Airfoil selection

[facewindu]

Hi guys
I'm a beginner in wing design and I really don't know how to choose the airfoil for my model.
My plane will be based on the Slo-V from Parkzone, but the weight will be slightly increased. I only want a plane with a very good L/D ratio and forgiving behavior for stall. I don't need to do crazy maneuvers.

I want to change the wings to fit my needs (overweight, energy saving)
My plane should fly at around 7 m/s (stall speed 4.7 m/s)
The aspect ratio will be around 6.3 (span 1.5 m)
My plane will weigh around 0.6 kg
I already chose a planform for my wings. Each wing will be made by 4 panels to give an approximate elliptical planform. I know the dimensions of each panel. I just have to find an airfoil, see if its thickness is big enough for my spars.
What do you think about twisting it a little to increase the stall behabior near the tips ? (is it easily feasible when assembling each part ?)

thanks !

[vintage1]

Go for clark -Y - its easy to build - and use whetaver depth you need for strength of spars etc.

Definitely put washout on it at that aspect ratio.

[raptor22]

Why is your planform so limited? If you want an efficient airplane with high L/D, a high aspect ratio is key.

--Alex

[Brandano]

Better LD doesn't always make for a bettter flying airplane, expecially for trainer types. Lower AR planes tend to behave better in a stall, and that's not something to sneer at when starting out. Even then, as Vintage points out, the AR is high enough already to require washout. And an elliptical planform goes some way to improve the LD as well (and to worsen the stalls, I might add)

[Sail 'n Soar]

Hi guys
I only want a plane with a very good L/D ratio and forgiving behavior for stall. I don't need to do crazy maneuvers.

IThe aspect ratio will be around 6.3 (span 1.5 m)

I already chose a planform for my wings. Each wing will be made by 4 panels to give an approximate elliptical planform.

What do you think about twisting it a little to increase the stall behabior near the tips ? (is it easily feasible when assembling each part ?)

thanks !

You appear to have a totally inconsistent set of requirements. If you want good L/D with forgiving stall behavior, avoid an elliptical planform like the plague - with our without washout. If you want similar L/D, but with forgiving stall behavior increase the AR to 7 or 8 with a constant cord wing. You can also increase the AR to ~7 with a simple taper with C,tip/C,root ~.6, put in about 5 degrees of washout and achieve better L/D AND still have great stall behavior.

Your AR 6.3 choice says you aren't all that worried about L/D anyway. How do you define "very good L/D"? Is this primarily an academic exercise or do you have a real goal in mind.

In terms of airfoil choice, go with vintage1's Clark Y recommendation; flat bottom for easy construction, reasonable L/D and Cl,max at representative model Reynolds Numbers.

But what ever you do, don't go for the 4-panel elliptical planform - unless your primary objective is a neat looking wing (vs. a good performing one).

[facewindu]

You can also increase the AR to ~7 with a simple taper with C,tip/C,root ~.6, put in about 5 degrees of washout and achieve better L/D AND still have great stall behavior.
I can try this, it seems to be a reasonable compromise.
By the way, do I need to put washout even if I don't have ailerons ? because my only control surfaces are on my tail (V-tail)
Your AR 6.3 choice says you aren't all that worried about L/D anyway. How do you define "very good L/D"? Is this primarily an academic exercise or do you have a real goal in mind.
It's part of an academic problem in which we have to reduce the energy consumption because we can only use one charge of the battery to perform our task (carry balls)
In terms of airfoil choice, go with vintage1's Clark Y recommendation; flat bottom for easy construction, reasonable L/D and Cl,max at representative model Reynolds Numbers.
Seems that everybody agrees on that. Thanks.

[BMatthews]

For a given wing area and weight the model will have to fly at pretty near the same airspeed regardless of what airfoil you choose. That's the plain and simple physics of it.

To get the model to fly as slow or nearly as slow as the Slo-V you need to make it with the same wing area and same or near to the same wing loading. Wanting it to be other than that is just not going to happen.

So the way to do this and to carry a load at the same time is to use advanced and intelligiently chosen building design and techniques to produce a Slo-V sized airplane that is lighter so that the payload won't adversely increase the wing loading much past the stock Slo-V's wing loading.

The good news is that the SloV uses one of those molded GWS tent airfoils and that it's made from relatively heavy materials and construction choices. It would not be hard to make a SloV sized model that is both more efficient and lighter. But it may be hard for you if you don't have any experience building with balsa and choosing to use free flight like structures for the building.

The other good news is that with brushless motors and lipo battery packs the power options can be a lot lighter and more powerful than the stock SloV.

So to work from there what experience with building and materials do you have?

I'm also going to run counter to the Clark Y suggestion in this case and say you will want a thin and undercambered airfoil such as a Benedek BE6356B or BE8405B. To see what they look like check out Profili2 at www.profili2.com . It's well worth the download even for the freebie version.

[portablevcb]

For a high lift cargo carrying platform that will stay at relatively slow speed I would agree with BM on the undercamber airfoil.

