# What airfoil design is needed for a 4 m wingspan soaring plane?

I’d like to build one that is heavier than a F3J plane and it will be similar to NASA’s soaring plane of the "Autonomous Soaring Plane" project. What kind of airfoil do I need? Simply scaling a normal airfoil would have bad results.

• I suggest you ask this on rcgroups.com in the glider forums. Lots of F3J and F3K designers hang out there. The glider crowd is generally very friendly and helpful. Commented Nov 24, 2014 at 6:23

First for those who wonder what F3J is: This is the FAI model aircraft category for thermal soaring.

Quick answer: This answer describes what resources are available on the Net.

Your choice of airfoil depends on the wing loading and the size of the model, because those will determine the Reynolds number of the flow over the wings. Soaring flight requires high lift coefficients, but maybe you also want low drag for sprints between thermals or to avoid being blown away by strong winds.

The description of the NASA Autonomous Soaring Plane gives 15 pounds of mass and a wingspan of 170 in. It was based on a model glider by RnR Products, and there I found the aspect ratio of 19.8 and the wing area of 1545 sq.in. These numbers are not very consistent, but close enough to get an idea.

Thermals are strongest in their center, so it is desirable to fly as tight a circle as possible. Low wing loading $\frac{m}{S}$ and a high lift coefficient $c_{l_{max}}$ drive the minimum turning radius $r_{min}$ down (at a given density of air $\rho$), and the formula is: $$r_{min} = \frac{\frac{n_{z_{max}}}{\sqrt{n_{z_{max}}^2 - 1}}}{c_{l_{max}}\cdot\frac{\rho}{2}\cdot \frac{S}{m}}$$

Here $n_{z_{max}}$ is the maximum load factor, and the picture below shows that little is gained by increasing $n_{z_{max}}$ beyond 3. The little DFS Habichts help to illustrate the bank angle.

With a wing loading of 6.8 $kg/m^2$ and a mean wing chord of just 23 cm the flight speed at a $c_l$ of 1.2 is 9.5 m/s. At sea level in a standard atmosphere the Reynolds number would be 150,000. The wingtip flies at a Reynolds number of below 100,000. In circling flight with a bank angle of 60° (equivalent to 2g, so you need twice the lift), the Reynolds number would increase to 212,000 for the mean chord and to 140,000 for the tip. If this sounds a lot to you, it isn't. This is a flow which needs highly specialized airfoils.

My recommendation is to try the UIUC airfoil database and select a specific model airfoil with low drag at high lift coefficients and Reynolds numbers down to 100.000. It should have high camber and a thickness of around 12% to keep wing weight down. If you have the lower range of Reynolds numbers covered, you need not to worry about the higher range. You might even consider to use different airfoils over the wingspan, depending on the minimum local Reynolds number. Ideally, you'll find a set of airfoils which have been optimized for exactly this purpose, like the DAE11 - DAE21 - DAE31 airfoils of the Dedalus human powered aircraft.