Thread Tools
Jul 22, 2010, 03:30 PM
Registered User
Thread OP

Looking for airfoil advice

I'm currently designing a UAV for a school project, but would like to ask some of you airfoil experts for some recommendations on airfoil selection.

The current design is a 2m span, 0.4 m^2 wing area (20cm chord, AR=10). Total weight is expected to be about 4 kg, putting the wing loading around 10 kg/m2 (33 oz/sq.ft.). The span and chord can be adjusted as required to lighten the wing loading if required. Cruise speed is 50-60km/h based on mission requirements.

From what I've calculated, cruise CL is around 0.1, which seems a bit low to me based on what I've seen elsewhere. So, it would seem to me that decreasing the wing size would be necessary, but that increases the wing loading, which already seems high (although I'm not quite sure what I'm basing this feeling on).

The aircraft should be as efficient as possible to maximize range/endurance, so I was looking at a glider airfoils, such as the E205 and SD7037, which seem to be fairly common. However, I'm not sure if these are suitable given my cruise speed and wing loading.

From what I know, I want to find an airfoil that will provide my cruise CL at minimum drag (Cd), with no sudden stall tendencies, minimum pitching moments to decrease tail size. What I'm looking for are a few suggestions to narrow my search.

The proposed manufacturing method is a CNC'd foam core with a carbon-fiber skin, and embedded spars (if necessary), so things like camber are not a problem, and the CF skin should be able to cope with a thin airfoil, but could cause problems if the TE is very thin.

Any help is appreciated...
Sign up now
to remove ads between posts
Jul 22, 2010, 05:37 PM
Registered User
I'd advise you to go look at CL vs alpha and drag polars in your aeronautics textbook. I know that in mine there's a good 50 or so airfoils plotted out back there, mostly NACAs.

Also: have you seen this website?
Jul 23, 2010, 01:30 AM
Registered User
John235's Avatar
Generally the modern glider airfoils are designed to be used with flaps. I think you will find they they do not necessarily offer the best performance at CL=0.1 without using some reflex flap.

You should also consider the landing speed of the model. At 33oz/sq' it will be like a large power model, more than a glider. That will be ok if you have an experienced pilot and an easy landing approach. An efficient glider airfoil and high aspect ratio may count against you unless you have some type of drag device to slow the model down on the final approach. You could end up with a very long and fast landing approach, so a lower aspect ratio and less efficient airfoil could help in slowing the aircraft on the landing approach.

I would suggest using an airfoil such as NACA2412. Even though it is an old airfoil, it will be quite efficient in the cruise situation with low CL at the speed you are looking at. The efficiency will be poorer at low speed, but I think that is actually a good thing because it will help slow the model down. If you want to reduce the stall speed of the model, perhaps you could increase the wing area a bit and maybe drop the aspect ratio to around 8.
Jul 23, 2010, 09:02 AM
Registered User
Thread OP
I would really like it to fly more like a glider than a large power model, which means decrease the wing loading. However, this will further decrease that CL value to something very small, below 0.1. Is that ok? It just seems out of the ordinary, where every airfoil selection process had their CL cruise around 0.3-0.4. I guess what is causing this is that we're trying to fly the airplane "unconventionally fast" for this size of airplane, since the cruise CL is the inversely proportional to the square of cruise speed. The cruise speed is a firm requirement in order to allow us to complete the mission in the alotted time.

Wind penetration is also important, so having a higher wing loading is good in this sense.
As for flaps/spoilerons, these shouldn't be a problem to implement.

Hawkwings - thanks for the link, looks like a good database. Basically I am looking for a couple airfoils to start looking at and compare. There are hundreds to choose from, and narrowing it down to one is difficult without a hint as to which ones to concentrate on.
Jul 23, 2010, 07:30 PM
Registered User
John235's Avatar
There's absolutely no problem using CL smaller than 0.1, it just means you need to study the airfoil properties in a different part of the curve, since you stated that drag is a major priority for you. It sounds like you are quite undecided still on the specifications for your aircraft so I wouldn't spend too much time for the "best ever" airfoil until you have nailed it down a bit better.

