Concept Report

Fixed Wing UAV

Team Members:

Alejandro Suarez

Dalton White

Dustyn V. Moizes

Jonathan White

Kyle Van Wagner

Pedro Lopez

Due Date: April 25, 2022

Senior Design I – 4700C

Tuesday & Thursday 1:30 pm - 2:45 pm

Advisor: James Hippelheuser

Team: Gold

Executive Summary

In this project, the challenge is to make a cargo fixed-wing UAV capable of carrying 25-35 fluid ounces of liquid while receiving thrust from a pusher propeller. The goal of this project is to compete against other teams to see which team can carry the cargo around the circuit the greatest number of laps. One of the requirements for this project is to have the payload be detachable and capable of holding the 25-35 fluid ounces of liquid. Despite this requirement, the UAV is expected to be able to fly with and without the payload attached. For this project to be possible, we must comply with both FAA regulations and SAE Aero Design Rules to ensure a safe and regulated UAV. The scoring is going to be done by taking the average number of laps done in two separate flights, while having the tie breaker being the UAV with the fastest average lap time. For our UAV, made from balsa wood, we decided to take a unique design with a cylindrical payload bay in the middle of the plane, with the motor in the back of the payload bay, and two booms connected to the bottom of the wings of the aircraft leading to the tail in the back. We will then have a large horizontal stabilizer connecting the two booms together and vertical stabilizers on top of the booms providing us with yaw stability. Using this unique design, we will be able to carry the liquid payload with great stability and endurance around the circuit and successfully completing our goal for this project.

Table of Contents

1. Problem Statement
2. Customer Needs Analysis
3. Critical Specifications
4. Technology Assessment
5. Competitor Benchmarking
6. Concept Generation and Selection
7. Engineering Analysis and Calculations
8. System Concept Evaluation Plans
9. Cost estimates
10. Significant Accomplishments and Future Work
11. Conclusions and Recommendations
12. References
13. Appendix

1. Problem Statement

A pusher-style-electric Unmanned Aerial Vehicle (UAV), will be designed to compete against others. It must fly as many laps as it can and land with and without a liquid payload. The aircraft must use electric propulsion and hold 25-35 fluid-ounces of water as its payload. The payload must also be detachable from the aircraft. The UAV with the most laps will take home the title for the best UAV made.

Some of the design problems that will be solved during the design of the aircraft include, but are not limited to: weight distribution, stability, material selection, mounting locations, design type (flying wing, cessna shape, etc), wing/airfoil shape and more. The design of the aircraft will revolve around the following functions: stable flight, maneuverability, propulsion, landing, liquid payload design, removable payload design, remote communication systems, and in case of emergency, shut-off systems. For structural analysis and optimization, HyperX and Nastran software will be used, and for best selection of airfoils and aerodynamic analysis: Ansys, Xfoil, and Xflr5 software. All of these functions will be taken into account when designing the UAV, and the most effective options will be taken in order to provide the best performing UAV. Critical specifications have been drafted and researched and located in the critical specifications section of this document (Part 3).

2. Customer Needs Analysis

        This form of the house of quality (located below) compares the customer-defined requirements (from the Project description) with the determined requirements for the plane as a whole. The critical requirements were chosen and defined by each of our component sections in the requirements table located in the critical specifications section (Part 3). The comparison allows us to prioritize goals and to satisfy the project goals (customer goals and needs) as a matter of course. For example, the payload capacity highlights how much the payload is going to ultimately affect each other component in the system, and the importance of focusing our efforts on designing around that. The customer needs is simply a UAV that is a pusher style aircraft capable of delivering a payload. The plane must also fly without the weight of the payload attached to it. Once this is accomplished, the customer needs will have been met.

In the table located in Part 3 are listed all the components that will constitute the UAV. Along with the list of components requirements, research for ranges of each component for the UAV to work properly and a list to verify these different ranges is provided. This table is essential to organize what will be needed to get for the project, and also helps with the selection of components that are compelling for the competition.

Fig 1. House of Quality (Continued on next page)

3. Critical Specifications

Listed below are the critical specifications and ranges for values of each main component. Located later in the document in Part 6 will be a more descriptive and accurate selection guide for each component. In Part 6 the team will elaborate on each component in depth and come up with a concept design with more accurate values.

Table 1. Component Specification

4. Technology Assessment

The pusher setup for this UAV is more unconventional compared to the typical pulling “tractor” setup and will provide additional problems that our group will have to overcome. The technology described below will highlight our steps to solve this design problem, and to clarify them.

