ABSTRACT 
 
 
 
 
Title of Document: COMPOSITE CONSTRUCTION OF AN 
UNMANNED AERIAL VEHICLE 
  
Degree Candidate: Evandro Gurgel do Amaral Valente 
Degree Year: Master of Science, 2006 
Directed By: Professor Norman Wereley 
Department of Aerospace Engineering 
 
 
 The purpose of this thesis is to develop an understanding of composite 
construction while building a systematic manufacturing and assembly process for the 
construction of an all-composite Unmanned Aerial Vehicle.  This thesis in intended 
for both beginners and advanced composite builders and documents the entire 
construction of a molded composite aircraft from CAD to Runway. 
 Several processes are discussed including CNC Plug Milling and Foam 
Cutting, Wet Lay-up, Tooling Manufacturing and Vacuum Assisted Resin Transfer 
Molding (VARTM ? Infusion).  Materials extensively used such as polyester 
resin/MEKP, gel coat, random directional matting, and carbon reinforcement are 
presented in an illustrated ?how-to? approach to building. 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
COMPOSITE CONSTRUCTION OF AN UNMANNED AERIAL VEHICLE 
 
 
 
By 
 
 
Evandro Gurgel do Amaral Valente 
 
 
 
 
 
Thesis submitted to the Faculty of the Graduate School of the  
University of Maryland, College Park, in partial fulfillment 
of the requirements for the degree of 
Master of Science 
2006 
 
 
 
 
 
 
 
 
 
 
Advisory Committee: 
Professor Norman Wereley, Chair 
Professor Darryll Pines 
Professor Jewel Barlow 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
? Copyright by 
Evandro Gurgel do Amaral Valente 
2006 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ii 
 
Acknowledgements 
?Tell me who you walk with and I?ll tell you who you are??  Well, these are my 
friends?..  
 
There have been many hands involved in the creation of this composite aircraft.  The 
undergraduate students whom I have had the pleasure of working with have 
repeatedly shown a level of commitment that exceeded my expectations many times 
over.   All of you have worked tirelessly and have created many positive memories 
for all of us to share.   Some of the undergraduate team members are shown in the 
group photo in Chapter 8.   
 
Their names, starting from top left, are: 
Gene Cook, Jolyon Zook, Joe Lisee, Nick Wilson, Ben Hoffman 
Bottom row, starting from left: 
Jon Graff, Evandro Valente (myself), Matt Kless 
Other friends have helped and participated in this project. These too deserve mention 
and my vote of gratitude. 
Chris Benic, Mike Chinn, Scott Jacobi, Lucas Parker, Brian Donelly and Holly 
Shurter 
 
One particular individual in this all-composite UAV project has given 1,000,000% (or 
maybe more).  My amount of admiration and profound respect for you, Gene Cook, 
can only be partially captured with the following adjective, ?leviathanesque?.  Gene is 
credited with all of the superb CAD renditions used in this document.  Gene, I wish 
you the best at MIT?your intelligence and talents are mirrored by your humility and 
professionalism, this is extremely rare. 
 
Thank you all for a job very very well done! 
 
As far as my graduate school days, other friends have been there through innumerous 
all-nighters and have given it their best along side myself.  To my grad friends and 
classmates, through thick and thin, from 9pm to 9am, from sweat to tears, from 
laughs to more laughs, we will be forever ?The Myers Crew? (Nathan Moulton, Assaf 
Barnoy, Allan Kruppa and Nick Tuozzo).    
 
I am happy to have all of you as my friends and look forward to a fruitful and 
rewarding career as your peer and friend. 
 
As for my parents and siblings, you have been supportive of my efforts since day one.  
There is only one four letter word that sums up the eternal bond that we have shared 
since we were born.  I am blessed in having you as my family.  Times have been 
tough, but we have been tougher. 
 
Lastly, I would like to thank the folks at the Aerospace Engineering Department at 
UMD.  You, the faculty and the staff, are no doubt number one. 
 iii 
 
Table of Contents 
 
 
Table of Contents......................................................................................................... iii 
List of Tables .............................................................................................................. vii 
List of Flowcharts and Gantt Chart............................................................................ viii 
List of Figures.............................................................................................................. ix 
Chapter 1: Introduction................................................................................................. 1 
1.1 Motivation..................................................................................................... 1 
1.2 Introduction to Composite Material and Construction ................................. 1 
1.2.1 Definition of Composite Materials ....................................................... 1 
1.2.2 Why Composite Construction............................................................... 2 
1.2.3 Manufacturing Techniques Overview................................................... 2 
1.3 Aircraft Construction Flowcharts and Gantt Chart....................................... 3 
1.4 Content of This Thesis.................................................................................. 9 
Chapter 2: CAD Drawing ........................................................................................... 10 
2.1 Overall Aircraft Drawings .......................................................................... 10 
2.1.1 Mission and Design Competition Parameters..................................... 10 
2.1.2 Aircraft Dimensions, Specs and CAD Drawings................................ 11 
2.2 Plug (Male Mold) CAD Drawings.............................................................. 19 
2.2.1 Sectioning the Aircraft........................................................................ 19 
2.2.2 Plug CAD Drawing............................................................................. 19 
Chapter 3: Plug Making.............................................................................................. 21 
3.1 CNC Machining and Plug Design............................................................... 21 
3.1.1 CNC Process ....................................................................................... 21 
3.1.2 CNC Limitations................................................................................. 21 
3.1.3 Plug Limitations and Design............................................................... 22 
3.1.4 The Flange .......................................................................................... 23 
3.1.5 The Offset ........................................................................................... 24 
3.2 Milling the Plug .......................................................................................... 25 
3.2.1 The Foam Block.................................................................................. 25 
3.3 Plug Preparation and Coating ..................................................................... 25 
3.3.1 Sanding ............................................................................................... 25 
3.3.2 Primer Coating.................................................................................... 27 
3.3.3 Polishing ............................................................................................. 28 
3.4 Materials & Tools Summary....................................................................... 28 
3.4.1 Materials & Tools Used...................................................................... 28 
3.4.2 Tools Clean-up.................................................................................... 29 
3.5 Advice......................................................................................................... 29 
3.5.1 De-molding Criteria............................................................................ 29 
3.5.2 Plug Surface Quality Trade-off........................................................... 30 
3.5.3 More Thoughts about the Plug Flange................................................ 30 
3.5.4 Damage and Storage ........................................................................... 32 
3.5.5 Thoughts on Project Management ...................................................... 33 
3.5.6 Machining Costs VS Manual Labor ................................................... 33 
 iv 
 
Chapter 4: Mold Making............................................................................................. 34 
4.1 Preparation .................................................................................................. 34 
4.1.1 Materials and Tools............................................................................. 34 
4.1.2 Labor and Task ................................................................................... 34 
4.1.3 Further Plug Preparation ? Release..................................................... 35 
4.2 Making the Molds....................................................................................... 35 
4.2.1 Mold Making ? Breakdown and Comments....................................... 35 
4.2.2 Phase I ? Gel Coating ......................................................................... 36 
4.2.3 Phase II ? Reinforcement Application................................................ 36 
4.2.4 Cure Time ........................................................................................... 40 
4.3 De-molding ................................................................................................. 41 
4.3.1 Releasing the Mold ............................................................................. 41 
4.3.2 Clean-up and Storage.......................................................................... 41 
4.4 Advice......................................................................................................... 42 
4.4.1 More on Labor, Tasks and Process..................................................... 42 
4.4.2 Prepare in Advance............................................................................. 45 
4.4.3 Thoughts on Mold Release ................................................................. 45 
4.4.4 De-molding ? Troubleshooting........................................................... 46 
4.4.5 Gel Coating ? Troubleshooting........................................................... 46 
4.4.6 PVA and Gel Coat Spraying Tips....................................................... 47 
4.4.7 Manage the Workspace....................................................................... 48 
4.4.8 Advice Summary ................................................................................ 48 
Chapter 5:  VARTM ................................................................................................... 49 
5.1 Preparation .................................................................................................. 49 
5.1.1 Materials and Tools............................................................................. 49 
5.1.2 Labor and Task ................................................................................... 50 
5.1.3 Further Mold Preparation.................................................................... 50 
5.2 Infusion ....................................................................................................... 52 
5.2.1 Process Description............................................................................. 52 
5.2.2 VARTM ? Breakdown and Comments............................................... 54 
5.2.3 Phase I - Dry Lay-up........................................................................... 55 
5.2.4 Phase I ? Inlet and Outlet Lines.......................................................... 58 
5.2.5 Phase II ? Resin Infusion .................................................................... 61 
5.3 De-molding ................................................................................................. 64 
5.3.1 Releasing the Part ............................................................................... 64 
5.3.2 Clean-up.............................................................................................. 67 
5.4 Advice......................................................................................................... 67 
5.4.1 Inlet and Outlet Ports ? Where to Place Them ................................... 67 
5.4.2 Work the Laminate ............................................................................. 68 
5.4.3 Working the Vacuum Bag .................................................................. 70 
5.4.4 Document Important Quantities.......................................................... 70 
5.4.5 Storage ................................................................................................ 70 
5.4.6 Summary ? VARTM Pros & Cons ..................................................... 70 
Chapter 6:  Wing Making ........................................................................................... 72 
6.1 Preparation .................................................................................................. 72 
6.1.1 Materials and Tools............................................................................. 72 
 v 
 
6.1.2 Labor and Task ................................................................................... 73 
6.1.3 Introduction and Foam Blank Preparation.......................................... 73 
6.2 CNC Foam Cutter ....................................................................................... 74 
6.2.1 Foam Cutter ........................................................................................ 74 
6.2.2 Wing Cutting....................................................................................... 75 
6.3 Foam Wing Preparation ? Structure Integration......................................... 81 
6.3.1 Internal Structure ................................................................................ 81 
6.3.2 Foam Joining....................................................................................... 81 
6.4 Wet Lay-up ................................................................................................. 82 
6.4.1 Process Description and Introduction ................................................. 82 
6.4.2 Wrapping the Wing............................................................................. 82 
6.4.3 Cure and Clean-Up ............................................................................. 86 
6.5 Moving Surfaces and Horizontal Stabilizer................................................ 86 
6.5.1 Ailerons and Elevator ......................................................................... 86 
6.5.2 Horizontal Stabilizer ........................................................................... 88 
6.6 Pre-Assembly.............................................................................................. 88 
6.6.1 Trim Wing to Size............................................................................... 88 
6.6.2 Moving Surface Hinging and Cutout.................................................. 91 
6.6.3 Final Product....................................................................................... 93 
6.7 Advice......................................................................................................... 96 
6.7.1 More on Labor and Tasks ................................................................... 96 
6.7.2 CNC Foam Cutter ? Hints for Successful Cutting.............................. 96 
6.7.3 The Hot Wire Bow.............................................................................. 97 
6.7.4 How to Ruin Your Work in Seconds .................................................. 97 
6.7.5 Summary ? Wet Lay-up Pros & Cons ................................................ 97 
6.7.6 Storage and Inventory......................................................................... 98 
Chapter 7:  Assembly................................................................................................ 100 
7.1 Preparation ................................................................................................ 100 
7.1.1 Tools and Materials........................................................................... 100 
7.1.2 Labor and Task ................................................................................. 100 
7.1.3 Assembly Flowchart ......................................................................... 100 
7.2 Machining and Joining Parts..................................................................... 101 
7.2.1 Machining ......................................................................................... 101 
7.2.2 Trimming, Trial Fitting and Cut Outs............................................... 101 
7.2.3 Joining............................................................................................... 105 
7.2.3.1 Joining the Fuselage.......................................................................... 105 
7.2.3.2 Joining the Hatch .............................................................................. 106 
7.2.3.3 Joining the Booms............................................................................. 106 
7.2.4 Before Joining ? Horizontal Stabilizer Cutout & Main Landing Gear
 107 
7.3 Internal Structure ? Fuselage and Hatch................................................... 112 
7.3.1 Internal Structural Components ........................................................ 112 
7.3.2 Materials used for Internal Structure ................................................ 112 
7.3.3 Nose Cutout ...................................................................................... 112 
7.3.4 Trial Fitting ....................................................................................... 113 
7.3.5 Fuselage and Hatch Internal Structure Installed ............................... 113 
 vi 
 
7.3.6 Vertical Stabilizer and Rudder Internal Structure............................. 114 
7.3.7 Wing and Fuselage Mating ............................................................... 115 
7.4 Assembly Interim ? Part I End, Part II Begin........................................... 116 
7.4.1 Vehicle Lay Out ? Trial Fit............................................................... 116 
7.5 Wiring and Servo Installation ................................................................... 118 
7.5.1 Boom Preparation ............................................................................. 118 
7.5.2 Running String.................................................................................. 120 
7.5.3 Installation of Servos ........................................................................ 120 
7.5.3.1 Rudder Servo Installation ................................................................. 120 
7.5.3.2 Elevator Servo Installation................................................................ 122 
7.5.3.3 Aileron Servo Installation................................................................. 124 
7.6 Final Vehicle Assembly ? Twin Boom Joined ......................................... 125 
7.6.1 Considerations Prior to Joining......................................................... 125 
7.6.2 Joining Subsections........................................................................... 126 
7.6.2.1 Twin Boom Joined with Stringing in Place ...................................... 126 
7.6.2.2 Twin Boom Joined with Servos and Wiring in Place....................... 126 
7.7 Moving Surfaces ....................................................................................... 129 
7.7.1 Moving Surfaces Final Assembly..................................................... 129 
7.8 Final Integration........................................................................................ 130 
7.8.1 Propulsion Hardware Installation ..................................................... 130 
7.8.2 Final Wiring and Electronic Hardware Installation.......................... 131 
7.8.3 Other Hardware Installation.............................................................. 131 
7.8.4 Spar Tube Screws/Tie-Down............................................................ 131 
7.9 Advice....................................................................................................... 132 
7.9.1 More on Labor and Task................................................................... 132 
7.9.2 Considerations Prior to Joining......................................................... 132 
7.9.3 Think First - Let the Solution Present Itself...................................... 133 
Chapter 8:  Conclusion ............................................................................................. 134 
8.1 Complete Aircraft ..................................................................................... 134 
8.2 Composite Construction Conclusion ........................................................ 139 
References................................................................................................................. 140 
 
 
 vii 
 
List of Tables 
Table 2.1. Vehicle Parameters and Specifications. 
 
Table 3.1. Material Trade-off for Plug Core. 
Table 3.2. Plug Materials Used. 
Table 3.3. Description of Tools. 
 
Table 4.1. Mold Making Materials. 
Table 4.2. Mold Making Tools. 
Table 4.3. Gel Coating (Phase I) Labor Distribution. 
Table 4.4. Reinforcement Application (Phase II) Labor Distribution. 
Table 4.5. Gel Coat Spray Troubleshooting. 
 
Table 5.1. VARTM Materials. 
Table 5.2. VARTM Tools. 
Table 5.3. Preparation & Dry Lay-up (Phase I) Labor Distribution. 
Table 5.4. Resin Infusion (Phase II) Labor Distribution. 
Table 5.5. VARTM ? Pros & Cons. 
 
Table 6.1. Wing Making (Wet Lay-Up) Materials. 
Table 6.2. Foam Cutting and Wet Wrapping Tools. 
Table 6.3. Wing Making Groupings and Tasks. 
Table 6.4. Wet Lay-up ? Pros & Cons. 
 
Table 7.1. Assembly Materials. 
Table 7.2. Assembly Tools. 
 viii 
 
List of Flowcharts and Gantt Chart 
Flowchart 1.1. Overall Vehicle Construction Flowchart ? collapsed. 
Flowchart 1.2. Chapters 2, 3 & 4 ? expanded. 
Flowchart 1.3. Chapters 5 & 6 ? expanded. 
Flowchart 1.4. Chapter 7 Part I (Low-Level Assembly) ? expanded. 
Flowchart 1.5. Chapter 7 Part II (High-Level Assembly) ? expanded. 
Gantt Chart 1.1 Suggested Vehicle Construction Timeline ? wintertime construction. 
 
Flowchart 4.1. Mold Making ? breakdown and comments. 
Flowchart 4.2. Mold Making Gel Coating (Phase I) ? detailed flowchart. 
Flowchart 4.3. Mold Making Reinforcement Application (Phase II) ? detailed 
flowchart. 
 
Flowchart 5.1. VARTM ? breakdown and comments. 
 ix 
 
List of Figures 
Figure 2.1. Competition Flight Pattern. 
Figure 2.2. Aircraft Dimensions. 
Figure 2.3. Detailed CAD Drawing ? fuselage structure. 
Figure 2.4. Detailed CAD Drawing ? wing center-section assembly and mounting. 
Figure 2.5. Detailed CAD Drawing ? wing outboard-section assembly. 
Figure 2.6. Detailed CAD Drawing ? boom and rudder assembly. 
Figure 2.7. Detailed CAD Drawing ? elevator assembly and mounting. 
Figure 2.8. Detailed CAD Drawing ? payload configurations. 
Figure 2.9. Fuselage Plug ? left half: front, isometric & side views. 
Figure 2.10. Boom Plug ? left half: front, isometric & side views. 
 
Figure 3.1. Sample Picture of CNC Milling ? rough cut. 
Figure 3.2. CNC Milling and Plug Design Limitations ? cutaway front view. 
Figure 3.3. Plug Flange and Offset. 
Figure 3.4. Sanding the Boom Molds. 
Figure 3.5. Sanding the Fuselage Molds. 
Figure 3.6. Primer Coated Plugs. 
Figure 3.7. Boom Plug on Plexiglas Base Plate. 
Figure 3.8. Fuselage Plug with Plywood Flange Extension. 
Figure 3.9. Surface Damage on Fuselage Plug. 
 
Figure 4.1. Strand Mat Sheet placed on bare Gel Coat. 
Figure 4.2. Boom Reinforcement Build-up. 
Figure 4.3. Fuselage Reinforcement Build-up. 
Figure 4.4. Finished Boom Mold ? reverse side. 
Figure 4.5. Finished Fuselage Mold ? reverse side. 
Figure 4.4. Boom Mold ? support 2? by 4? frame. 
Figure 4.7. Cracked Gel Coat due to Uneven Thickness. 
 
Figure 5.1. Fuselage Mold (Right Half) Sprayed with PVA ? tape along flange border 
removed.  
Figure 5.2. Boom Mold (Right Half) Being Sprayed with PVA ? with tape along 
flange border. 
Figure 5.3. Vacuum Infusion Process.  
Figure 5.4. VIP Setup.   
Figure 5.5 .Boom Mold (Left Half) ? dry lay-up. 
Figure 5.6. Red Mesh Infusion Medium ? used in vertical stabilizer and rudder. 
Figure 5.7. Airfoil Cutout ? infusion medium custom cutout with ?walls?.  
Figure 5.8. Carbon/Infusion Medium/Carbon Sandwich ? worked into boom. 
Figure 5.9. Boom ? inlet and outlet ports detail, no vacuum applied. 
Figure 5.10. Boom ? vacuum bag sealed, no vacuum applied. 
Figure 5.11. Boom ? ready for infusion, vacuum applied.  
Figure 5.12. Fuselage ? outlet port detail and ?rabbit ear?. 
 x 
 
Figure 5.13. Fuselage ? ready for infusion, vacuum applied. 
Figure 5.14. Resin Infusion in Progress ? right boom half.  
Figure 5.15. Boom Resin Infusion Complete. 
Figure 5.16. Resin Infusion in Progress ? right fuselage half.  
Figure 5.17. Fuselage Resin Infusion Complete.  
Figure 5.18. Boom ? de-molding process (from top left to bottom right). 
Figure 5.19. Boom ? de-molded part with release film. 
Figure 5.20. Boom Mold and Part ? PVA release rinsed off with water. 
Figure 5.21. Fuselage Mold and Part ? with PVA release and outlet port attached. 
Figure 5.22. Working reinforcement into mold cavity. 
Figure 5.23. Vacuumed part ? check for bridging. 
 