So far the only specific requirement you stated so far is a desire to maximise the cruise efficiency. On that criteria, I can offer 3 more airfoils for you to input into your deisgn process. E207, SD6060, NACA1410. Each of them have different characteristics, but its hard to say if any are really suited to your project at this stage.
Last edited by John235; Jul 23, 2010 at 07:37 PM.
Jul 23, 2010, 11:16 PM
Registered User
Well if you're flying with a tiny CL, that means tiny induced drag, awesome. Just watch that parasite drag though.

Are you trying to maximize range, endurance, or speed? Or some combination of the three?
Jul 24, 2010, 03:47 AM
B for Bruce
BMatthews's Avatar
I think you need to go back and check your Cl value calculations. Using Foilsim for a quick check I'm seeing your 32'ish mph cruise at the weight you gave as being more like .5'ish.

And I hate to break it to you but there is no way your UAV will be sailplane like at that much area and size and at that weight. In fact 4kg for a 2 meter "glider" is really very overweight to the point of being intensely overweight.
Jul 24, 2010, 03:31 PM
Registered User
Thread OP
Thanks again for the advice. The intended use is for a "search and rescue" type mission, basically flying a grid pattern over a defined area. We are given an alotted time (about 1 hour) to complete the mission. Based on camera resolution, cruise altitude, field of view, etc, we get our required cruise speed to complete the mission on time (50-60km/h). So, there's not much room to play with flight speed. Increasing the speed too much makes it difficult to spot targets.

I don't expect it to be a "glider" per se, but want be able to cruise efficiently. I figured sailplane airfoils would be a good place to look. The less power we need, the less battery weight we need. On the other hand, good stall characteristics and the ability to slow down for landings (which will be in unprepared fields) are also important.

We have a general estimate for the aircraft weight of around 4kg. I say estimate because it will depend on how much funding we can get to buy the equipment we need. 4kg is heaviest-case scenario. After looking at similar UAVs, such as the Aerovironment Puma, the wingspan for this size of aircraft is around 2.5m, with an aspect ratio of about 10. So, increasing the wingspan will help decrease the wing loading.

DOH! I found the reason for my really low CL....I was using kg as my lift force instead of Newtons, hence being off by a factor of 10 (9.81 actually). Wow, that's embarassing...

Okay, so with a weight of 4 kg, span of 2m, AR=10, cruise at 50km.h (14 m/s), I get a CL of 0.8. That makes much more sense. Changing the span to 2.5m, I get CL = 0.53. That's a bit better...
Jul 24, 2010, 08:52 PM
B for Bruce
BMatthews's Avatar
If you're looking to play with some various sizes and get some quickie answers do look into Foilsim. Google for it and it'll take you to the NASA website where it lives. A highly useful and informative tool for basic lift vs speed and area of a plane at any size. It's far nicer to use that than a page and a half of equations to do the same thing....

Getting back to your project I would suggest from a lot of years of flying models that a 4 kg 2 meter model is not going to want to land at a speed consistent with survival in a lot of unprepared fields. At that mission weight I think that I would aim for more like a 2.5 to 2.8 meter span with the same 10:1 aspect ratio. That will lower the wing loading enough to make those unprepared field landings a little more doable. You may also want to look into providing flaps to aid in slowing and steepening the UAV's approach at a more managable speed that will also greatly aid in shortening the landing skid. Even to the point of looking at adding a small drogue chute to further slow and steepen the landing approach for the last two hundred feet of approach once you're commited to landing. Both the flaps and the chute would greatly aid in boosting the chance of a non damaged arrival and quicker re-deployment as a result. Incorporating an energy absorbant nose skid would further increase the odds of an undamaged arrival. Also along this survivable landing design I would plan on a fairly deep fuselage with a long nose along with nylon shear bolts for the wing so that if it should hook up in a bush on landing that the nylon bolts will shear away and allow the wing and fuselage to part company easily. Along this line the wing to fuselage joint should be a smooth one with easy break away electrical connections to the wing mounted servos to allow for better survival chances and a quick return to operational status via replacement of the nylon shear bolts.

Best of luck with the project.
Jul 24, 2010, 09:12 PM
Registered User
No body is going to believe this but--. I have Compufoil and have tried a kazillion difference airfoils from the Drela to some really exotic unknown ones over the years.

So full circle - I just designed and built a new 48" sport plane. I used a Clark Y and this plane has the best handling of any thing I built in the past several years.

Thick airfoil makes a super strong wing. While the thin airfoils are faster they also are stall prone and goosie.

