SAE Aero East 2017 Final Report

Final Paper

SAE Aero East 2017 Final Report

Old Dominion University

Mechanical & Aerospace Engineering Department

MAE 435 Project Management II - Fall 2016

December 4th, 2016

Project Advisor: Dr. Drew Landman

Team Members: Frank Dixon, Coleman Gordon, Kathy Hansen, Thomas Houck, Kevin Schesventer, Gerald Short, Marquis Ward, Zhangsiwen Xiao, Joseph Zongolowicz, Nick Montana

HONOR PLEDGE

“I pledge to support the honor system of Old Dominion University. I will refrain from any form of academic dishonesty or deception, such as cheating or plagiarism. I am aware that as a member of the academic community, it is my responsibility to turn in all suspected violators of the honor system. I will report to Honor Council hearings if I am summoned.” By attending Old Dominion University you have accepted the responsibility to abide by this code. This is an institutional policy, approved by the Board of Visitors.”

1

SAE Aero East 2017

Final Report

Table of Contents

Table of Contents …………………….…………………………………………………….. ii

List of Figures ...... ………………………….…………………………………………………...iii

List of Tables...... ………………………….……………………………………………………iv

Abstract…………………………………………………………………………………… v

Introduction…………………………………………………………………………………… 1

Methods…………………………………………………………………………………… 3

Results…………………………………………………………………………………… 9

Discussion…………………………………………………………...……………………...12

Conclusion………………………………………………………………………………….. 14

Appendices………..………………………………………………………………………… 16

References………………………………………………………………………………….. 23

List of Figures

Figure 1: Basic Wing-Tail Model in XFLR5 ...………………………………………………..…5

Figure 2: Fuselage Structure ....………………………………………………………………….. 6

Figure 3: Proposed Main Landing Gear Designs. A-Torsion Bar, B-Straight Axle, C-Trailing Arm with Torsion Springs, D-Bracket with Extension Springs, E-Short Torsion Bar, F-Damper Bracket, G-Absorbing Trailing Arm ……………………………………………………..……....8

Figure 4: Thrust Comparison at 0 mph …………………………………………...... …………...9

Figure 5: Fuselage Tail Boom ...………………………………………………………………... 10

Figure 6: Wing Deflection Data from the 2016 Wing ...………………………………….……..11

Figure 7: Tail Reinforcement Comparison………………………………………………………12

List of Tables

Table 1: SAE Aero Design 2017 Aircraft Configuration ………………………………..…….. 10

Table 2: Landing Gear FEA Results .…………………………….………….……………….… 12

Abstract

To make a competitive radio controlled aircraft that meets the requirements of the SAE Aero Design Series competition, the 2017 team focused on carrying the most payload by evaluating the motor/propeller package, aircraft configuration, and the airframestructure. Several motor/propeller packages were considered to maximize power, efficiency, and thrust without adding significant weight. The aircraft configuration was chosen based on real-world low power to weight ratio aircraft and then optimized for the specific mission requirements of the competition. The configuration was analyzed using vortex lattice methods and the lift, moment, and drag forces calculated to determine if the mission requirements could be met. The structure of the plane’s fuselage and wing were examined in order to reduce the amount of unnecessary weight. To better distribute landing impact loads through the airframe, finite element analysis was performed on six main landing gear designs to determine whether they would support the target gross weight with the appropriate factor of safety. Construction of the aircraft was guided with a complete SOLIDWORKS model of the competition plane and plywood holding jigs.

1

SAE Aero East 2017

Final Report

Introduction

The SAE Aero Design Series is an international collegiate competition that tasks teams to design, build, and fly an original radio-controlled (R/C) aircraft to meet specific performance and mission requirements. ODU has participated in the competition’s East Division Regular Class since 2014, finishing eighteenth, thirteenth and fifth, respectively. This year’s competition features a new challenge for the regular class, teams are to design and build an electrically-powered aircraft that is capable of lifting as many passengers, in the form of tennis balls, and luggage, in the form of steel plates, as possible.The plane must be able to takeoff in less than 200 ft, land in less than 400 ft, and is not to exceed 55 lb gross weight or use any fiber-reinforced plastics (FRP) [1].The dimensional restrictions present in previous years have been removed, resulting in the need to create a brand-new design for the 2017 competition.

The competition rules restrict total power provided to the motor to 1000 W, severely limiting the available thrust that can be reasonably achieved[1].A higher thrust leads to higher speeds, which results in more lift. Therefore, selecting a motor/propeller package requires careful analysis to determine which combination provides the most thrustwhile within the power range, and without adding too much weight [2]. Based on the types of motors used and size of the aircraft, simulations can be used for this analysis so that physical testing is not necessaryfor all possible combinations[3].The current propeller/motor package has been unchanged since being initially selected for the 2015 aircraft. Due to changes in the aircraft design, it is currently unknown whether that package is still the optimal selection for the 2017 aircraft.

