Thundergull

[aerogant]

I came across Earthstar Aircraft's Gull series of light planes and was highly impressed. The thing that puzzles me the most is how they are able to get such a slow stall speed out of a small wing and a reasonably high weight. I've contacted the designer and he insists that stall speeds were measured at gross weight. In every review of this plane that I have read, the stall speed given was really close to the specs given by the company. I know they use a 40% plain flap. Any thoughts on how they get so much out of that wing?
The specs are given here:

http://www.thundergull.com/specifications.htm

I get an overall Clmax of almost 3.1 when running the numbers. How on earth does it get this value? Thanks.

 

[mstull]

Aerogant,

I too have been very impressed with Earthstar's planes, since I first saw one at the EAA Arlington, WA airshow a couple decades ago. Their fully cantilevered wing is stiff and strong... the best and most efficient design in the U/L industry. I've heard it called "the hardest working wing in the industry."

Titan aircraft copied or bought the wing design, and is using it on their planes. And I heard some other company recently copied it. People who have flown the planes are extremely impressed with their crisp handling. And they require less engine horsepower because of the lack of struts and their aerodynamic efficiency.

I even used some of that thinking in my designs... make an efficient airframe that only needs a small engine to perform well. For legal Part 103 U/Ls, that means putting more of the weight limit into the airframe, and less in the engine. That leads to crisp handling, strong airframes. Doing the oposite, leads to floppy winged designs.

As far as the stall speed... a fully sheeted (on the upper surface) wing, that has a good airfoil will have a significantly lower stall speed than one that isn't. I've seen STOL wing tips on some Earthstar planes. That helps a shorter wing pass the legal stall speed limit. And shorter wings have a better roll rate, giving a very sporty feel.

Every aircraft design is a compromise. Most U/L pilots want a lot of fun for their dollar. So many U/L designers find the cheapest, easiest way to make an aircraft that is safe to fly. For a legal U/L, most everything has to be compromised for weight. Many want the reliability of a Rotax twin. So everything else on the plane needs to be minimized/compromized to make the weight limit.

I've seen airfoils that are severely compromised, usually to save weight and/or expense. A few common compromises I've seen: No leading edge sheeting (very common); an airfoil that is completely flat on the aft half (for example Kolb); an airfoil with an extra bulge for the rear spar (for example MiniMax).

It's really hard to make a fabric covered wing that has a perfect airfoil. The fabric dips between the ribs. Designers can make the leading edge sheeting come back farther to minimize this problem. Then there's the problem of the aft edge of that sheeting, creating a slight corner. We can add more ribs or false (half) ribs. But that adds weight.

Another problem with passing the stall speed limit is in FAR Advisory Circular 103-7. The wing area prescribed in that Circular is based on the poorly carried out, fabric wings that were common when it was written. It does allow less area, if you officially demonstrate and document the actual stall speed.

A manufacturer, like Earthstar, can go through that procedure, and pass out copies of the document to their customers. For the rest of us, it is often easier to just use the formula in the Circular, and have extra wing area, and the resulting extra low stall speed.

I have made efficient enough wings that stall slower than the Circular's formula suggests. For example, an under-cambered wing has a similar effect as having flaps down, as far as stall speed. The Circular doesn't make the distinction.

Remember that we have to be able to prove that our plane is legal. The FAA doesn't have to prove that it isn't. And usually the FAA won't question your plane's legality until after you crash. So there's no way to demonstrate its stall speed after the fact. Either you have the required documentation on file, or you'd better have the wing area in the Circular.

Looking on the positive side, the extra wing area tends to make the plane climb and glide real well. But on the negative side, the extra long wing tends to have a dismal roll rate. My next U/L wing will be fully sheeted. I ordered the materials for it and will start construction as soon as they arrive. I'm trying progressive spoilerons with this design to try to get a decent roll rate. It will have the extra wing area as prescribed in the Circular.

 

[lake_harley]

I've done a search for the wing area formula you referred to regarding AC 103-7 and didn't find the formula. Could you post a link or the formula?

Thanks

Lynn

 

[George Sychrovsky]

http://www.airweb.faa.gov/Regulatory_and_Guidance_Library/rgAdvisoryCircular.nsf/0/8cf133ad5a5ba4b3862569de005bd75a/$FILE/Appx1-4.pdf

 

[lake_harley]

George....Thanks for the link! I printed it out. The information seems simple enough to make it useable and understandable to a novice like myself. It's good to know what would be asked if the situation ever presented itself.

From just a couple quick "what if" examples I ran throught the formula it takes a lot of wing area to get the stall speed down without proving you can indeed do it with less. In fact, a couple of the ultralight plans I have, that are "103" plans, may not even be able to pass.

