Symmetrical airfoil for lightweight 3D trainer?

[ADB2]

Hi,
I am building yet another iteration of my Owl-RT 3D trainer. The wingspan is 806mm (31.7") and root chord is 250mm (9.84"), tip chord is 196mm (7.72"). The model AUW including a 3S LiPo is 280g (9.9oz).
Wing loading is 15.6g/dm2 or 5.10oz/sq.ft.
I have tried two KF airfoils (easy to build and good performance, but some issues I am still sorting out) until now and am still researching the matter, but I would like to try a built-up symmetrical airfoil for comparison.
My design constraints are:

Easy to build using 3mm Depron skins.
Good for 3D (whatever that means!) and aerobatics.
Thick enough so I can use a 9mm wood spar.
Tolerant of building inaccuracies.
Tolerant at high AoAs/slow speed flight.

Any suggestions would be very much welcome. BTW I am a total beginner when it comes to aerodynamics, so please if you can explain your choice, that would be great! :)

[Brandano]

When you hover the profile won't really matter much, if you aren't looking for high speeds or good glides, the profile won't matter much.. give it a fairly thick wing with nicely rounded leading edges, anything behind the maximum thickness is pretty much irrelevant, though you might want to add turbolators to get consistent air flow over it. And perhaps some stall strips on the inboard LE to mitigate any tendency to spin, though on a straight wing this is unlikely. Just have a look at the older fun-fly designs, the wing profile usually looks like something drawn with a compass rather than out of a set of coordinates.

[ADB2]

When you hover the profile won't really matter much...
Thanks, I fully agree with you about the airfoil not mattering much in a hover! :)
I was thinking about the airfoil behavior at high AoAs and near stall, as I can't seem to get a clear explanation of thick vs. thin and sharp LE vs blunt LE for a 3D design.
I think I'll be trying a modified NACA 0009 profile at the tips with a sharper LE, with constant spar height (i.e. getting thinner at the root), and see how that flies.

[JetPlaneFlyer]

I'd go with a well radiused 'blunt' LE... This gives better stall behavior than your suggested sharp LE

[ADB2]

I'd go with a well radiused 'blunt' LE... This gives better stall behavior than your suggested sharp LE
Hi,
Yes, that's what I thought too, so as you can see on the picture above the wing has a well-rounded LE.
But then I read that the airfoil for a 3D model must be able to stall "cleanly" (don't ask me what a dirty stall is, or a clean one, I have no idea), and that a sharp LE would provide this "clean" stall.
My guess is that a sharp LE will cause the wing to stall at a well-defined AoA over a wide speed range, whereas a blunt LE, while stalling at higher AoAs than the sharp LE, will stall at different AoAs depending on speed.
This is just a guess and I have no idea how this changes the "feel" of a 3D model in flight, when doing 3D maneuvers.
I am willing to experiment with different airfoils exactly to sort out this kind of claims. :)

[JetPlaneFlyer]

Generally a airfoil with a rounded /blunt LE will have a wider operating range than a sharp edged one so they should be better for your purpose... However funny things happen at very low Re numbers where conventional airfoil theory can break down.

The 'can't fail' method: Take a look at a successfull model that has the flying properties you are looking for and copy it's airfoil ;)

[ADB2]

...
The 'can't fail' method: Take a look at a successfull model that has the flying properties you are looking for and copy it's airfoil ;)
Yep, I tried going that route too. However it seems that there is no general consensus in this field except the blanket statement that "flat sheet airfoils are good enough for indoors 3D" or similar statements to that effect. :(
Also checking on the few existing designs with similar wing loading that use airfoils and reproducing them is not going to help me understand why they fly the way they fly, unfortunately.

[bwalt822]

All comments are made below are in a general sense, there are always exceptions.

Thin vs Thick: The thinner, the less drag there is as well as less lift at a give AoA

Blunt vs Sharp TE: Sharp TE will stall at lower AoA.

For a given airfoil, stall is determined by AoA until you get down to really low reynolds numbers. I'm not sure if RC aircraft fall into those numbers or not.

If you want a "gentle" stall, go you could put some washout in the wings. Washout means to purposly warp the wingtips downward. This causes the AoA at the tips to be less than at the roots. This means the wing will first stall at the roots and as the plane slows down the stall will progress outwards more gently and you will still have some aileron control.

[bwalt822]

Here is an interesting graph that I found. I havent seen a lift vs AoA plot for the full 180 degrees of motion but here it is.

http://www.aerospaceweb.org/question/airfoils/q0150b.shtml

Stall by the conventional definition is that first very sharp dip. Lift then starts increasing again due to the wing deflecting more air downwards. It seems that a lot of 3D flight is done around this second round peak. You will notice how fast drag shoots up after the first dip.

[ADB2]

Hi,
Thx, these are indeed very interesting graphics. I noticed one thing though, it's that they were drawn for Reynolds numbers above the typical ones for RC models.
AFAIK for RC models Reynolds numbers in the range of 30 000 to 300 000 are applicable. Probably 3D maneuvers like harriers are executed at the low end of this range. So I am not sure how exactly these graphics would apply.

[bwalt822]

Thats true, they are for very high Reynolds numbers. And at a low enough Re # a sphere will not have separation. I have no theory to back this up and very well could be wrong but my intuition tells me that the initial stall would not be as harsh at low a Re #.

Relatively, even at RC model Re #'s, a flat airfoil will stall sooner. My intuition also tells me that after stall, it doesn't really matter what type of airfoil you have since most of the lift of an airfoil at 45 degrees AoA probably comes from air deflection and not pressure difference.

[JetPlaneFlyer]

If you want a "gentle" stall, go you could put some washout in the wings.
Not really a good idea for a '3D' aerobatic model because such a model needs inverted performance to be just as good as 'right way up' performance ( washout becomes washin when inverted :( )

I'd agree with bwalt822's last post that for manoeuvres that are done with the wing beyond the stall AoA then airfoil selection is irrelevant, which is probably why a flat plate works as well as anything in these conditions.

[ADB2]

Hi,
Thanks guys for your comments and tips. I agree with you both that beyond the stall AoA, airfoils behave essentially like a flat plate, so a built-up wing with an airfoil would behave essentially like a flat sheet wing.
I have also come to the conclusion that for a 3D model I really want a symmetric airfoil and wing, as dealing with asymmetrical behavior when the model is in flight increases the pilot's workload during 3d maneuvers, which are already a handful by themselves.
I am thinking that a thin airfoil is OK because with a very low wing loading, I don't need gobs of lift, and I would actually prefer that the model fly level even when accelerating, and not need to trim the elevator at different speeds.
As you can see in the top view of the Owl-RT, the LE is straight, and since the TE is tapered, the wing is actually forward swept, like on the Extra. I also build with a constant thickness spar, so the wing gets a thicker % airfoil at the tips than at the root. I believe that as a consequence of these two factors, I haven't noticed any serious tip stall tendencies in the Owl-RT, upright or inverted.
I still have no idea about the relative advantages of sharp/blunt LEs for 3D flight, but since the models I have built until now had blunt LEs, I am willing to try a sharp LE for a change! :)
I'll be building two models probably, one with a symmetrical airfoil and another one with a flat sheet wing, just to see how different they behave in flight!

[bwalt822]

Good point on washout being bad for inverted flight. I forgot about that