This unmanned aircraft is required to fly with a payload of 25-35 fluid ounces, and a liquid payload requires extra consideration in the design for sloshing. The CFD analysis for the wing and airfoil will be done using XFLR5 and XFOIL. The craft must be a fixed wing style UAV and cannot contain any multi-rotor or VTOL capabilities. A fixed wing aircraft will use the wings and airfoil shape to generate the lifting force. Also, the fixed wing aircraft has much greater takeoff weight capabilities, and a much further range in comparison to a multi-rotor drone. There are numerous wing styles that are frequently used in the design of fixed wing UAVs, with the most beneficial to us being the conventional setup. The symmetry of the wing, size of the wing, the amount of camber, and the thickness of the airfoil all play important roles in the amount of lift and drag generated. The camber of the airfoil has a very large effect on the amount of lift and drag produced. The more cambered an airfoil is, the more lift it will generate due to a higher-pressure differential and thus increasing the total lifting force. Most importantly, a cambered airfoil will generate lift at a zero angle of attack, unlike a symmetrical airfoil. The airfoil selected for this UAV is the Clark Y airfoil. This airfoil has a maximum thickness of 11.7% at 28% of the chord length and a maximum camber of 3.4% at 42% of the chord length.  In a test alongside other commonly used airfoils for UAVs, such as the Eppler 423 and Selig 1223, the Clark Y showed the highest Cl/Cd ratio, and had a stall angle of about 13°. The aspect ratio of the wing also needs to be taken into consideration when choosing the airfoil and wing design. The aspect ratio of an aircraft is the ratio of the wing span to mean chord length, given by the equation, AR = b2s. Where b is the wingspan, and s is the wing area. With a 5ft wingspan and 12-inch chord length, our aspect ratio comes out to 5. All of the fluid simulation done in this project will be using XFLR5 and XFOIL. XFOIL is a program used for 2-D, viscid and inviscid analysis of subsonic airfoils. The program uses a high-order panel method and a fully coupled viscous/inviscid interaction method when finding the lift and drag characteristics, as well as finding and displaying the boundary layer separation. A standard use of n=250 number of panels, and a leading to trailing edge density of 1 will be used for all analysis in XFOIL for this project. XFLR5 allows for analyzing aircraft wings, elevators, and fins in 3-D. This program was built off of XFOIL, so the same specifications will apply to the analysis. The simulation in XFLR5 was done using the lifting line theory explained in detail later in this report.

5. Competitor Benchmarking

When comparing this UAV to others, it can be seen that this design is fairly new. Most UAVs are designed with up front propulsion while ours is designed with the pusher motor. The UAV the team designed is designed for payload delivery. The way the whole design was made is different from most other UAV’s you see on the market. The newly designed tail for extra stability will help keep the UAV in stable flight with or without payload. The super lightweight design will help make it more efficient and provide a longer flight time. Once fully prepared and assembled, it will become a UAV of its own kind which is a goal of the team from the beginning. The team did not just want to replicate another UAV that has been made before. In comparison to other teams, we believe that we are right on track with the rest of the groups and have also made some outstanding achievements with the design that others have not done with theirs. When our team comes to the competition, we will be providing a UAV to provide sufficient satisfaction to the customer’s needs. This will be more than adequate with the other teams and possibly be the UAV that takes home the first-place title. The team is also up to date with all assignments and is ready to start final design and build in the fall term.

6. Concept Generation and Selection

Airfoil Analysis

Analysis was done on three select airfoils (Eppler 423, Clark Y, and Selig1223) to determine the best option for our UAV. This selection was narrowed down from a larger list of 10-15 airfoils, all selected for their known characteristics of low Reynolds number flight in RC drones. Key factors in our choice include optimizing the Cl and Cd such that the takeoff speed was within reason in order to achieve flight, and maximizing total lift. In the beginning stages of wing design, it was decided between the group to set a wingspan and find the other wing characteristics based on the wingspan. We chose a wingspan of 5 feet or 1.52 meters. This led us to a chord length of 1 foot, approximately 12% of the wingspan, which is increased from the 5 - 6% suggested in the early stages of research. The analysis was done at speeds of 10 m/s, 15 m/s, and 20 m/s to cover a wide range of speeds during flight. This led us to Reynolds numbers of approximately 215,000, 322,000, and 429,000, respectively. Shown below in Figure 2 are the XFLR5 simulation results done over a range of angles of attack from -5° to 20°. According to the XFLR5 plots, for a Reynolds number of 200,000, the Clark Y airfoil has an approximate maximum Cl/Cd value of 73.2 at ɑ=4.3°, the Selig 1223 has an maximum Cl/Cd value of 73.6 at ɑ=3.75°, and the Eppler 423 has a maximum Cl/Cd value of 73.7 at ɑ=9.25°. For a Reynolds number of 500,000, the Clark Y airfoil has an approximate maximum Cl/Cd value of 98.7 at ɑ=3.75°, the Selig 1223 has a maximum Cl/Cd value of 98.8 at ɑ=4.5°, and the Eppler 423 has a maximum Cl/Cd value of 123.4 at ɑ=5.5°. The stall angles for all 3 airfoils at both Reynolds numbers of 200,000 and 500,000 are very close to the same value of 13°. From the information provided below, we decided that the Clark Y was the best airfoil for our application. This selection was chosen because the max Cl/Cd value was at a low angle of attack and we are more capable to manufacture the Clark Y airfoil than the other 2.