Thanks

Lynn

 

[aerogant]

Thanks for the reply Mstull. I have been focusing my attention on slotted flaps, vortex generators, slats, and any other device which might help get big lift out of small wings. To think that a simple 40% plain flap can do that really blows my mind. I've been leaning towards the thin plywood over foam ribs approach. I'd like to think that it should be a little more accurate than fabric covering. How would you approach estimating the lift calculations using such an airfoil? It is my understanding that this airfoil is a modified 23012 w/ a drooped nose and reflexed aft section.

 

[mstull]

Lynn,

Yes, traditionally, a lot of so called U/L planes, kits, and plans weren't really legal. When we could count on the FAA giving U/Ls the blind eye, people took advantage of that in every way. We call them "fat" U/Ls now. But half a dozen years ago, they were just U/Ls. I'm still waiting to hear about a big FAA crackdown. So far it's just a lot of scare tactics.

There were a few planes that were designed specifically to comply with the Circular's formula, but wouldn't really fly that slow. I'm trying real hard not to name names.

You're right. It does take an awful lot of wing area to pass the stall speed limit, without flaps. I used flaperons on my biplane, so I could design the plane smaller. It works. But I miss the climb rate and glide ratio of a big, efficient wing. I wanted to fly my biplane around Moab. But it just doesn't have enough lift to climb well at that altitude. That was my first experiment with flaps.

There's a lot of things that the formula doesn't take into account. It doesn't say how much chord the flaps have to have, or how much they have to deflect. I doesn't account for wing aspect ratio, even though a higher aspect ratio wing will stall much slower. It also doesn't account for different wing tips. You can gain a lot of lift with trick wing tips, especially with large chords.

It's pretty challenging designing a legal U/L wing without flaps. If you add more span than 30', you have to add more weight and/or drag to properly support the ends of the very long wing. So you reach a point where you lose more than you gain. I consider 32' the absolute limit, with the type of structure I use.

If you use a shorter span and longer chord, you end up with a poor aspect ratio, and a pretty extreme reynolds number. But a shorter span wing structure can be made much lighter. You end up with a decent roll rate. 5' chord is my limit.

It really helps if you can keep your dry weight well below the 254# limit. But that usually means having to use single cylinder engines, that often have extreme vibration, and/or poor reliability. And vibration can cause terrible reliability problems.

The wing I'm building this winter will have a 30' span and 4.5' chord, with no flaps. It should climb and glide great. I've never experimented with spoilerons. So we'll see how the roll rate comes out. I love experimenting. This will be my 5th wing.

Good luck with your project.

 

[mstull]

Aerogant,

That's an interesting approach, using slotted flaps and other devices to increase lift to minimize the wing area. Those things all work. I'd sure add some kind of trick wing tips to that list. But none of those things are in the Circular's formula. So to take advantage of them, you'd have to officially demonstrate and document its stall speed, like it says in the second part of AC 103-7. I'd be interested to hear how that goes... finding a couple officials willing to sign your document... using a radar gun on a perfectly calm day?

I wouldn't have the patience to make slotted flaps that are aerodynamically accurate enough. I assume you mean ones that actually move/deflect. And I don't know if the structure to fabricate and support them would add more weight than they're worth. I NEVER use the flaps on my biplane, other than to try them once. I think most pilots of truly legal U/Ls that have flaps, don't use them. If you want to have lift enhancing devices that are really useful, and you plan to measure and document your stall speed, I would skip the trick flaps.

Light weight slats might also be difficult to fabricate and support. One of the keys to a good U/L design is to keep it SIMPLE and LIGHT. You can gain more performance with lighter weight, than with any other modification.

You can really lower your stall speed with a deep under-camber. But it's trickier to cover and fabricate. My under-cambered wing stalled at around 24 mph. So you can gain as much with under-camber as with many other devices put together. The goal is to make the included angle of the trailing edge extremely small. And under-camber doesn't even add drag, like some other devices do. I love the way that wing glides. You can throttle back and float.

Another way to get more lift is to use Junkers style ailerons. I used them on 3 of my wings. They are very powerful. But I struggled with flutter with them. They have to be very rigidly supported, and may need to be counter-balanced.

The reduced wing area allowed in the Circular with short, plain flaps is very arbitrary. I'm not sure how the FAA estimated it. Notice it says "less than 50% span flaps". What if you just had 10% span flaps?