Figure 6.1. CNC Foam Cutter. 
Figure 6.2. Laptop Screen ? foam cutter software running. 
Figure 6.3. Bare Foam Wing Center Section ? all cuts completed. 
Figure 6.4. Wing Center-Section ? interface mounting thru-holes marked. 
Figure 6.5. Center Wing-Section ? mounting thru-hole being cut. 
Figure 6.6. Center Wing-Section ? mounting thru-hole done, detail view top right. 
Figure 6.7. Center Wing-Section ? front support-pin cutout, detail view top right. 
Figure 6.8. Center Wing-Section ? notch being cut for spar end caps, detail view top 
Figure 6.9. Outboard Wing Construction ? explode view (see Figure 2.5). 
Figure 6.10. Outboard Wing-Section ? wet lay-up preparation. 
Figure 6.11. Outboard Wing-Section ? wet lay-up, detail view top right. 
Figure 6.12. Outboard Wing-Section ? reinforcement strip wet lay-up. 
Figure 6.13. Outboard Wing-Section ? wet out complete, foam core ready to wrap. 
Figure 6.14. Outboard Wing-Section ? wrapped and ready to receive weights. 
Figure 6.15. Elevator ? hinge support cut out, detail view bottom left. 
Figure 6.16. Untrimmed Elevator Wrapped in Carbon ? detail view top left. 
Figure 6.17. Outboard Wing-Section ? trimming the ends. 
Figure 6.18. Outboard Wing-Section ? sanding the ends. 
Figure 6.19. Outboard Wing-Section ? trimming the trailing edge. 
Figure 6.20. Outboard Wing-Section ? trimming complete, next wing on queue. 
Figure 6.21. Elevator Hinge Installation. 
Figure 6.22. Elevator Hinged and Balsa Support Blocks Prepped. 
Figure 6.23. Ailerons Hinged with Balsa Blocks Glued. 
Figure 6.24. Aileron Hinge Detail. 
Figure 6.25. Aileron Hinging and Preparation Complete. 
Figure 6.26. Center Wing-Section Complete. 
Figure 6.27. Entire Wing Complete ? outboard and center sections. 
Figure 6.28. Horizontal Stabilizer and Elevator Complete. 
Figure 6.29. Wing Blank Being Cut with the Hot Wire Bow. 
Figure 6.30. Organized Foam Cores and Finished Wings.  
 
Figure 7.1. Main Landing Gear Assembly (see Figure 2.6). 
Figure 7.2. Trimming Tools and Parts. 
Figure 7.3. Boom Trimming in Progress ? inscribed line detail view, top right. 
 xi 
 
Figure 7.4. Fuselage Ready for Hatch Cutout ? fine sanding, detail view top left. 
Figure 7.5. Rudder Cutout ? free hand cutting, detail view bottom right. 
Figure 7.6. Rudder Cutout Complete for all Boom Halves. 
Figure 7.7. Fuselage Joining ? taped seam, detail view top right. 
Figure 7.8. Fuselage Joined. 
Figure 7.9. Horizontal Stabilizer Cutout ? top left and bottom right detail views. 
Figure 7.10. Landing Gear Interface ? detail view bottom right. 
Figure 7.11. Installation of Main Strut Interface (from top left to bottom right). 
Figure 7.12. Boom Joining ? resin being injected (close-up). 
Figure 7.13. Boom Joining ? resin being injected. 
Figure 7.14. Left and Right Boom Pair Injected with Resin. 
Figure 7.15. Nose Cutout for Propeller and Cooling. 
Figure 7.16. Internal Structure Trial fit with Simulated Cargo (see Figure 2.8). 
Figure 7.17. Hatch and Fuselage with Internal Structure Glued in Place. 
Figure 7.18. Vertical Stabilizer and Rudder Hinging Balsa Strip (see Figure 2.6). 
Figure 7.19. Fuselage and Center Wing-Section Mating (see Figure 2.4). 
Figure 7.20. Aircraft Major Components Laid-Out. 
Figure 7.21. Aircraft Major Components Laid-Out.  
Figure 7.22. Load Transferring Spar and Assembly. 
Figure 7.23. Spar Hole Template in Place ? ready to locate spar hole. 
Figure 7.24. Spar Hole Marked and Drilled ? detail view top right. 
Figure 7.25. Wire Access Hole Marked and Drilled ? detail view bottom left. 
Figure 7.26. Rudder Servo Installation ? detail view top left (see Figure 2.6). 
Figure 7.27. Rudder Servo Installation Complete ? detail view top left. 
Figure 7.28. Servo Installation (from top left to bottom right). 
Figure 7.29. Elevator Servo Installed ? ready for installation wing foam-core. 
Figure 7.30. Hand Made Wire Access Tunnel ? detail views diagonal corners. 
Figure 7.31. Twin Boom Sections Ready to be Joined. 
Figure 7.32. Twin Boom Sections Ready to be Joined. 
Figure 7.33. Twin Boom Assembly Joined. 
Figure 7.34. Rudder and Tail Wheel Installed. 
Figure 7.35. Almost Completed Aircraft ? elevator not installed yet. 
Figure 7.36. Motor Mount Assembly. 
Figure 7.37. Wing Bolt Clearance Hole Made. 
 
Figure 8.1. Completed Aircraft ? nose close-up. 
Figure 8.2. Completed Aircraft ? Maryland flag reflection. 
Figure 8.3. Completed Aircraft ? runway access ramp. 
Figure 8.4. Completed Aircraft ? on take-off. 
Figure 8.5. Completed Aircraft ? flyby over runway. 
Figure 8.6. Completed Aircraft ? group picture. 
 
 
 1 
 
Chapter 1: Introduction 
1.1 Motivation 
The motivation for this thesis is to document the construction methods required for 
the manufacture of an all-composite aircraft.  The goal is to create a recipe-like 
manuscript that not only develops the systematic procedures, but also passes 
invaluable experience to the reader.   
 
The methods used for composite manufacturing promise substantial results, yet, 
heavily rely on one?s ability to both carefully and patiently follow the described 
instructions.  To increase first-time success rate, the author is particularly motivated 
towards sharing both good and bad experiences in the Advice Section of each 
Chapter.  The beginner will benefit from this document by understanding and 
following the step-by-step construction.  Meanwhile, advanced builders can use this 
document as a point of reference to refresh or refine their own composite construction 
background.  
1.2 Introduction to Composite Material and Construction 
1.2.1 Definition of Composite Materials  
A composite material is made up of two constituents:  the reinforcement and a binder 
called the matrix.  The reinforcement and the matrix, when combined, work in 
concert to offer mechanical properties far superior than the components by 
themselves.  The matrix has two roles in the composite material: 1) transfer loads to 
the structure, i.e. reinforcement and 2) shield the reinforcement from external and 
environmental hazards.   
 
The reinforcement is the structural component of the composite often responsible for 
its anisotropic property.  This means that a composite material is purposely designed 
to transmit loads along a preferred fiber direction.  As a result, while properties along 
the fiber are superior, load-carrying ability in the transverse direction is minimal.  
Therefore, a composite structure can be specifically designed to carry loads in a 
particular or multidirectional path(s).  Furthermore, a composite laminate may 
employ different materials for each layer; glass fiber ? e-glass, s-glass, c-glass; 
carbon fiber ? graphite and organic fiber ? aramid.  
 
Composites contain materials that are not only chemically different, but also 
generally mechanically separable.  This additional definition further narrows the 
types of materials that qualify to be a composite.  The composite materials discussed 
and used in this project fall under the polymer matrix and fiber reinforcement class of 
composites.  One other term can be applied to composites as defined above.  Since a 
 2 
 
composite, made of fiber and matrix, is made up of two or more materials, it is 
considered inhomogeneous.  The material properties like density and internal 
structure are varied from point to point within a given sample.     
1.2.2 Why Composite Construction  
Since the first modern application of composites, a glass-reinforced fishing pole 
constructed in 1945, this new class of materials have revolutionized the way products 
are made.  Although well established and backed by hundreds of years of usage, 
metals have been slowly replaced by composites in several industries.  Advancements 
in technology have pushed the physical limits and requirements for materials. Metals 
are in no way being extinct from modern manufacturing methods, but rather sharing 
the workspace with composites that have unique characteristics. 
 
Although the Aerospace Industry has used aluminum and titanium alloys that 
outperform steel, requirements like corrosion and fatigue resistance or high specific 
tensile stiffness have created a niche for composite materials.  Early uses for 
composite were restricted to nonstructural applications since material properties were 
not completely known and were still under testing.  As the decades progressed, the 
advents of new reinforcement fibers and construction techniques have inspired new 
applications for composites.  These applications have combined both aesthetics 
appeal and structural properties into one package. 
 
Typical overall advantages to composite materials are: 
head2right High strength to weight ratio 
head2right High stiffness to weight ratio 
head2right Low density 
head2right Environmental resistance 
head2right Design versatility 
head2right Chemical resistance  
head2right Quick part turnaround 
1.2.3 Manufacturing Techniques Overview 
Manufacturing techniques for composite construction vary greatly and rely mainly on 
one key trade-off, target mechanical properties versus manufacturing costs.  This 
trade-off profoundly influences the selection and application of one manufacturing 
method over another.  Desired levels of output in production also drive the decision-
making process to favor one method over another.  Methods that offer quick 
turnaround at low costs often fall short in long-term applications, part quality and/or 
labor efficiency.  On the other hand, methods of manufacturing with a higher start-up 
cost and/or skill requirement tend to be more advantageous for long-term high-quality 
high-output applications. 
 
 Typical manufacturing methods include manual wet lay-up, manual prepreg lay-up 
(with autoclaving), vacuum bagging, filament winding, pultrusion, vacuum assisted 
resin transfer molding (VARTM), resin transfer molding (RTM) with matched molds, 
 3 
 
resin film infusion (RFI) and more.  The manufacturing techniques utilized in this 
project are wet lay-up, vacuum bagging, and VARTM. 
1.3 Aircraft Construction Flowcharts and Gantt Chart 
Each chapter develops a major construction phase of the build (Flowchart 1.1).  The 
contents of each chapter are organized into Flowcharts 1.2 to 1.5.  A suggested 
vehicle construction timeline, during winter months, is illustrated in Gantt Chart 1.1. 
 Flowchart 1.1. Overall Vehicle Construction Flowchart ? collapsed. 
 4 
 
 Flowchart 1.2. Chapters 2, 3 & 4 ? expanded. 
 5 
 
 Flowchart 1.3. Chapters 5 & 6 ? expanded. 
 6 
 
 Flowchart 1.4. Chapter 7 Part I (Low-Level Assembly) ? expanded. 
 7 
 
 Flowchart 1.5. Chapter 7 Part II (High-Level Assembly) ? expanded. 
 8 
 
 Gantt Chart 1.1 Suggested Vehicle Construction Timeline ? wintertime construction. 
 9 
 
1.4 Content of This Thesis 
This thesis concentrates on developing the expertise necessary for fabrication, 
manufacture, assembly and integration of an all-composite aircraft.  The sequentially 
ordered chapters represent the different stages of construction.  
 
Chapter 2 introduces the CAD drawings generated during the design phase of this 
aircraft.  From the overall design of the aircraft, the design of the plugs is realized. 
 
Chapter 3 takes the CAD drawings of the plugs and turns them into real objects.  The 
plugs are milled then prepped for mold making. 
 
Chapter 4 develops the steps necessary to create molds from the plugs.  The high 
quality molds are used to create the various composite parts of the aircraft. 
 
Chapter 5 contains a systematic procedure for manufacturing aircraft parts using 
Vacuum Assisted Resin Transfer Molding.  The high quality parts are the result of a 
vastly efficient molding technique that benefits from the vacuum infusion process. 
 
Chapter 6 introduces the wet lay-up method of construction for wing making.  The 
wings and horizontal stabilizer of the aircraft have a foam core wrapped in a single 
ply of carbon reinforcement. 
 
Chapter 7 encompasses the entire assembly and integration phases of the build.  The 
parts manufactured in previous chapters are joined piece by piece to create a fully 
integrated product.  The first half of the chapter details the low-level assembly of 
parts.  The second half of the chapter finalizes the assembly and vehicle integration. 
 
Chapter 8 presents the completed aircraft.   
 
 
 10 
 
Chapter 2: CAD Drawing 
2.1 Overall Aircraft Drawings 
2.1.1 Mission and Design Competition Parameters 
The mission profile and aircraft requirements, in accordance with the ONR/AIAA 
Design/Build/Fly Competition Rules, defined the design criteria for successful 
entries.  The vehicle discussed in this report was built to satisfy two of the three 
missions listed in the competition rules.  The payload configuration and storage inside 
the aircraft is shown in Figure 2.8.  Each mission involved flying the pattern shown in 
Figure 2.1.   
 
 Figure 2.1. Competition Flight Pattern. 
 
The first mission required 96 tennis balls to be flown for at least two minutes.  The 96 
balls could be flown at once or in multiple sorties.  This aircraft was designed to carry 
48 tennis balls in its cargo compartment, thus, required to fly two separate flights of 
two minutes each to score points successfully. 
 
The second mission required three flights with three different payloads.  Each 
payload was flown once around the course without an airborne flight duration 
requirement.  The payload was flown once then changed, after landing, by the ground 
crew.  The payloads were 48 tennisballs, two 2-Liter Soda bottles filled with water 
and, lastly, a wood block whose dimensions were specified in the competition rules. 
 
The overall rules required all entries to have a propulsion system that used either a 
brushless or a brushed motor with a NiCad or NiMh battery pack weighing no more 
than three pounds. 
 11 
 
2.1.2 Aircraft Dimensions, Specs and CAD Drawings 
The vehicle parameters and dimensions are given in Table 2.1 and Figure 2.2, 
respectively.  The airfoils used for the wing and horizontal stabilizer are 
NACA633618 and NACA0012, respectively.  Figures 2.3 to 2.8 contain detailed 
CAD drawings of the aircraft.  
 
Outboard Wing Section 
Outboard Span 46 in 
Section Reference Area 455.4 in2 
Section Mean Aerodynamic Chord 9.9 in 
Root Chord 12 in 
Tip Chord 7.8 in 
Taper Ratio 0.65 
Entire Wing  
Wing Span 113.75 in 
Wing Reference Area 1171 in2 
Wing Mean Aerodynamic Chord 10.3 in 
Aspect Ratio 11.05 
Empennage 
Horizontal Tail Area 196 in2 
Vertical Tail Area 144 in2 
Horizontal Tail Volume 5.11 ft2 
Vertical Tail Volume 3.57 ft2 
Performance Parameters 
CLmax 1.3 
L/Dmax 22.17 
90 ft (max weight) Takeoff Distance 
50 ft (empty weight) 
42 ft/s (max weight) Stall Speed 
32 ft/s (empty weight) 
Aircraft Empty Weight 13 lbs 
Electronic Systems 
Motor & Gear Ratio NeuMotor 1515/2Y/6.7/F, 6.7:1 
Speed Controller 77 Amp Jeti Opto 
Propeller APC 24 x 12 
Battery Pack 21 Cells, CPB3300SC 
Receiver Futaba FP-R148DP 
Aileron Servos Hitec HS-85MG+ Mighty Micro 
Rudder Servos Hitec HS-225MG+ Mighty Mini 
Elevator Servos Hitec HS-645MG Ultra Torque 
Table 2.1. Vehicle Parameters and Specifications. 
 12 
 
 Figure 2.2. Aircraft Dimensions. 
 13 
 
 Figure 2.3. Detailed CAD Drawing ? fuselage structure. 
 14 
 
 Figure 2.4. Detailed CAD Drawing ? wing center-section assembly and mounting. 
 15 
 
 Figure 2.5. Detailed CAD Drawing ? wing outboard-section assembly. 
 16 
 
 Figure 2.6. Detailed CAD Drawing ? boom and rudder assembly. 
 17 
 
 Figure 2.7. Detailed CAD Drawing ? elevator assembly and mounting. 
 18 
 
 Figure 2.8. Detailed CAD Drawing ? payload configurations. 
 19 
 
2.2 Plug (Male Mold) CAD Drawings 
2.2.1 Sectioning the Aircraft  
There are many patterns available for sectioning the aircraft.  The available options 
directly reflect the complexity of the surface curvature and overall design of the 
aircraft.  The easiest partitioning scheme for a closed shape is to split it into two 
halves.  Most often, a two-section split results in what is intuitively called left/right or 
top/bottom halves.   
 
The design criteria of sectioning a plug in two halves imposes restrictions on the 
complexity of the plug in question.  This is because there is an overall construction 
limitation in molded composite manufacturing.  A plug whose shape does not allow 
de-molding is not suitably partitioned.  In another words, the restriction says that the 
plug and mold must be able to separate from each other.  If the desired plug shape is 
highly complex, the plug may have to be sectioned into various pieces that are 
guaranteed to de-mold.  This subject is developed in detail in the Chapter 3.1.3. 
 
The booms and fuselage for this construction were designed for a left/right plug-
partitioning scheme.  Therefore, no other plug-partitioning scheme sectioned the 
plugs as efficiently as the vertical cut along the center of the fuselage and boom.     
2.2.2 Plug CAD Drawing 
The aircraft was sectioned into parts that would make up the various plugs.  To 
simplify construction, the aircraft was purposely designed so the fuselage and boom 
molds consisted of only left and right sections.  The drawings were modified 
accordingly to generate the CAD representation suitable for plug milling.  The CAD 
drawings of the left-hand plugs sent out for CNC milling are found in Figures 2.9 and 
2.10.  The right halves of the plugs are mirror images of the ones displayed below.  
Two very important features in plug design are the flange and the offset.  Design 
considerations and CNC milling for the plugs are discussed in detail in the next 
chapter.    
 
 20 
 
 Figure 2.9. Fuselage Plug ? left half: front, isometric & side views. 
 
 Figure 2.10. Boom Plug ? left half: front, isometric & side views. 
 21 
 
Chapter 3: Plug Making 
 
3.1 CNC Machining and Plug Design 
3.1.1 CNC Process 
To create the plugs for the aircraft, Computer Numerical Control (CNC) Milling was 
extensively utilized.  This particular project used a milling machine that had 3 degrees 
of freedom: X, Y and Z axis translation.  More sophisticated CNC milling machines 
may have 5 degrees of freedom where the cutting tool is manipulated in not only 
translation but also angular direction.   
 
The CAD drawings discussed in the previous chapter were saved as Initial Graphics 
Exchange Specification (IGES) files that the CNC milling machine recognizes.  The 
machine shop took the CAD drawings in IGES format and ran a Computer Aided 
Machining Software (CAM) that calculated the path of the cutting bit.  During this 
process, the machinist was able to further diagnose the integrity of the drawings and 
estimate the cutting time/cost. 
3.1.2 CNC Limitations 
It was important to correspond with the machine shop and establish the dimensional 
limitations of the milling machine in two regards: maximum XYZ cut travel lengths 
and cutting table size/interior enclosure.  The size of the table can be a limitation 
because most jobs require clamping the blank block of material.  The clamps 
themselves reduce the actual cutting space of the table (Figure 3.1). 
 
 Figure 3.1. Sample Picture of CNC Milling ? rough cut. 
  
 
 22 
 
Since the mill uses rotary bits, a radius is unavoidable when milling inside corners 
(Figure 3.2 B & F).  If a specific radius of an inside cut is required, the mill will have 
to past over the same area several times with an incrementally smaller radius bit.  
Thus, increasing both machining time and cost.  A sharp internal cut corner is very 
difficult to achieve unless an end mill can be used for the desired cut (Figure 3.2 B).   
 
Considerations for CNC machining are evident in Figure 3.2 and are described in the 
following two cases: 
1) Examples A, D, E & F require CNC milling machines with capabilities 
beyond XYZ Degrees of Freedom. 
2) Examples B & C require CNC milling machines with XYZ degrees of 
freedom only.    
  
 Figure 3.2. CNC Milling and Plug Design Limitations ? cutaway front view. 
 
3.1.3 Plug Limitations and Design 
As discussed earlier, the plug CAD drawings must contain curves that the CNC 
machine can handle depending on the degrees of freedom of the machine.  Angled 
depth/insert cuts must be avoided if the CNC milling machine has only translational 
degrees of freedom.  However, even if a five DOF CNC milling machine is available, 
complex or deep depth cuts should still be avoided because the female mold that will 
be extracted from the plug must have a straight separation path to de-mold.  Deep cuts 
into the plug, although straight, can pose difficulties during separation of the mold 
from the plug.  The arrows in Figure 3.2 show the direction that the mold must be 
pulled to be detached from the plug.  Also in Figure 3.2, the mold is drawn in blue 
and the plug is drawn in orange. 
 
Limits on plug design are exemplified in Figure 3.2 and are described in the 
following cases: 
 
A) A plug may require that the mold be ?pulled? at an angle.  Although this is 
possible in theory, in practice this can be much more difficult.  The user 
must understand how to de-mold the part correctly as to avoid damage and 
 23 
 
unnecessary wear on the mold.  A warning sign accompanies this case to 
indicate that this is possible, yet, requires careful consideration. 
B) A well-designed plug should have an intuitive de-molding process.  This 
mold can be separated straight up without being caught in any other 
feature of the plug.  As mentioned earlier, the insert cut should be 
discussed with the machinist.  It is recommended to have a shallow cut 
depth so the mold can release easily and the part to be generated from the 
mold can tolerate the complex curvature and right angle.  (This is seen in 
the boom molds where the airfoil cutout is located.  This complex inside 
cut causes bridging, a phenomenon described in Chapter 5). This case 
receives a check mark. 
C) This example does not violate the de-molding process.  This case receives 
a check mark. 
D) Two features on the plug render this case unfeasible.  Firstly, given the 
overall shape of the plug, the blister, given its location, on the side makes 
de-molding impossible.  However, there are cases where features similar 
to the blister can be used.  Secondly, the contour of the circular shape 
?tucks in? at the base.  The mold will not separate due to this undesired 
contour.  In this case, the contour is exaggerated for demonstration 
purposes.  In reality, minimal invasion (in the order of 1/64th of an inch) of 
the contour at the base/flange will create de-molding inconvenience or 
failure. 
E) This case fails by inspection.  No part can ever be obtained from this plug.  
This plug requires a different partitioning scheme all together to work. 
F) The angled cut imposes a de-molding direction parallel to it.  However, 
the overall shape of the mold requires an upward de-molding direction.  
This case fails because the mold must be pulled in two directions at the 
same time (which is not possible).             
3.1.4 The Flange 
The flange is the flat area around the curved surface of either the plug or the mold.  
The flange of the plug was transferred to the mold during wet lay up and played an 
important role in the manufacturing of parts.   
 