The requirement to carry tennis ball passengers has led to the fuselage being enlarged. Additionally, the lack of dimensional constraints means that a longer, thinner wing can be used which produces lift more efficiently than previous years. The increased dimensions also mean that drag reduction will have more of an impact on performance than previous years. All these considerations must be accounted for, along with the limited thrust available, in developing the new design.

With an increased payload comes increased structural loads on both the airframe and the landing gear. Previous ODU aircraft have been designed based on a 2 g load condition, or approximately 60 lb. This gave a factor of safety (FS) of 2 for the 2016 aircraft. Through experience, it appears this FS is too high to produce a sufficiently lightweight aircraft. Therefore, the 2017 airframe was designed around a FS of 1.5. This, in turn, has led to a redesign of the major internal structural components of the fuselage and wing to produce a lighter-weight airframe. In order to ensure the airframe will be able to withstand aerodynamic and static loads with the required FS, structural strength data must be analyzed from previous designs [4]. However, no such data exists for the previous airframes.

Previous ODU SAE Aero teams used a truss style fuselage. This style is made up of structural members that give the fuselage its shape and strength. A semi-monocoque style fuselage uses bulkheads connected with longerons to give it its shape while the strength comes from the covering. Whereas a truss style fuselage must be rectangular, a semi-monocoque can be made into any shape, potentially lowering the overall drag. No structural data for a semi-monocoque fuselage exists so it must be tested to determine if it is as strong as the truss style.

The 2016 landing gear resulted in damage to the bottom of the fuselage. Therefore, it became necessary to redesign the landing gear to properly transmit the landing forces into the fuselage, especially atan increased gross weightand newly designed fuselage[5, 6].Increasing the strength of the existing solid axle design means increasing the weight;[4], which would negatively impact the design goals of the 2017 aircraft. Therefore, alternate designs must be evaluated to determine if shock absorption can be added at a lower weight penalty than increased axle size while also minimizing drag.

The goals for the 2017 SAE Aero East team were to design, build, test, and analyze an R/C aircraft capable of carrying 60 passengersand their luggagearound a rectangular circuit while meeting the competition requirements, and to deliver the aircraft by December 2016.
Methods

In order to determine what design changes would be most impactful, the team looked at issues from previous teams and analyzed the 2016 competition aircraft. This analysis provided the following direction: thrust could be increased by selecting a different motor and propeller; the wing and fuselage were likely too strong and could be redesigned to reduce weight, which required structural testing to determine the optimal design; the tail was not stiff enough to prevent warping when the covering was applied; and the landing gear needed to be redesigned to better transmit landing forces into the fuselage to prevent damage to the airframe. The new rules also made the team reconsider the overall design of the aircraft (fuselage size, wing length, tail height, etc.).

A custom design tool [4] was created using Microsoft Excel (Microsoft Corp, Redmond, WA) The Excel-based design tool is capable of calculating the appropriate shape and size of the aircraft components, creating a detail weight analysis, and predicting the planes climbing performance. The weight analysis was converted in to a weight budget which determined the maximum weight constraints for the structural design teams. The size and shape information from the design tool was used to create a CAD model of the Outer Mold Lines of the aircraft (the outer shape), which was then used to create detailed models of the internal structure of the aircraft. These models were used to generate cut files for the CNC laser cutter to produce the structural components required for construction. Finally, the different CAD subassemblies were used as a guide for construction and provided accurate weight and inertial predictions for the finished plane.

Motor and propeller data were collected from the 2016 plane such as weight, battery type, wing area, etc., and, using the online calculator eCalc (Solution for All Markus Mueller, Tann Switzerland),the theoretical results: static thrust, maximum input power, and thrust to weight ratio were recorded [7].Different motors(not exceeding 1000W) and propellers were picked to increase thrust, thus lift, while the controller, battery, and overall plane characteristics were held constant and the calculations were rerun. Potential combinations (Table A3.1) were recorded and later compared to data received from wind tunnel testing [7]. Additionally, the program XROTOR (Massachusetts Institute for Technology, Mark Drela, Cambridge, United States)was used to model a few of the propellers to run further analysis by varying different parameters.The combination that yielded the most thrust was then modeled using SOLIDWORKS(DassaultSystémes SOLIDWORKS Corp., Waltham, MA) to be included in the final aircraft CAD model.

In order to meet the payload and passenger count goals, the total wing area required to produce sufficient lift was determined for a designed cruise speed of 25 mph (Eq. 1)

whereL is the total lift generated, ρ is the atmospheric density, V is the freestream velocity of air, S is the planform area of the wing, and CLis the aircraft lift coefficient. The airfoil selected for the wing was the Eppler 423 (E423) due to its high lift coefficient (Cl) and thick trailing edge for ease in construction. For the horizontal and vertical tail (HT, VT), symmetrical National Advisory Committee for Aeronautics (NACA) airfoils were selected (NACA 0009 for the HT and NACA 0007 for the VT)[4]. The size of the wing and tail were then determined through the Excel-based design tool.

Once the general dimensions had been determined, the wing-tail only aircraft was modelled in XFLR5 (Techwinder, to obtain the aerodynamic performance data for the airplane. First, the airfoil coordinate data were obtained from the UIUC Airfoil Database[8],evaluated using the Direct Foil Analysis tool to obtain required aerodynamics coefficients (CL, CM,CD) and then fed to the 3D aircraft design analysis tool.

