Airfoils
Airfoil selection is one of the earliest and most important decisions made when desiging your UAV. The airfoil will play a large role in determining its aerodynamic performance characteristics and capabilities. As such, care should be taken to choose an airfoil that properly meets the performance requirements for your mission. Relevant requirements may relate to stability, lift and drag characteristics, and manufacturability. These and other important concepts will be discussed.
Airfoil Geometry
To start, let’s discuss the different parts of an airfoil. As you can see below, an airfoil typically has a thin, long profile. This provides large amounts of lift with minimal drag. The front (or forward most part) is called the leading edge, while the back (or aft most part) is called the trailing edge. If you were to draw a straight line from edge to edge, this would be called the chord line and its length is the chord length (or just “chord”).
NACA Airfoils
The NACA four-digit series provides a convenient way to demonstrate some additional airfoil parameters. Each digit describes a different parameter of the airfoil. Let’s use the NACA 4412 airfoil as an example. The first two digits describe the magnitude and placement of the camber, respectively. In this case, there is a 4% maximum camber at 40% of the chord length from the leading edge. The last two digits give the maximum thickness, which is 12% of the chord length here. Most small UAV’s have a thickness-to-chord ratio of about 8-14%.
To explain camber, we refer to Figure 2. The camber line lies exactly (vertically) between the upper and lower surfaces of the airfoil at any point. Camber is the relative distance between the chord line and camber line, measured in chords. Generally, a more cambered airfoil will provide more lift with decreased aerodynamic stability.
Lift and Drag Performance
One of the most important considerations when selecting an airfoil is the lift and drag characteristics. These and a few other metrics will indicate the airfoil’s performance. We will focus first on lift. Keep in mind that with airfoils we are calculating a theoretical coefficient of lift (Cl). The coefficient of lift is a nondimensional parameter that helps us to understand the general lifting capabilities of an airfoil (see equation below). This is different than lift itself, which is simply the force acting on an aircraft perpendicular to the free stream. Lift is explained in further detail in fluid mechanics.
There are a few key characteristics you are looking for that will be shown on your lift and drag polars. Let’s look at the polar that plots the coefficient of lift as a function of angle of attack (α) for various airfoils. The first thing you should notice is the general shape of the polar. There is at first a linear relationship between the Cl and angle of attack. As you increase the angle of the airfoil relative to free flow, the coefficient of lift will increase with a slope of about 2π (Cl/α in radians). At around 10-12 degrees, the airfoil begins to stall, meaning that large amounts of drag are beginning to occur and lift is being lost. This is represented on the polar by a peak and eventual downward curve (see Figure 3).
Some airfoils have a higher peak than others, signifying a higher maximum Cl. This is advantageous because it allows the UAV to take off with smaller lifting surfaces and increases overall aerodynamic efficiency. Here, the airfoil represented by the dark blue line obviously has the highest Cl max (about 1.6). Another interesting feature is the shape of the peak. A gentler downward curve represents an airfoil with a more forgiving stall development. For example, lift will be lost more gradually on the airfoil represented by the green line than the light blue line, which drops sharply and will lose lift very quickly after beginning to stall at Cl max.
Let’s take a look at another polar. This one shows the coefficient of lift (Cl) as a function of the coefficient of drag (Cd). This is useful mostly in helping us to determine the maximum lift over drag ratio, or L/D. If you draw a straight line from the origin upward, then pivot it until it touches the polar, the slope is the airfoil’s theoretical maximum L/D. The lift over drag ratio is a common way to determine the optimum flight scenario for your UAV because it provides the highest efficiency in terms of lift and drag.
This last polar is more concerned with the general stability of the airfoil. Here, we plot the coefficient of the pitching moment (Cm) as a function of the angle of attack. Notice that all of the lines are generally flat, but have varying magnitudes. This magnitude has to do with the general lateral stability of the airfoil. A higher magnitude will result in a less stable airfoil, but is usually accompanied by better lifting capabilities. In any case, this data should be used as more of a tie-breaker when selecting an airfoil rather than a key requirement.
