Boundary Layer

capturing.

Based on the above discussion and the detailed discussion on prism meshing in section

29.2.5, we can safely establish that the simulation was able to accurately capture boundary

layer phenomena such as ‘no slip condition’, ‘velocity profile generation’, ‘viscous drag

capturing’ and ‘flow separation due to adverse pressure gradient’. Furthermore, the

statement is corroborated by the representation of the velocity profile of the boundary layer

over a wing section in Fig 29.56, which matches the schematics of the boundary layer in

Aerodynamics textbooks such as Anderson [65].

Friction in the boundary layer causes friction drag as well as pressure drag due to a

phenomenon called flow separation. Flow separation is discussed in great detail in

Anderson [65]. In summary, at lower angles of attack, once the flow crosses the point of

minimum pressure, which is also the point of maximum velocity (visible in Fig 29.57 and

Fig 29.61 above the airfoil’s maximum thickness point), the flow begins to slow down and

recover its pressure which then leads to it facing an adverse pressure gradient. Since this

adverse pressure gradient is very low at low AOAs, i.e. 00, as in Fig 29.57 and Fig 29.61,

the flow remains attached to the airfoil. However, at a higher AOA, the adverse pressure

gradient is high; in other words, the gap between freestream and lowest pressure becomes

so high that the adverse pressure gradient becomes severe, and the flow separates or begins

to separate from the surface, depending upon how high the AOA is. The contrast between

an attached and separating boundary layer is quite visible in Figs Fig 29.63 and Fig 29.64,

representing AOAs 00 and 120, respectively. Fig 29.64 also reveals recirculation into the

region of separation where small opposite arrows represent the low-speed eddies which

175
inhabit the region of separated flow and low pressure, a region of low pressure which

ultimately leads to a larger net pressure drag on the airfoil due to a lower pressure being

exerted at the rear of the airfoil, as has been observed and discussed in sections 29.4.1 and

29.5.2.1.

Fig 29.55 Wall Y+ (Same for all cases)

Fig 29.56 Close-up Boundary Layer Visualization at AOA = 120

176
Fig 29.57 Boundary layer Visualisation for AOA = 00 Fig 29.58 Boundary layer Visualisation for AOA = 120

Fig 29.59 Close-up of Boundary Layer Velocity Fig 29.60 Close-up of Boundary Layer Velocity

Contour Visualisation at AOA = 00 Contour Visualisation at AOA = 120

177
Fig 29.61 Velocity Vectors around wings at AOA = 00 Fig 29.62 Velocity Vectors around wings at AOA = 120

Fig 29.63 Close-up of Boundary Layer Velocity Fig 29.64 Close-up of Boundary Layer Velocity

Vector Visualisation at AOA = 00 Vector Visualisation at AOA = 120

178
29.5.2.3 Investigating CP and Flow Separation at different Span-wise Locations at

High AOA

A geometric twist of -30 was given in the design phase for two reasons, to approach

elliptical lift distribution as explained in section 15.2 and to avoid tip stalls at a high AOA

so that aircraft can retain aileron effectiveness. This section is aimed at verifying whether

the twist actually helps avoid tip stall or not.

Looking at Fig 29.65, Fig 29.66 and Fig 29.67, which represent the wing spanwise stations

8 ft, 30 ft and 56 ft, respectively, it is clear that even at a high AOA of 120, the twist is

reducing flow separation as the wing span increases.

Observing the chordwise pressure coefficient distributions in Fig 29.68 for the

aforementioned spanwise locations, the relationship between flow separation and adverse

pressure gradient seems to be at play. For example, at 8 ft, the adverse pressure gradient

after crossing the point of minimum pressure near the LE is the highest of the three stations;

the gradient can become less severe as the span location increases, i.e., flow separation

becomes less likely as we go farther along the span.

The disadvantage of geometric twist is also visible in Fig 29.68, where the twist causes a

reduction in the local angle of attack perceived by the airfoil and leads to a reduction in the

section lift produced therein as the spanwise location of the section increases.

179
Fig 29.65 Flow at wing location Y = 8 ft

Fig 29.66 Flow at wing location Y = 30 ft

Fig 29.67 Flow at wing location Y = 56 ft

180
 Fig 29.68 Comparison of chord-wise Cp distributions at different spanwise locations

29.5.2.4 Vortex Formation

This section shows Figs which attempt to capture the vortex formation and downwash

phenomenon for the AFV-21. These velocity contours were made behind the wing,

empennage and far behind the aircraft till the end of the first density box developed to

capture wing wake (shown in Fig 29.9) for the AOAs 40 and 120 to show the effects of

higher lift generation on vortex ansd downwash sizes.