There are computer simulators, like X-Plane that can do all your aerodynamic calculations. I heard there's even a free one. I've always used a Got 387 upper surface, because it is so well proven for U/Ls, and allows plenty of room for spars. It's a very stall resistant airfoil. Most of my wings have used flat bottoms. One was deeply under-cambered.

Even relatively thin plywood is fairly heavy. It has to be thick enough not to dip or lift between the ribs. Be sure the glue and ribs are strong enough that the plywood doesn't rip off at high Gs. I have used it for leading edge sheeting, with mixed success. (I'm much happier with the carbon ones I've made) Be sure to seal it completely, lest it absorb humidity and expand, warp, and buckle. So the sealer adds more weight. I think it would be too heavy to cover both the top and bottom wing surfaces with that. So you'll probably want to fabric cover the bottom surface.

It would be expensive... But you could sheet with balsa. I've had more success with that for leading edge sheeting than plywood. For the sharply curved part near the front, you can use thicker balsa boards that are sanded to the curve on one side.

I'm not quite ready to talk about the sheeting material I ordered for my next wing... mostly because I don't know if it will work. I should know this winter. It could be pretty revolutionary if it works.

Incidentally. On my biplane, I used no ribs at all on the bottom wing surface. The (shrunk Stits) fabric stays flat just fine at legal U/L speeds. I hope some of this helps you.

 

[Dana]

I've seen airfoils that are severely compromised, usually to save weight and/or expense. A few common compromises I've seen: No leading edge sheeting (very common); an airfoil that is completely flat on the aft half (for example Kolb); an airfoil with an extra bulge for the rear spar (for example MiniMax)...

A flat bottom airfoil isn't so bad... the old Clark Y was nearly flat bottomed (and many modify it to make it completely so) and it's still a pretty good airfoil for all around low speed performance.

There were a few planes that were designed specifically to comply with the Circular's formula, but wouldn't really fly that slow. I'm trying real hard not to name names.

The Kolb Firefly is "103-7 legal" (if you can manage to keep it under the 254 lb limit). They even publish both the 103-7 "calculated" and "actual" speeds. It has flapperons, and they added a second wing strut when only one is necessary just for the extra drag to keep the top speed down.

I never could see whey they needed flaps on the Kolbs... until I flew my Ultrastar (which doesn't have flaps). I suspect they put the flaps on for glide angle control rather than for stall speed, since they have so little side area that slips don't do very much... and it's a floater.

...a higher aspect ratio wing will stall much slower...

Aspect ratio has an effect on induced drag, but not on stall speed (except perhaps for very minor Reynolds number effects).

BTW, your monoplane looks pretty cool, but the seating position (foot position in particular) looks pretty awkward.

-Dana

In general, liberalism consists of A & B getting together to see what they can make C do for poor old D.

 

[mstull]

Dana,

I agree completely about flat bottom airfoils. It's a no brainer compromise that I use most of the time. The Got 387 is nearly flat on the bottom too.

I was talking about the upper surface, as far as compromises. And yes, there are some truely legal U/Ls. I didn't say that the compromises were so detrimental that a Kolb couldn't really pass the stall speed limit. This thread is about how the Thundergull could be legal with such small area. And one of the reasons is its perfect airfoil. (The Kolb was not the name I was trying not to name.)

One of the reasons the aspect ratio has such a large effect is because of wing tip losses. Very few U/Ls have tapered wings (for good reasons. But that's another thread). So our wing tips have a relatively huge chord. At our slow air speeds and extreme angles of attack near stall, the pressurized air under the wing has plenty of time to ease around the wing tip to the lower pressure air on top. There are things we can do with wing tips to discourage this. But that's another thread. The longer the chord, the greater the tip losses.

So picture the last 2' of each wing not having any lift with a 5.5' chord (at stall speed). On a 20' total span, that's a wopping 20% of total lift gone. (And you're still getting drag from that part.) On a 32' span with a 4' chord, figure the last 1.5' not having any lift. That's less than 10% of the total lift lost. (These numbers are not at all accurate... they're just to help explain one of the reasons aspect ratio has a large effect on wing efficiency and lift at stall.)

The high reynolds number (of a large chord wing) has a huge effect at the extreme angles of attack an U/L can achieve. The air will break away and stall the aft portion of the wing when the boundary layer gets too thick and turbulent.

Another problem with a high chord wing is that the air flowing over the top of the wing is not pulled down parallel to the upper surface. The boundary layer gets so thick near the trailing edge, that you lose a couple degrees... again at the extreme angles of attack near stall.

These reasons have way less effect in cruise.