Comparing Airfoils
When selecting an airfoil, it is useful to compare at least 10-15 different airfoils to have sufficient numbers for a good selection. This can be done using XFLR5’s airfoil direct analysis feature or another program. It is good practice to create plots with an angle of attack ranging from -4 to 16. Some airfoils may require a larger range to see the full curve before the stall point is reached.
One additional consideration that should be made is the general shape and curvature of the airfoil. If lift and drag characteristics alone are considered, you may be stuck with an airfoil that is thin, cambered, and difficult to manufacture. A thin airfoil will often be difficult to create with the foam cutter, and will experience frequent breakage. As an example, let us consider the two airfoils in Figures 6 and 7. In this case, the DAE-31 airfoil has a higher Cl max and better stall and drag characteristics. The NACA 4412 airfoil has somewhat comparable characteristics, but is significantly thicker (especially at the trailing edge) and easier to manufacture. This makes it the preferred selection in most cases for small UAVs.
The UIUC Airfoil Database
When it comes to getting data for UAV airfoils, the UIUC database is likely the best public resource available. It boasts a collection of over 1,500 different airfoils with various characteristics and applications. You can view plots of the different airfoil shapes as well as download a text file containing the data itself (typically 50-100 points). Many of the airfoils appropriate for small UAVs are labeled “low Reynolds number airfoil”. This essentially means that they are designed for aircraft that fly at relatively low speeds. A couple of good series to begin looking at are the Eppler and Selig series.
Appendix: Formatting Airfoil files for XLFR5
In order to be able upload airfoil data files to XFLR5 for analysis, there is a specific format that the data points need to be in. Take a quick look at the above airfoil plots again. Notice the normalized coordinates. In XFLR5, an airfoil must be plotted starting at the trailing edge (x = 1), go over the top to the leading edge (x = 0), then curve back around the bottom and end at the trailing edge again. Many of the airfoils in the UIUC database are already formatted this way, but some need to be switched around a bit. The ideas covered here are specifically for Windows users, but will have relevance to students using Macs. Let’s look at a typical mixup situation, as shown below.
0.000000 0.000000
0.012500 0.019300
0.025000 0.031700
0.050000 0.051300
0.075000 0.066400
0.100000 0.078000
0.150000 0.093400
0.200000 0.101300
0.250000 0.104400
0.300000 0.104800
0.400000 0.100200
0.500000 0.090500
0.600000 0.077100
0.700000 0.061000
0.800000 0.042800
0.900000 0.022900
0.950000 0.012400
1.000000 0.001600
0.000000 0.000000
0.012500 -0.005000
0.025000 -0.004200
0.050000 -0.001000
0.075000 0.002800
0.100000 0.006800
0.150000 0.014500
0.200000 0.021700
0.250000 0.028200
0.300000 0.033300
0.400000 0.038500
0.500000 0.038600
0.600000 0.035000
0.700000 0.028600
0.800000 0.020200
0.900000 0.010000
0.950000 0.004400
1.000000 -0.001600
Notice that the points begin at the leading edge, move over the top to the trailing edge, and then jump back to the front to do the bottom. Not only will this result in a line right through the middle of the airfoil, but XFLR5 will be unable to work with the data file for analysis. The first thing you want to do is copy the data points from the text file given on the UIUC site, and then paste them into an Excel spreadsheet. At first, all of the data will be in one column. In order to easily rearrange the data, we need to split it into two columns. This can be done using the Text to Columns function under the “Data” tab. Under the data type, select “Delimited”, then push “Next”. Since this data is space delimited, check the box for “Space” and then click “Finish”. With a few minor adjustments, your data should be in two columns now.
The next step is to sort the first section that goes over the top. We want this to go from back to front instead of front to back. This can be done using the Sort function, also under the “Data” tab. Highlight the first section, then click the button to sort. You will want to sort by Column A (which is the x positions) from “Largest to Smallest”, then click “OK”. Your data should now match up with zeros in the middle. Go ahead and delete any empty rows and the extra row of zeros. Your data is now properly ordered.
As a last step, we need to put the data points into a DAT file type. To do so, copy the data from Excel into a text editor such as Notepad, and then click File»Save As. Go ahead and name your file, making sure to change the file type dropdown to “All Files”. Then simply put “.dat” at the end of your file name and save. Now your airfoil is ready to be uploaded to XFLR5!