It may appear attractive to minimize the flange so there is less material used in the 
plugs.  The material and CNC milling costs are significant, but product quality and 
ease of manufacturing depend on how much flange is available to work with.  For 
reasons discussed in Chapter 5, a good mold should have wide flanges.  A good rule 
of thumb is to allow no less than 5 inches of flange around the entire perimeter of the 
contour curve.   
 
In Figure 3.3, the flange of both the plug and the mold is represented in red, the mold 
contour or mold cavity in blue, and the plug offset in dashed orange and red. 
 
 
 24 
 
 Figure 3.3. Plug Flange and Offset. 
3.1.5 The Offset 
As observed in Chapter 2.2.2, the complex curves of the plugs have an offset from the 
surface plane of the flange.  In another words, the contoured surface was raised a 
certain amount above the flange.  Thusly, the mold extracted from the plug was 
deeper by the offset amount, as observed in Figure 3.3.  The offset was there because 
of three choices made during plug design:  
 
1) Creating an offset allowed the machine shop to mill a radius on the corner 
edge of the perimeter of the contour.  Having a radius along the perimeter 
of the cavity provided some relief on the fiberglass/carbon cloth during 
dry lay-up of the cloth, and facilitated vacuum bagging/resin transfer.  The 
details of Resin Transfer Molding are discussed in Chapter 5.      
2) The offset was necessary based on the method chosen for cutting the 
flange from the part extracted from the mold.  This is discussed in Chapter 
7.2.2. 
3) To save weight, a manufacturing strategy was chosen that allowed the 
parts to be of highest quality and finish; thusly, not requiring painting and 
retaining the original carbon composite ?look?.  As a result, the joining 
line of two parts would not be hidden by putty and paint.  The offset 
allowed the flange to be trimmed in a systematic and repeatable fashion.  
Trimming and Joining are developed in Chapter 7.   
 25 
 
3.2 Milling the Plug 
3.2.1 The Foam Block 
Multiple sheets of high-density foam were used as the core material for the plugs. 
Other materials, like aluminum, could have been used for the plugs.  Important 
factors considered for choosing foam over metal for the plug are discussed in Table 
3.1. 
 
 Foam Metal/Aluminum 
Material Cost Reasonable Expensive 
Machining Cost Reasonable Very Expensive 
Weight Reasonable Heavy 
Durability Fair Very Good 
Reparability Good Moderate 
Reshape/Sanding Possible Very Difficult 
De-molding Cycles 1 - maybe 2 Multiple 
Table 3.1. Material Trade-off for Plug Core. 
 
The material cost, machining cost and weight of aluminum were the main reasons for 
choosing the foam.  However, it was noted that one important factor made aluminum 
very desirable despite the higher costs.  An aluminum plug would be much more 
resistant to the wear and tear due to cycling the plugs.  In another words, if multiple 
sets of molds had been desired, then aluminum molds would have been the safest 
choice.  Since this project only called for one set of molds, the high-density foam core 
remained the more justified choice. 
3.3 Plug Preparation and Coating 
3.3.1 Sanding 
Sanding was required in mostly all preparatory phases of manufacturing.  The use of 
the power sander made the job quicker.  However, the power sander quickly sanded 
away too much of the foam.  Therefore, as a cautionary measure, power sanding was 
done ONLY on the flange areas.  The contoured areas of the plugs were sanded solely 
by hand with 600grit or finer sand paper (Figures 3.4 & 3.5).  The sanding of the bare 
foam plugs was dry unlike wet sanding utilized later on. 
 
The bare foam was porous and easily damaged.  It was important to keep any sharp or 
heavy objects clear from the vicinity of the plugs.  Objects like pencils, pens, 
wristwatch, keys, screwdrivers and even cell phones could ding or stab the foam plug.  
This kind of senseless damage created more work.  Any unwanted features due to 
damage had to be puttied and re-sanded.  Higher density foam would have been more 
 26 
 
resistant to surface rashes due to accidents.  However, the use of higher density foam 
tooling boards would increase the project costs.  
 
 Figure 3.4. Sanding the Boom Molds. 
 
 Figure 3.5. Sanding the Fuselage Molds. 
 27 
 
 
Note:  The flange of both the fuselage and the boom plugs were insufficient, a lesson 
learned the hard way, as discussed in the Advice Section of this Chapter and Chapter 
5.  Also, note the lines on the foam plugs.  These lines are also discussed in the 
Advice Section of this Chapter.  
3.3.2 Primer Coating 
The primer coating provided a surface seal for the plug.  The primer was sprayed on 
the plug and hardened to a semi-glossy finish (Figure 3.6).  The spraying of the 
primer coating required patience and skill with the spray gun.  A careful spray job 
required hardly any sanding once the primer dried.  Conversely, any accumulation or 
sags in the primer created the necessity for wet sanding.  From this point forward in 
the construction, wet sanding was used with either water or liquid polishing wax.   
 
 Figure 3.6. Primer Coated Plugs. 
 
The sanding process was repeated several times, but with recursively finer grit.  In 
most cases, wet sanding paper between 1000 and 2000 Grit was used.  Two key 
?don?ts? were always kept in mind while wet sanding; do not sand down to bare foam 
and renew the water on the both plug and sand paper often.  Sanding the primer coat 
 28 
 
off completely created the necessity for patchwork.  The purpose for changing the 
water often was to remove the sanded primer dust mixed with the water from the 
surface of the plug.  The entire reason for sanding with finer grit was to remove the 
sanding scratches made from the previous grit.  Not changing the contaminated water 
negated the purpose of using the finer grit, to reduce the ?scratched surface look?.  
3.3.3 Polishing 
Since some areas of the primer coated plug were sanded, the scratches made by the 
sand paper had to be eliminated as much as possible.  The polishing process was 
divided into 3 stages: 1) Fine Grit Sanding, 2) Liquid Wax Buffing and 3) Hand Wax 
Coating.   
 
The fine-grit sanding step called for wet sanding with 3000 and 4000 Grit sanding 
pads.  This process used a pneumatic sanding gun and a ?thin? liquid wax instead of 
water.    
 
The buffing step used a pneumatic polishing gun and a thick liquid carnauba wax to 
polish the plug.  Since this liquid wax dried after application, it was necessary to 
apply and buff the plug in stages.  Once the entire plug was buffed, the dried liquid 
wax was removed with a clean terry cloth.  In some cases, this step was repeated if 
the results were not achieved.  It was very important to keep clean both the buffing 
pads and the terry cloth.  Any dust or dirt on either the buffing pad or the terry cloth 
resulted in scratches being made on the surface.  
 
Lastly, a final coat of carnauba liquid wax was applied; however, this time a terry 
cloth was used instead of the pneumatic polishing gun.  An even coat of wax was 
spread by hand over the plug surface and allowed to dry.  This step removed some of 
the minutest scratches left from the buffing pad.  The resulting surface was mirror-
like and had little to no scratches. 
 
It is important to note that the sanding and buffing instructions in this chapter also 
apply for the molds as well.  Once the molds were made from the plugs, the same 
sanding procedure described above was utilized as necessary depending on how the 
molds looked.  The superior quality of the surface of the plug transferred to the mold 
being cast from the plug.   
3.4 Materials & Tools Summary 
3.4.1 Materials & Tools Used 
The materials used during the plug making process are listed in Table 3.2.  Basic 
supplies like latex gloves, stir sticks, tyvek suits, eye protection, mixing cups, water 
bucket, acetone bucket, etc. are not explicitly discussed.  The author assumes that the 
reader is aware of such necessities.  
 
 29 
 
 
Materials Description 
General Materials Mixing cups, stir sticks, latex gloves, eye protection, masks, etc 
Packing Tape Tape used for masking areas 
Contact Adhesive Spray adhesive used to bond foam block to base plate 
Primer Polyester coating used for sealing and priming the plug surface 
MEKP Catalyst Hardener used for polyester based systems: gel coat and resin 
High Density Foam Closed cell high density tooling board foam 
Liquid Shine Gel coat clean and shine compound 
Machine Glaze Polishing Compound 
Liquid Wax Surface Wax in liquid form  
Table 3.2. Plug Materials Used. 
 
Several tools were necessary to prepare the plugs for casting the female molds.  The 
tools are listed in Table 3.3.  These tools along with others were repeatedly used in 
various phases of the construction as developed in later chapters.  Nevertheless, one 
tool not listed below that was of overwhelming importance was patience.   
 
Tools Description 
Sand Paper/Pad Abrasive sheet/pad used for adjusting surface quality and features  
Buffing Pads Polishing Pad used for removal of surface scratches due to sanding 
Digital Scale Instrument used for weighing polyester primer 
HVLP Spray Gun High Volume Low Pressure Siphon Gun used for spraying primer 
Pneumatic Sander Pneumatic Device used for efficient sanding of large areas 
Pneumatic Polisher Pneumatic Device used for efficient polishing of large areas 
Paint Booth Closed ventilated area reserved for spraying 
Compressed Air High pressure air line 
Table 3.3. Description of Tools. 
3.4.2 Tools Clean-up 
Aside the buffing pads, which were cleaned with warm water, all other tools were 
cleaned with acetone as necessary.  Any tool that had contact with polyester or/and 
MEKP hardener was thoroughly cleaned immediately with acetone.  The acetone was 
poured in the siphon of the spray gun and sprayed until the acetone sprayed clear.  
Often, the spray gun was taken apart and cleaned as well.  Clean up was done quickly 
since polyester materials mixed with hardener had a finite working time before 
setting.  Tools were squirted with acetone or dipped in an acetone bath before being 
wiped.  
3.5 Advice 
3.5.1 De-molding Criteria  
Sections 3.2 and 3.3 discuss several limitations and considerations for plug design 
and CNC milling.  The cases shown in Figure 3.2 are cutaway views and represent 
 30 
 
specific body stations of a given plug along its span.  If a plug fails the de-molding 
criterion at any given body location along the plug then most likely the mold will not 
separate from the plug.   
 
The de-molding direction should be as intuitive as possible even if this means 
sectioning the body into more then 2 pieces.  It is desirable to have as few molds as 
possible; however, some shapes cannot be split into left and right halves unless 
designed/drawn that way.  Creating multiple plugs and molds for a complex shape is 
sometimes the best alternative.  After all, a mold and plug that does not separate from 
each other results in total loss.    
 
In summary, the main criterion for plug design is its ability to generate a mold that 
separates successfully and retains the desired features of the plug.  This sounds 
simple, but be mindful not to overlook it.  Furthermore, the parts for the aircraft will 
come from the molds.  In the end, the ease of manufacturing of parts directly tracks 
back to choices made or overlooked during the plug design.    
3.5.2 Plug Surface Quality Trade-off 
The process of preparing/polishing the plugs can be tedious and repetitive.  However, 
every effort at increasing the quality of the plugs results in a superior surface quality 
of the mold.  Often, the difference between fair and superior surface quality translates 
to two or three more passes with the pneumatic polisher and carnauba liquid wax.  If 
the scratches are not vanishing during polishing than most likely, the plug was not 
sanded with fine enough grit or there is dirt on the buffing pad. 
 
If the quality of the mold is poor, every part pulled from this particular mold will 
require attention.  This ?attention? often results in hours of sanding.  Evidently, it is 
beneficial to spend the time creating plugs that provide superior quality to the molds 
and, consequently, to every part thereof.    
3.5.3 More Thoughts about the Plug Flange 
As mentioned earlier, taking measures to reduce material (high-density foam board) 
and machining costs was not advantageous.  In this project, due to a limited supply of 
foam, the plugs were machined with very small flanges.  Not only was the surface 
area of the flange small but also its thickness was insufficient (Figure 3.4 & 3.5).  The 
foam plug was mounted on a baseboard for support and additional flange area.   
 
Since the thickness of the flange was insufficient, the bare-foam flange was fragile.  
The surface tension created by the primer caused the thin flange to lift off from the 
base plate in both boom and fuselage plugs.  The contact cement used to glue the 
foam to the board was not strong enough to keep the plug flange from lifting off.  The 
first attempt at fixing the problem was to use Plexiglas rather than ply wood for the 
baseboard.  Although the contact spray glue worked better on Plexiglas, it still did not 
solve the problem (Figure 3.7).   
 
 31 
 
To solve the problem, new molds with adequately sized flange would have had to be 
machined.  However, the problem was bypassed by laying down plenty of packing 
tape along the flange edges.  The resulting mold lost some aesthetics along the flange 
area and posed problems during de-molding.  The uneven flange and the step along 
the flange made de-molding the boom mold fairly trying. 
 
 Figure 3.7. Boom Plug on Plexiglas Base Plate. 
   
 
Figure 3.8. Fuselage Plug with Plywood Flange Extension. 
 32 
 
 
After having dealt with the booms and knowing that the fuselage plug flange had 
similar problems, a different approach was taken.  The fuselage plywood baseboard 
was left in place and additional plywood was added.  The additional layer of plywood 
added flange area in the same plane as the foam flange, thus, the fuselage plugs did 
not have the ?step? that the boom plugs did (Figure 3.8).  The additional plywood 
layer was installed with its own frame for support.  Packing tape was used to seal the 
bare plywood and cover the gap along the foam/plywood edge (Figure 3.8).      
 
These problems took many days to solve and, to some extent, reduced the quality of 
the flange of the plugs/molds.  Despite a few surface imperfections in the flange, the 
molds extracted from the plugs still performed as expected.  Nevertheless, this ordeal 
highlighted the importance of not cutting cost and corners.   
3.5.4 Damage and Storage  
After many days working on the plugs, the most frustrating event is to damage or ruin 
it carelessly.  A sunless climate-controlled environment free of overhead objects that 
can fall on the plug is the ideal storage location.  Heavily trafficked areas are highly 
discouraged for use as storage.  Curious hands tend to damage and scratch the surface 
that took much labor to perfect.  Special care must be given to plugs that have not yet 
had molds extracted.  Lastly, the plugs must lie on a flat surface.  In time, the entire 
plug will warp under its own weight if not supported evenly on all corners.   
 
 Figure 3.9. Surface Damage on Fuselage Plug. 
 
The damage shown in Figure 3.9 occurred after the molds were made.  Therefore, 
these scratches did not interfere with the progress of the project or the quality of the 
fuselage mold.  The scratches were a result of mishandling the plugs during transport 
to storage. 
 33 
 
3.5.5 Thoughts on Project Management  
There are many materials necessary for the construction of the plugs, molds and parts.  
Running out of items like latex gloves, stir sticks or mixing cups can create delays in 
production.  Similarly, other items like polyester gel coat and resin can only be 
ground shipped since these are considered hazardous materials.  The shipping of these 
hazardous items alone can take two weeks.  Often times, it is best to order the 
hazardous supplies by themselves.  There is no need to have non-hazardous supplies 
take so long to arrive.   
 
It is very important keep track of supplies and inventory.  The longer plugs or molds 
sit due to a shortage in materials the more prone they are to sustaining careless 
damage.  This is because the plugs/molds/parts sit in the active work area where other 
unrelated activities may damage them. 
3.5.6 Machining Costs VS Manual Labor  
As observed in Figures 3.4 & 5, there are visible lines running the length of the milled 
foam.  The lines are created by the step-over rate used by the CNC machine shop.  
The CNC milling process was done in stages where several ?rough? cuts were made 
followed by ?finer? cuts.  It was possible to have the machine shop mill the plugs to 
perfection, but required more milling machine time.  To reduce machining cost, the 
foam plugs were machined to a point where a person could hand sand the small 
ridgelines.  Had the plugs been made of aluminum, there would be no choice other 
than to mill the metal to perfection.  Evidently, hand filing an aluminum plug, 
although not impossible, poses many problems and risks. 
 
Hand sanding the lines did reduce CNC milling costs, but took a few days to sand.  If 
the foam had been of higher density, it would have taken more time to hand sand.  In 
this case, the choice is clearly between available money and time.  Electing to hand 
sand the step-over lines is the only safe money saving alternative (when using foam).  
As discussed earlier, saving material and machining costs by reducing size of the 
blank foam block (minimizing flange area) is highly discouraged.   
 34 
 
Chapter 4: Mold Making 
4.1 Preparation 
4.1.1 Materials and Tools 
The materials used for mold making are listed in Table 4.1. 
 
Materials Description 
General Materials Mixing cups, stir sticks, latex gloves, eye protection, masks, etc 
Strand Mat Random directional chopped strand mat 
Gel coat Glossy polyester mold surface coating 
Gel coat Additive Gel coat thinner and enhancer 
Resin Polyester resin used for reinforcement of matting 
MEKP Catalyst Hardener used for polyester based primers, gel coats and resins 
Release Polyvinyl Alcohol (PVA) release film  
Liquid Shine Gel coat clean and shine compound 
Machine Glaze Polishing compound 
Liquid Wax Surface Wax in liquid form  
Wood 2? by 4? 
Table 4.1. Mold Making Materials. 
 
The tools used for mold making are listed in Table 4.2. 
 
Tools Description 
Cutting Instruments Shears and/or sharp bladed tools for cutting cloth 
Buffing Pads Polishing Pad used for removal of surface scratches due to sanding 
Digital Scale Instrument used for weighing polyester primer 
HVLP Spray Gun High Volume Low Pressure Siphon Spray Gun 
Dump Gun Cup gun used to spray polyester gel coat and resin  
Pneumatic Polisher Pneumatic Device used for efficient polishing of large areas 
Paint Booth Closed ventilated area reserved for spraying 
Compressed Air High pressure air line, adequate fitting, nozzle, etc. 
Table 4.2. Mold Making Tools. 
4.1.2 Labor and Task 
The mold making process was a coordinated group effort organized in two parts.  The 
first phase of this process required only three individuals.  For the second phase of the 
process, it was found that the ideal number of individuals was six.  Although it was 
possible to include more pairs of hands in either phase, management of too many 
people was cumbersome as the room became crowded.  Once proficient, it was 
possible to perform phase two tasks with just four individuals; however, this required 
 35 
 
a highly experienced group, thorough preparation and, most importantly, knowledge 
of the various quantities to mix.  
 
The subdivisions of tasks for Phases I and II are listed in Tables 4.3 and 4.4, 
respectively.    
 
Labor Task Description 
Sprayer Sprays gel coat and issues commands to other members 
Mixer Mixes pre-measured gel coat batch and supplies sprayer as queued 
Observer Aids spray effort, prevents mishaps and points out bare areas 
Table 4.3. Gel Coating (Phase I) Labor Distribution. 
 
Labor Task Description 
Sprayer Sprays resin, issues commands 
Rollers (2) Places mat and rolls reinforcement mat 
Cutter Cuts strand mat 
Mixer Mixes pre-measured gel coat and supplies sprayer as queued 
Observer Aids spray and mat placement efforts 
Table 4.4. Reinforcement Application (Phase II) Labor Distribution. 
 
The Advice Section of this Chapter contains further information on Labor and Tasks. 
4.1.3 Further Plug Preparation ? Release 
Once the plug was polished and waxed as described in Chapter 3, the PVA release 
was applied to the plug.  The release was sprayed on with a high volume low-pressure 
siphon gun.  The release was applied with either repeated light coats or one ?just 
right? coat.  In due course, different sprayers developed their own preferred 
methodology: one ?thick?, two ?medium? or three ?light? coats.  The sprayed release 
was allowed a couple of hours to dry.  The result was a glossy film barrier that 
protected the plug from the, yet to be sprayed, gel coat.  The Advice Section of this 
Chapter contains several important thoughts on this matter. 
4.2 Making the Molds 
4.2.1 Mold Making ? Breakdown and Comments 
The molds were made following the sequence described in Flowchart 4.1. 
 36 
 
 Flowchart 4.1. Mold Making ? breakdown and comments. 
 
4.2.2 Phase I ? Gel Coating 
Once the PVA release was sprayed and dried, the gel coat was applied evenly with 
the HVLP spray gun.  Although of same make and model, this HVLP was reserved 
for gel coating only.  An additive was mixed into the gel coat before canalization.  
The purpose of the additive was to: 1) thin the gel coat so the HVLP gun could spray 
it, and 2) enhanced the luster and surface quality of the gel coat once dried.   
4.2.3 Phase II ? Reinforcement Application 
The random directional strand mat served as a support structure for the gel coat and, 
overall, the mold.  The first strand mat layer was of lighter weight than the following 
layers.  The ?oz strand mat was desirable as the first layer because it was easier to 
roll out the trapped air bubbles.  The first layer of matting had to support the gel coat 
completely.  Any portion of the gel coat that was not in direct contact with the initial 
 37 
 
layer could crack or collapse once the mold was placed under vacuum.  A defect of 
this kind could risk the molds ability to release from the part being made not to 
mention produce faulty parts.  This problem did not occur with this project.  
Additional layers of reinforcement used 1.5oz weight strand matting.      
 