For your application use the Clark Y. No surprises and you won't be sorry.
Jul 25, 2010, 12:22 AM
Registered User
John235's Avatar
Originally Posted by Aaron 1
For your application use the Clark Y. No surprises and you won't be sorry.
^^ I was going to post the same recommendation, now the CL requirement has been updated.
Jul 25, 2010, 09:00 AM
Registered User
Thread OP
Thanks again for the advice guys. It really is appreciated. I had my eye on the Clark Y from the beginning, but wanted to know if there was anything more efficient. I want to have something built within the next two months, so I don't want to spend too much time worrying about the ultimate airfoil. I've used Xfoil a bit, but I'll check out Foilsim.

I'm pretty set on the 2.5 span now that I looked at other designs. It also gives a little more room in case things are a bit heavier than we expected. It will be a 2 or 3 piece wing, so the extra span isn't a big deal for transport. I've been thinking about a net recovery for landings in rough terrain, but durabilitiy is definitely important. Good advice on the "frangible" joints. I think we have some Kevlar laying around that might be handy.

I've been going back and forth between a 3 and 4 channel setup, i.e. ailerons or no ailerons. It appears that many commercial UAVs in this size range have only rudder, with dihedral at the wing tips to allow rolling - like an EasyStar, which will be our autopilot testbed. The obvious drawbacks are the lack of flap/spoiler function, and lack of independtent roll control. The advantage is a simpler control system, less servos, and less weight. Plus the wing can be installed and removed with no need to connect any wiring. I'm leaning toward the aileron option, but we may try both wing types just to compare. Does anyone know of a resource for choosing a proper dihedral for a 3ch setup? I'm sure I could figure out the math, but a more "experienced-based" approach would be valuable.
Jul 25, 2010, 01:45 PM
Ascended Master
Sparky Paul's Avatar
Ailerons permit better cross-wind pointing with the help of the rudder.
The extra servos don't add that much weight.
For your search pattern, flying a straight line will be important.
A tricycle gear is best.
The imager.. will it be pointable?
Will the operator be able to see where it's looking, or just guess?
Net recovery will result in damage.
Jul 25, 2010, 01:45 PM
B for Bruce
BMatthews's Avatar
The way to go with the dihedral is that if it looks like a 3 channel model then it'll fly OK. The more the dihedral the stronger the rolling action from a given yaw angle will be. But if you overdo it then the rudder gets too touchy and the design tends to react to turbulence more strongly. Just make it so it sort of looks like the Easy Star and you'll be golden. If all this makes it sound like there is no formula to find an ideal angle that is because this is definetly the case. The range of angle that produces acceptable results is very wide. Just don't make it too flat and don't make it look like a gliding pigeon.

For a simple area coverage where course changes are mild and well planned you could do a LOT worse than the self stabilzing effect of a dihedral wing. It then becomes one axis that you don't need to worry about controlling. The only issue would be the camera image if it is tied to the plane's airframe. The constant upsetting from turbulence will generate nausea in the strongest stomach. But even ailerons would not help that. You'd still have lots of possible image steadiness issues with any sort of fixed platform camera. A stabilized mounting frame for a camera would be an ideal solution.
Jul 25, 2010, 10:41 PM
Registered User
John235's Avatar
I previously searched for a better optimised version of the Clark y. I wasn't able to find anything that was really better overall. I found that increasing the thinkness of SA7036 to around 11% gave a useful increase in lift in Xfoil simulation at low reynolds numbers around 200k. It has lower drag than the clark-y, but the maximum CL isn't quite as good. For your poject I would stay with the Clark y because it is a well proven airfoil and the extra CL might be useful too.

Quick Reply

Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Help! Looking for a 3D EPP Plane w/airfoil <32" Wingspan typhoonmaster 3D Foamies 3 Apr 30, 2010 08:38 PM
Help! Looking for cord for SN26 airfoil RicVaughn Slope 11 Dec 04, 2009 10:31 PM
Question Looking for a thermal glider with AG40-AG43 airfoil pstoehr1964 Thermal 18 Aug 28, 2009 08:10 PM
Discussion Looking for airfoil advice Sleedo Slope 16 Jan 26, 2007 12:38 AM
Discussion Looking for Airfoil Advice for a Wingeron Design Utahrd Slope 11 Nov 20, 2006 05:38 AM