Figure 2. Xflr5 Lift and Drag Plots

The XFLR5 analysis of the airfoil was verified by using another software called XFOIL. This program shows the change in coefficient of lift, CL, Coefficient of Drag, CD, and Coefficient of Momentum for a range of angle of attack from -5 to 20 degrees, at a different Reynold’s Number, Re, value (1*10^5 and 2*10^5). The comparison of the CL, CD and CM values are shown in graphs 1, 2 and 3 for Re 1*10^5 and graphs 4,5 and 6 for Re 2*10^5. The values to make those graphs are shown in the appendix tables number 7 through 12.

Graph 1. Coefficient of Lift vs Angle of Attack (Re = 1*10^5)

Graph 2. Coefficient of Lift vs Angle of Attack (Re = 2*10^5)

Graph 3. Coefficient of Drag vs Angle of Attack (Re = 1*10^5)

Graph 4. Coefficient of Drag vs Angle of Attack (Re = 2*10^5)

Graph 5. Coefficient of Momentum vs Angle of Attack (Re = 1*10^5)

Graph 6. Coefficient of Momentum vs Angle of Attack (Re = 2*10^5)

Wing Analysis

The wing was designed in XFLR5 using the 5-foot wingspan and 12-inch chord length defined above. The wing has a planform area of 0.465 m2 and an aspect ratio of 5. The aspect ratio is directly proportional to the wingspan of the UAV, and will determine the gliding performance of the UAV. The analysis for the wing was done using the lifting line theory (LLT) in XFLR5 at velocities of 10 m/s, 15 m/s, and 20 m/s. The LLT solver uses a linear, inviscid method utilizing thin airfoil theory. This considers the airfoil as a lifting line that increments the vortices shed along the span trail in the direction of the free-stream velocity. When these thin-plate theory equations are linear, they can be solved analytically, but in general, the lift curve is non-linear, so analytical solutions are no longer viable. The 3-D wing models were designed using the plane analysis in XFLR5. This was done with all 3 of the airfoils. Shown below are the Cl vs Cd, Cl vs Alpha, and Cl/Cd vs Alpha plots generated at a velocity of 20 m/s. Some of the lower angles of attack would not converge after 200 iterations when using a velocity of 20 m/s.

Figure 3. Clark-Y’s Xflr5 wing analysis

Figure 4. Eppler 423’s Xflr5 wing analysis

Figure 5. Selig 1223’s Xflr5 wing analysis

Graph 1. Clark-Y’s Xflr5 wing analysis

        Green - Clark Y

Gray - Eppler 423

Purple - Selig 1223

The interior wing design, including the ribs, spars, and stringers was done using Wing Helper. Wing Helper is a 3-D CAD software for designing and exporting all characteristics of an RC wing. The program defines the wing as a set of trapezoid or free-form panels, and lets the user define standard parameters such as wingspan, tip and root chord length, sweep, dihedral, tip twist, etc. For our design we did not utilize any wing sweep, dihedral, winglets, or wing offset. Our UAV will be a relatively slow flying, low Reynolds number RC UAV, so none of these parameters, besides an upward dihedral to increase roll stability, would be of any use to us. Wing sweep is used on jets flying in the transonic regime to delay the formation of shockwaves which induce more drag on the wing. Winglets help reduce wing drag by decreasing the size and thus strength of wingtip vortices. Both of these parameters prove to be useless in our low-speed flight regime. A slight upward dihedral could prove to be useful in increasing roll stability of our UAV, but we ultimately decided against this for ease of manufacturing and symmetry. We decided to use laser cut balsa wood as our interior wing structure. These parts will be cut out and assembled before we wrap the whole wing in monokote. This was chosen over the use of a high-density foam, for increased strength and less flexibility. As stated above, we decided to choose the Clark Y airfoil for our UAV, so the ribs are made in the shape of the Clark Y airfoil and are 2mm thick. We decided to use 8 ribs per-wing, and all ribs besides the outer 2 have cutouts in them called lightning holes. This will decrease the overall weight of the wing without losing much strength or stability. These lightning holes in the ribs will also allow us to run the servo wires and connectors through the wing and to the fuselage for control of the ailerons. There are 3 spars running along the main section of the wing. These are rectangular spars with webbing between each rib. This will increase the strength of the wing more than just having spars with no webbing. The last 25% of the trailing edge of the wing is cut away in a shape called an “aileron cutter K-form” allowing space for the aileron control surface to be attached to the main section of the wing with Robart pin hinges. This shape will allow the aileron to move freely up and down without interfering with the main wing section. The aileron section of the wing will have one circular spar running through the front, and one thicker spar at the trailing edge. The interior wing sections will be laser cut out of a sheet of balsa wood and then assembled and wrapped with monokote. Monokote is a fire-proofing material we will use to “shrink-wrap” over the wing with a heat gun.