The plugs were measured and sectioned into three even parts.  Sectioning the ply into 
three separate pieces allowed the rollers to concentrate their effort on a smaller 
region.  It was possible to blanket the mold with one continuous ply; however, this 
would pressure the rollers to eliminate/roll out trapped air in record time.  For this 
reason, as a cautionary measure, the spraying of resin and placement of matting was 
done piecewise.  The resin was sprayed with a dump gun capable of wetting the 
surface very quickly.  The dump gun used 90PSI; thus, delivered large quantities of 
resin in each continuous burst. 
 
 Figure 4.1. Strand Mat Sheet placed on bare Gel Coat. 
 
In the case of the boom molds, the front section contained the wing mount cutout, the 
middle section contained the boom extension and the rear section contained the 
rudder.  The mat sheets were cut oversized so each sheet overlapped at least 1 inch.  
The overlapped mat sheets provided additional stiffness to the mold.  The mat sheets 
were placed in the following order: middle, tail and front.  This order was chosen 
based on difficulty.   
 
The middle portion of the boom was the easiest to cover while the front of the boom 
required many relief cuts and patches due to the complex curvature.  Once an entire 
layer of reinforcement was placed, the process was repeated four more times.  The 
first layer used the lighter weight matting while the subsequent three layers used the 
heavier matting as mentioned earlier.  Figure 4.1 shows the first strand mat sheet 
 38 
 
being placed on the bare gel coat cover.  The technical drawing in Figure 4.2 shows 
the sequence that the layers were applied.  The green layer represents the PVA release 
and the red layer represents the tooling gel coat (Figures 4.2 & 4.3). 
 
 
Figure 4.2. Boom Reinforcement Build-up. 
 
In similar fashion, the fuselage mold reinforcement was placed in sections starting at 
the middle then the ends.  The fuselage mold was easier to work with since its 
curvature did not contain internal cutouts.  The technical drawing in Figure 4.3 shows 
the sequence that the layers were applied.  Since the fuselage flange was wide, 
additional layers were included.    
 
 39 
 
 Figure 4.3. Fuselage Reinforcement Build-up. 
 
 
The reinforcement side of the molds using this process are shown in Figures 4.4. and 
4.5.  It is possible notice the locations of the overlap in the random directional 
matting sheets. 
 
The Subsection entitled ?More on Labor, Tasks and Process? provides a 
comprehensive flowchart highlighting the actions and instructions for Phases I and II 
of mold construction. 
 
 40 
 
 Figure 4.4. Finished Boom Mold ? reverse side. 
 
 Figure 4.5. Finished Fuselage Mold ? reverse side. 
 
4.2.4 Cure Time  
Cure times are mentioned in Table 4.5 of this chapter.  However, the molds only 
reached a full cure after 24 hours.  In all cases, although the molds were dry to the 
touch within six hours, the resin had not yet reached the full cure time suggested by 
the manufacture.  To ensure complete cure, the molds were allowed a minimum of 
three days curing time.  The next chapter develops the techniques used for resin 
infusion and part manufacturing.  
 41 
 
4.3 De-molding 
4.3.1 Releasing the Mold 
After waiting at least eight to ten hours, the mold was separated from the plug.  The 
primary tool used for de-molding was compressed air.  Initially, compressed air was 
burst between the mold and plug without forcing the two apart.  The air was injected 
along the flange of the mold all the way around.  By placing one hand on the mold, it 
was possible to feel the rushing air inside and judge how much of the mold was 
detached.  A good release cover allowed the mold to ?pop? loose almost on its own.   
 
The boom molds were more difficult to release because of the features it contained.  
The airfoil cutout on the boom required patience and methodical air injection.  It was 
important to slowly lift the mold and plug apart straight up as the two detached.  Any 
asymmetrical lifting force on the mold caused the mold to jam and resist de-molding.  
Furthermore, the ridgeline created by the plug along the flange onto the mold posed 
problems.   The step on the perimeter of the flange made it difficult to focus the 
compressed air blast in any one desired place.  The problems surrounding the flange 
were discussed in detail in Chapter 3. 
 
The fuselage molds detached with significant ease.  The most important point to 
remember was to inject the compressed air evenly around the entire perimeter of the 
mold.  The fuselage ?popped off? after repeated application of the compressed air. 
4.3.2 Clean-up and Storage 
The de-molding process was somewhat messy since the PVA release film flaked off 
into many pieces.  The PVA release flakes were either swept with a broom or wiped 
with a damp cloth.  The release is water soluble, so a mop was also used for cleaning 
the floor.  The mold was rinsed with a garden hose and dried with a clean cloth 
without scratching the interior mold surface.  The plug was wiped with a damp cloth 
and set aside for storage. 
 
As described in Chapter 3, the tool clean-up involved some variation of rinsing, 
wiping or bathing the tool into acetone.  The HVLP spray gun used for gel coating 
(Phase I) and the dump gun used with resin (Phase II) were both cleaned by spraying 
pure acetone.  The HVLP spray gun used with PVA release was rinsed with water.  
Water was also sprayed to clean the inside of the gun used in the PVA application.   
 
 42 
 
 Figure 4.6. Boom Mold ? support 2? by 4? frame. 
 
In accordance with Chapter 3 storage instructions, the plugs were relocated shortly 
after the molds were released.  A wood frame was constructed for the molds (Figure 
4.6).  The frame was important because it provided a stable platform for the mold to 
rest on.  Furthermore, the frame prevented the mold from warping under its own 
weight over time.   
4.4 Advice 
4.4.1 More on Labor, Tasks and Process 
In both steps to mold making, the person doing the spraying is the group leader.  The 
coordinated effort of the group occurs around the progress of the person spraying.  It 
is not advisable to mix the catalyst to all portions of gel coat or resin at once.  Since it 
takes time to spray, the portions are mixed as the job progresses. 
 
For phase 1, all three of the group members should be able to maintain their hands 
clean at all times.  The first batch of gel coat is mixed and handed to the sprayer.  As 
the sprayer gets down to ? of gel coat in the spray-cup, the command is issued to the 
mixer to prepare the next batch.  Once the sprayer runs out, the mixer has a fresh 
batch on standby.  The observer serves as an aid to the spraying effort.  The observer 
constantly looks for bare spots or areas of concern.  Furthermore, the observer 
ensures that nothing touches or drags on the plug while the sprayer concentrates on 
spraying.  The Phase I detailed flowchart is provided in Flowchart 4.2.   
 
 43 
 
 Flowchart 4.2. Mold Making Gel Coating (Phase I) ? detailed flowchart. 
 
The hose supplying pressure to the spray gun can easily drag on the plug if both 
sprayer and observer are not attentive.  Another common mishap occurs when 
clothes, especially the sprayers sleeve, touches the partially sprayed plug.   During 
Phase I, objects touching or dragging on the plug can create two problems: 1) the 
quality of the gel coat surface is compromised or 2) the PVA release film is detached 
or frayed.  If the release film is damaged, the mold may be permanently attached to 
the plug.   
 
Phase 2 of the mold making process requires more discipline.  The sprayer is still the 
group leader and has to manage three more helpers.  The sprayer should make every 
effort to keep both hands uncontaminated.  More cleanup is required if the sprayer is 
messy.  The observer performs the same duties as in phase 1, but with one additional 
duty.  While the sprayer is reloading, the observer aids the rollers by making relief 
cuts to the mat as requested by the roller.  The mixer performs the same task as 
described in Phase 1.  However, the mixer, in this phase, is working with polyester 
resin rather than polyester gel coat.   
 
The cutter is responsible for cutting the strand mat.  Several pieces of strand mat were 
pre-cut during setup.  However, patches and other pieces may need to be cut on the 
spot.  To minimize cleanup, the cutter works only with dry materials and tools and 
should not contaminate either hand with resin.  The rollers have two duties: placing 
the strand mat and rolling out bubbles.  It takes practice to learn how to perform this 
function without contaminating either hand.  The Phase II detailed flowchart is 
provided in Flowchart 4.3. 
 44 
 
 Flowchart 4.3. Mold Making Reinforcement Application (Phase II) ? detailed flowchart. 
 45 
 
4.4.2 Prepare in Advance 
Preparing in advance is a key factor for the successful completion of any composite 
project.  Poor preparation may result in wasted materials, damaged tools, loss of parts 
and significant stress.  Every detail must be satisfied prior to mixing the catalyst with 
the resin.  The work may seem slow during preparation, but goes quickly once the 
resin working time is counting down.        
4.4.3 Thoughts on Mold Release 
The spraying of PVA release was found to be more difficult than anticipated.  In 
some cases, the plug had to be sprayed several times.  To achieve repeatedly 
successful applications the PVA release, some level of skill was required on the part 
of the person doing the spraying.  If a PVA spray job was bad, the plug was 
wiped/rinsed with water, dried and re-sprayed.  If the HVLP spray gun is atomizing 
the PVA while instantly generating ?cobwebs?, the room temperature is below ideal.  
This can be important to note during the winter months. 
 
The PVA release is often considered cumbersome to use because it is destroyed with 
each cycle.  Thusly, a new release coat must be applied with each use.  However, the 
film forming PVA release does provide a water-soluble barrier between the mold and 
the plug.  This can provide some degree of fault tolerance as far as detaching two 
parts since there is a measurable gap, in this case, separating the plug from the mold. 
 
Wax/paste release products are available and commonly used in industry.  These 
waxes vary in chemical content and, when buffed, create a non-stick hard shell on the 
surface.  This kind of product can and will work; however, familiarity with the 
product and its application is mandatory.  A possible drawback to wax or paste based 
release systems is build up.  The wax can accumulate in places both filling in features 
and/or changing the surface contour.  A wax surface coat does require maintenance 
and will require re-application and buffing after 3 or 4 cycles.    
 
Agent release products are found in liquid form and, most often, wiped on with a 
clean cloth.  These release agents can be very effective, yet, tend to be quite costly.  
Commonly, the release agent will be highly flammable, have a potent odor and 
require plenty of ventilation during use.  Most often, the ?what you pay is what you 
get? rule does apply to these products. 
 
In general, all external release products follow one rule: more is not always better.  
For example, spraying too much PVA results in sags and runs while too much wax 
results in accumulation/build up.  Agent release products also have an ideal method of 
application.  Too much release agent can cause crystallization on the surface of the 
mold.  The crystals negate the efforts made during sanding and polishing creating a 
rough surface.  Furthermore, the tiny crystals grab the intended counterpart and 
render it impossible to de-mold.  A heavy application of either PVA film or release 
 46 
 
agent can suffer from uneven dry-out where the top surface is dry while the undercoat 
of the product is still liquid. 
 
Further general considerations for external release products are chemical 
compatibility, temperature range and permanence.  Some release products are only 
chemically compatible with a specific family of epoxies, gel coats and/or resin 
systems.  The temperature range is related to the required cure cycle.  Epoxy or resin 
systems that require autoclaving will also require high-temperature release 
compounds.  Lastly, most waxes and release agents are semi-permanent or permanent 
and may require a special ?strip? agent to recondition the surface to original state.  
Otherwise, a permanent release agent can only be removed by sanding it away 
completely. 
  
In summary, the best way to become comfortable with a release system is to practice 
using it with a trial piece.  A good supplier will be willing to share experience and 
pointers about the various release agents.  A trustworthy vendor with good customer 
service can make the difference between success and frustration.  Lastly, any recently 
made plug or mold does have a break-in period.  Often times, it takes repeated release 
applications or overall cycles for a plug or mold to create the best parts, so, be patient.    
4.4.4 De-molding ? Troubleshooting 
Using compressed air is the least evasive method of separating the mold from the 
plug or, as explained later, the mold from the part.  It was found that using multiple 
compressed air nozzles worked better than just one.  The rushing air entering from 
opposite locations in the plug generated a lifting cushion.  This was so effective that 
the use of two or more nozzles became standard practice. 
 
Some cases required more effort and called for the use of high-pressure water.  The 
PVA is water-soluble and gets flushed when rinsed.  This can create just the 
necessary room to ?wiggle? the pieces loose from one another.  Some arm strength 
was also applied in these stubborn cases.  However, there is always some margin of 
risk when using brute force to separate two pieces.  Forceful intervention should only 
be used in desperate situations, in which case, a decision might have to be made 
wether to preserve the plug or the mold.   
4.4.5 Gel Coating ? Troubleshooting   
As mentioned before, the gel coat must be applied evenly over the entire plug.  Prior 
to starting Phase I, it was evident that Phase II (Reinforcement Application) of the 
boom molds would be time consuming.  The airfoil cutout in the boom plug would 
require many relief cuts and custom patchwork as far as the placement of 
reinforcement matting.  Therefore, in an effort to facilitate the Phase II efforts, the 
airfoil cutout in the boom plug was filled more than halfway with gel coat.  At the 
time, this seemed to be an attractive proposition; however, it did more harm than 
good.   
 
 47 
 
The gel coat has a shrinkage property sometimes specified by the manufacture.  In 
actuality, the amount of shrinkage may vary from job to job and most companies 
provide this and other information as reference parameters.  The shrinkage of the gel 
coat as it cures can vary due to storage conditions, age, catalyst to gel coat mixture 
ratio, room conditions and overall application/methodology employed by the user. 
 
 Figure 4.7. Cracked Gel Coat due to Uneven Thickness. 
 
The catalyst initiates an exothermic reaction that releases quite a bit of heat and, 
thusly, hardening the resin.  Filling the airfoil cutout created a concentration of heat in 
that area of the plug.  The gel coat deformed/shrank while curing and caused cracks to 
occur on the surface (Figure 4.7).  This was catastrophic but, fortunately, did not 
damage the plug itself.  The gel coat was carefully peeled off.  The PVA release was 
washed off and re-sprayed.  The next time the gel coat was applied, the sprayer was 
mindful to apply a constant coating no thicker than ?th of an inch.   
 
In making the best out of this situation, the solid gel coat with the shape of the airfoil 
was set aside and later used as a template.  The custom cut patches of strand mat used 
to reinforce the inside of the airfoil cutout was cut using the trace made from the solid 
gel coat airfoil.  This gel coat airfoil was also used as a cutting template in Chapter 5.      
4.4.6 PVA and Gel Coat Spraying Tips 
Both PVA and gel coat should be sprayed systematically.  The sprayer should 
maintain the spray nozzle both perpendicular to and 10 inches away from the surface.  
It is important to spray through the ends/edges of the part.  This means that the return 
pass should be initiated outside the piece being sprayed.  Notice Figure 4.1, the gel 
 48 
 
coat was sprayed on the brown paper covering the table.  To ensure that the entire 
mold piece had a level gel coat, the operator sprayed beyond the edges/flange before 
coming back.  This was particularly important when spraying PVA release.  If the 
corners of the flange are starved of release this can make de-molding very difficult.  
During preparation, the worktable was covered with either newspaper or brown paper 
so spraying onto the table was not a problem. 
 
Although more forgiving than the PVA release, the spraying of gel coat does require 
attention.  Troubleshooting guidelines for gel coat application are listed in Table 4.5. 
Removing and re-applying the entire gel coat cover is the best solution to fixing 
cracks and/or tackiness problems on the surface.  Any doubts as to the quality of the 
applied gel coat cover needs to be addressed immediately.  Risking continuing on to 
Phase II can be a gamble too costly to take on.   
 
Problem Cause 
Low Gloss Incorrect catalyst ratio, under cure or poorly mixed gel coat 
Uneven Coat Thickness Excessive surges during spraying ? operator error 
Tackiness Insufficient catalyst present, under cure conditions, poor mixture 
Sags Excessive thickness and/or air assist, spray nozzle too close to mold 
Cracks Shrinkage, uneven gel coat thickness applied 
Table 4.5. Gel Coat Spray Troubleshooting. 
4.4.7 Manage the Workspace 
The workspace needs to be clear of all potential obstacles no matter how innocent it 
appears.  All furniture not involved with the process should be cleared. Common 
hang-ups and accidents involve chairs, extension cords, and/or the compressed air 
hoses. 
 
The working time can be 20 or 30 minutes, but all it takes is 10 seconds for the gel 
coat or resin to set.  Not having the correct tools and materials handy commonly adds 
stress to the workplace, especially when the resin working time is nearly expired.  It 
is wisest to use these precious working time minutes doing meaningful work rather 
than running around after a new box of latex gloves or negotiating objects that are 
constantly in the way.      
4.4.8 Advice Summary 
Processes described both above and below promise many desirable outcomes.  
However, the integrity or lack thereof ones maintains with the work is the decisive 
factor as to its outcome.  It is difficult to predict if something will come out great or 
superb; yet, easy to predict if it will result poorly or unsatisfactorily.  It is always best 
to error in the side of caution than to remediate the consequences.  
  
 49 
 
Chapter 5:  VARTM 
 
5.1 Preparation 
5.1.1 Materials and Tools  
The materials used for Vacuum Assisted Resin Transfer Molding (VARTM) are 
listed in Table 5.1. 
 
Materials Description 
General Materials Mixing cups, stir sticks, latex gloves, eye protection, masks, etc 
Spiral Tubing Tubing used to vacuum air and resin from part, ?in. or ?in. 
T & Y Connectors Connector fitting for tubing, ?in diameter 
Vacuum Hose Vacuum rated tubing, ?in. diameter 
Hose Clamp Clamp used to secure hose and connectors 
Tacky Tape Vacuum tape used for sealing vacuum bag to mold 
Vacuum Bag Impermeable film cover used to seal mold during vacuum application 
Peel Ply Barrier film/fabric that is removable post cure 
Resin Polyester resin used for reinforcement of matting 
MEKP Catalyst Hardener used for polyester based primers, gel coats and resins 
Carbon Cloth 3K 2x2 Twill 5.7oz/sq Carbon Reinforcement 
Infusion Media Honeycomb infusion mat and diamond shape infusion mesh 
Release Polyvinyl Alcohol (PVA) release film  
Liquid Shine Gel coat clean and shine compound 
Machine Glaze Polishing compound 
Liquid Wax Surface Wax in liquid form  
Table 5.1. VARTM Materials. 
 
The tools used for VARTM are listed in Table 5.2. 
 
Tools Description 
Cutting Instruments Shears and/or sharp bladed tools for cutting cloth 
Measuring Equipment Measuring tape (preferred) and/or yard stick 
Buffing Pads Polishing Pad used for removal of surface scratches due to sanding 
Digital Scale Instrument used for weighing polyester resin 
HVLP Spray Gun High Volume Low Pressure Siphon Spray Gun for PVA 
Resin Trap Container that prevents resin from going into vacuum pump 
Pneumatic Polisher Pneumatic Device used for efficient polishing of large areas 
Paint Booth Closed ventilated area reserved for spraying 
Air Injection Wedge Wedge shaped air injection nozzle 
Compressed Air High pressure air line, adequate fitting, nozzle, etc 
Vacuum Pump Vacuum pump with adequate fittings, hose, etc 
VARTM Tools. 
 
 
 50 
 
5.1.2 Labor and Task 
The vacuum infusion process (VIP) required several people.  To management of the 
available personnel was made efficient by dividing the process into two parts: 
preparation/dry lay-up and infusion.  The labor and task descriptions are provided in 
Tables 5.3 and 5.4.    
 
Labor Task Description 
Sprayer Pre-PVA prep, handles molds, and sprays PVA  
Sprayer Helper Handles molds, aids in PVA preparation and spraying 
Dry Lay-up Crew Cuts materials, reinforcement cloth and infusion dissipation core Locates cut materials into mold and assembles vacuum bag (2 to 3) 
Workspace Support Clean-up & Prep Workspace, organize tools 
Preparation & Dry Lay-up (Phase I) Labor Distribution. 
 
Labor Task Description 
Mixer Mixes pre-measured resin batch 
Vacuum Operator Manage vacuum pump 
Infusion Support Assists infusion process (4 to 6 people) 
Resin Infusion (Phase II) Labor Distribution. 
 
Although both phases could be carried out with as little as three people, it was found 
that multiple hands working together increased the chances of making a successful 
part.  During the preparation phase, typically four to five individuals were sufficient.  
However, the infusion phase required more helpers on-site.  The Advice Section of 
this Chapter contains further information on Labor and Task. 
5.1.3 Further Mold Preparation  
The molds were prepped as necessary depending on their quality once de-molded 
from the plug.  In this stage, the mold was polished and waxed as described in 
Chapter 4.  Any polishing work done to the molds followed the same steps and 
procedures used for the plugs. 
 