A finite wing has a higher drag and lower lift than a two-dimensional airfoil; the reason is

the formation of wingtip vortices and the downwash component of velocity in the vicinity

of the wing. Since the air can move in a spanwise direction in finite wings and there is

higher pressure below the wing, the air tends to leak upwards from the tip. This flow

establishes a circulatory motion that trails downstream of the wing and forms a vortex. The

vortices drag the surrounding air with them and develop a velocity component in the

downward direction at the wing, and the downward component is called downwash. Major

observations from the following Figs are:

181
1. All Figs confirm the formation of vortices behind the wing tips and downwash

regions behind the wing.

2. Comparison between contours of AOA 00 and 120 shows that the vortex and

downwash regions formed for 120 are larger, more distinct and have sharper

gradients; this contrast is especially visible between Fig 29.69 and Fig 29.70. The

obvious reason for this difference is more lift generation at the 120 AOA.

3. The vortices become larger as they move farther back from the wing, but the

velocity gradients reduce as well. This change is especially notable in the Figs

associated with the 120 AOA.

Fig 29.69 Downwash region and vortex behind wing at AOA = 40

182
 Fig 29.70 Downwash region and vortex behind wing at AOA = 120

Fig 29.71 Downwash region and vortex behind empennage at AOA = 40

Fig 29.72 Downwash region and vortex behind empennage at AOA = 120

183
Fig 29.73 Downwash region and vortex at the end of density box at AOA = 40

Fig 29.74 Downwash region and vortex at the end of density box at AOA = 120

184
 30. FINITE ELEMENT ANALYSIS (FEA)

Finite element analysis was carried out ‘only’ for the wing due to the lack of available time

and computational resources. Moreover, the FEA of the whole aircraft required a Fluid

Structure Interaction (FSIs) model for the whole aircraft’s internal structure, which was not

possible due to the high computational power required to complete the simulation of over

500 body elements in the assembly with a mesh of over 50 million elements. The analysis

was carried out on ANSYS Workbench 2021 R1.

30.1 Geometry

The wing structure generated in SOLIDWORKS, also shown in section 24.2, was

converted to the stp format and then exported to SPACECLAIM, an inbuilt CAD tool for

ANSYS. The geometry was then simplified, stringers were removed as they required

extremely refined mesh, and the team did not have access to the computational resources

needed for such meshing or solution. The geometry was then split into separate parts using

the split body function so that ANSYS does not apply the same meshing techniques all

over the wing structure, which almost always results in a failed mesh. Finally, the ribs were

split into three parts separated by the front and rear spars, and the geometry module

automatically generated a total of 168 contact regions between the different bodies.

185
 Fig 30.1 Wing Structure Geometry in ANSYS Mechanical

30.2 Material

The ribs and spars were assigned Al 7075-T6 for reasons discussed earlier in section 24.2

and shown in section 24.5. The FEA focused on finding the maximum stress bore by the

spars and ribs under the positive limit load factor by using this material.

30.3 Meshing

Global meshing controls were used to generate a mesh with adaptive sizing ‘off’ and only

proximity ‘on’. Program-controlled inflation was used with a smooth transition to generate

a fine mesh with an average element quality of 0.73 and 0.6 million elements with 4.3

million nodes. A more refined mesh was also generated with an average element quality of

0.87, but the computer stalled during the simulation due to a high number of small elements

(about 10 million) and low processing power; therefore, the team decided to go forward

with the former mesh.

186
 Fig 30.2 Wing Structure Mesh

Fig 30.3 Wing Structure Mesh in Close-up

It is apparent from Fig 30.3 that the various corners and curves in the geometry were

efficiently captured using the previously explained mesh settings.

30.4 Boundary Conditions and Loading

A single boundary condition was applied to the structure. First, fixed support was added to

the root rib to model the wing as a cantilever beam. Then, a pressure loading diagram was

187
obtained from MATLAB using the total wing loading diagram shown in section 24.1,

whereby the pressure loading in lb/ft2 was plotted against the wing span in ft. The pressure

loading was then exported to an xml file and imported to ANSYS. It was assumed that the

front spar bears 75% of the total pressure while the rear spar bears about 25% of the total

pressure. This assumption was based on Schrenk’s chordwise lift distribution. The loadings

were applied to the front and rear spar on the base of the wing. The forces acted upwards

in the y-direction and the pressure varied along the z-axis.