Figure 6a. Clark-Y’s Ribs top view

Figure 6b. Clark-Y’s Ribs diagonal view

Figure 6c. Clark-Y’s Ribs side view

Payload selection

The selection for payload design stems from the different types of containers that could be designed to hold in the necessary amount of liquid payload, which ranges from 25-35 fluid ounces. The quantity of liquid ranges in weight from 1.63lb-2.28lb. There are also different mechanisms that could be used to secure the payload in place while allowing for the payload to be removed, which results in different permutations for the payload design. The payload must also be built such that it can be fitted onto or inside of the fuselage. Critical components for the design of the payload are weight, manufacturability, size, cost, security, stability and reliability. Below is a chart to show how the different mounting systems chosen can be related and also how they can be compared with a scoring system to help the team choose the appropriate system.

Mounting system

Table 2. Selection of build type (1 worst, 10 best)

The mounting system needs to be developed such that the payload is able to be removed, the UAV is able to fly without the payload attached and it is securely in place during flight. Another crucial parameter is the time it takes to remove the payload from the UAV. This will allow for the team to be able to run the UAV in a timely manner during the flight competition and possibly have the best time out of the teams. Three ways of mounting the payload are using clips, friction holds, or screws. It is also possible to use more than one of these mounting systems at once. This system also needs to be able to hold the weight of the liquid payload as well, considering that the weight of the liquid ranges from 1.63lb-2.28lb and also need to be able to hold the weight of the container as well.

Using clips to mount the payload would offer an adequate solution to the design problem. It would be more stable than only using friction hold and would allow for the easy removal of the payload, unlike the screws which would have to be removed to remove the payload. This would cut down on removal time tremendously and is a very safe design. Clips can be ordered and secured to the payload to ensure that it will not move. This method scores the highest in the chart and can be used with a combination of both screws and friction to provide the most secure setup.  

Using friction hold would allow for the payload to be removed easily, but does not provide the security that the other methods offer. The payload is not completely fixed in place and could separate from the fuselage given the right amount of force. If this happens, the center of gravity can be shifted causing the stability of the UAV to fail and also cause collateral damage to the fuselage. If this were to happen, the UAV will most likely crash due to no control. It would work great in perfect flight but if the UAV encountered turbulence, the payload could cause the damage. Also, if the plane were to yaw or roll too much, the shift of weight can incur as well.

The screws would be the most secure method to hold the payload but would not allow for the quick removal of the payload from the fuselage. This needs to be considered since the UAV needs to be able to fly with and without the payload and the payload will be removed and inserted multiple times. This will create issues such as new holes needing to be drilled or the screws could start to strip after multiple uses. This is a great addition to using some other mounting systems though to supply additional support.

Stability

Table 3. Selection of build type (1 worst, 10 best)

The stability of the payload needs to be ensured due to the sloshing of the liquid payload. The payload must hold a payload of 25 to 35 fluid ounces, meaning that the container will likely not be filled fully and will have slosh, leading to a changing center of mass and forces on the walls of the payload and thus fuselage. In order to reduce these effects, different systems can be put in place. Below is a description of the different systems and why each one should be considered.

A plunger system can be used to create a container of variable volume. The plunger could be extended or pressed in to make sure that the liquid payload has just enough room to fit into the container without excessive movement throughout flight. The plunger would also allow for easy removal of liquid from the container when necessary. Once the plunger is removed, the fluid can then be released. This plunger system can be incorporated into a container such as a long tube thus the tube can still be then mounted at the center of gravity to ensure that the stability or controls of the UAV do not change as well. A concept for this design can be seen in Figures 7 and 8, in which the ridges in the plunger are designed to fit rubber gaskets that would stop any water from leaking and secure the plunger in place.  

Baffles can also be used to prevent sloshing, although it will have a lesser effect than the plunger. Baffles break up the container into different sections in order to reduce the forces exerted by liquid motion while having space for the liquid to move between sections and stabilize the center of mass. If the UAV goes into a roll or yaw this prevents the heavy force of sloshing as stated above, but the fluid can still flow into different areas and will cause the center of gravity to change. This will create an unstable flight until the fluid transfers back to its original positions.

A sponge could be used to absorb the liquid inside of the container and keep it from moving inside of the payload. This system would be the most stable during flight but it presents a new challenge when removing the liquid form the container. A plunger system would likely be necessary to compress the sponge to make it release the liquid payload. Even once this happens, there will be residual liquid left in the UAV and will not be considered “completely empty”. The sponge will also take up lots of room and lead to the need of a large container to hold the liquid payload.

Test tubes kept in a tray is also a candidate. Multiple small containers put on the UAV will allow for all of the containers to be full. This will lead to a stable flight with no shift in the center of gravity. This idea however will add quite a bit of weight and also will be extremely time consuming when adding the liquid payload or removing the liquid payload.

From the research done above on the mounting options and the container/stability options, we can likely conclude what will be the most appealing candidate for the payload selection. Using a plunger system in a container of any shape will work the best to hold the liquid steady. This will more than likely be held down by clips that are secured to the fuselage with screws. From our research we can see this may be the best selection but depending on other selections, from other parts of the UAV, we can always adjust if necessary.