The PVA release was sprayed in similar fashion as previously done on the plugs.  
However, one additional step was required.  The border of the flange was masked 
with clear packing tape.  Once the mold was sprayed and the sprayer approved the 
job, the packing tape was peeled off.  It was important to peel off the packing tape 
before the PVA fully dried.  If the packing tape is not removed readily, the dried PVA 
will peel off with the packing tape as it is removed.  Thusly, the spray job may be 
ruined and require re-application.  The spraying of PVA followed the same procedure 
as described in Chapter 4. 
 
The sprayer and helper performed the following actions in sequence: moved the mold 
to the spray booth, wiped off any surface dust, masked the flange edges, sprayed PVA 
release, inspected spray job, and removed the packing tape.  If the spray job was 
unsuccessful, the existing film was washed off with water.  The mold was then dried 
 51 
 
and prepped for another PVA application.  A successful PVA application resulted in a 
shiny film free of sags, bare spots and pools (Figures 5.1 and 5.2).   
 
 Figure 5.1. Fuselage Mold (Right Half) Sprayed with PVA ? tape along flange border 
removed.  
 
The fuselage mold was further prepped with a pinstripe line.  The white line was 
located inside the mold with a template.  The PVA was sprayed on top of the 
pinstripe.  The line created a slight impression on the fuselage part.  The hatch cutout 
was made following the line imprint (Chapter 7).  Once successfully sprayed, the 
packing tape was removed and the fuselage mold was allowed 2 hours to dry 
completely (Figure 5.1). 
 
 
 52 
 
 Figure 5.2. Boom Mold (Right Half) Being Sprayed with PVA ? with tape along 
flange border. 
5.2 Infusion 
5.2.1 Process Description 
The Vacuum Infusion Process (VIP) uses lower than atmospheric conditions on the 
laminate to propel the catalyzed resin.  The liquid resin, while under vacuum, wets 
the dry reinforcement inside the sealed mold.  In the case of VARTM, the resin is 
infused by vacuum while creating a single-sided finished part.  
 
The single sided Vacuum Assisted Resin Transfer Molding used in this project 
utilized both resin injection molding and vacuum bagging processes (Figure5.3).  The 
laminate retained the exact contour and surface quality of the mold on the finished 
side.  The vacuum bag side of the laminate did not retain a repeatable finish since it 
did not have direct contact to a counter mold.    
 
The single sided resin infusion molding process allowed the reverse side to be 
finished such that it was suited for gluing other laminates post-infusion.  The use of a 
porous fabric, called Peel Ply, allowed the reverse side to be constantly rough; 
therefore, surface ?ready for later assembly (Chapter 7).    
 53 
 
 Figure 5.3. Vacuum Infusion Process.  
 
  
 
Figure 5.4. VIP Setup.   
 54 
 
 
The Infusion Setup is illustrated in Figure 5.4.  The resin is drawn into the part by 
means of suction provided by the vacuum pump.  The resin trap protects the vacuum 
pump by allowing out flowing resin to drip into the bucket.  The gage on the resin 
trap is used to monitor the vacuum pressure in the system.    
 
Other Resin Transfer Molding (RTM) methods use molds on both sides to provide 
consistent finish in both the exterior and interior faces of the part.  Furthermore, parts 
that are required to have a specific thickness everywhere may also need to have molds 
on both sides. However, the creation of matched male and female molds double the 
time to production time and cost.   
 
5.2.2 VARTM ? Breakdown and Comments 
The parts were made following the sequence described in Flowcharts 5.1 and 1.3. 
 
 Flowchart 5.1. VARTM ? breakdown and comments. 
 55 
 
5.2.3 Phase I - Dry Lay-up 
The dry lay-up phase, as the name implies, represents all portions of the process prior 
to the infusion of the catalyzed resin.  There are many important steps between the 
spraying of PVA release and the infusion of resin.  Any skipped or abbreviated 
actions in this phase may result in loss of part and/or mold.   
 
While the PVA release film dried, two plies of carbon cloth and one ply of infusion 
medium was sized and cut.  The necessary materials and tools were organized and 
laid out in the designated work area. The mold, with a successfully applied and dried 
PVA film cover, was removed from the paint booth.  If other molds needed PVA, the 
spray crew continued on to the next job. 
 
The double-sided vacuum seal tape, called tacky tape, was placed along the outside 
edges of the mold where the clear packing tape once was.  It was found, through 
experience, that placing the tacky tape entirely on top of the PVA release caused 
leaks to occur during the vacuum bagging process.  This was the reason why masking 
the perimeter of the flange was important.  However, the presence of the unprotected 
border of the flange posed new concerns.   
 
Two solutions were found for this problem. Firstly, having a sufficiently large flange 
provided a gap between the reinforcement cloth and the tacky tapeline.  Therefore, 
resin did disperse onto the unprotected edges of the mold since it only traveled where 
cloth was present.  Secondly, in places where the flange was minimal, the tacky tape 
was slightly overlapped with the PVA release film.  This allowed the infused resin to 
touch the tacky tape while not contacting any bare gel coat.       
 
In the case of the boom mold, the tacky tape, with reverse adhesive side still covered, 
was overlapped with the release film line.  The pre-cut infusion media and carbon 
reinforcement were placed in the mold.  The vacuum bag was cut oversized 
widthwise and placed on the mold (Figure 5.5). 
 
The infusion dispersion media between the carbon plies was custom cut to match the 
features of the boom.  Two different media were used in the fabrication of booms.  In 
order to reduce weight in the tail of the aircraft, the rudder was made with a 
lightweight infusion medium.  However, to increase stiffness, the entire boom and 
vertical stabilizer was manufactured with a thicker infusion medium.  Figures 5.6 ? 8 
show the placement of the infusion media for the boom.  The red infusion mesh was 
sized to fit inside the vertical stabilizer while overlapping with the hexagonal infusion 
medium.  The rudder was cut from the vertical stabilizer as discussed in Chapter 7.  
The airfoil-shaped cut at the nose of the boom in Figure 5.7 was made with the gel 
coat template mentioned in Chapter 4 (the gel coat template can be seen on the table, 
top left, in Figure 5.8). 
 
 56 
 
 Figure 5.5. Boom Mold (Left Half) ? dry lay-up. 
 
 Figure 5.6. Red Mesh Infusion Medium ? used in vertical stabilizer and rudder. 
 57 
 
 
 Figure 5.7. Airfoil Cutout ? infusion medium airfoil cutout with ?walls?.  
 
 Figure 5.8. Carbon/Infusion Medium/Carbon Sandwich ? worked into boom. 
 58 
 
Small weights were placed inside the boom cavity to assist in the placement of the 
cloth and infusion media.  By using the weights, one person could work the 
reinforcement in place without the assistance of more hands (Figure 5.8).  
Meanwhile, another worker attached the vacuum bag to the tacky tape.  Once the long 
side of the bag was adhered to the tacky tape, a small cut was made for the ?T? 
connector used as the suction port.  
 
The dry lay-up process for the fuselage was similar to the boom.  However, the 
fuselage was less cumbersome to fabricate because it did not contain any internal 
features like the airfoil cutout of the boom.  To reduce fuselage weight, the red mesh 
infusion medium was used instead of the hexagonal dispersion cloth.  The mesh did 
require relief cuts so it could conform to the curvature of the fuselage.  In some places 
where the relief cuts resulted in an overlap, the excess mesh was trimmed so it only 
intersected itself.    
5.2.4 Phase I ? Inlet and Outlet Lines 
The resin distribution line and the suction port were strategically located to maximize 
wet out.  The distribution setup varied for the boom and fuselage.  Due to the feature 
and size difference between the boom and fuselage, the inlet spiral tubing and outlet 
suction port required different arrangements. 
 
 
Figure 5.9. Boom ? inlet and outlet ports detail, no vacuum applied. 
 
 59 
 
 Figure 5.10. Boom ? vacuum bag sealed, no vacuum applied. 
 
For the booms, the resin flowed along a radial path toward the suction port. The spiral 
tubing partially curled around the nose of the boom while almost entirely curled 
around the rudder.  The suction port was essentially a sink located in the center of the 
mold.  The spiral tubing used was ?-inch diameter while the ?T? connector was ?-
inch diameter.  This allowed the spiral tubing to be one continuous piece that slid 
through the connector.  The ?T? connectors were prepped with a ring of tacky tape at 
the stem.  Once a hole was made on the vacuum bag and the connector was pushed 
through, the tacky tape ring sealed the bag. Small patches of peel ply were placed 
under the ?T? connectors to facilitate their removal once the part was cured (Figure 
5.9).  After the spiral tubing, ?T? connectors and peel ply were in place the vacuum 
bag was sealed (Figure 5.10).     
 
The evacuation of air required several helpers since the reinforcement cloth, spiral 
tubing, and ports needed to be held in place during suction.  In the sharp contours of 
the boom, the cloth was tucked in and held in place with a plastic putty spreader.  It 
was necessary to apply and release the vacuum several times to eventually work the 
cloth sandwich into place.  The laminate tucked in some more with each cycle that the 
vacuum was released and re-applied.  Eventually, the laminate took the exact shape of 
the boom and was ready to receive the resin (Figure 5.11).  Lastly, the entire piece 
was scrutinized for leaks along the tacky tapeline.   
 60 
 
 Figure 5.11. Boom ? ready for infusion, vacuum applied.  
 
 Figure 5.12. Fuselage ? outlet port detail and ?rabbit ear?. 
 61 
 
 Figure 5.13. Fuselage ? ready for infusion, vacuum applied. 
 
The fuselage used a different inlet and outlet setup.  The inlet spiral tubing was placed 
inside the fuselage; therefore, peel ply was placed along its entire length.  Without the 
peel ply, the spiral tube would be impossible to remove cleanly, hence, adding weight 
to the aircraft fuselage.  The perimeter of the fuselage was lined with the outlet spiral 
tubing.  A ?Y? connector was used on the outlet port since it laid flat on the mold 
flange (Figure 5.12).  The fuselage mold was sealed and vacuumed in similar fashion 
as the boom mold (Figure 5.13).  The fuselage also required several vacuum and 
release cycles until the laminate fully contoured to the mold.     
5.2.5 Phase II ? Resin Infusion 
The vacuum-sealed boom and fuselage were injected with 1.25lbs of catalyzed resin 
each (Figures 5.14 & 5.16).  It was possible to track the progress of the resin by the 
color of the carbon cloth.  The carbon laminate turned dark as the resin soaked it.  
Once full wet out was achieved, the inlet hose was clamped off and the part was 
allowed 10 hours to cure (Figures 5.15 & 17). 
 
 62 
 
 Figure 5.14. Resin Infusion in Progress ? right boom half.  
 
 Figure 5.15. Boom Resin Infusion Complete. 
 63 
 
 
 Figure 5.16. Resin Infusion in Progress ? right fuselage half.  
 
 Figure 5.17. Fuselage Resin Infusion Complete.  
 64 
 
5.3 De-molding 
5.3.1 Releasing the Part 
The primary tool used to release the part from the mold was compressed air with a 
wedge shape nozzle.  The compressed air was injected evenly around the flange 
causing the part to ?pop? off the mold (Figure 5.18).  To remove the PVA release, the 
mold and part were rinsed off with water (Figures 5.19 & 20).  The mold was re-
prepped for another molding cycle.  The fuselage part was de-molded in similar 
fashion (Figure 5.21).  The resin inlet and outlet ports were removed, the part was 
rinsed with water and the fuselage was ready for trimming.   
 
Troubleshooting for de-molding the part consisted of the same procedures as 
discussed in the Advice Section of Chapter 4.  
 65 
 
 
 Figure 5.18. Boom ? de-molding process (from top left to bottom right). 
 66 
 
 Figure 5.19. Boom ? de-molded part with release film. 
 
 Figure 5.20. Boom Mold and Part ? PVA release rinsed off with water. 
 67 
 
 Figure 5.21. Fuselage Mold and Part ? with PVA release and outlet port attached. 
5.3.2 Clean-up 
The PVA release film on both mold and part were washed off with water.  The used 
vacuum bag material, tacky tape, spiral tubing, peel ply, resin mixing bucket, ?T? & 
?Y? connectors and vacuum hose lines were discarded.  The resin trap was opened 
and the hardened resin was discarded.  The mold was either prepped for another 
infusion cycle or transported to storage. 
5.4 Advice 
5.4.1 Inlet and Outlet Ports ? Where to Place Them 
It is important to realize that the catalyzed resin tends follow the path of least 
resistance during infusion.  Once the resin reaches the suction port, it will prefer to 
flow in that particular path; therefore, decreasing wet out dispersion rate in other 
portions of the laminate.  A symmetric location of inlet and outlet resin ports may not 
be ideal depending on the features and size of the part.  There are many possible 
setups for infusion and some might require multiple in and out ports along with the 
use of spiral tubing in one, two or all ports.  In this project, for example, two 
completely different successful setups were used to permeate resin into the laminate.  
 
 68 
 
For a given part, where locations of ports are not trivial, it may take some trial and 
error to find the ideal setup.  It is highly advised to perform trial and error cycles 
using e-glass rather than carbon.  This will reduce the cost associated with the trials 
and will allow the team to practice with less apprehension towards failure.  Software 
that predicts the flow of resin during the VARTM process is available.  However, 
licensing for such products can be costly.  Acquiring such software is a calculated 
trade-off between project size, cost, time/deadline, required first-time success and 
required production rate.  
 
Simple tricks can be used to manipulate the speed of wet out.  One method of 
controlling infusion involves tilting the mold so gravity can aid or prohibit the flow.  
Raising the catalyzed resin container above or below the mold is another method of 
controlling wet out.  It is important to note that the resin flows slower in 
reinforcement cloth that does not have infusion medium as part of the laminate.  
However, the overall single limiting factor that dictates the dynamics of the resin 
during infusion is its viscosity.  Suitable epoxy or polyester matrix systems used for 
infusion should have a viscosity no greater than 1000 mPa-s or 1000 centipoise (cps).  
In this project, pigment-coloring additives were found to increase viscosity of the 
catalyzed polyester resin; thus, impairing the resins ability to wet out the laminate.    
5.4.2 Work the Laminate 
It is imperative to work the laminate completely into the cavity of the mold.  Failure 
to do so typically results in two kinds defects: blistering or resin bridging.  Trapped 
air on the surface of the mold diminishes the aesthetics of the part while trapped air 
between plies significantly degrades structural properties of the laminate.   
 
Bridging is a natural occurrence inherent to Vacuum Assisted Resin Transfer 
Molding.  It occurs when the laminate does not fully compress into the mold surface, 
especially in molds whose features do not readily allow the cloth to negotiate 
complex curvatures.   During infusion, gaps are filled creating visible resin-rich 
pockets on the surface.  The resin rich pockets not only add unnecessary weight to the 
part but are also subject to cracking, especially if the surface is subject to flexural 
stresses.  Although impossible to avoid along edges with tight radii, it can be 
minimized by systematically working the cloth into place by means of applying and 
releasing vacuum as described earlier in this Chapter.  This requires patience and 
several workers (Figures 5.22 & 5.23).   
 
The bridging phenomenon can be observed in Figure 5.20 at the leading edge of the 
airfoil cutout.  The cloth is unable to accommodate the change of curvature, 
especially in sudden 90-degree transition.  The resin accumulates along the entire 
edge of the airfoil cutout creating a solid rim.  Taking the time to deal with areas 
prone to bridging is highly advisable since resin rich pockets increase the weight 
while scarcely, if at all, enhancing structural properties of the part.  
 
 
 69 
 
 Figure 5.22. Working reinforcement into mold cavity. 
 
 
Figure 5.23. Vacuumed part ? check for bridging.  
 70 
 
5.4.3 Working the Vacuum Bag 
The vacuum bag has to accommodate the curvature of the mold without ripping.  To 
aid the elongation demand placed on the vacuum film, pleats or ?rabbit ears? are built 
into the seal (Figures 5.12. & 5.13).  The vacuum bag film is widthwise cut oversize.  
The extra material is folded and sealed with tacky tape along the border.  The rabbit 
ear provides the extra vacuum bagging material necessary to drape the mold 
successfully.  Pleats are strategically located to prevent bridging of the vacuum bag.  
Places where the bag does not conform to the cavity jeopardize the entire bagging 
process.  The unsupported vacuum bag may stretch beyond its limits and burst during 
vacuuming.    
5.4.4 Document Important Quantities 
Enough resin has to be mixed to fill both the ?-inch inlet and ? inch spiral tubing, 
while wetting out the entire part.  The depletion of the catalyzed resin reservoir while 
not achieving complete wet-out posed two problems: 1) more resin had to be mixed 
?on the fly? and 2) air was injected into the part; thus, introducing air bubbles or 
pockets in to the laminate.  This mistake could compromise both quality and utility of 
the part.  
 
It is better to have resin left over in the mixing bucket than to run out during infusion.  
The next time a part is infused, reconsider the resin batch amount based on how much 
was left over.  To do this, simply weigh the hardened resin and subtract 80% of this 
weight from the weight of the original batch.  The 20% margin is there for insurance 
so there is always some resin left over.  It is impossible not to lose some resin, but 
possible to minimize the waste.   
 
The trial runs, necessary to learn the locations of inlet and outlet ports, are ideal for 
figuring out important quantities.  Such quantities are dimensions of reinforcement 
cloth, infusion medium, and vacuum bag, lengths of spiral tubing, and vacuum 
tubing, and, finally, resin amount for infusion.      
5.4.5 Storage 
Storage considerations for mold and parts follow the same guidelines as previously 
discussed in Chapter 3.  However, the parts require particular attention for these are 
fairly subject to warping.  Recently de-molded parts may be dry to the touch yet not 
fully cured.  If left inadequately supported, the part may permanently warp or bend 
out of shape. 
5.4.6 Summary ? VARTM Pros & Cons 
The pros and cons associated with VARTM are summarized in Table 5.5.  This 
manufacturing method offers superior productivity and repeatability.  However, the 
process is unforgiving to mistakes, often resulting in total loss.  There is no size limit 
 71 
 
on the part being created.  Thus, VARTM is a preferred method for large-scale parts 
and production.   
 
The greatest advantage of VARTM is the unlimited setup time.  Resin infusion only 
takes place after the dry reinforcement is successfully vacuumed and conformed into 
the mold.  The resin intake is minimal (compared to manual wet lay-up) since the 
mold is sealed and already at vacuumed conditions when infusion begins.  This 
reduces weight while maximizing structural properties of the composite. 
 
Good Fiber to Resin Ratio 
Less Resin Waste than Other Methods 
Consistent Resin Usage 
Unlimited (Dry) Setup Time 
Minimal Human Exposure to Resin and Vapors 
Good Parts Reproducibility 
Special Reinforcement Easily Added During Lay-up 
Shorter Production Time than Wet Lay-up 
Identical Parts Created Every Time 
PR
OS
 
Quick Turn Around 
Mold Design is Critical - requires skill and experience 
Resin Bridging in Radii or Edges 
Reinforcement Movement During Vacuuming 
Potentially Complicated Setup 
Easy to Ruin Part 
Requires Perfect Vacuum Seal Every Time 
CO
NS
 
Trial and Error Process 
Table 5.5. VARTM ? Pros & Cons. 
 
 
 
 72 
 
Chapter 6:  Wing Making 
 
6.1 Preparation 
6.1.1 Materials and Tools 
The materials, used for Foam Cutting and Wet Wrapping, are listed in Table 6.1. 
 
Materials Description 
General Materials Mixing cups, stir sticks, latex gloves, eye protection, masks, etc 
Hot-Wire Nickel-Chromium (Nichrome) Wire for Foam Cutter 24 WG 
Foam Low density foam, 8ft tall x 2ft wide x 4in thick 
Tacky Tape Vacuum tape used for sealing vacuum bag to mold 
Vacuum Bag Impermeable film cover used to seal mold during vacuum application 
Breather/Bleeder Cloth used in vacuum bagging that allows airflow  
Release Film High gloss non-perforated release film (not PVA) 
Vacuum Port Thru-bag vacuum connector 
Hinge Materials High quality hinges, balsa stock, servo control horn 
Spar Tubing Carbon-fiber composite spar tubing 
Structural Inserts Composite laminate insert ? part of structure of wing 
Carbon Cloth 3K 2x2 Twill 5.7oz/sq yard Carbon Reinforcement Cloth 
Resin Polyester resin used on carbon 
MEKP Catalyst Hardener used for polyester based primers, gel coats and resins 
Spreader Squeegees or ?slicks? for spreading catalyzed resin during wet lay-up 
Packing Tape Tape used to secure release film 
Sanding Paper Sanding paper grits 220 to 400 
MDF Board Compressed wood-dust board 
Wood Dowel Non-tapered wood dowel 
Solvent Denatured Alcohol 
Table 6.1. Wing Making (Wet Lay-Up) Materials. 
 
