Author Topic: wing thickness? (Read 16363 times)

Gordon Tarbell · « **on:** October 25, 2009, 08:48:36 AM »

On a stunt design with a w.s. of around 55" how thick should the high point of the root rib be if the cord length (not including flap measurement) is 10.000"? And same question for tip rib of 8.375" . Also hinge line distance of 17" would be about right? Power supplied by ST 46 with bristunt ABC P&L set and head supplied by same (might try saito 40 also later )

John Miller · « **Reply #1 on:** October 25, 2009, 09:35:42 AM »

Great question Gordan.

With the parameters you've established, you may be looking at a 20% thickness at the root. This percentage is governed by the wing loading, and available power.

For the tip, I would use a 21 or 22% thickness, and move the high point forward about 5% further than the root location. This is to help delay the stall at the tip during hard manuevers.

As for the hingeline distance, 17" inches is fine, but it depends on other factors as well. The nose legnth needs to be long enough to mount the engine, and a tank large enough to fly the pattern with. The tail should be long enough to balance the nose without adding weight. Errors should be towards being slightly nose heavy, as it requires less additional weight to balance.

There is a formula we can use to figure the balance, but some weights have to be estimated, such as the total weight of the tail group when finished.

The basic Algebraic formula is; (A times B) minus (C times D) = 0, Where A= the weight of the center of mass of the area in front of the CG. B= the distance from the center of mass, in front of the CG. C = the weight of the center of mass behind the CG. D = the distance from the center of mass, behind the CG.

It seems that there may be a sq function in there, but the answer would include a square root in the solution. I believe that would cancel out.

Gordon Tarbell · « **Reply #2 on:** October 25, 2009, 11:19:19 AM »

OK so if cord length is 10" then the aifoil should be at least 2" thick at the high point. and 1.8425" thick for the high point on the tip rib. Also I am hoping that as long as the flaps are not overly long cord wise they will not dictate a change in wing thickness. I just see so manty ultra fat airfoils at the contests these days and want to be sure I don't go too thin on wing thickness and get a plane that needs to fly 70 mph to work. The flaps will probably be flat 5/16" for quick easy building. I also wondered if it is normal to do the calcs. with or with out the flap size.

John Miller · « **Reply #3 on:** October 25, 2009, 04:51:34 PM »

Gordan, the chord legnth includes the flap chord in your figures. For instance, 10 inches from leading edge to flap hingeline, plus, oh, say 3 inches flap chord at the root, equals 13 inches.

So, 20% root chord times legnth, 13 inches, equals 2.6 inches thick at the root. 18% would equal 2.47 inches thick. 16% would equal 2.08 inches thick.

Ty is correct about the entry radii of an airfoil. The blunt NACA 4 diget has the greater radii, but it presents a higher potential for induced drag.

Going to extremes the other way, a sharp pointy airfoil, think of a 1/4 sq set so the point is at the extreme leading edge. Such an airfoil may penetrate turbulence, and groove in level flight, but as soon as the airfoil is rotated so that the angle of attack changes to creat the lift needed to negotiate a turn, the airflow seperates, and the wing begins to enter into a stall.

So often times somewhere between these two examples lies the ideal airfoil for the job at hand.

I've had good success with a particular airfoil known as NAC 63A modified. It is as thick as the 4 digit airfoil, but the entry radius is smaller, while still not being to the point that it stalls as the pointed airfoil does. It penetrates better in wind and turbulence, grooves well, and still doesn't enter into stall territory. at the speeds and angle of attacks we use.

It may be said that the most important part of our airfoils is the entry profile. The part from the leading edge to possibly just aft of the high point. All that's behind it, many believe, does almost nothing for us at our reynolds numbers.

I'm not sure this is always the case, as many feel that the Pathfinder airfoil has some positive characteristics that are attributed to it's particular airfoil. Indeed, many combat ships and several full scale aerobatic designs use about the same airfoil.

Like any exercise, when designing a stunt plane, we need to make some basic choices, and then use what works best around those choices.