Fig 30.4 Pressure Loading for Front Spar

188
 Fig 30.5 Pressure Loading for Rear Spar

Fig 30.6 Boundary and Loading Conditions for Wing FEA

30.5 Solution and Results

Using the boundary conditions and loadings described in the previous section, the FEA

was solved for total deformation, equivalent stress and safety factor.

189
30.5.1 Total Deformation

The mid-wing incurred a maximum total deformation of 29.4, whereas spar portions near

the root did not experience much deformation since they were fixed at the root.

Fig 30.7 Total Deformation of Wing Structure

30.5.2 Equivalent Stress

The wing experienced maximum stress of 28226 psi near the root due to the high-pressure

loading. The yield strength of the Al 7075-T6 is about 73000 psi. Theoretically, it was

found in section 24.2 that the wing experienced maximum stress of about 54000 psi at the

root; therefore, it was verified by the FEA that the wing would be able to withstand the

high loads at maximum g’s using the current materials. Moreover, a less costly material

with much lesser yield strength, such as Al 2024-T3, with a yield strength of 42100 psi,

can be selected in later design iterations to fabricate wing ribs which would drive down

production costs significantly.

190
 Fig 30.8 Equivalent Stress on Wing Structure

Fig 30.9 Close-up showing Maximum Equivalent Stress near Root

191
30.5.3 Safety Factor

Fig 30.10 Safety Factor for Wing Structure

The following formula defines safety-factor:

Local Equivalent Stress

S. F =
Material Yield Strength

A minimum safety factor of 1.2846 was obtained from the FEA from the above Fig, thereby

confirming that the local equivalent stress did not exceed the material’s yield strength and

that the material would not fail in such conditions during maximum g’s. Although the wing

was designed against a safety factor of 1.5 a safety factor of 1.3 is also appreciable

considering that the stringers were not modelled for the FEA due to lack of computational

resources.

192
 31. SUSTAINABLE DEVELOPMENT GOALS

Sustainable Development Goals or SDGs are set of goals adopted by the United Nations

General Assembly on the 25th of September 2015. The resolution aims to eradicate poverty

through the collaboration of stakeholders and countries [66], [67]. The 17 goals are to be

achieved over 15 years, that is, by the end of 2030. The goals that can be facilitated through

our FYP are explained in depth below.

31.1 Good Health and Well-Being

Among the seventeen sustainable development goals established by the United Nations,

good health and well-being are among the most important; Good health of every individual

in a society is a fundamental right. The main aim of this goal is to increase the life

expectancy within society and to eradicate life-threatening diseases.

Forests play a vital role in keeping the environment healthy. They provide oxygen to

humans and absorb the emitted carbon dioxide, increasing life expectancy and making the

environment human friendly.

Extermination of forests caused by factors such as wildfire can put human health and well-

being at risk since we directly benefit from forests. However, wildfires can be controlled

using firefighter aircraft such as the AFV-21, consequently saving forests and humans

alike.

31.2 Clean Water and Sanitation

Availability of clean water for drinking and adequate sewage disposal is one of the

Sustainable development goals. The annihilation of forests leads to desertification which

causes water scarcity. Furthermore, the unavailability of water leads to poor sanitation,

193
which paves the way for chronic diseases; this is also linked to the previously discussed

goal, i.e., good health and well-being. In addition to deforestation by humans, wildfires are

also taking away huge chunks of forests, especially now due to global warming. The

burning of forest leaves the fertile land arid and barren. The AFV-21, being a firefighter

aircraft, can help stop the advance of wildfires.

31.3 Climate Actions

Climate is changing, and the temperature of the planet is constantly rising. This climatic

change due to global warming is alarming and can cause a severe increase in the number

and severity of wildfires, as we have seen recently in Australia and Brazil.

Measures must be taken urgently to combat the climatic change, and any impact caused by

the changing climate must be tackled. Although reversing climate change may remain a

debate and a prolonged process for the foreseeable future, immediate actions to save the

forests include the EIS of aircraft like the AFV-21 specifically designed to control

wildfires.

31.4 Life Below Water

The goal is to conserve the ocean water and save the life beneath the water. Unfortunately,

the emission of hazardous carbon dioxide due to the burning of forests has accelerated the

acidification of oceans which in turn has affected marine life catastrophically.

Marine life can be protected, and ocean resources can be sustained by controlling the

wildfires, a goal which can be facilitated using aerial firefighting aircraft.