I use a Clark Y (or a modified version that is completely flat bottom) type on almost anything that will fly in a 'standard' manner, ie, scale type airplanes, for all the reasons stated above.

charlie

[JetPlaneFlyer]

Would this be a university project by any chance?... It sounds similar to a contest a university student I talked to a couple of weeks ago was involved in here in the UK.
If I remember correctly a 'standard' motor and battery is supplied to all competitors and you have to build an airframe of 1.5M span max. The object is to transport a load of balls round a course in the shortest time possible (or as many balls as possible in a set time... I can’t remember)

Assuming I'm correct, it’s a race to move as many balls as quickly as possible, on the energy available from one battery. Ideally therefore you need a model that will fly fast, be manoeuvrable, easy to fly, carry a big load and be efficient. In the real world with a fixed power source there will be a trade off between speed and load carrying ability. Generally a model that can carry a lot of load at moderate speed and make fewer round trips will use less power than a model that goes like stink but has low load capacity and therefore has to do more journeys to move the same load from A to B. Assuming that you can design a fuselage capacious enough to carry lots of balls then I'd tend toward big load carrying capability over outright speed, the slightly slower 'heavy lifter' model will also be easier to fly than the hot rod, and if you have to make fewer journeys you have a statistically lower chance of a mishap.
I'd maximise the wing area, which with span limited by the rules, means using a wide chord and low AR, use a parallel chord wing for simplicity, no washout should be required on a low AR parallel chord but perhaps a little washout (no more than two deg) may help reduce induced drag slightly and give 'extra safe' stall behaviour without reducing the ultimate lifting capacity too much.
Choice of airfoil is tough, the BE airfoil suggested above probably will give greatest outright lifting capacity but will be draggy when speeded up. I'd probably stick with the suggested Clark-Y simply for ease of building. If you have time try building a few alternate wings and test them back to back.
Probably the most important factor would be to keep it simple and build light, every gram saved in the airframe will be an gram extra load capacity.

Good luck

[MarkusN]

I'd probably stick with the suggested Clark-Y simply for ease of building. Seconded. Also for controllability at higher speeds.

[mnowell129]

For a given wing area and weight the model will have to fly at pretty near the same airspeed regardless of what airfoil you choose. That's the plain and simple physics of it.

I think this is a little misleading. This assumes that all airfoils have the same Clmax, which they don't (for example with multiple slotted flaps the Clmax may be >> 1).
It also ignores the fact that the Cd at that Clmax is the same for all airfoils, which it isn't.
As this is for a design competition with fixed power and the goal is to carry the largest payload, efficiency (Cl/Cd) is definitely important.
I think you should do some preliminary sizing on the aircraft, get an estimate on the reynolds number and start looking at airfoils that, at that reynolds number, have a high Clmax with gentle stall behavior and lowest Cd numbers. My guess is that Clark Y is not going to be the best choice. As a start, I'd go to the Selig numbers for low speed airfoils.

[BMatthews]

I think this is a little misleading. This assumes that all airfoils have the same Clmax, which they don't ......

A little misleading perhaps but I was trying to simplify it by assuming a range of airfoil options from the ClarkY to the thinner undercambered airfoils where the Cl max are not all that different. And to point out that the wing loading is going to play a larger part in determining the speed and load ability of the model. Of course ideally both factors along with other things would be optimized.


If this is the same rules as JPF has shown us then that alters the picture. In that case my gut guess is that there's a happy middle ground where the model would carry a happy amount at a reasonable speed. A speed high enough that the model has good airflow over the surfaces to encourage prompt and crisp control but not so fast that it's intimidating to fly in tight confines. The model would be built as light and "airy" as practical with a quick access fuselage for ball loading and unloading but one that is decently aerodynamic to help avoid energy consuming drag so that the electrons can be used to lift the model rather than push it.

[mnowell129]

...but I was trying to simplify it by assuming a range of airfoil options from the ClarkY to the thinner undercambered airfoils where the Cl max are not all that different.

I understand your intent but my opinion is that I don't think its something you can ignore.

Picking two sections that were convenient to me:
Eppler 374 (A clarkYish section) has a Clmax of about 0.8 at Re = 60000
Selig S-2091 has a Clmax of 1.3 also at Re 60000.

Both have similar drag numbers at Clmax (about 0.04, yielding section L/D's of 20:1(E374) and 32.5:1(S2091)).

This gives the 2091 a 62% ((1.3-0.8)/0.8) increase in lift over the E374, something you can't ignore.
Further since the drag numbers are the same the Selig has a 62% better L/D, thus the Selig section aircraft will lift 62% more payload for the same propulsive thrust.

I think airfoil selection for this project is very important.

[BMatthews]

I'll have to dig it up but I seem to recall the Clark Y as being good up to around 1.2 or so.

Of course if the Re is around 60K then all bets are off and you're quite right about that being a very significant amount.

I've tried to find a ClarkY curve and google isn't helping me much so I can't say what the Cl would be. In a couple of hours I'll have my old Selig Soartech in my hands and can look at the curves.