Jim Thomerson · « **Reply #4 on:** October 25, 2009, 07:15:48 PM »

Om the 1948 Aeromodeller Annual there are pages and pages of aerodynamic stuff. The focus is, I think, mostly on free flight. There is a short discussion of airfoil thickness. There is a rough suggestion that for greatest efficiency is that % thickness divided by two gives a minimum chord in inches. So a 20% thick wing should have at least a 10 inch chord, and a 15% thick wing should have at least a 7.5 inch chord. so the smaller the wing, the thinner it should be. George Aldrich once told me that a 1/2A stunt airplane should have a thin wing. i have cheerfully ignored his advice. But maybe this is why we hear of 1/2A's with blank wings flying well. Something to think about anyway.

minnesotamodeler · « **Reply #5 on:** October 26, 2009, 03:10:54 AM »

In my experience, even at 1/4A level (.020), a moderate built-up airfoil (say, 15%?) flies better than a flat plate.
Re: Pix below. In every case the builtup wing flies better maneuvers.

Gordon Tarbell · « **Reply #6 on:** October 26, 2009, 06:47:19 AM »

OK guys , I think I know where I want to go with this first. With your info I have a starting point . Thanks, Gordon

Erik Janssen · « **Reply #7 on:** October 26, 2009, 08:09:25 AM »

A wing profile has a drag coefficient, a typical value for an aerobatic profile we use at a lap speed of 5,2 sec is Cd = 0,014
All calculations in metric system so we can easily compare drag * speed = Watts and Volts * Amps = Watts too so you can easily compare.

The drag of a wing is a function of air density, speed (quadruple), frontal area and this Cd factor.
For an aerobatic wing like the Stiletto of Les McDonald this is:
F = ½ density * speed^2 * Cd * Frontal area
Density of air is 1,2 kg/m^3
Speed is 25 m/s
Cd = 0,014
Frontal area is 1,5 m span * 0,054 m thickness
F = ½ 1,2 25^2 0,014 1,5 0,054 = 0,42 Newton

The amount of Watts that is needed to fly this force around is 0,42 N * 25 m/s = 10,6 Watt. From electrical logging I know that a .46 model drags about 12 Newton in level flight.

Thicker profiles have greater drag, higher alpha’s create higher drag. When using software from Profili diffenrences can be calculated between profiles, there are differences in drag but thick wing profile has approx 50% more drag than a thin profile.

So wing thickness has an influence but not as big as I expected. Thick wings can be very successful, Paul Walker is a good example.
I want to prove this by building an Electric Impact with a wing thickness of 65mm where a Stiletto has a wing thickness of 54mm.

It is for next year so I can only predict the outcome.

Balsa Butcher · « **Reply #8 on:** October 26, 2009, 08:44:04 AM »

Or you could do it the old fashioned way, choose an airplane of appropriate size with a proven contest track record. Build the wing, change the wingtips to suit, fashion a fuselage styled to your liking but maintaining well proven "numbers" and you have your new world beater!

Alan Hahn · « **Reply #9 on:** October 27, 2009, 08:11:35 PM »

Quote from: Erik Janssen on October 26, 2009, 08:09:25 AM

A wing profile has a drag coefficient, a typical value for an aerobatic profile we use at a lap speed of 5,2 sec is Cd = 0,014
All calculations in metric system so we can easily compare drag * speed = Watts and Volts * Amps = Watts too so you can easily compare.

The drag of a wing is a function of air density, speed (quadruple), frontal area and this Cd factor.
For an aerobatic wing like the Stiletto of Les McDonald this is:
F = ½ density * speed^2 * Cd * Frontal area
Density of air is 1,2 kg/m^3
Speed is 25 m/s
Cd = 0,014
Frontal area is 1,5 m span * 0,054 m thickness
F = ½ 1,2 25^2 0,014 1,5 0,054 = 0,42 Newton

The amount of Watts that is needed to fly this force around is 0,42 N * 25 m/s = 10,6 Watt. From electrical logging I know that a .46 model drags about 12 Newton in level flight.

Thicker profiles have greater drag, higher alpha’s create higher drag. When using software from Profili diffenrences can be calculated between profiles, there are differences in drag but thick wing profile has approx 50% more drag than a thin profile.