Fig 7. Payload

Fig 8. Plunger in

Material Selection

Table 4. Selection of material type (1 worst, 10 best)

The material of the Payload container plays a crucial part in the design of the UAV. Depending on which material is selected, it can alter the way the UAV flies in many ways. The weight needs to stay low; it needs to be a durable container to ensure that there are no faults with spillage, and it needs to be able to be made or bought fairly easily. This will help us choose the final material of the payload container.

The plastic choice for material comes out with the highest score for selection. The reason for this is that plastic is very light and can be manufactured really easily. If the container is not bought, the team will be able to make the container from a 3-d printer which is super helpful. Being able to produce the container will be really cost effective and make it easier to design the container to fit in the UAV design that we choose.

The glass option is not the worst idea ever but can lead to a few issues. This would have to be bought. This would drive up the price due to glass options usually being pricier. Glass is also a very fragile material. If the plane were to land hard or experience hard jolting, the glass could crack and cause spillage of the liquid. This would be a major failure in the UAV.

For carbon, this would be a great pick if the cost of carbon materials weren't so high. If the team were able to produce a container made of carbon, the cost could get driven down. The reason for carbon being one of the best is it is extremely lightweight. This will reduce the weight of the UAV and also not make such a drastic change in weight when removing the payload from the UAV.

The best material for selection will be a pvc pipe container. This will be the easiest to make. The team will be able to produce a plunger style lid and also form a cylindrical design due to the pipe being round. This will fit smoothly into the fuselage of the UAV. The pvc pipe is also a very light material and also very malleable. This will make for a very slight change in weight as well when the payload is removed to not affect the flight of the UAV as badly. This material also comes in at the cheapest cost for all materials and can be easily purchased at any hardware store.

Motor, Propeller & Battery Selection

When choosing a propulsion system for a UAV, there are many different factors that go into it. First of all is the force of the motor itself, and there are a few different methods of determining this metric.

First is to look at the maximum Wattage that the motor can handle. This will tell you not only the strength of the motor, but also how much flight time you’ll get out of a battery. A proportion we have found that seems to be somewhat accurate is 9.76 W/oz, with an assumption of total weight of the aircraft of 5lbs. for the weight of the UAV and a targeted thrust to weight ratio of at least 80%, that gives us an estimated 625W, at least, for the appropriate maximum wattage of a motor.

Another parameter to look at when choosing a motor is the Kv value of the motor, This is essentially the rpm/V. So with a given Voltage from the battery, you can determine what will be the rpm of the propeller. Ideally, our Kv value is within 500-1000kv.

Lastly, we need to consider what propeller we are going to use in combination with the engine. This one is a bit tricky since there is not a really good metric for how to determine the size and pitch of the propeller, but it is hugely important. Thankfully most motors come with a user manual that recommends a range of propellers that are appropriate for that engine.

With all these values, you will be able to use other programs, such as: https://www.ecalc.ch/ (fig. 9) to try to model the plane, motor, propeller, controller, and battery combinations to try to determine which one is best for our UAV. Considering all the given information and the simulation used, we were able to come to the conclusion for two engine recommendations in combination with a 4000mAh to 10000mAh battery:

        E-flite Power 32 (recommended for 3.5 - 6lbs.)

https://www.horizonhobby.com/product/power-52-brushless-outrunner-motor-590kv-4mm-bullet/EFLM4052A.html#

• Smaller motor, good if we have a lighter than expected plane
• Recommended Prop Range: 11x7 to 14x10

E-flite Power 52 (recommended for 5 - 7.5lbs.)

https://www.horizonhobby.com/product/power-32-brushless-outrunner-motor-770kv-3.5mm-bullet/EFLM4032A.html#

• Bigger motor with more torque
• Recommended Prop Range: 14 x 7 - 15 x 8

Fig 9. Eclac.ch

Fuselage Design

Initially the fuselage selection process involves determining the method by which the plane will sustain the stresses and forces involved in the aircraft flight modes. The choices of skin are compared below in the figure to determine the best case for the design. The total effect of drag on the craft is calculated last, for further use in developing the UAV, and to determine required thrust to counteract the opposing drag. Ultimately, the use of monokote over a set of load-bearing struts to form a skin will be the method employed for this design, to save on weight and manufacturing cost. The design selection process was detailed and shown in table 5.

Table 5. Selection of Fuselage (1 worst, 10 best)

Each type of Skin

Monocoque, involves putting stress on the whole frame or skin of the aircraft, as the skin sustains the load for the flight

Non-monocoque, structural design allows for the stress to be carried by trusses and struts, with a skin wrapped around. This method will likely be employed in our design as the second-best option to us.

A semi-monocoque, or a mixed build structure is recommended since it allows a balance of load carrying durability (important for payload carrying planes) and saves on weight in equal measure.