Tools Description 
Cutting Instruments Shears and/or sharp bladed tools (X-Acto Knives) 
General Tools Pliers, screw drives and Dremel 
Measuring Equipment Measuring tape, metal yardstick (or longer), T-square, digital scale 
Sanding Blocks Support for sanding paper 
Cutting Aids Steel straight edge and large cutting mat 
Foam Cutter CNC Foam Cutter with wing cutting software 
Variac & Wiring Hot wire power supply ? provide current to hot wire 
Lay-up Room Well ventilated clean room that contains a wide cutting table 
Vacuum Pump Vacuum pump with adequate fittings, hose, etc 
Power Sander Woodshop power sander 
Other Tools Wood clamps/grips, misc. lead weights 
Table 6.2. Foam Cutting and Wet Wrapping Tools. 
 73 
 
 
The tools used for Foam Cutting and Wet Wrapping are listed in Table 6.2. 
6.1.2 Labor and Task 
The division of labor did not occur so much in phases; rather, the effort was divided 
into three main sections and one post-processing section: 1) CNC Wing Cutting, 2) 
structure integration, 3) wet lay-up and pre-assembly.  It is just as possible to have 
one group perform all the work as having three distinct groups performing the 
piecewise tasks in sequence.  The advantage to having multiple groups was that it 
allowed an assembly line type progression to occur.  The wing cutters concentrated 
on making wing cores, while structure integration team outfitted the wing sections 
and, lastly, the wet lay-up crew applied the composite skin.  The post-processing 
actions required just one person and mostly entailed trimming and hinge installation 
duties. 
  
The labor and task descriptions are provided in Tables 6.3.    
 
Labor Task Description 
Group 1 - Wing Cutters CNC hot wire operators (2 to 3) 
Group 2 - Structures  Install internal structure (2 to 3) 
Group 3 - Wet Lay-up Wet wrap and vacuum bagging (3 to 4) 
Post Processing Trimming and Hinging 
Wing Making Groupings and Tasks. 
 
The Advice Section of this Chapter contains further information on Labor and Task. 
6.1.3 Introduction and Foam Blank Preparation 
The construction methods developed in this Chapter involve CNC foam cutting, wet 
lay-up and vacuum bagging processes.  The composite-skinned wings have a foam 
core with embedded carbon tubing for structural support and carry-thru.  The foam 
sheets were cut to length according to the various parts (wings) being manufactured.  
To minimize waste, the cuts were arranged to maximize the number of blanks created 
per sheet.  For example, from a given sheet, one outboard wing, two horizontal 
stabilizers and one wing center-section were generated.   
 
The wing section cut from the foam blank was called the foam wing-core.  The 
remaining foam was called the foam ?beds?.  The wing section, wether in bare foam 
or wrapped in composites, was stored inside the beds at all times.  Prior to wet lay-up, 
the beds had a role just as important as the foam wing-core themselves (as explained 
later).  It was important to oversize the blanks chord-wise.  In another words, the span 
dimension of a blank matched the span of the wing-core while the chord of the blank 
was oversize by a few inches.  The oversize in the chord dimension was necessary so 
the wing beds did not collapse (buckle) during vacuum bagging, thus, potentially 
crushing the wing and causing defects on its surface.     
 74 
 
6.2 CNC Foam Cutter 
6.2.1 Foam Cutter 
The CNC foam cutter required some practice and considerable attention to operate 
successfully.  As with any machining tool, undesirable events could occur at any time 
and needed to be stopped immediately.  The risks, however, are far less than the ones 
involving milling, drilling or turning.  As a cautionary measure, every attempt was 
made to remain clear of the hot wire during operation.  The hot-wire, if set on high 
heat, can burn skin upon contact. 
 
 Figure 6.1. CNC Foam Cutter. 
 
The machine consists of two towers with independent X and Y translation capability 
Figure (6.1).  The towers are mounted on two rails, which allow the adjustment of 
separation between the towers.  It is important to have the ability to move the towers 
closer to each other since a (needlessly) long hot wire will arch during cutting.  In 
most cases, the distance between the towers was adjusted such that the foam blank 
barely fit between the moving carriages.  The issues of wire arching/lagging are 
discussed further in the Advice Section of this Chapter.  
      
The towers operate by means of a stepper motor installed in each rail.  The corkscrew 
driven by the independent stepper motors translate the carriage either in the plane of 
 75 
 
the table or perpendicular to the table. The stepper motors are connected to a control 
box that is interfaced with a laptop.  The CNC Foam Cutter operator uses software on 
the laptop to setup and execute the cut (Figure 6.2).  The wing that was cut in Figure 
6.2 was tapered in both leading and trailing edges, had spar hole cuts (main spar and 
rear alignment spar) and weight saving cuts both ahead and behind the spar.   
 
 Figure 6.2. Laptop Screen ? foam cutter software running. 
 
Note: the example in Figure 6.2 is not a wing cut used in the project.  The example in 
Figure 6.2 was chosen because it demonstrates the machine?s ability to create wing 
taper and internal features. 
6.2.2 Wing Cutting 
The CNC cutter was programmed to cut the wings according to the sizing parameters 
listed in Table 2.1 and specified in Figure 2.1.  The wing, in its entirety, consisted of 
the two (left and right) outboard sections and one center-section.  Internal cuts were 
made in both outboard and center-sections as illustrated in Figures 2.3 and 2.4. The 
center-wing also featured cuts necessary to interface the center-wing/boom system to 
the fuselage.  The horizontal stabilizer was a simple NACA0012 wing cut without 
any internal cuts.   
 
Actual wing cutting required practice and attention.  Although details about the 
operation of the CNC Foam Cutter are not explicitly developed, the preparation for 
 76 
 
the center wing-section is discussed below since it contained the most number of 
embedded structural and interfacing parts.  The procedures described for the center 
wing section apply for both outboard wings and horizontal stabilizer.   
 
The desired, final product bare foam, center wing-section is shown in Figure 6.3.  
(This is how it was done?) 
 
 Figure 6.3. Bare Foam Wing Center Section ? all cuts completed. 
 
The foam blank was set on the cutting workspace and the wing center-section 
program was initialized.  The internal cuts were made first, then the airfoil/wing cut 
and, finally, the structural/interfacing cuts.   
 
The process began by making a (strategically located) ?in. diameter hole in the foam 
blank.  The wire was fed through this hole/tunnel and connected to either column.   
The wire was switched on and the cut was executed.  At the end of the internal cut, 
the wire was snipped off in either end of the blank foam and the towers were moved 
to a clear location on the table.   
 
A new wire was installed and the towers were moved to the designated airfoil cut 
start point.  The wire was switched on and the outer wing cut was made.  At this 
point, the foam blank is in three pieces: a wing core with internal cuts, an upper wing 
bed and a lower wing bed.  Figure 6.3 shows the center section sitting in its lower 
bed.  At this stage, the cuts that are yet to be made are the fuselage interface/mounting 
thru-holes and the spar end caps notch.  The wing was carefully measured and 
marked with the locations of the cuts (Figure 6.4).   
 
 77 
 
 Figure 6.4. Wing Center-Section ? interface mounting thru-holes marked. 
 
It was important to measure and double measure before doing anything.  In this 
example, a mistake was found.  The left rear mounting thru-hole was misplaced.  The 
error was corrected and new markings were made on the wing as noticed in Figures 
6.5 & 6.6.  The wrong marking was labeled ?no good?.   
 
Design side note.  The fuselage longitudinal axis ran parallel to the airfoil chord 
line.  The designed angle of incidence was implemented during vehicle design 
and milling of the boom plugs.  The airfoil-shaped slots in the booms were 
machined (milled) to the required angle of incidence.  However, since this airfoil 
is highly cambered, it generates substantial lift even at zero angle of attack.  
Therefore, the angle of incidence required for level flight at cruise speed is very 
shallow.  As a result, the wing is almost unnoticeably pitched up when the aircraft 
is viewed at a distance.  The wing cutout in the bottom of the fuselage was made 
such that it naturally sat on the center section-wing with both fuselage 
longitudinal axis and chord line parallel.  This facilitated the assembly of the 
fuselage and center wing-section mounting as explained in Chapter 7. 
 78 
 
 Figure 6.5. Center Wing-Section ? mounting thru-hole being cut. 
 
 Figure 6.6. Center Wing-Section ? mounting thru-hole done, detail view top right. 
 79 
 
 Figure 6.7. Center Wing-Section ? front support-pin cutout, detail view top right. 
 
To make the cuts, the center wing-section was oriented such that the hot wire ran 
perpendicular to the chord line of the airfoil (Figures 6.5-7).  The wire was fed thru a 
hole made with a pin and the cut was executed (Figure 6.6).  The front alignment pin 
for the center wing-section was cut next (Figure 6.7).  At this point, the preparation of 
the bare foam center section was almost complete.  The wing was re-oriented so the 
notches for the spar end caps could be made (Figure 6.8).   
 
Caution. Notice in Figure 6.8 the jagged trailing edge.  A good wing will have a 
perfect trailing edge (TE) fresh from the CNC machine.  The defects found in the 
trailing edge were caused due to careless handling of the foam wing.  The trailing 
edge is especially fragile since it is so thin at the TE and prone to tearing.  
Typically, this damage is not a problem since the composite cloth will cover the 
imperfections.  Furthermore, resin will fill the spots where the foam is missing.  
However, if a missing piece of the trailing edge is large enough, the reinforcement 
cloth will not hide the blemish and, probably, will be weak at that location.  
Damage on the leading edge (LE) tends to be more serious.  The reinforcement 
cloth will form to the contour of the wing and will not hide foam core defects.  
The leading edge may also be weakened because the reinforcement cloth may be 
prone to buckling in the locations where the foam core is already indented.  The 
bottom line is, do not damage the foam core.    
 80 
 
 Figure 6.8. Center Wing-Section ? notch being cut for spar end caps, detail view top 
right. 
 
Once the wing had all the necessary cuts (Figure 6.3), it was ready for the integration 
of internal structure. 
 81 
 
6.3 Foam Wing Preparation ? Structure Integration 
6.3.1 Internal Structure 
The parts used for the structure of either outboard or center sections were sized and 
cut separately.  Both outboard and center sections had a fixed spar sleeve with end 
caps.  The carbon tubing, used for the spar sleeves, was cut to size and lightly sanded 
at the ends.  The reason for sanding the ends of the spar tube is that resin adheres 
better to the roughened surface.  The end caps were custom cut with a thru-hole for 
the fixed carbon sleeve.  After, the spar sleeve was slid in place, the tips of the sleeve 
(on the outside) were wetted with catalyzed resin and the end caps were placed.   
 
Caution.  When gluing the spar sleeve and end caps it is extremely important to 
prevent excess resin from dribbling into the spar.  Resin-drops in the sleeve will 
prevent the (long) carry-thru internal spar from sliding through. To help prevent 
resin contamination inside of the sleeve, a couple of sheets of paper towels were 
rolled into a plug and jammed in each tip.  Once the sleeves and end caps were 
placed and glued, the paper towel plugs were removed and a flashlight was used 
to inspect the inside of the spar sleeve.  If any resin seeped into the spar sleeve, it 
was cleared away with either Denatured Alcohol or Acetone by squirting it into 
the tube.  Acetone rinses away resin very effectively, but it also damages the 
foam; therefore, denatured alcohol was the preferred choice.     
 
The rear support inserts in the foam wing-core were custom cut and fitted.  The holes 
for the fuselage mounting bolts were centered on the block, which was already cut 
and test fitted in the foam wing-core.  If the block fitted snuggly and the bolthole was 
correctly made, the blocks were sanded to match the curvature of the airfoil/wing.  
Likewise, the front pin support was also sized, cut, drilled and shaped to match the 
curvature of the wing, in this case the leading edge.  The drill hole made in the front 
support block was intended for the alignment pin/dowel.  These ?blocks? were cut 
from a medium density fiber (MDF) board.  The MDF was used because it sanded 
easily with a power sander. (Before reading further, refer back to Figures 2.3. and 2.4 
and take note of what has been accomplished so far.)  
6.3.2 Foam Joining 
The center wing-section did not require foam-to-foam joining because it was 
designed as one piece.  However, the outboard wing was cut in parts; therefore, 
needed gluing/joining together.  In addition, the foam beds of the outboard wings 
were also glued to form one continuous upper and lower bed.  To attach the pieces 
together, acetone-free contact glue was sprayed to the ends of the wing pieces then 
mated together.  The result was a continuous outboard foam wing-core with matching 
upper and lower beds (Figure 6.10 ? note the wing beds on the table in the back).   
 82 
 
6.4 Wet Lay-up 
6.4.1 Process Description and Introduction 
The wet lay-up (hand lay-up) process involved the manual distribution of catalyzed 
resin onto a cloth.  Typically, in molding processes using wet lay-up, the catalyzed 
resin is poured into the mold then the reinforcement cloth is introduced to the wet 
surface.  A roller is used to aid in the spreading and impregnation of the resin onto the 
reinforcement.  Once the user has laid and rolled all of the reinforcement as desired, 
the part is allowed to cure.   
 
The curing process can be open-molded or vacuum-sealed with bagging material or 
other kinds of semi-rigid to rigid coverings.  The application of the vacuum bag 
promotes good consolidation between layers and aids in the removal of trapped air.  
The mold making technique described in Chapter 4 is in a sense a modified hand lay-
up process.  The random directional matting (used in Chapter 4) was both manually 
placed and rolled; however, the catalyzed resin was applied via spray gun and 
allowed an open cure. 
 
For wing making, the wet lay-up process is again modified to fit the application.  The 
catalyzed resin was introduced to the reinforcement and evenly dispersed.  The 
wetted cloth was then wrapped around the wing and placed inside the foam beds.  The 
wrapped wing while packaged inside its beds was subjected to pressure in one of two 
ways: 1) 20lbs weights were stacked on top of the upper bed or the entire bed and 
wrapped wing assembly was placed inside a vacuum bag.  
6.4.2 Wrapping the Wing 
The wing center-section and the horizontal stabilizer were wrapped in a single ply of 
carbon cloth.  The wet lay-up process for these two sections was relatively easy to 
perform since these had constant chord (non-tapered).  For the most part, symmetric 
three-dimensional shapes with constant dimensions are easier to wrap because these 
shapes can be flat-wrapped.  However, since the outboard wing had a trailing edge 
taper, it was more complex to wrap-up.  The taper (span-wise change in chord) 
caused the wing to change dimension not only in the chord length, but also in airfoil 
thickness.   
  
The CNC wing cutting procedures discussed in Section 6.2 focused on the central 
wing section since it contained the most number of features.  For this section, the 
outboard wing-section is the subject of interest since it required the most attention.  
The internal constructs and external application of the reinforcement is illustrated in 
Figure 6.9.  The outboard wing-section had reinforcement ?cuffs? in three locations: 
the wingtip, the wing root and the foam joint where the spar sleeve ended (Figure 
6.9). 
 
 83 
 
 Figure 6.9. Outboard Wing Construction ? explode view (see Figure 2.5). 
 
 Figure 6.10. Outboard Wing-Section ? wet lay-up preparation. 
 
The outboard wing-section was cut, assembled and ready to receive the reinforcement 
skin.  Once the carbon cloth was sized and cut, the work area was prepped for the wet 
 84 
 
lay-up process.  An oversized sheet of release film, taped by the corners, was placed 
on the cutting table.  The dry carbon reinforcement was carefully laid on of the 
release film.  A silver marker was used to mark important locations on the cloth, like 
the location of the foam core wing and locations of the carbon strips (cuffs).  The 
additional carbon reinforcement strips were placed on top of the ?big? carbon sheet 
after it was impregnated with catalyzed resin. 
 
The actual wings and surfaces of the aircraft used carbon cloth; however, the wet lay-
up example shown below uses E-Glass with a resin dyed red.  The author has 
purposely done this to facilitate the reader?s ability to see the resin since the contrast 
is red on white.  The release film used is green and is contrasted by the black cutting 
mat.   
 
The dyed catalyzed resin was first poured on the reinforcement cloth then distributed 
with a spreader (Figures 6.11).  The target fiber volume fraction was about 35%.  In 
another words, the overall composite skin, when cured, had a weight distribution of 
35% cloth to 65% resin.  Next, the reinforcement strip was placed then wetted out 
(Figure 6.12). 
 
 
Figure 6.11. Outboard Wing-Section ? wet lay-up, detail view top right. 
 85 
 
 Figure 6.12. Outboard Wing-Section ? reinforcement strip wet lay-up. 
 
 
Figure 6.13. Outboard Wing-Section ? wet out complete, foam core ready to wrap. 
 
Once the hand lay-up was complete, the wing foam-core was ready to be wrapped.  
The leading edge of the wing was placed on the center of the impregnated 
reinforcement cloth.  To facilitate the wrapping process, the bottom bed was placed 
under the wing (under the release film).  The cloth was worked around the leading 
edge and wrapped over the top surface of the wing towards the trailing edge.  
Wrinkles created by the manipulation of the cloth during wrapping were smoothed 
out by hand with a soft cloth.  
  
 86 
 
To apply pressure, the packaged wing was either placed in a vacuum bag or pressed 
down with ten 25lbs weights.  When using weights, a scrap Plexiglas sheet was 
placed on top of the encased wing (Figure 6.14).  The purpose of the Plexiglas sheet 
was to prevent the weights from crushing the foam bed, thus, applying uneven 
pressure.  The 25lbs weights proved to produce parts of equal quality as those cured 
with vacuum bagging.  The process was made simple and cheaper by using the 
stackable weights. 
 
 Figure 6.14. Outboard Wing-Section ? wrapped and ready to receive weights. 
6.4.3 Cure and Clean-Up 
All aircraft parts manufactured with wet lay-up were allowed to cure overnight or for 
at least 8 hours.  No special clean up was required for this process; dry materials 
(vacuum bagging materials, release film, etc) were discarded after use along with 
used gloves, stir sticks, mixing cups, etc.   
6.5 Moving Surfaces and Horizontal Stabilizer 
6.5.1 Ailerons and Elevator 
The CNC foam cutting and wet lay-up methods used for the moving surfaces 
(ailerons and elevator) and horizontal stabilizer are similar to those used for the 
wings.  The foam was CNC cut and wrapped with carbon reinforcement similarly to 
the wings; however, the preparation involved varied drastically.   
 
The elevator was a non-tapered cut easily handled by the CNC cutter.  The elevator 
was then repositioned so the balsa-block hinge support cutouts could be made.  The 
balsa blocks were custom cut and sanded to the shape of the elevator (Figures 6.15 & 
6.16).   
 87 
 
 Figure 6.15. Elevator ? hinge support cutout, detail view bottom left. 
 
 Figure 6.16. Untrimmed Elevator Wrapped in Carbon ? detail view top left. 
 88 
 
The location of the balsa blocks was recorded since, when wrapped in carbon, the 
blocks could not be seen.  The ailerons were manufactured in similar fashion to the 
elevator.  However, due to the taper of the wing, the aileron cut was more 
complicated and required much more attention than the elevator.  
6.5.2 Horizontal Stabilizer 
The horizontal stabilizer was a solid foam core wing wrapped in carbon 
reinforcement.  The wet lay-up for this part was simple since the airfoil was both 
symmetric and non-tapered.  The installation of the elevator to the horizontal 
stabilizer was the same as the installation of the ailerons to the outboard wing-
sections.  This process is discussed in the Section 6.6.2. 
6.6 Pre-Assembly 
6.6.1 Trim Wing to Size 
The trimming process did not vary from part to part.  All wings and moving surfaces 
were trimmed and sanded in similar fashion.  The first step was to reduce the excess 
reinforcement at the tips of the wing.  Most of the excess material at the tips was 
removed with the Dremel then hand-sanded the remaining materials with a straight 
block down to the foam core.   
 
In some cases, depending on how much excess material was present, the operator 
elected to make two passes with the Dremel.  In Figure 6.17, the operator was making 
the first pass; then, a second pass was made to trim the excess material closer to the 
foam.  Once enough material was removed, the manual dry sanding started (Figure 
6.18). 
 
 Figure 6.17. Outboard Wing-Section ? trimming the ends. 
 89 
 
 Figure 6.18. Outboard Wing-Section ? sanding the ends. 
 
The trailing edge was trimmed based on the desired tip and root chord lengths; it was 
hand cut with a sharp razor along a steel/aluminum straight edge.  The straight edge 
was secured to the wing by means of 3 small double-sided tape strips. 
 
 Figure 6.19. Outboard Wing-Section ? trimming the trailing edge. 
 
 90 
 
Hints: 1) Do not use too much double sided tape, it makes the removal of the 
straight edge difficult and can result in damage to the composite skin or cause de-
lamination of the skin from the foam core.   
2) Place the straight edge on the wing surface and not on the part being 
trimmed, this is important because if the razor slips, it scratches/cuts the piece 
being cut off. 
 
Cutting the cloth by hand required several passes with the knife (Figure 6.19).  
Moreover, several replacement blades were used since the tip dulled quickly.  The 
razor technique was preferred for cutting the trailing edge because it produced the 
sharpest TE quality possible; sharp enough to slice up unsuspecting fingers or hands 
(Figure 6.20). 
   
 Figure 6.20. Outboard Wing-Section ? trimming complete, next wing on queue. 
 