The overall aerodynamic coefficients were obtained once the models were analyzed over an angle of attack (A.o.A.) range of -4degree to 10 degrees using the Vortex Lattice Method (VLM) solver under Standard Temperature and Pressure (STP) (Figure A3.1). The results were then used to compare static stability of the configurations (CM), to calculate the theoretical total lift using the lift equation (Eq. 1), and evaluate total drag using the drag equation (Eq. 2)

whereD is the total drag force. The lift and drag coefficients, along with the thrust data from the motor and propeller testing, were then entered in to the Excel-based design tool and the overall size and shape of the aircraft was calculated.

A balsa fuselage was modeled in SOLIDWORKS to evaluate the effects of modifying the truss structures.It consists of a Pratt truss[9]main box that holds the payload and electronics as well as attaches the motor, landing gear, and wing, and a Warren truss [9]tail boom that attaches the tail (Figure 2).To minimize weight, most vertical members were eliminated from the 2016 fuselageaside from the perimeter and certain high stress areas such as at the wing attachment points and above the landing gear. A test fuselage was constructed and subjected to load testing to measure structural strength of the payload bay and the tail boom. This involved supporting the fuselage with makeshift landing gear on a flat, level surface, and placing weights across the top of the payload bay and tail boom. Weights wereslowly added until a joint or a structural member failed. The fuselage was then repaired and reinforced at the location of the failure and retested. To account for potential joint failure, thin plywood gussets were glued over the joints that repeatedly failed during the load testing [10]. These gussets assisted in transferring the load between structural members andallowed for more gluing surface area at the joints. The same tests were then conducted on a covered semi-monocoque fuselage so the data could be compared.

Two representative wing sections were built so that deflection and the load carried by the skin could be determined. The wings were comprised of ribs for shape, an I-beam spar for bending strength, longerons for connectivity, and a “d-tube” to provide torsional resistance [9] and to define the leading edge curvature (Figure A3.2).The wingswere made entirely of balsa except for the spruce inboard spar caps.Using a holding jig to ensure component placement, a wing with 3” spaces between ribs and a wing with 6” spacing were built. A constant taper ratio of 1 was used for inboard 3/16” thick ribs and 0.8 for outboard 1/8” thick ribs.The entire structure was assembled using cyanoacrylate adhesive. Each wing section was built and attached to a block at the fuselage position to secure it during testing. To test each wing structure and its skin covering strength, the wingswere mounted upside down and a distributed load was applied along the span [4, 10] and deflections were measured along the main spar including the tip via a dial gauge. Each wing was thencovered with Ultracote and the testrepeated. The deflection measurements wererecorded to compare the bending resistance with and without the skin (Table A3.2) to determine what percentage of the load is carried by the skin. The deflection of the wing can be calculated using the equation for deflection of a cantilever beam (Eq. 3)

where is the maximum deflection, is the maximum intensity load, l is the length, E is the Modulus of Elasticity, and I is the moment of inertia [4].Using the same mount, the representative wings were subjected to a torsional test by suspending loads near the tip of the wing. As with the deflection test, the wings were tested with and without covering in order to assess the reinforcement of the covering. The wings remained covered and were tested to failure to determine their FS.

The tail was redesigned to have a larger span and one large vertical stabilizer instead of three smaller ones. The new tail configuration was modeled in SOLIDWORKS to evaluatetorsional resistance. A representative 2016 tail was constructed. Bending deflection was tested by placing weights along the top of the spar, and torsion tested by suspending weights from the tips of the trailing and leading edges, and measuring angles of torsion. The tail frame was tested two additional times, once with diagonal cross members added between each rib and once with a diagonal shear web between each rib. The deflections were then compared to analyze how much reinforcement the modifications provided.

SOLIDWORKS and Finite Element Analysis (FEA) were used to analyze sevenlanding gear designs (Figure 3).To simplify the process, bolt connections were assumed to be fixed geometry, small tabs were used in place of wheels, and the rod, arms, and cross member were modeled as separate, but connected, bodies [6].The vertical force of 100 lbf, FS 1.75, was obtained from qualitative observations of the 2016 plane [11] while the side force of 40lbf, FS 1.6, was obtained through Newton’s second law of motion(Eq. 4) where m is the mass and a is the acceleration.

(4)

These conditions will indicate the landing gears’ ability to withstand a landing hard landing as well as a landing on the side of the wheel.The torsion spring was successfully used in previous designs and was therefore not included in the FEA analysis.

Finally, holding jigs were designed using SOLIDWORKS for each of the major airframe structures and referenced off the component assembly model. Each jig was cut from a sheet of plywood using a 3-axis CNC router and consists of component holders, connecting bars to secure the jigs in the proper relative orientation, and a common base to mount the jig when in use.The jigs were used to align the ribs and structural members when constructing the wing and tail to increase accuracy and repeatability in construction. This also provides the ability to make dimensionally consistent repairs.