So wing thickness has an influence but not as big as I expected. Thick wings can be very successful, Paul Walker is a good example.
I want to prove this by building an Electric Impact with a wing thickness of 65mm where a Stiletto has a wing thickness of 54mm.

It is for next year so I can only predict the outcome.

Erik,
Your numbers seem pretty low to me, in terms of thrust and power. For example, in my measurements with a Nobler and a Vector, I need about ~240 watts (measured) input power to the motor to maintain level flight. If I add motor/ESC efficiencies of ~70-80% to this, plus prop efficiencies of another ~70% (guestimated from PropCalc)==total ~50, that still comes to~120 watts. Also Prop Calc tells me that at 24m/s, this particular prop is making 4.5N of thrust.

These numbers (power and thrust) are about a factor of 10 above your numbers. I am not sure if I have accounted for all the effects except airframe drag.

Anyway if this informations tends to be too esoteric for the main, feel free to post some of your stuff in the engineering section!

Erik Janssen · « **Reply #10 on:** October 28, 2009, 03:43:32 AM »

I measured the number of watts needed to fly level in flight, 300 Watts for my .46 converted model. I did not bother about motor/ prop efficiency but wanted to have an idea of how important the drag of the wing is.

The number of Watts is the Force multiplied by the movement per second, so 300 Watts divided by 25 m/s = 12N, due to efficiency it will be lower, I agree.
This is not enough to prevent the model from picking up speed in the loops as the weight of the model (gravity is always 100%) is more than this 12N

I calculated the drag of the wing and was surprised by the low amount of drag Cd 0,014 for a Stiletto and 0,021 for a Walker style profile. 0,42N to 0,63N The 12N drag comes from lines, induced drag, wheels etc.

So my conclusion is that using a thick profile will increase my power consumption very little and I can use the same setup.

Unless I mis interpreted the numbers somewhere, if I missed a factor of 10 somewhere then I might get into trouble. Doing the same will not increase my knowledge about what works best so I will take te risk and build a thick wing model for next year.

phil c · « **Reply #11 on:** October 28, 2009, 10:54:22 AM »

I don't think you missed anything Erik. The drag of a wing is always quite small, at least until you start making lift in a maneuver. As you pointed out, it is really inconsequential compared to the rest of the drag coefficient. The lines are the biggest contributor of drag by far.

PerttiMe · « **Reply #12 on:** October 28, 2009, 12:00:42 PM »

As far as I have understood from other threads... the thick and blunt airfoils tends to work better in corners: they do not stall as easily and carry more speed through a sharp corner, than a slim airfoil. Power for level flight and round maneouvers is not really a problem.

Alan Hahn · « **Reply #13 on:** October 28, 2009, 02:15:34 PM »

Erik,
I misunderstood your original post---not realizing you were only talking about the wing.

I claim the total drag is ~4-5 N (in my previous comment), which is in the ballpark of what you are saying if you put in the efficiencies.

Alan Hahn · « **Reply #14 on:** October 28, 2009, 04:10:21 PM »

Out of curiosity, I calculated the drag due to a 0.015" diameter line whose tip is moving at 24m/s, and the value comes out near 1N. Given that most of us fly with 2 lines, and ignoring any interference between the two, then the total line drag is ~2N. I grabbed the formula from Martin Hepperle's site
(http://www.mh-aerotools.de/airfoils/control_line_aero_3.htm)

He ignores the curvature of the flying lines, and I ignored some details he mentions, but 2N (about a quarter of a pound) is in the right ballpark. Which is to say the lines are about 4 times as draggy as the wing that Eric calculated. Then from my guestimate of ~5N to fly my Nobler and Vector in level laps, roughly half of the thrust is being taken up by the remainder of the airframe.

Well that's interesting to realize. Thanks Erik!

Igor Burger · « **Reply #15 on:** November 02, 2009, 01:43:02 PM »

Quote from: Erik Janssen on October 26, 2009, 08:09:25 AM

For an aerobatic wing like the Stiletto of Les McDonald this is:
F = ½ density * speed^2 * Cd * Frontal area
Density of air is 1,2 kg/m^3
Speed is 25 m/s
Cd = 0,014
Frontal area is 1,5 m span * 0,054 m thickness
F = ½ 1,2 25^2 0,014 1,5 0,054 = 0,42 Newton

Sorry, I did not see this thread earlier, but wonder where it comes from?