I spent months designing and adjusting that cockpit (that I use on all my planes) so it would be totally comfortable. It's like reclining in your favorite easy chair. It's way more comfortable than the pilot's seat on most GA planes. (I guess it must be more comfortable than it looks I did move the foot rest bar forward a couple inches on its latest iteration... mostly for taller pilots. All the seat parts are curved and padded for comfortable support.

 

[lake_harley]

Rather interesting. As I read the formula, flaps with length of 49% of the wingspan buys you the "up to 50% flap" factor, and 51% flaps gets you the next. That's rather a no-brainer I'd say. In any case, unless I'm way off, it would take 110-120 Sq. Ft. of wing area to pass the test formula without being way under 254# empty, or proving what your wing can do.

 

[mstull]

Lake,

That's for sure. But you'll save some weight and have better aerodynamics if you use flaperons. What I'm proposing is have flaperons over 50% of span, and no separate flaps. That way the root 1/3 (or whatever) of the wing span can have a perfect airfoil that comes all the way to a point at the trailing edge, with no flap gap, hinges, or flap spar.

Now the brain power begins... to come up with a simple, light, control system that combines the aileron and flap controls. I looked at other planes to get ideas for the flaperons on my biplane, and couldn't find any that I liked. So I invented my own very light, simple system.

The MiniMax has a clever, simple system, that only works on full span ailerons. I looked at a Kitfox 4, and couldn't belive the complexity and weight of the mechanism, with many large bellcranks, shafts, pushrods, arms, and Heim bearings. I'm sure it was designed that way for good reasons. But it definitely isn't ultra light.

 

[PTAirco]

I personally like the approach of getting the required stall speed using a smaller wing working at a higher lift coefficient. The smaller dimensions really help in keping bending moments etc down to a manageble level or, put it another way, you can get more strength for a given weight of structure. My current project uses an airfoil designed to work with a Junkers-type flap. It is not undercambered, but has a distinctly downturned T.E.. It was specifically designed for ultralights. I intend to use flaperons in the sense that that the surface will be deflected far enough to create enough of a Cl to meet the stall requirements and then fixed at that angle, leaving it to work as ailerons only. If we didn't have the max. speed limit it might be worth making it a 'proper' flaperon, but I don't think it will be requierd for this. The deflection needed is only about 10 -12 degrees by my best guesstimate.
The small size of the wing (4.5' x 12' wing panels, plus tips, plus the seperate flaperons) allows me to use a more conventional wing structure with single C-beam spars and a skinned D-box L.E., all in metal with fabric covering, braced by a single strut.

 

[lake_harley]

PTAirco....in an effort to learn something, please expand on your wing not being undercambered but having a distinct turned-down TE. I would have thought that to be one and the same?

I'm curious, what thickness aluminum are your ribs and leading edge? Looks like nice work. I'd bet you wouldn't mind a bit more room for construction since it looks like you're a bit tight there, unless there's a lot more behind the camara.

Lynn

 

[aerogant]

Mstull, you have mentioned trick wingtips at least twice. Would you please shed some light on these? Thanks .

 

[PTAirco]

PTAirco....in an effort to learn something, please expand on your wing not being undercambered but having a distinct turned-down TE. I would have thought that to be one and the same?

I'm curious, what thickness aluminum are your ribs and leading edge? Looks like nice work. I'd bet you wouldn't mind a bit more room for construction since it looks like you're a bit tight there, unless there's a lot more behind the camara.

Lynn

Well, I suppose it has undercamber near the rear, unlike the early Kitfox airfoil where the undercamber starts right at the L.E..

My ribs are 0.020", the D-Box skin is 0.016" , webs are 0.040" (Overlapping near the strut joint). 3/4" x 3/4" x 1/8" angles of 2024 for the caps. All very straightforward stuff. There is also a drag brace (1.25" x0.058" tube) which isn't fitted yet in the photo.

My workshop is actually luxurious by my standards, I just have too much stuff!

 

[mstull]

Aerogant,

I don't consider myself particularly knowlegable about trick wing tips, other than knowing they can make a huge difference... like 10% to 20% more lift. Most modern (non-U/L) planes have them... one kind or another. One reason you don't see more on U/Ls is their expense of development and manufacture. To work up to their potential, their shape needs to be complex and precise.

Then there's the structural issues of how to mount and support them, particularly to a fabric covered wing. None of these things are impossible. For one-off designers, it's usually not worth all the effort to make the 4 large, complex shaped molds to lay up the trick tips. My first wing had Horner tips. Maybe you can find some droop (STOL) tips, manufactured for another U/L, that will fit your airfoil and chord. Some kits offer/offered a droop tip option. Two that come to mind... Earthstar and MiniMax.