Construction side note:  The foam beds have been invaluable during the entire 
wing lay-up process.  The surface quality of the foam beds, or lack thereof, is 
reproduced onto the wing surface during curing.  Hence, the preservation of the 
foam beds is just as important as the preservation of the bare-foam wing-core 
themselves.  However, once the foam core is wrapped and cured, the wing beds 
serve a different purpose.  The beds do not have to remain unharmed.  As seen in 
Figures 6.17 ? 20, the foam bed is used as support for the wing.  During the 
trimming of the trailing edge, the blade digs into the wing bed once it cuts through 
the composite laminate, thus, the wing beds will endure some damaged.  From 
 91 
 
this point forward, the purpose of the beds is to provide a work support for the 
wings and to encase the wings when stored or transported.  
6.6.2 Moving Surface Hinging and Cutout 
The moving surfaces were first outfitted with the hinges.  The hinges were installed in 
the center of the balsa inserts installed prior to the wet lay-up.  The hinges were 
carefully installed so the travel of the hinge arms were oriented in the correct 
direction.  Measurements were used during the installation of the hinges, but the best 
tool for the job was the ?eyeball?  (Figure 6.21).  Once properly aligned, the hinges 
were glued in place. 
 
 Figure 6.21. Elevator Hinge Installation. 
 
The wings and stabilizer were trimmed so the moving surfaces could be installed.  
Similarly to the trimming of the trailing edge, the moving surface cut-out was 
performed with a bladed instrument and straight edge.  The exposed foam of the wing 
was sanded with a round block whose curvature matched the curvature of the moving 
surface (Figure 6.24).  This allowed the round leading edge of the aileron/elevator to 
fit perfectly with the wing/stabilizer.     
 
The balsa support blocks for the wings/stabilizer were sized, cut and installed to the 
exposed hinge arm.  At this point, the blocks were not glued to the hinge arm (Figure 
6.22).  The wing/elevator was prepped to receive the balsa support blocks.  Once the 
entire process was done and the blocks fit snuggly, the balsa hinge blocks were glued, 
in the proper orientation, to the exposed hinge arm  (Figure 6.23). 
 92 
 
 
 Figure 6.22. Elevator Hinged and Balsa Support Blocks Prepped. 
 
 Figure 6.23. Ailerons Hinged with Balsa Blocks Glued. 
 
The moving surfaces were also outfitted with the servo control horn.  A horn was 
installed to each moving surface at its center (Figures 6.22 & 6.23).  The servo 
control horn is attached to the servo push rod as explained in Chapter 7.  Finally, the 
moving surface was test fitted to the wing/stabilizer (Figure 6.25).  
 93 
 
 Figure 6.24. Aileron Hinge Detail. 
 
 Figure 6.25. Aileron Hinging and Preparation Complete. 
6.6.3 Final Product 
The complete wing sections and horizontal stabilizer are shown in Figures 6.26 - 28 
(with moving surfaces in place).  
 94 
 
 
 Figure 6.26. Center Wing-Section Complete. 
 
 Figure 6.27. Entire Wing Complete ? outboard and center sections. 
 95 
 
 Figure 6.28. Horizontal Stabilizer and Elevator Complete. 
 
Construction side notes:   
1) The elevator in Figure 6.28 is longer than the horizontal stabilizer with 
good reason.  For that matter, the horizontal stabilizer is also span-wise oversized.  
The reason for over sizing the stabilizer is because it will be inserted into the 
boom, as discussed in Chapter 7.  Similarly, the elevator is oversized because the 
exact length to cut it depends on the rudders.  The spacing between the booms 
provides the upper limit on the length of the elevator; however, the amount of 
rudder ?throw? (i.e. rudder deflection) further restricts the length of the elevator 
because of the twin boom setup.  The result is that both rudder and elevator must 
operate freely without any risk of binding with each other.  For now, the elevator 
is left oversized.  
2) The moving surfaces have the hinges glued in, but are not permanently 
installed.  Although assembled, the moving surfaces are not glued into place.  To 
eliminate any chances of damaging the moving surfaces and their hinges, the final 
gluing of the moving surfaces was done after the entire aircraft was wired, 
integrated and fully assembled.  At this point, both ailerons and elevator are 
removable (Figures 6.27 & 6.28).   
 96 
 
6.7 Advice 
6.7.1 More on Labor and Tasks 
The most important point about division of labor at this stage (Chapters 6 & 7) of the 
construction is to identify individual talents/skills.  For example, group members with 
machining skills are encouraged to perform tasks that involve machining.   From a 
teaching standpoint, it may appear attractive to divide the work evenly so everybody 
learns a bit about everything.  However, in practice this is hardly the most optimized 
group setup.  Individual talents can vary so much that it is often safest to restrict some 
specialized activities to individuals that are either comfortable with or have prior 
experience in performing the specific task. 
 
In cases where members have minimal prior experience, the entire process has to be 
learned from scratch.  For example, cutting foam wings with the CNC cutter is often a 
novelty.  However, once a particular subset of individuals is expert at CNC cutting, 
these workers should be the only ones to perform this duty.  If the need arises for a 
second group to cut wings, at least one member in the new group should be borrowed 
from the original group.  This is important because it ensures repeatability in the 
process while minimizing common errors.   
 
As the project nears the assembly phase, mistakes can cause substantial setbacks 
because parts of the airplane are near completion.  For example, alignment problems 
or erroneous location of a spar hole can completely scrap an entire subsection of the 
aircraft.  For this reason, the workforce, when subdivided, needs to stay consistent.  
Individuals who have specialized in a particular function must execute it from start to 
finish.    
 
In summary, known individual strengths/skills must be exploited.  It is too risky to 
allow inexperienced hands to perform critical tasks this late in the construction, 
especially when knowledgeable/experienced individuals are on hand.  Once team 
members learn and become comfortable with a particular duty, these individuals or 
group focus on that portion of the build exclusively.   
6.7.2 CNC Foam Cutter ? Hints for Successful Cutting 
The ideal conditions for cutting foam are under low heat and low speed.   This 
guarantees good resolution during cutting, repeatability, and prevents the wire from 
lagging in the center.   The wire, while lagging during cutting, will round sharp 
corners and cause chord variance along the span of the wing.  Lag in the wire is not 
corrected by increasing the heat; rather, is corrected by decreasing the cut speed 
setting on the CNC machine.  Just be patient.     
 97 
 
6.7.3 The Hot Wire Bow 
The 8?x 2?x 4? foam boards were marked and cut into foam blanks intended for wing 
cutting.  This procedure was primarily performed with the Hot Wire Bow Cutter 
(Figure 6.29).  Making these simple gravity cuts with the bow was advantageous 
since it alleviated the work queue for the CNC Foam Cutter. 
 
 Figure 6.29. Wing Blank Being Cut with the Hot Wire Bow. 
6.7.4 How to Ruin Your Work in Seconds 
There is no other chemical more harmful to foam than Acetone.  It is extremely 
important to keep any Acetone or Acetone-Based fluids away from the foam blanks, 
foam beds, foam wing-cores, etc.  The Acetone will dissolve the foam in seconds.  If 
there is any doubt about chemical compatibility with the foam, just read the label.  
However, the best test is to try some of it on a scrap foam piece.  If the chemical 
reacts, lock it up, warn everyone about it, and keep it away!     
6.7.5 Summary ? Wet Lay-up Pros & Cons  
In a general sense, wet lay-up processes have similar advantages and disadvantages 
no matter the variation.  The pros and cons associated with the customized wet lay-up 
(hand lay-up) process for wing making are summarized in Table 6.4.   
 
This process, when used in wing making, can be very effective.  The quality of the 
part can vary from technician to technician; nevertheless, the method can generate 
high quality high output when performed carefully and methodically.  As suggested 
 98 
 
before, individuals (team members) that demonstrate proficiency in particular areas 
are encouraged to work on those same tasks during the build.  By having the same 
group constructing the wings from start to finish, from a quality control aspect, the 
process repeatability and consistency in overall finish of the wings/surfaces are 
achieved. 
 
Low Capital Investment (compared to other methods) 
Simple and Straightforward Method  
Good Turn Around with Experienced Technicians 
Negligible Molding Costs since Foam Beds are Used 
Foam Beds Perfectly Match the Wings  
PR
OS
 
Easy to Setup 
Labor Intensive (wrapping can be tricky) 
Human Exposure/Contact to Resin 
Quality Consistency Requires Experienced Technician 
Reproducibility Requires Experienced Technicians C
ON
S 
Easy to Ruin Part 
Table 6.4. Wet Lay-up ? Pros & Cons. 
6.7.6 Storage and Inventory  
The completed wing and horizontal stabilizer sections were always stored inside the 
foam beds.  In turn, these were organized in a shelving system where the work-in-
progress and the finalized work were kept separate (Figure 6.30).  
 
 99 
 
 Figure 6.30. Organized Foam Cores and Finished Wings.  
 100 
 
Chapter 7:  Assembly 
7.1 Preparation 
7.1.1 Tools and Materials 
The materials, used for vehicle assembly, are listed in Table 7.1. 
 
Materials Description 
General Materials Mixing cups, stir sticks, latex gloves, eye protection, masks, etc 
Resin Polyester resin used on carbon 
MEKP Catalyst Hardener used for polyester based primers, gel coats and resins 
Packing Tape Tape used to secure release film 
Sanding Paper Sanding paper sheets grits 220 to 400, sand paper roll 8in. wide 
Syringe Syringe used to inject resin 
Solvent Denatured Alcohol 
Table 7.1. Assembly Materials. 
 
Tools Description 
Cutting Instruments Shears and/or sharp bladed tools (X-Acto Knives) 
General Tools Pliers, screw drives, drill bit set, etc 
Measuring Equipment Measuring tape, metal yardstick (or longer), T-square, digital scale 
Sanding Blocks Support for sanding paper 
Height Gage Precision table height gage with inscriber tip 
Wood & Metal Shop Shop drill press, band saw and power sander 
Dremel Handheld rotary cutting tool with various cutting bits 
Rotozip Table mounted rotary cutting tool 
Shop Vac Vacuum for dust clean-up 
Level Cutting Surface Marble/Granite cutting top 
Other Tools Wood clamps/grips, misc. lead weights 
Table 7.2. Assembly Tools. 
 7.1.2 Labor and Task 
The division of tasks for the assembly phase was grouped based on individual skills 
and abilities.  The assembly itself was divided into two parts.  The difference between 
the two parts was the level in which assembly and integration was occurring.  Part I 
focused on the finalization of individual pieces of the aircraft.  Part II was a higher 
level of assembly where the pieces came together to make the whole.  The Advice 
Section of this Chapter contains further information on Labor and Task. 
7.1.3 Assembly Flowchart 
See Figures 1.4 (Part I) and 1.5 (Part II) for vehicle assembly flowchart. 
 101 
 
7.2 Machining and Joining Parts 
7.2.1 Machining 
Two pieces required special machining for this project.  These pieces were used in the 
main landing gear installation (Figure 7.1).  One piece was permanently installed in 
the boom as the interface for the removable gear strut assembly.  The second piece 
was a transition part that allowed the strut extension to thread into the fixed internal 
gear support. 
 
 Figure 7.1. Main Landing Gear Assembly (see Figure 2.6). 
7.2.2 Trimming, Trial Fitting and Cut Outs 
All of the molded parts required trimming.  The first step to trimming was the 
removal of the planar surface of the flange.  This was done using a band saw.  The 
machinist cut away much of the flange while leaving enough material for the entire 
piece to sit flat on the granite-cutting top.  The pieces of the flange cut by the band 
saw were kept and used as internal structure for the fuselage. 
 102 
 
 
 Figure 7.2. Trimming Tools and Parts. 
 
 
Figure 7.3. Boom Trimming in Progress ? inscribed line detail view, top right. 
 
 103 
 
For the pieces to mate to one another, the remainder of the flange had to be removed.  
This was done with a rotary tool using a diamond blade disk.  The cutter was fixated 
on the granite such that the blade was oriented in a parallel plane ? inch above the 
granite surface.  Prior to cutting, a line was inscribed on the part with a height gage 
that had an inscriber tip.  The line served as a guide during trimming (Figures 7.2 & 
7.3).   
 
The trimming procedure for the flange was the same for the booms and the fuselage.  
This trimming process was found to be very efficient and repeatable.   The end result, 
when trimmed correctly, required little post processing (fine sanding) and mated right 
away with its the corresponding half. 
 
In cases where the trimmed parts needed further leveling to mate properly, the parts 
were sanded on a continuous sanding paper strip stapled to a butcher-top table (Figure 
7.4).  The parts were sanded and mated to perfection. 
 
 Figure 7.4. Fuselage Ready for Hatch Cutout ? fine sanding, detail view top left. 
 
After trimming the various parts and mating them together, the pieces were ready to 
receive their respective cutouts.  In the case of the fuselage, the hatch cut out had to 
be made.  This cutout was performed by hand using the lines imprinted on the 
fuselage during VARTM.  The hatch cutout guideline can be seen in Figure 7.4. 
 104 
 
 Figure 7.5. Rudder Cutout ? free hand cutting, detail view bottom right. 
 
 Figure 7.6. Rudder Cutout Complete for all Boom Halves. 
 
 105 
 
The rudder was cut off from the vertical stabilizer by hand with the Dremel (Figure 
7.5).  The rudder was sized and marked with a straight edge and a pointer.  The 
rudder was then cut free hand following the line just as previously done with the 
fuselage hatch.  The complete set of rubber halves, sufficient for two airframes, is 
shown in Figure 7.6. 
7.2.3 Joining 
The joining processes for the fuselage and hatch were similar.  The booms, however, 
used a slightly different technique. 
7.2.3.1 Joining the Fuselage 
The sanded and mated fuselage halves were trial fitted very carefully to ensure a flush 
seam.  The fuselage halves were held together by packing tape on the outside.  The 
tape was carefully placed so no air bubbles or wrinkles were present at the seam line 
(Figure 7.7). 
 
 
Figure 7.7. Fuselage Joining ? taped seam, detail view top right. 
 
The internal surface of the taped fuselage halves was sanded along seam (Figure 7.7).  
It was important to roughen up the inside surface so the joining reinforcement strip 
could adhere better.  A carbon strip was wet laid inside the fuselage along the entire 
seam.  An oversized peel ply sheet was placed on top of the reinforcement strip to 
remove excess resin.  After full cure, the peel ply was removed and the fuselage was 
 106 
 
in one piece (Figure 7.8).  The clear packing tape was also removed from the outside 
perimeter, revealing a flush and gapless seam that required no additional finish.   
 
 Figure 7.8. Fuselage Joined. 
7.2.3.2 Joining the Hatch 
The fuselage hatch was joined similarly to the fuselage.  The hatch halves were mated 
and joined together via the wet lay up of a reinforcement strip with peel ply.  The 
hatch halves were taped together along the outside surface, sanded in the inside 
surface and, lastly, joined with a reinforcement strip (Figure 7.18). 
7.2.3.3 Joining the Booms 
The boom halves were joined by means of injecting catalyzed resin into the mated 
halves.  The initial preparation involved was similar to that of the fuselage and hatch.  
The boom halves were matched and mated with packing tape along the seam.  As 
previously done, the packing tape held the two halves together and was carefully 
placed such no bubbles or wrinkles were present on the seam itself.  
 
Once prepped, a small batch of resin was catalyzed and poured into a large diameter 
syringe.  The catalyzed resin was injected into the closed part and allowed to flow 
internally along the seam.  The excess resin drained out the front of the boom 
(Figures 7.12 ? 14).  However, the preparation for this particular piece required more 
than fine sanding and mating with packing tape.  The main landing gear interface 
needed to be housed in the landing gear fairing as well as the horizontal stabilizer 
cutout.  The necessary preparations for the booms are discussed in the next subsection 
(7.2.4).      
 
 107 
 
7.2.4 Before Joining ? Horizontal Stabilizer Cutout & Main Landing Gear 
The cutout for the horizontal stabilizer was made on the designated internal boom 
halves.  The location of the horizontal stabilizer was measured and marked 
accordingly.  A bare foam stabilizer was used as a template to aid in the location of 
the stabilizer.  The foam stabilizer was shaped at the ends to the contour of the 
boom/vertical stabilizer.  By doing this, the foam horizontal stabilizer sat 
perpendicular to the surface.   
 
A silver permanent ink marker was used to mark the airfoil shape on the 
boom/vertical.  The horizontal stabilizer centerline was also drawn and measured to 
ensure proper angle of attack.  The cutout was made inside the marked profile with a 
Dremel as to undersize the slot/fit (Figure 7.9).  The cutout was carefully sanded to 
snuggly fit the composite wrapped horizontal stabilizer (Figure 7.9 ? top left and 
bottom right detail views).  See Figure 2.7 for technical drawing of elevator assembly. 
 
 Figure 7.9. Horizontal Stabilizer Cutout ? top left and bottom right detail views. 
 
Construction side note:  It was very important to note that the horizontal stabilizer 
cutout established a left and right designation.  The cutouts were made on the 
internal facing boom halves of the twin boom configuration.  Prior to cutting, any 
boom half pair could have been a left or right boom.  It was very important to 
keep track of the work and not end up with two left or two right booms.   
 
The next step was to install the landing gear assembly as illustrated in Figures 7.1 and 
shown Figure 7.10.  To do this, the gear interface was embedded before the boom 
halves were permanently joined.   
 108 
 
 
 Figure 7.10. Landing Gear Interface ? detail view bottom right. 
 
The first step was to trial fit the landing gear interface to the boom halves.  In most 
cases, this required ?dremeling? the inside of the gear-strut fairing with a small 
sanding tube attachment.  Once the fairing was properly sanded, the interface sat in 
place flush to the inner surface.  Measurements were taken to ensure proper location 
of the interface and markings were made on the boom with a silver sharpie.  The gear 
strut and wheel were screwed on so a visual inspection of the orientation and 
alignment of the overall assembly could be made.  The strut was removed and top and 
bottom holes in the interface were covered with packing tape.  The holes for the 
fixating screws of the interface were made with a hand drill.  The hole pattern was 
then transferred to the opposite boom half and the boom was taped together.  The 
installation of the main gear strut following the above instructions is shown in Figure 
7.11. 
 
Construction side note:  The top and bottom-treaded hole in the interface was 
taped shut so the injected resin did not plug the hole during joining.  This was 
very important to remember because hardened resin could plug the hole or fill the 
internal thread, thus, making it very difficult to attach the rest of the landing gear 
strut.  The border of the boom halves was roughened with sand paper in the inside 
as well.  For the same reason mentioned before, the roughened surface provided 
the resin a ?grip? on the surface.     
 109 
 
 Figure 7.11. Installation of Main Strut Interface (from top left to bottom right). 
 110 
 
 Figure 7.12. Boom Joining ? resin being injected (close-up). 
 
 
Figure 7.13. Boom Joining ? resin being injected. 
 
 111 
 
Once the horizontal stabilizer profile cutout was made and the main gear strut 
interface installed, the boom was ready for joining.  The boom halves were mated 
(taped) together and the screws of the interface were tightened with a nylon lock nut.   
 
At this point, the boom was joined with resin.  The catalyzed resin was injected into 
the horizontal stabilizer cutout (Figure 7.12).  The boom was held at a 45-degree 
angle over a wastebasket.  The excess resin drained out of the main gear fairing where 
the gear interface was installed (Figure 7.13).  The resin wet out could be observed 
through the clear packing tape along the seam as it ran down the boom.  Once the 
working time of the resin expired, the booms were allowed to cure in a safe location 
(Figure 7.14).  The resin was known to have reached its gel stage because the 
remainder inside the syringe was both hot and solidified.     
 
 Figure 7.14. Left and Right Boom Pair Injected with Resin. 
 
After full cure, the booms were permanently joined.  The tape was not removed until 
the entire aircraft was fully assembled.  The packing tape was left on until the last 
moment so it could protect the finished surface during further handling and assembly.   
 
 112 
 
7.3 Internal Structure ? Fuselage and Hatch 
7.3.1 Internal Structural Components 
The structural components of both fuselage and hatch are detailed in Figure 2.2. 
7.3.2 Materials used for Internal Structure 
The internal structures of both fuselage and hatch were made from the flange material 
leftover from the fuselage and boom parts.  A template was made and the outline of 
various internal parts was traced on the flange pieces.  The individual structural 
members were cut with a band saw. 
7.3.3 Nose Cutout 
The nose of the fuselage needed to be cut to expose the motor shaft.  Moreover, the 
cutout was made such that there was enough clearance between the spinner back-plate 
and the cowling.  In an effort to simplify the motor installation, the nose cutout also 
included the cooling cutout (Figure 7.15).     
 
 
Figure 7.15. Nose Cutout for Propeller and Cooling. 
 113 
 
7.3.4 Trial Fitting  
The structural pieces were hand sanded to match the internal contour of either 
fuselage or hatch perfectly.  Several markings were made in both structural parts and 
fuselage/hatch.  The markings were used as a visual guide during gluing.  Figure 7.16 
shows the internal structure in place, but not glued.  A string was taped in either end 
at the top of the fuselage to serve as a centerline.  The motor was installed on the 
firewall and placed in the fuselage (Figure 7.16).   
 