Either way I'd like to stress to facewindu that even the best airfoil selection doesn't mean you can ignore all the other factors. It's the whole package that counts. Light weight, good airfoil and a "clean" airplane. Oddly enough the basic square wing with perhaps a raked back wing tip works really well at our slower model flying speeds. So at least you can simplify the wing construction.

[vintage1]

I am by no means an aerodynamicist of any note, but what I HAVE noticed from building a few old timers is this.

If you want to run on the lowest possible power, use the largest possible wing.

Sink speed times weight is power needed to stay aloft: sink speed is airspeed times D/L.

So you can either improve L/D OR simply lower airspeed. A BIG wing is the easy way to do that.

High L/D large aspect ratio wings tend to be vicious. Thats a fact. Maybe a super design can help, but IME thats pretty much the way it is. Undercamber - its like flaps, the model flies slower, but it can be draggy at higher speeds and WILL snap at you.

Plus these floaty things area pain to land. My ideal low power model is slow with a fairly low L/D ratio,. That gives it slow sink rate due to its speed, but a steep glide ANGLE so it an be brought down in tight spaces. Its also got either a lot of washout or a fairly low aspect ratio, so it doesn't snap in tight turns.

Sailplanes are not becessarily what you want. Its important for those to be able to trade height for speed and speed for height, to get from one thermal to another. With an engine, all you want is low speed and low sink rate to do load carrying with. Big light wing. It will be deep to make it strong, and light, rather than for any crucial aerodynamic reason.

[mnowell129]

Either way I'd like to stress to facewindu that even the best airfoil selection doesn't mean you can ignore all the other factors. It's the whole package that counts.

I agree 100%.

[kral27]

for my MISS2 by simply building a NACA 6409 wing at a span of 61 inches and a chord of 9.5" (approx as I build in metric and am running short on time so if you want the details feel free to ask).

To make a long story short, this arrangement flew very well and the control throws just need a little more extras in it if you wish to give more authority.
Tollerance is high to stalls and roll out as gentle as the original wing. Tight turns are exceptional near the ground and have noticed that it takes the tricky breezes just a little nicer than the original wing.

[JetPlaneFlyer]

I understand your intent but my opinion is that I don't think its something you can ignore.

Picking two sections that were convenient to me:
Eppler 374 (A clarkYish section) has a Clmax of about 0.8 at Re = 60000
Selig S-2091 has a Clmax of 1.3 also at Re 60000.

Both have similar drag numbers at Clmax (about 0.04, yielding section L/D's of 20:1(E374) and 32.5:1(S2091)).

This gives the 2091 a 62% ((1.3-0.8)/0.8) increase in lift over the E374, something you can't ignore.
Further since the drag numbers are the same the Selig has a 62% better L/D, thus the Selig section aircraft will lift 62% more payload for the same propulsive thrust.

I think airfoil selection for this project is very important.

I'd like to jump to the defence of the good old Clark Y here.
The Clark Y actually has almost identical Cl max and Cl/Cd as the S-2091. Take a look at the attached graphs. Cl max for both airfoils at Re=60,000 is 1.3.

Both the above airfoils perform far better in terms of Cl max than the low camber E274, but that’s to be expected. The E-274 wins hands down on inverted performance which is why it's a popular choice for sports models but that's not an issue here. Using the E-274 as a 'ClaykYish' airfoil is highly misleading because the two foils really have little in common in terms of geometry or performance.

The S-2091 does admittedly have a marginal performance edge on the Clary-Y, but in practical terms the Selig is let down by its wafer thin TE which would be all but impossible to do in balsa rib/spar construction. The Clark Y is also a little thicker overall which means it could be built lighter for the same strength as a Selig wing.

Given a choice I’d take the Clark Y over the Selig due to ease of construction.

[facewindu]

Hi guys, sorry to answer just now. I didn't know that so many people were involved.
Thanks everybody for your answers.
Anyways, my Re number will be around 100k. (is it close enough to 60k to use your conclusions ?)

A clarkY seems a good choice (easy to build and good perf for my use).
For the planform, I will use a rectangular one near the fuselage and add a tapered panel towards the tip to improve a little bit the perfo.
dimensions of my wings will be 1.5m*0.25cm (or close).
my gross weight will be around 615g (1.37 lbs) and it won't change.
With a wing area of 0.33 m^2 (3.54 ft^2), my wing loading is 1.8 kg/m^2 (0.38 lbs/ft^2)
How does it sound ? should I increase my area ?

[MarkusN]

dimensions of my wings will be 1.5m*0.25cm (or close).
How does it sound ?
If you take out either that c there or the decimal point it sounds good :D

For a wing of this size not too much taper at the tips. I'd say goint down to about 20 cm at the tips would give a wing that's not too critical yet.

[Brandano]

I'd build a biplane, with wing struts. I have a feeling that having a lighter structure and more wing area will ofsset the increased drag. Seems to work fine for the AN2

[JetPlaneFlyer]

Anyways, my Re number will be around 100k. (is it close enough to 60k to use your conclusions ?)


Here are the 100k Re graphs... they look very similar to the 60k

[facewindu]

thanks JetPlaneFlyer for these graphs.