Methodology

The selection of using a non-monocoque structural design, while not the highest ranked option in the chart, was made due to the necessity of manufacturing the system in parts. A semi-monocoque system of design would be a method employed in most modern aircraft design, but due to constraints in manufacturing and ease of design, the non-monocoque method is used for our plane.

Balsa is easily shaped into the required trusses versus using a block of material and cutting into a solid shape. A solid block of material used as the fuselage would be wasteful in terms of total weight of the aircraft, necessitating the use of struts and tresses. The usage of Balsa as a material for the fuselage is due to being the most readily available material for use by senior design students, while also being lightweight and easily workable.

Ideally, we want to minimize drag as well, so the outside frame can be rounded off, whether by hand or cut by machine. The coefficient of drag for the fuselage was calculated using the material provided by our advisor to determine Cdmin (fuselage). This is done for each part of the aircraft to determine the total effect of drag on the plane. A sample calculation will be provided for in the appendix. The ultimate weight of the frame can be decreased as needed to fine tune the center of mass of the aircraft, and reduce drag as much as possible.

Structural analysis/CFD will be performed once the flight conditions are determined to optimize this process, and make sure that the form of the craft can withstand the flight conditions. This involves using CFD to determine the resultant drag force on the craft once the prototype is finalized in Solidworks. As it stands, we have the frame and the airfoil/wing/tail designed in Solidworks as a whole assembly.

With All of the above in mind, an initial model was made in Solidworks to get a sufficient idea of the aircraft to be made. Figure 10 is shown below, with some center of gravity values calculated by Solidworks located in the appendix.

Fig 10. Solidworks design Fuselage and Wings

Tail Selection

The Main focus of the tail is to provide stability for the aircraft, and thus is designed last to provide a correct amount of stability and ensure that the aircraft can maintain trimmed flight. It is important to design the tail last as it is an integral part of the balancing of the plane. In order to facilitate this, the tail design is subject to change depending on final analysis results and testing. Each tail could be designed using the equations in figure 11 to optimize their sizing initially.

Figure 11. Formulas for tail design

Where:

S_W = Wing Area

C_MAC = Mean Aerodynamic Chord

b = wing span

S_HT = Horizontal Tail Area

S_VT = Vertical Tail Area

L_HT = Length between the aerodynamic centers of the wing and horizontal tailplane

L_VT = Length between the aerodynamic centers of the wing and vertical tailplane

One initial design selection for the tail is to have two booms connected to a rear mounted tail like below in Figure 12.

Fig 12. Initial tail design

Another is to use a 3-pronged tail with one vertical stabilizer and two horizontal ones on the sides. This design path would lessen the effect of the weight of the stabilizers on the center of gravity, and is the lightest option of the three design paths.

The final design consideration for the Tail is a T-Type tail mounted at the rear akin to a cargo style plane, which would provide a large vertical stabilizer with two horizontal ones attached at the top.  The major disadvantage of this sort of design is the weight which would greatly move our center of gravity towards instability.

Table 6. Tails comparison

With these specific advantages/disadvantages in mind, It is likely our team will either go for the boom type style aircraft or a 3-pronged style tail for the aircraft to ensure stability. The sizing of the boom style will be tricky to stabilize, but allows for decent control, whilst the 3-pronged tail will provide stability but allow for manipulation to better stabilize the aircraft.

Controls Selection

While the controls themselves do not need selection since radio controls are fairly ubiquitous, there are a few things that need consideration. The aircraft itself will be manipulated and controlled through the use of a controller that is connected to a receiver on the aircraft. This receiver will then be powered by an onboard battery and then connected to each of the servos manipulating the ailerons rudder and elevators. This setup will look fairly like figure 13. In addition, an emergency cutoff switch is installed to ensure safety is considered in case of a failure.

Fig 13. Controls 

The battery itself will be considered based upon its ability to draw power and provide it to all the components of the system. Based upon the engine choice, at the minimum we need to provide for ~800 watts of power for the engine itself plus some for the servos, the ESC, and a receiver.

Landing

Landing is obviously very important, with that in mind there are several options to go about designing a landing method for our aircraft. We can choose to land the aircraft through a variety of methods, but some of the most viable are a wheel configuration or skids. The basic requirements these wheels need to follow include withstanding impact, rolling across an airstrip for landing/takeoff while holding aircraft weight, minimizing drag impact, and maintaining configuration under stress/ repeated loading.

        In terms of selecting wheels/vs skids, there are several factors at play. The drag impact of skids/skis is much less than wheels on an aircraft, but they degrade with use, while wheels are good for many repeated light impacts. Another consideration is their effect on aircraft stability, the skids would have to be sized so as to not throw the center of gravity off, while the wheels have a comparatively lower weight in comparison. The best of these options for us is going to be determined once the weights/ center of mass is more solidified. When selected, the landing materials are going to be bought, procured online/using the senior design inventory. There are several options available to us through the senior design inventory, and landing gear/skids are not too expensive.