Obviously, you'd want to choose tips that decrease stall speed (like STOL tips), not ones that make the wing more efficient for high speeds. I'm sure you can find some info on-line. And I'm sure there's aerodynamics books and computer programs about them.

There are a couple of wing tip types that are detrimental. Avoid making the wing tips out of round tubing... even when there's both upper and lower wing surface tubings at the tip. The rounded shape greatly encourages the air from under the wing to ease around the tip to the top. So make the corners sharp... like the tips on a Piper Cherokee... if at all possible.

Don't use a flat, vertical plate, that extends above, ahead, and/or below the tip, sometimes called a fence. It does a great job of keeping the air where it belongs, BUT it makes the aircraft much more prone to spin, and much harder to recover from a spin. It's kind of like making your vertical stabilizer and rudder too small. (It gives me the willies just to think about it.)

A friend sent me this picture of a plane with progressive spoilerons, like what I plan for my new wing. The plates rotate. Mine won't stick up that high.

I'll be interested to hear what tips you decide to use.

 

[lake_harley]

Well, I suppose it has undercamber near the rear, unlike the early Kitfox airfoil where the undercamber starts right at the L.E..

My ribs are 0.020", the D-Box skin is 0.016" , webs are 0.040" (Overlapping near the strut joint). 3/4" x 3/4" x 1/8" angles of 2024 for the caps. All very straightforward stuff. There is also a drag brace (1.25" x0.058" tube) which isn't fitted yet in the photo.

My workshop is actually luxurious by my standards, I just have too much stuff!

Thanks for the explanation and also the info on the wing components. You think you have too much stuff? You should see my mess.....er.....collection of things.

Lynn

 

[olgol]

I came across Earthstar Aircraft's Gull series of light planes and was highly impressed. The thing that puzzles me the most is how they are able to get such a slow stall speed out of a small wing and a reasonably high weight. I've contacted the designer and he insists that stall speeds were measured at gross weight. In every review of this plane that I have read, the stall speed given was really close to the specs given by the company. I know they use a 40% plain flap. Any thoughts on how they get so much out of that wing?
The specs are given here:

http://www.thundergull.com/specifications.htm

I get an overall Clmax of almost 3.1 when running the numbers. How on earth does it get this value? Thanks.

First, let me state for the record that admire ThunderGull airplanes myself and consider them the best example of light airplanes. Well engineered and designed. But...

Lets all be realistic here. Clmax=3.1 is simply not possible with a plain flap or even with a slotted flap. Airfoil Clmax can approach 3.0 values with a double slotted flap. Most airliners have airfoil Clmax under 3.0 with a well engineered slotted flap. When we start talking about CLmax of a wing with a partial span plain flap, optimistically we can say 1.5-1.8 maybe. This means airfoil Clmax of 2.0-2.2 - a very high value.

Two points to consider, when somebody says they "measured" their stall speed.

1) how did they measure it? Indicated airspeed can be near zero at stall angles of attack. This does not mean the true airspeed is near zero. The difference can be significant, on the order of several knots/mph, possibly well over 10mph. To properly measure any airspeed, not to mention stall airspeed, one has to use some elaborate calibrated equipment, e.g., speed probes that are dragged by a long cable, or self-orienting swiveling pitot tubes that point into the airflow. CALIBRATED is the key word. Usually when somebody starts talking about their stall speed, it's the indicated stall speed, since that is the only value one can get sitting in an airplane.

2) what speed did they measure - power on or power off? again, a big difference, possibly well over 10mph, especially on a relatively high powered airplane. Even idling engine produces some thrust and helps to reduce the stall speed. FAA limits the power OFF speed.

So the real answer to your question "how they are able to get such a slow stall speed out of a small wing" is - they can't. The laws of physics cannot be bent.

 

[orion]

Good answer olgol (and welcome) - took the words out of my mouth. And you're right, I've seen some significant error sources in flight testing several airplanes, all having to do with the placement of the pitot device. And yes, the error can be as much as ten knots.

And to Mark above, I got to fly the Wren conversion of a Cessna 182 about two years ago up in Alaska. In short, I wasn't impressed. The installation is relatively complex, dramatically increasing the cost of each annual, and heaven forbid you ding a wing.

The double slotted flap can generate good lift however since the system is also connected to the ailerons, in order to provide safe roll control the entire installation has a limited downward movement (too much movement down and the ailerons might stall). As such, not only does it not deliver as much CLmax as it might but it also delivers very low drag on approach. As a result, while it can slow down well, it cannot come in over high obstacles since even a slight increase in speed (as a result of pushing the nose over) will send you floating down the runway into the next county. Slipping does improve this a bit, but only a bit.