194
31.5 Life on Land

The goal is to conserve terrestrial life, protect the ecosystem, combat desertification, halt

wildfires and safeguard the forests. Such a goal can be pursued using a firefighter aircraft

like the AFV-21 to protect land resources like crops and forests from fires.

195
 32. CONCLUSION

The introduction of a purposely built Large Airtanker (LAT) is necessary to counter the

increasing number and intensity of wildfires across the globe. The AIAA RFP outlines the

requirements for a suitable, purposely built replacement to the aging fleet of converted

aircraft currently serving as LATs.

This report details the trade studies and analyses utilized to develop the AFV-21 in

compliance with RFP requirements and major FAA regulations. Also demonstrated is the

iterative and evolutionary nature of the procedure followed in the design effort as

parameters such as MTOW and AR change from 159535 lb and 9.0 in class 1 design to

153284 lb and 7.6 in class 2 design; and as the complexity and intricacy of the methods

used, increase with increasing rigidity and decisiveness in the configuration and its

geometric parameters.

With a conventional high wing configuration, the AFV-21 has a 380 nm design radius with

full payload, a 6000 gal retardant capacity, a 2740 nm ferry range, and a drop speed of 131

kts in clean configuration. The combination of a large wingspan of 122.08 ft, high camber

of the NACA 653-618, and two of the latest LEAP-1A high bypass turbofan engines

ascertains exceptional low speed performance as well as efficient cruise. EIS is expected

in 2030 with a unit price of $ 93.7 million.

196
 REFERENCES

[1] AIAA, “Request for Proposal Responsive Aerial Fire Fighting Aircraft,” 2021.
[2] S. Doerr and C. Santín, “Wildfire: A Burning Issue for Insurers?,” [Online].
Available: https://assets.lloyds.com/assets/pdf-risk-reports-wildfire-final/1/pdf-
risk-reports-Wildfire-FINAL.pdf.
[3] “Data Explorer | MTBS.” https://apps.fs.usda.gov/lcms-viewer/mtbs.html (accessed
Dec. 04, 2021).
[4] CAL FIRE, “Firefighting Aircraft Recognition Guide,” [Online]. Available:
www.fire.ca.gov.
[5] K. Hoover and L. A. Hanson, “Wildfire Statistics,” [Online]. Available:
https://fas.org/sgp/crs/misc/IF10244.pdf.
[6] F. Tiernan and E. O’Mallon, “Australia’s 2019-20 bushfire season | The Canberra
Times | Canberra, ACT.”
https://www.canberratimes.com.au/story/6574563/australias-2019-20-bushfire-
season/ (accessed Feb. 08, 2022).
[7] “Russia forest fire damage worst since records began, says Greenpeace.”
https://www.rferl.org/a/russian-forest-fires-greenpeace/31467967.html (accessed
Feb. 08, 2022).
[8] “Area burned by forest fires in Canada 2000-2020 | Statista.”
https://www.statista.com/statistics/553520/area-burned-of-forest-fires-canada/
(accessed Feb. 08, 2022).
[9] “Spain fire: Thousands flee blaze near Costa del Sol town - BBC News.”
[10] M. Carlowicz, “Fire Consumes Large Swaths of Greece,” Aug. , Accessed: Feb. 08,
2022. [Online]. Available: https://earthobservatory.nasa.gov/images/148682/fire-
consumes-large-swaths-of-greece.
[11] H. N. Jong, “2019 fires in Indonesia were twice as bad as the government claimed,
study shows.” https://news.mongabay.com/2021/12/2019-fires-in-indonesia-were-
twice-as-bad-as-the-government-claimed-study-shows/ (accessed Feb. 08, 2022).
[12] N. Sönnichsen, “Portugal: wildfire area burned 2021.”
https://www.statista.com/statistics/1265367/area-burned-by-wildfire-in-portugal/
(accessed Feb. 08, 2022).
[13] D. Wang, “How Are Planes Decommissioned and How Much Value Can Be
Salvaged?” https://www.flexport.com/blog/decommissioned-planes-salvage-value/
(accessed Nov. 27, 2021).
[14] “Neptune Today - Neptune Aviation.” https://neptuneaviation.com/neptune-today/
(accessed May 05, 2022).
[15] “Pakistan Aeronautical Complex Kamra - amf-history.”
https://www.pac.org.pk/amf-history (accessed May 02, 2022).