7. Engineering Analysis and Calculations

The wing and airfoil analysis was done in XFLR5 and XFOIL software. XFOIL is a program used for 2-D, viscid and inviscid analysis of subsonic isolated airfoils. The program allows the user to specify and adjust the airfoil characteristics, set a fixed or varying Reynolds number, set a forced or free laminar boundary layer transition to a turbulent one, and find the movement of the transition point with varying angles of attack.  It also allows the user to read and write airfoil coordinates and polar save files. The program uses a high-order panel method and a fully coupled viscous/inviscid interaction method when finding the lift and drag characteristics, as well as finding and displaying the boundary layer separation. As long as the boundary layer remains attached to the airfoil, very good approximations of lift and drag can be predicted just beyond CLmax. A standard use of n=250 number of panels, and a leading to trailing edge density of 1 will be used for all analysis in XFOIL for this project. The 3 airfoils selected for analysis were the Clark Y, Eppler 423, and Selig 1223. The analysis was done using the viscous solver in the “.OPER” menu of XFOIL, used Reynolds numbers of 215,000, 322,000, and 429,000, which correspond to speeds of 10 m/s, 15 m/s, and 20 m/s respectively, with a 12 inch chord length. Just like XLFR5, XFOIL uses the lifting line theory (LLT) based off of thin airfoil theory to estimate the airfoil as a lifting line that increments the vortices shed along the span trail behind the wing in a straight line in the direction of the free-stream velocity. The “classic” LLT is a linear relation between the lift coefficient, Cl, and the angle of attack, ⍶. When these thin-plate theory equations are linear, they can be solved analytically, but generally, the lift curve is non-linear, so analytical solutions are no longer viable.

        XFLR5 is a program that is derived from Drela’s XFOIL, but has a much more user-friendly GUI. The analysis for the wing was done using the lifting line theory (LLT) in XFLR5 at velocities of 10 m/s, 15 m/s, and 20 m/s. The 3-D wing models were designed using the plane analysis in XFLR5. This was done with all 3 of the airfoils. Shown below are the Cl vs Cd, Cl vs Alpha, and Cl/Cd vs Alpha plots generated at a velocity of 20 m/s. Some of the lower angles of attack would not converge after 200 iterations when using a velocity of 10 m/s, 15 m/s, or 20 m/s.

8. System Concept Evaluation Plans

As a result of this design conceptualization and the work we have done to start realizing this project, we’ve accomplished several major goals over the course of this project. Firstly, we’ve established the design of this fixed-wing aircraft and developed an initial model to work towards going forwards with the analysis and iteration stages. In addition, we’ve gained a marked improvement in understanding of the design process as a whole and how to better implement our ideas into a real-world application.

        In the future, especially with this design going forward, we plan to further design and complete a prototype by the end of next semester and bring it forwards to the design competition, and aim to win or perform admirably. In addition, our goal is to maximize our efforts and come forward with a clear, optimized, and well-built plane that can manage each of the tasks of the competition with relative ease.

9. Cost estimates

The cost for the UAV can be broken down into each of the different selection categories. For the payload, if made out of the PVC pipe and using O rings, the cost can be estimated. The cost for the payload will also have to factor in the mounting. For all these parts, we estimate that the payload will account for about $25 of the overall cost of the UAV. For the wing, tail, and fuselage, this we will have to factor in the price for the material to build. For the surface area calculated above, the cost will come out to about $200. As far as the remote, controls, motor, and battery, this will be received from the UCF inventory and will not cost the team anything. Some wire may have to be purchased as well. For miscellaneous parts and goods, the team estimates that it will cost around $100. The propulsion and battery setup will also be received from the UCF inventory and not cost anything. The last bit would be landing gear. This will add up to about $35 with the current fixed selection. The overall cost of the UAV will run the team somewhere between $350-$450 depending on new additions that we add later on or a change in the design.

10. Significant Accomplishments and Future Work

This project has solidified the use of low speed, low Reynolds number airfoil and wing simulation and analysis. This also includes application of 3-D modeling in CAD software such as SOLIDWORKS and Wing Helper. With the use of computer simulation of the Clark Y, Selig 1223, and Eppler 423 airfoils and then comparing it to accurate already published models was fulfilling to learn. This same feeling of accomplishment was attained when comparing the 3-D simulation of and design of the wing with theoretical values calculated by hand, and being able to verify that the math was solid.

All members of our group are in the Aerospace Engineering discipline, so this work will likely contribute heavily to future work we will do in industry. Also, with this background knowledge of fixed wing UAV design, manufacturing, and testing, they will be invaluable to us in the future. Designing the interior and exterior wing sections, along with the fuselage and tail sections were all big accomplishments for each of the members of our group responsible. The hand calculations of estimated lift and drag on the wing section will be extremely valuable to use in the future, for being able to explain why we picked specific components for the project.

11. Conclusions and Recommendations

In this project our group is going to make a UAV to carry a 25-35 fluid ounce payload and use a pusher propeller to try and get as many laps possible in our competition. For this, we must prioritize high lift to carry the payload and low drag for better efficiency.

