Help! Ultra-low Speed UAV Airfoil Selection

[Yavuz]

Hi, I'm designing an ultra low speed UAV for a dogfight locking compatition.So I have to keep flying at very low speeds while being pretty maneuverable (and stable since camera for locking should be stable and OIS banned).
Total weight of UAV estimated around 3kg, also we can go up to 3mt wingspan and 1mt cord lenght.
I found washout and wing fences will help with stall.Also, reflexed airfoils will solve most of the stability issues, generate lift at high aoa and keep laminar flow nice and clean.So I found MH45 but I'm not that sure about that. I'm interested in your experiences and suggestions.Thanks in advance!

 

[Piotrsko]

One of the ancient caveats in all my airfoil books says their wind tunnel data is not reproducible at scale sizes or speeds. SO, intelligent airfoil profile choice for models becomes a guessing game. Most just pick one they think they like, typically flat bottom because they are easier to build on a table but I have seen bulbous symmetricals and huge undercambers. Large flat thin boards fly remarkably well, are easy to build, but not structurally strong in bigger sizes. Foam board relies on the strength of the paper covering.

There is no substitute for wing area other than thrust. More of either is better at slow speed but walking speed is kinda the lower limit where you transition to prop thrust for flight.

 

[Yavuz]

One of the ancient caveats in all my airfoil books says their wind tunnel data is not reproducible at scale sizes or speeds. SO, intelligent airfoil profile choice for models becomes a guessing game. Most just pick one they think they like, typically flat bottom because they are easier to build on a table but I have seen bulbous symmetricals and huge undercambers. Large flat thin boards fly remarkably well, are easy to build, but not structurally strong in bigger sizes. Foam board relies on the strength of the paper covering.

There is no substitute for wing area other than thrust. More of either is better at slow speed but walking speed is kinda the lower limit where you transition to prop thrust for flight.

Thank you,this one will produced with carbon fiber, so strenght is not a problem.I see undecambers like S1223, but i feel like it will introduce a lot of drag. I found FX 63-137, which has a pretty high Cl and performs well at high AOA as far as I can understand. What do you think about it?

 

[clolsonus]

There is no substitute for wing area other than thrust. More of either is better at slow speed but walking speed is kinda the lower limit where you transition to prop thrust for flight.

I didn't spend time finding the best resource, but here is a link discussing how to compute stall speed: https://www.quora.com/How-do-you-calculate-the-stall-speed-of-an-aircraft and it appears to be correct.
If you simplify the formula to just think about the relationship of values you can control in your design, the stall speed is proportional to:
sqrt[ (mostly_constant_value * aircraft_weight) / (wing_area * Clmax) ]
So if you look at the formula: as aircraft weight increases, stall speed increases. As wing area or ClMax increase, stall speed decreases. A 50% change in any of these values would yield the same stall speed change as a 50% change in any of the other values.
For one of my airplanes the stall speed comes out at about 19 mph. If I improve any of these things under my control (weight, wing area, Clmax-via-airfoil) by 50%, then the stall speed drops to about 16 mph.
So ask yourself which is easiest to improve by 50% (or 20% or whatever percent) in your design: 50% wing area increase? 50% weight reduction? or finding an airfoil with 50% better Clmax?
If your goal is lowest stall speed, I would suggest your best energy should be spent optimizing wing area and weight in your design, and then you can circle back around later and try to find a better airfoil, and with that (maybe?) squeeze a few more percent out of the stall speed ... but at that point, the improvements may not be discernible with whatever instrumentation you might have to measure these things in flight. (pitot speed gets really noisy on the low end ... and there's more math I could pull up to show why.)
[ Edit: everyone seems to want to spend the first 1/2 of their project energy by evaluating and picking some optimal airfoil ... and I want to push back against that ... if it was my project I'd start with a clark-y and then do all the other stuff, and come back later and see if it's worth trying to find a more exotic airfoil. ]

 

[Merv]

....I see undecambers like S1223, but i feel like it will introduce a lot of drag....

You are correct, undercambers fly very slowly but do produce a lot of drag. There is no wing shape that does everything well. Each airfoil is a compromise of strengths and weaknesses.

The Clark Y is a good compromise worthy of consideration. If you want slower flights, consider a Clark Y with a slight undercamber.

 

[Piotrsko]

Oh mama, you went and invoked the ancient aero gods with your clark-Y comment. An airfoil that doesn't do anything well except generate drag.
Ok for purpose, a proper selected undercambered doesn't necessarily produce more drag at or near stall than a high angle of attack efficient something which is kinda the point. At much higher speeds yuppa, it literally sucks. Probably why you don't see them much on modern jet aircraft, or, to think about it, modern gliders. But you do see all sorts of camber changing stuff hanging down when modern stuff lands, they can afford all the monkey motion brackets that make that camber change work. Birds ditto. Btw fastest falcon uses undercambered thin airfoil, which explains the whole Gottengen series.

still not a fan of computer simulation. I have had situations where the CAD said something would do the task and reality said "No Thanks, go away" and vice versa. Still a great tool to avoid wasting materials if you aren't at the edges of the envelope.

 

[Yavuz]

Oh mama, you went and invoked the ancient aero gods with your clark-Y comment. An airfoil that doesn't do anything well except generate drag.
Ok for purpose, a proper selected undercambered doesn't necessarily produce more drag at or near stall than a high angle of attack efficient something which is kinda the point. At much higher speeds yuppa, it literally sucks. Probably why you don't see them much on modern jet aircraft, or, to think about it, modern gliders. But you do see all sorts of camber changing stuff hanging down when modern stuff lands, they can afford all the monkey motion brackets that make that camber change work. Birds ditto. Btw fastest falcon uses undercambered thin airfoil, which explains the whole Gottengen series.

still not a fan of computer simulation. I have had situations where the CAD said something would do the task and reality said "No Thanks, go away" and vice versa. Still a great tool to avoid wasting materials if you aren't at the edges of the envelope.

Thanks, I feel clark-y in same way.It flys at normal planes anyway, but its not ideal for anything as far as I understand.I don't trust CFD too, so we will make a prototpe out of very thin (2 layer) shell with UW PLA and sections + acrylic pipes.
I came up with this. FX 63-137 for wing itself, MH45 for prop-wash area,winglets,wing fences and 5 degree washout with 5 degree aoa,1.7m^2 wing area.AIs says it can fly at 4m/s (I suck at physics ).