197
[16] M. B. O’Connor and J. M. Lincoln, “Aviation-Related Wildland Firefighter
Fatalities — United States, 2000–2013.”
[17] International Civil Aviation Organization (ICAO), “Remotely Piloted Aircraft
System (RPAS) Concept of Operations (CONOPS) for International IFR
Operations,” [Online]. Available:
https://www.icao.int/safety/UA/Documents/RPAS CONOPS.pdf.
[18] T. H. Cox, C. J. Nagy, M. A. Skoog, and I. A. Somers, “Civil UAV Capability
Assessment.”
[19] R. Austin, Unmanned Aircraft Systems. .
[20] “PROCEDURES AND CRITERIA for the Interagency Airtanker Board(IAB),”
2013.
[21] D. Raymer, “Aircraft Design: A Conceptual Approach, Sixth Edition.”
[22] J. Roskam, Airplane Design. .
[23] D. Howe, “Aircraft Conceptual Design Synthesis.”
[24] USDA Forest Service, “Wildland Fire Management Aerial Application Study.”
[25] I. H. Abbott, A. E. Von Doenhoff, and L. S. Stivers, “REPORT No. 824 -
SUMMARY OF AIRFOIL DATA.”
[26] D. Scholz, “Aircraft Design 8: High Lift Systems and Maximum Lift Coefficients,”
Hambg. Open Online Univ., [Online]. Available: http://www.fzt.haw-
hamburg.de/pers/Scholz/HOOU/AircraftDesign_8_HighLift.pdf.
[27] G. Dimitriadis and O. Leonard, “Lecture 2: Aircraft Propulsion G.” [Online].
Available: http://www.ltas-cm3.ulg.ac.be/AERO0023-
1/ConceptionAeroTurbomachine.pdf.
[28] “CFM LEAP vs P&W GTF » AirInsight.” https://airinsight.com/cfm-leap-vs-pw-
gtf/ (accessed May 03, 2022).
[29] J. D. Mattingly, W. H. Heiser, and D. T. Pratt, “Aircraft Engine Design.”
[30] R. D. Schaufele, “The Elements of Aircraft Preliminary Design.” .
[31] L. M. Nicolai, G. E. Carichner, J. a Schetz, and L. M. L. Malcolm, Fundamentals of
Aircraft and Airship Design Volume I — Aircraft Design Library of Congress
Cataloging-in-Publication Data. .
[32] C. Gologan, “A Method for the Comparison of Transport Aircraft with Blown
Flaps.”
[33] “Fuselage Upsweep Drag.”
http://aerodesign.stanford.edu/aircraftdesign/drag/upsweepdrag.html (accessed Jan.
14, 2022).
[34] J. D. Anderson, Aircraft Performance and Design. .
[35] B. Rodriguez, N. P. Lareau, D. E. Kingsmill, and C. B. Clements, “Extreme
Pyroconvective Updrafts During a Megafire.”

198
[36] M. H. Sadraey, Aircraft design: a systems engineering approach. .
[37] The Goodyear Tire and Rubber Company, “GOODYEAR TIRE BOOK 2018.”
[38] R. H. Daugherty, “A Study of the Mechanical Properties of Modern Radial Aircraft
Tires,” Accessed: Apr. 29, 2022. [Online]. Available:
https://www.researchgate.net/publication/24318512_A_Study_of_the_Mechanical
_Properties_of_Modern_Radial_Aircraft_Tires.
[39] E. Torenbeek and H. Witternberg, “Synthesis of Subsonic Airplane.” .
[40] J. Roskam, “Methods for estimating Stability and control Derivatives for
Conventional Subsonic Airplanes.” .
[41] “FP7-4 : INTRODUCTION TO RELIABILITY AND FAULT TOLERANCE
Course-related Internet Resources,” pp. 4–6, 1996, Accessed: Apr. 08, 2022.
[Online]. Available: http://homes.et.aau.dk/yang/course/IRFT/.
[42] I. Moir and A. Seabridge, Aircraft Systems Mechanical, electrical, and avionics
subsystems integration. .
[43] “Boeing 777 - program supplier guide | Airframer.”
https://www.airframer.com/aircraft_detail.html?model=B777 (accessed Mar. 20,
2022).
[44] C. B. Watkins, “Integrated modular avionics: Managing the allocation of shared
intersystem resources.”
[45] J. W. Ramsey, “Integrated Modular Avionics: Less is More - Aviation Today.”
https://www.aviationtoday.com/2007/02/01/integrated-modular-avionics-less-is-
more/ (accessed Mar. 20, 2022).
[46] “Flight Controls and Autopilots.”
https://aerospace.honeywell.com/us/en/learn/products/cockpit-systems-and-
displays/flight-controls-and-autopilots (accessed Mar. 20, 2022).
[47] X. Kang, “Optimal Tank Design and Directional Dynamic Analysis of Liquid Cargo
Vehicles under Steering and Braking.”
[48] A. Dasgupta, “Effect of Tank Cross-Section and Longitudinal Baffles on Transient
Liquid Slosh in Partly-Filled Road Tankers,” [Online]. Available:
http://eprints.uanl.mx/5481/1/1020149995.PDF.
[49] “Home — Air Tractor’s Fire Response Dispersal System, Gen III.”
https://www.frdsgeniii.com/home (accessed Mar. 20, 2022).
[50] NTT Advanced Technology Corporation, “HIREC : Super Hydrophobic Water
Repellent Coating.” http://www.hirecpaint.com/ (accessed Mar. 20, 2022).
[51] R. C. Nelson, Flight Stability and Automatic Control. 1998.
[52] M. C. Niu, Airframe Structural Design. .
[53] O. Schrenk, “A Simple Approximation Method for Obtaining the Spanwise Lift
Distribution.”
[54] T. H. G. Megson, Aircraft Structures for Engineering Students. .