With this in consideration, we took a look at 3 different airfoils. The first is the Selig 1223, which has amazing lift, but really high drag and is really hard to manufacture. We then took a look at the Eppler 423 which was similar to the Selig 1223, but was easier to manufacture. And after analyzing and comparing the 3 airfoils together we came to the conclusion that the Clark-Y airfoil has the best mixture between, high lift, low drag, and ease of manufacturing. We then decided that we are going to make the interior wing out of a sheet of balsa wood that is lazer cut and then assembled and wrapped with monokote. This makes for a strong, yet light wing design.

For the payload, we have 3 different mounting options: clips, screws, and friction holding. Clips would be more secure than the friction, and more accessible than the screws . Screws are the most secure, but also take more time to add and remove the payload. And friction being by far the most accessible, but provides little to no security of the payload. After looking at our options, we decided that clips are the best method of securing the payload to our plane. To reduce the sloshing of the payload, there are a few different methods such as: plungers, baffles, sponges, and test tube trays. Each of these methods have their own advantages and disadvantages, but we chose the plunger as our method of reducing slosh in the payload bay. Lastly, is the material of our payload, which we found our best option to be made out of PVC piping due to its ease of use, weight, and accessibility.

Next thing we looked at was the motor and propeller combination. We used our estimated weight for the UAV (5 pounds) and programs like ecalc, to try and pick out a motor and propeller that suited our aircraft. We came to the agreement that the E-flite Power 32 & E-flite Power 52, along with their recommended propellers, were going to give us enough lift to weight ratio for our aircraft to fly stable for the competition.

Finally, we looked at the tail of the aircraft where we decided to use a unique design with 2 booms connected to the bottom of the wings, going back to the tail where there is going to be a horizontal stabilizer connecting the two booms. We then placed two vertical stabilizers on the booms which provide us with yaw stability and control.

With all these parts together, this brings us to our final design of our aircraft that was modeled using SolidWorks, we will eventually be able to run a finite element analysis (FEA) on the model to determine where the loads are, and how they are distributed and also to give us a good idea of what our plane will look like. Using all the components we have chosen, along with good testing, manufacturing, and assembly, we will be able to put together a plane that flies and carries a 25-35 fluid ounce payload successfully around the circuit. FOTO DE ABAJO ANADIRLA

12. Appendix

Fig 14a. Weight Estimations for the fuselage, ~chord length 1ft, wing span 5ft

Figure 14b. Weight estimations Panel

Table 7. Clark-Y Xfoil analysis (Re = 1*10^5)

Table 8. Clark-Y Xfoil analysis (Re = 2*10^5)

Table 9. e423 Xfoil analysis (Re = 1*10^5)

Table 10. e423 Xfoil analysis (Re = 2*10^5)

Table 11. S1223 Xfoil analysis (Re = 1*10^5)

Table 12. S1223 Xfoil analysis (Re = 2*10^5)

ABET Matrices

Table 13. Aeronautical and/or Astronautical Topics Utilized in this Senior Design Project

13. References

Clark Y. (n.d.). Airfoil Tools. http://airfoiltools.com/airfoil/details?airfoil=clarky-il

Drela, M. (2000, December 11). XFOIL 6.99. MIT.edu.

Eppler 423. (n.d.). Airfoil Tools. http://airfoiltools.com/airfoil/details?airfoil=e423-il

Fahad, E. (2021, March 13). RC Plane Designing Calculations, Making, and Flight Test. Electronic Clinic. https://www.electroniclinic.com/rc-plane-designing-calculations-making-and-flight-test/#:~:text=So%2C%20for%20this%20particular%20RC%20plane%20the%20Aspect%20ratio%20is%205

Karthik, M. A., Adithya, H. V., Purantagi, A. M., Hari, R. K., & Kumar, C. V.(2018, May). Analysis and Selection of Airfoil sections for Low Speed UAV's. International Journal of Latest Engineering Research and Applications. http://www.ijlera.com/papers/v3-i5/7.201805099.pdf

Online aerospace resources. AeroToolbox. (n.d.). Retrieved April 24, 2022, from https://aerotoolbox.com/

Selig 1223. (n.d.). Airfoil Tools. http://airfoiltools.com/airfoil/details?airfoil=s1223-il

WingHelper. (n.d.). Wing Helper Home. https://www.winghelper.com/default/XFLR5. (2003, November 4). XFLR5.tech. http://www.xflr5.tech/xflr5.htm

XFLR5 Analysis of foils and wings operating at low Reynolds numbers. (2009, October). College of Engineering - Purdue University. https://engineering.purdue.edu/~aerodyn/AAE333/FALL10/HOMEWORKS/HW13/XFLR5_v6.01_Beta_Win32%282%29/Release/Guidelines.pdf

“eCalc - Reliable Electric Drive Simulations.” eCalc - Reliable Electric Drive Simulations, www.ecalc.ch. Accessed 24 Apr. 2022.