 Figure 7.16. Internal Structure Trial fit with Simulated Cargo (see Figure 2.8). 
 
7.3.5 Fuselage and Hatch Internal Structure Installed 
Once properly aligned and marked, the internal structure was glued in place.  This 
was done by wet laying strips of carbon cloth along the structure and the 
fuselage/hatch.  Figure 7.17 shows a hatch being joined while at the same time having 
its internal structure installed.  A completed fuselage is seen in the background of 
Figure 7.17.  The front alignment pins of the hatch were glued in place and the 
matching holes of the pins were drilled in the fuselage firewall. 
 
 114 
 
 Figure 7.17. Hatch and Fuselage with Internal Structure Glued in Place. 
7.3.6 Vertical Stabilizer and Rudder Internal Structure 
The rudder halves were joined in a similarly fashion to the booms.  Catalyzed resin 
was injected into the mated and taped rudder halves.  Balsa inserts were sized and 
shaped to fit in both the vertical stabilizer and the rudder.  The balsa strips ?closed? 
the boom and rudder and provided material for the hinges to be installed in to (Figure 
7.18). 
 
Construction side note:  The booms were covered with packing tape  to protect its 
surface from scratches and/or resin contamination.  When working with resin in 
other portions of the boom, it was very easy to unknowingly contaminate the latex 
gloves and make finger print on objects.  To remove the risk of ruining the glossy 
carbon finish, the entire boom(s) was covered with packing tape (Figure 7.18).  
The balsa strip installed in the vertical stabilizer was glued flush to the end of the 
boom while the balsa strip glued in the rudder had its rounded side exposed. 
 
 115 
 
 Figure 7.18. Vertical Stabilizer and Rudder Hinging Balsa Strip (see Figure 2.6). 
 
7.3.7 Wing and Fuselage Mating 
The fuselage was mounted on the center wing-section via a pin connection in the 
front and two mounting bolts in the rear of the wing (Figure 2.3).  The first step was 
to fabricate the front pin support that was glued inside of the fuselage.  The fuselage-
mounting block, made from MDF Board and shaped to the leading edge curvature, 
was drilled prior to installation.  This hole received the alignment dowel that was 
installed in the leading edge of the wing (Figure 7.19 bottom left detail view). 
 
The rear boltholes had already been made on the support blocks that were embedded 
in the center wing-section.  During wet lay-up, the carbon skin covered the boltholes 
(Figure 2.3).  The boltholes were located and the carbon skin was carefully drilled 
out.  The bolt holes were transferred to the fuselage by means of a pointer bolt fitted 
into the holes (Figure 7.19 top left detail view).  In another words, a bolt with a sharp 
point was placed through the center wing and was used to etch the location of the 
bolts on to the fuselage.  The inscribed bolt locations were checked both visually and 
with measurements (Figure 7.19).  Lastly, the rear bolt pattern was drilled through the 
fuselage, thus, allowing the fuselage and center wing-section to be fastened together 
(Figures 7.20 & 7.21).    
 
 116 
 
 Figure 7.19. Fuselage and Center Wing-Section Mating (see Figure 2.4). 
7.4 Assembly Interim ? Part I End, Part II Begin 
7.4.1 Vehicle Lay Out ? Trial Fit 
Several major components of the aircraft were now manufactured and partially 
assembled.  At this point of the construction, the various parts were laid out side by 
side on the floor.  Doing this not only allowed the visualization of the finished 
product, but also boosted the team?s enthusiasm.  All of the major components 
available were trial fitted and checked against each other (Figures 7.20 & 7.21).   
 117 
 
 
 Figure 7.20. Aircraft Major Components Laid-Out. 
 
 Figure 7.21. Aircraft Major Components Laid-Out.  
 
 
 118 
 
7.5 Wiring and Servo Installation 
7.5.1 Boom Preparation 
The booms had to have the spar and wiring holes drilled prior to wiring.  The load 
transfer spar ran inside the spar sleeves, which were installed in the wing outboard 
sections and the center section during wing fabrication (Figure 7.22).  The load 
transfer spar was a single-piece carbon composite tube and the only structural 
member to cross the boom.  
 
 Figure 7.22. Load Transferring Spar and Assembly. 
 
The two holes were done on a drill press.  A foam wing template was used to locate 
the spar hole (Figure 7.23).  This process was done by hand with a small piece of the 
spar tubing and a drill bit inside the spar (Figures 7.23 & 7.24 detail view).  The drill 
bit was hand turned and etched a small center point on the boom.  Based on this 
marking, the spar hole was drilled with the adequate spade bit on the drill press.  The 
template guaranteed a perfect spar hole alignment with the outboard wings and center 
wing section.  Furthermore, the process was repeated with success with the remaining 
booms. 
 
The wire access hole was drilled with a hole saw bit.  This hole did not require a 
specific/repeatable alignment.  However, the template was used to locate the wire 
 119 
 
access hole in a reasonable ?ball park? (Figure 7.25).  Once the spar and wire access 
holes were complete, the boom was ready for wiring.   
 
 Figure 7.23. Spar Hole Template in Place ? ready to locate spar hole. 
 
 
Figure 7.24. Spar Hole Marked and Drilled ? detail view top right. 
 120 
 
 
 Figure 7.25. Wire Access Hole Marked and Drilled ? detail view bottom left. 
7.5.2 Running String 
A weight, small enough to pass through the small cavities/holes, was attached to the 
end of the string and acted like a plum bob.  The malleable string and weight were 
dropped through the part and rescued (?fished?) out the other end.   The string was 
cut such that at least 12 inches was exposed out either ends.  The loose ends were 
then secured with packing tape. 
 
In the case of the outboard wings, the servo nest had to be made first then connected 
with a wiring ?tunnel?.  Similarly, the horizontal stabilizer also had its servo cavity 
made then its wiring access tunnel burrowed.  The stringing and wiring for the wing 
structures are developed below.    
7.5.3 Installation of Servos 
7.5.3.1 Rudder Servo Installation 
The rudder servo was located/hidden immediately below the horizontal stabilizer 
(Figure 2.5).  To install the rudder servo a cutout was made that allowed the servo to 
be seated in the lower portion of the vertical stabilizer.  The servo wire extension was 
taped to the end of the string and ?threaded? through the boom (Figure 7.26).  The 
rudder servo was then secured with its mounting hardware (Figure 7.27). 
 121 
 
 
 Figure 7.26. Rudder Servo Installation ? detail view top left (see Figure 2.6). 
 
 
Figure 7.27. Rudder Servo Installation Complete ? detail view top left. 
 
 122 
 
7.5.3.2 Elevator Servo Installation 
The elevator servo installation followed the reverse sequence of steps as the rudder 
installation.  In the cases of the elevator and ailerons, the servos were first located and 
housed then wired.  The reason for wiring after installation will soon be apparent. 
 
The servo was located such that the control arm was connected to the (span-wise) 
center of the moving surface.  The servo was installed at the center of the moving 
surface to minimize twist of the aileron/elevator when deflected.   
 
Once located, the servo outline was traced onto the wing surface.  Using a straight 
edge and a sharp X-Acto knife, the outlined servo shape was cut out from the carbon 
skin; thus, once peeled away exposed the foam core.   
 
The bare foam was removed with the Dremel and the servo was trial fitted in place.  
The wire tunnel was then drilled through the wing foam core to allow access to the 
servo wire (the wire tunneling and stringing procedures are shown in the next 
subsection).  The servo extension wire was connected to the string and pulled through 
the horizontal stabilizer (Figure 7.28). 
 
Before sinking the servo in place, the servo was turned on.  The servo direction and 
deflection was adjusted accordingly (Figure 7.29).  Finally, the servo was buried into 
its tight foam housing.  Only after the entire aircraft was completely assembled was 
the elevator servo pushrod connected to the elevator control horn. 
 
The servo, once sunk in place, was secured with packing tape.  However, just as 
easily, a well-placed decorative decal can secure the servo and hide it from view.   
   
Construction side notes:  It is of paramount importance to create the servo nest as 
snuggly as possible to the servo.  To sink the servo into the foam core correctly 
requires practice and patience.  This procedure is easy to do, but equally easy to 
mess up.   
 
The depth of the servo bay is equal to the physical dimension of the servo 
depending on the orientation the servo is being installed.  In another words, the 
servo should be flush with the wing/stabilizer skin surface.  The servo, once sunk 
in place, was secured with packing tape.  However, a well-placed decorative decal 
can be used to secure the servo in place as well, thus, almost completely hiding it 
from view.  
 
As mentioned before, the moving surfaces are only permanently glued in place 
once the entire vehicle is completely assembled.  Moreover, the servos are only 
connected to their respective surface once the moving surfaces are glued in place.    
     
 
 123 
 
 
 Figure 7.28. Servo Installation (from top left to bottom right). 
 
 124 
 
 Figure 7.29. Elevator Servo Installed ? ready for installation into wing foam-core. 
7.5.3.3 Aileron Servo Installation 
The aileron servo was installed in similar fashion to the elevator servo.  However, the 
tunneling process required to allow wiring connection to the servo is described in this 
section.   
 
 The aileron servo was located and nested in place accordingly.  A solid ? rod with a 
sharp point was used to tunnel the foam up to the servo cavity.  To complete this 
process successfully, the operator visualized the path necessary to reach the servo 
nest (Figure 7.30).  The sharp-tipped rod was inserted into the wing at the root chord 
and hand turned.  The tunnel was slowly burrowed until the sharp tip was visible in 
the servo bay.  The tunneling rod was then carefully retracted from the wing. 
 
A second smaller diameter rod, with the string attached to its tip, was inserted into the 
wire access tunnel (Figure 7.30 top right detail view).  The string was captured and 
taped to the aileron servo wire extension (Figure 7.30 bottom left detail view).  The 
string was pulled back, from where it came from, while at the same time wiring the 
servo extension inside the wing.  
 
 
 125 
 
 Figure 7.30. Hand Made Wire Access Tunnel ? detail views diagonal corners. 
 
7.6 Final Vehicle Assembly ? Twin Boom Joined 
7.6.1 Considerations Prior to Joining 
The rudder and elevator servos were at this point placed and wired in.  The next step 
in the assembly was to join the twin boom section of the aircraft.  Two aircrafts were 
constructed and both received a slightly different treatment.  The choice here was to 
either leave the servos in place or remove them prior to joining the twin boom body.  
The risk involved was that excess resin could run inside the boom and glue the string 
or wire permanently in place.  As a result, the glued string could not be used to thread 
wire through or the glued wire extension could not be swapped if necessary.   
 126 
 
7.6.2 Joining Subsections 
7.6.2.1 Twin Boom Joined with Stringing in Place 
For the first airframe, the team sided with caution and removed the servos prior to 
joining.  While removing the servo extensions, the string was threaded through in the 
booms and horizontal stabilizer again (Figure 7.31).  At this point, the twin boom 
assembly was carefully aligned and permanently glued into one piece.  After the resin 
cured, the servo extensions were rethreaded through the booms and the servos were 
reinstalled.  The joining/gluing procedure is detailed in the next subsection.  
 
 Figure 7.31. Twin Boom Sections Ready to be Joined. 
7.6.2.2 Twin Boom Joined with Servos and Wiring in Place 
The second airframe inspired more confidence; therefore, the twin boom section was 
joined with the servos and wire extensions in place.   The initial step was to fully wire 
the twin boom section.  The servo wire extensions were pulled through the wire 
access port drilled on the center top surface of the center wing section (Figure 7.32). 
 
The tips, soon to be inserted into the booms, of the center wing section and horizontal 
stabilizer were roughed with a 400 grit sand paper.  The sanded surface was lightly 
coated with catalyzed resin and the twin boom assembly was mated into one piece.   
 127 
 
 Figure 7.32. Twin Boom Sections Ready to be Joined. 
 
The most important details were to make sure that the two boom shafts were parallel 
to one another (from a top view) and the vertical stabilizers parallel to one another 
(from a front view).  A precision metal ruler was used to measure the distance 
between the seam of the booms both at the front and at the rear of the twin boom 
assembly (Figure 7.33).  
 
Once the assembly was allowed to set, a bead of resin was laid down along the seam 
of the inserts.  This resin bead gave the twin boom assembly a finished look and also 
sealed the joint line, thus, preventing dust/dirt from entering the twin boom assembly 
(Figure 7.33 bottom right frame & 7.34 left frame).  Since holes or slits along the 
joint seam can generate noise, the resin breadline also prevented the aircraft from 
whistling during flight. 
 128 
 
 Figure 7.33. Twin Boom Assembly Joined. 
 129 
 
7.7 Moving Surfaces 
7.7.1 Moving Surfaces Final Assembly 
The moving surfaces were permanently glued in place.  The manufacturing and 
assembly work for the ailerons and elevator were completed in Chapter 6 and now 
were ready to be permanently installed.  The first step was to glue the ailerons to the 
wings and install the servo push arm. 
 
The next step was to install the rudders permanently.  The hinges were located and 
glued to the rudder as done before for the other surfaces.  A chrome sticker was 
placed along the hinging balsa strip to provide a finish to the hinge line (Figure 7.34).  
The tail wheel was held in place via an axle with a collar attached to either end 
(Figure 2.5).  Lastly, the rudders were connected to the servo (Figure 7.34 & 35). 
 
 
Figure 7.34. Rudder and Tail Wheel Installed. 
 
The elevator was sized to fit in the span between the two rudders.  Moreover, the 
elevator was trimmed so the rudders, when deflected either left or right, did not 
interfere with the elevator actuation.  The control arm for the elevator servo was 
installed after the elevator was glued in place. 
 
 130 
 
 Figure 7.35. Almost Completed Aircraft ? elevator not installed yet. 
7.8 Final Integration 
7.8.1 Propulsion Hardware Installation 
The motor was fitted to the adjustable motor mount and secured (Figure 7.36).  The 
motor was then installed to the firewall and connected to the speed controller.  The 
main battery was secured to the rear-facing side of the firewall.  Other minor 
electrical hardware was also installed like the main battery fuse holder and the main 
on/off switch.  The propeller and spinner were secured to the motor shaft. 
 
 Figure 7.36. Motor Mount Assembly. 
 131 
 
7.8.2 Final Wiring and Electronic Hardware Installation 
The receiver was installed in the rear of the fuselage and its antenna ran outside the 
aircraft towards one of the vertical stabilizer.  All of the servo extensions were 
connected to the receiver in their proper channels.  Both the receiver and receiver 
battery were housed in foam and secured to the aft bulkhead. 
7.8.3 Other Hardware Installation 
The main landing gear struts were also screwed in place and adjusted such that the 
wheels were parallel the aircraft?s longitudinal axis and to each.  The hatch front 
alignment pins were glued and the rear hatch release was installed (Figure 7.35). 
7.8.4 Spar Tube Screws/Tie-Down 
The outboard wings were secured by means of a ?-inch nylon bolt screwed at each 
end of the spar.  A hole was made through the spar sleeve and half way through the 
carry-thru spar (Figure 7.37).  The bolt threaded to the internal carry-thru spar and 
held the two outboard wings in place.   
 
 
Figure 7.37. Wing Bolt Clearance Hole Made. 
 132 
 
7.9 Advice 
7.9.1 More on Labor and Task 
The labor and tasks for this phase of the construction are interchangeable.  From a 
project management point of view, three individuals can finish the project just as well 
as twenty can.  The question is, how soon does the airplane have to be in the air?   
 
The root branches in Figure 1.4 (Machining, Joining Parts and Internal Structure) can 
be developed independently.  However, there is still some interplay between the 
different sections, but progress can be made in each branch.  For example, while 
small parts are machined, trimming and trial fitting of vehicle parts can occur and 
fuselage internal structures can be manufactured. 
 
In conclusion, the most efficient way to assemble the aircraft is to finalize each part 
independently and then piece them together in sequence.   It is vital that all 
individuals understand what exactly he/she is working on and how it fits with the 
parts someone else is working on.  The speed in which the assembly phases 
progresses depends on the number of skilled individuals working in various parts at 
the same time, communication between individuals and minimal operator/worker 
errors/mistakes.  
7.9.2 Considerations Prior to Joining  
Before committing pieces together, either in Part I or II, careful consideration must be 
given to ensure a (future) smooth assembly progression.   The most important event 
that must occur before joining parts is the realization of what needs to be done prior 
to joining.  One of the most damaging mistakes made during joining is the realization 
that parts that should have been embedded were not installed and ?now it?s too late?.  
This kind of mistake can completely ?kill? project momentum, cause on-the-spot 
design changes and/or junk entire vehicle subsection(s).   
 
The only way to be truly immune to this problem is to have experienced workers.  
However, if this is not the case, then one has to settle for the next best option.  If time 
is not too short, it is advisable to assemble the B-Grade vehicle first.  In another 
words, chances are many parts were made in Chapters 5 & 6 and some parts are better 
in quality than others are.  The variance in quality is there because of the learning 
curve involved, especially for first-timers.   
 
The parts that are satisfactory, but otherwise not the best ones, can be used in the first 
go-around of assembly.  This creates a safety net in case mistakes are made or issues 
overlooked.  This kind of cautious approach to assembly alleviates the ?zero mistake 
tolerance? mindset, thus, allowing the novice builder to learn the process and gain 
experience without jeopardizing the best-looking parts.  
 
 133 
 
The approach mentioned in the previous three paragraphs was put to use in this 
project.  Due to the wide-ranging variance in experience between group members, the 
first vehicle assembled was made up of parts that were somewhat less ?gleaming?.  
To differentiate between the various parts, a non-permanent silver marker was used to 
label all of the pieces.  The writing on the pieces is seen in various pictures in this 
Chapter. 
7.9.3 Think First - Let the Solution Present Itself 
The title of this section may seem innocent, but it speaks mountains.  Another likely 
title for this section may have been ?Fools Rush In?.  The point here is that when a 
problem surfaces, it must be given adequate contemplation.  At this stage in the 
construction, a misstep can ruin much of the work and delay the assembly process.  
During this project, when an undesirable event happened, often times the best 
solution was not the first solution that came to mind.  The assembly of the airplane is 
connected to itself in many levels and solving one thing without thinking about the 
rest can, by far, be the least desirable course of action. 
 
It is definitely exciting to see all of the pieces come together to form a final product.  
Although very close to the end, it is important to maintain quality control.  The 
details, tirelessly given during the beginning of the construction, must be upheld 
down to the last screw.  In general, once the aircraft is ready for flight it will look as 
good as the least detailed/finished part visible.  
 134 
 
Chapter 8:  Conclusion 
 
8.1 Complete Aircraft 
The completed aircraft is shown in Figures 8.1 to 8.5.  The final product is not 
painted or coated with gloss.  The reflective carbon composite skin is a result of the 
workmanship and attention given to the various manufacturing techniques described 
in previous Chapters.  Decals were applied to airframe to enhance its appearance and 
aid the pilot?s visual of the airplane during flight.  Figure 8.6 is a team photo. 
 
 
 
 
 
 
Figure 8.1. Completed Aircraft ? nose close-up. 
 
 
 
 
 
 135 
 
 
 
 
 
 
 Figure 8.2. Completed Aircraft ? Maryland flag reflection. 
 
 
 
 
 
 
 
 136 
 
 
 
 
 
 
 Figure 8.3. Completed Aircraft ? runway access ramp. 
 
 137 
 
 Figure 8.4. Completed Aircraft ? on take-off. 
 
 138 
 
 
 
 
 
 Figure 8.5. Completed Aircraft ? flyby over runway. 
 
 
 139 
 
 
 
Figure 8.6. Completed Aircraft ? team picture. 
8.2 Composite Construction Conclusion 
The composite construction methods were customized to meet the needs of the 
individual parts of the aircraft.  Each method utilized has its pros and cons.  
Furthermore, each method required different levels of care and skill sets. 
 
However, the construction techniques developed in this document, when followed 
attentively, generated a product with superior quality given its cost.   
 
 140 
 
 
References 
 
Mallick, P.K., ?Composites Engineering Handbook?, Marcel Dekker, Inc., 1997. 
 
Bunsell, A. R., Renard, J., ?Fundamentals of Fibre Reinforced Composite Materials?, 
IOP Publishing Ltd., 2005. 
 
Gay, Daniel, Hoa, Suong V., Tsai, Stephen W., ?Composite Materials: Design and 
Application?, CRC Press LLC, 2003. 
 
Astrom, B. T., ?Manufacturing of Polymer Composites?, First Edition, Chapman & 
Hall, 1997. 
 
Gutowski, Timothy G., ?Advanced Composites Manufacturing?, John Wiley & Sons, 
Inc., 1997. 
 
Strong, A. Brent, ?Fundamentals of Composites Manufacturing: Materials, Methods 
and Applications?, First Edition, Society of Manufacturing Engineers, 1989. 
 
Mazumdar, Sanjay K., ?Composites Manufacturing: materials, product and process 
engineering?, CRC Press LLC, 2002.