199
[55] J. L. Decker, “Horizontal Tail Loads in Abrupt Pull-Ups from Level Flight.”
[56] E. F. Bruhn, “Analysis and Design of Flight Vehicle Structures.” .
[57] S. Gudmundsson, General Aviation Aircraft Design: Applied Methods and
Procedures. 2013.
[58] G. E. Carichner and L. M. Nicolai, Fundamentals of Aircraft and Airship Design
Volume 2 - Airship Design and Case Studies. .
[59] L. M. Nicolai and G. E. Carichner, Fundamentals of Aircraft and Airship Design
Volume I, vol. I. 1801.
[60] “Pakistan Aeronautical Complex Kamra - JF-17 Thunder Aircraft.”
https://www.pac.org.pk/jf-17 (accessed May 02, 2022).
[61] R. B. Koopan and K. Laney, “Business Jet Aircraft Industry: Structure and Factors
Affecting Competitiveness. Investigation No. 332-526.,” [Online]. Available:
https://www.usitc.gov/publications/332/pub4314.pdf.
[62] “Pakistan Aeronautical Complex Kamra - Aircraft Manufacturing Factory.”
https://www.pac.org.pk/amf (accessed May 02, 2022).
[63] “(23) In CFD, why prism layers are needed? - Quora.” https://www.quora.com/In-
CFD-why-prism-layers-are-needed (accessed Jul. 18, 2022).
[64] “Mastering Ansys CFD (Level 2) | Udemy.”
https://www.udemy.com/course/mastering-ansys-cfd-part-
ii/learn/lecture/28014400#reviews (accessed Jul. 18, 2022).
[65] J. D. Anderson, Introduction to flight, vol. 19, no. 1. .
[66] A. Baeyens and T. Goffin, “European Court of Justice,” Eur. J. Health Law, vol. 22,
no. 5, pp. 508–516, 2015, doi: 10.1163/15718093-12341375.
[67] United Nations, “Work of the Statistical Commission pertaining to the 2030 Agenda
for Sustainable Development,” Glob. Indic. Framew. Sustain. Dev. Goals targets
2030 Agenda Sustain. Dev., vol. 11371, no. July, pp. 4–25, 2017, [Online].
Available:
https://upload.wikimedia.org/wikipedia/commons/9/9d/A_RES_71_313_E.pdf.

200
 APPENDICES

Appendix – A (Softwares)

Following is the list of softwares used, links where they can be downloaded / purchased,

and files generated by the team:

1. OpenVSP : http://openvsp.org/download.php

2. XFLR5 : https://sourceforge.net/projects/xflr5/

3. MATLAB : https://www.mathworks.com/products/matlab.html

4. DATCOM + Pro : http://www.holycows.net/datcom/

5. AAA (Advanced Aircraft Analysis) : https://www.darcorp.com/advanced-aircraft-

analysis-software/

6. Ansys Workbench : https://www.ansys.com/academic/students/ansys-student

7. Codes and Input files generated by the team :

https://www.dropbox.com/sh/6qy2q94spnx1kwc/AADE8joApGeCmRMbFV_rw_-

ca?dl=0

201