By my evidence, the drag of wing of usual srunter is approximately 1.5N.

Erik, where you got that number Cd = 0,014 from? and why you do not use usual formula for calculating of drag? If I use drag of NACA 0018 what is airfoil of thickness very close to out ususal wings, then I see it is cd=0.01. And if I take my wing area 0.45m^2

Then drag of the wing is:
F = ½ density * speed^2 * Cd * WING AREA
F = ½ * 1,2 * 25^2 * 0,01 * 0,45 =~ 1,7 Newton

To the other posts: Yes, the wing drag is relatively low so it really does not matter if we use one or two % more or less, while the lift depends on thickness much more.

Air Ministry . · « **Reply #16 on:** November 03, 2009, 12:01:15 AM »

apparently 30 % thick would have good lift, but'd look a bit odd .

Erik Janssen · « **Reply #17 on:** November 03, 2009, 12:10:13 AM »

Drag is calculated with frontal area, not surface.

0,01 or 0,014 in in the same range

Howard Rush · « **Reply #18 on:** November 03, 2009, 01:15:55 AM »

The convention for airplanes is to use wing area as the area in the definition of lift and drag coefficients.

Erik Janssen · « **Reply #19 on:** November 03, 2009, 01:23:20 AM »

ok,

that helps, so increasing the wing thickness from a Cd of 0,014 to 0,021 will add 50% to the 1,7N

that is great as I want to increase my airplane drag from 7 to 8N, so if this adds 0,85 that is perfect.

thanks

Serge_Krauss · « **Reply #20 on:** November 03, 2009, 01:23:50 PM »

As usual, I don't know how well my scans will come out, but these from NACA TR 586 of 6/24/36 show that not everything is obvious or simple. Our Reynolds numbers are usually half a million or less. You'll see that profile drag varies with aoa and that the thick sections enjoy not just maximum lift advantages in his area. Look at the behavior of the profile drag coefficient of the NACA 0012 section compared with the 0018 and see whether these profile drag coefficients are proportional to thickness. These measurements were made after NACA discovered that their RN's were faulty and had taken steps to come up with valid "effective" Reynolds Numbers, taking into account tunnel wall effects and turbulence from supports. So this data should be valid. Yeah, I know, "We don't need no steenkin' math!" But I've already posted this material in words in archives that everyone should visit from time to time - 'lots of info there.

SK

Alan Hahn · « **Reply #21 on:** November 03, 2009, 02:14:14 PM »

<snip>....
Reynolds Numbers, taking into account tunnel wall effects and turbulence from supports. So this data should be valid. Yeah, I know, "We don't need no steenkin' math!" But I've already posted this material in words in archives that everyone should visit from time to time - 'lots of info there.

SK

Well we are waiting to be banished!

[/quote]

Howard Rush · « **Reply #22 on:** November 03, 2009, 05:42:41 PM »

I found TR 586 in 1962. That first chart won me a lot of combat matches.

Serge_Krauss · « **Reply #23 on:** November 04, 2009, 12:26:20 AM »

Quote from: Alan Hahn on November 03, 2009, 02:14:14 PM

<snip>....
Reynolds Numbers, taking into account tunnel wall effects and turbulence from supports. So this data should be valid. Yeah, I know, "We don't need no steenkin' math!" But I've already posted this material in words in archives that everyone should visit from time to time - 'lots of info there.

SK

Well we are waiting to be banished!

Yeah, it sure seems that way sometimes. Wait! We're already out here in the "south forty"...ever since I posted beginning algebra notation.

Seriously though, the graphs contain particularly pertinent info I've found nowhere else. Three views in 12 1/2 hours...?

SK

Alan Hahn · « **Reply #24 on:** November 04, 2009, 08:38:39 AM »

A lot of people have a real aversion to graphs--probably a reaction to some unpleasant exam back in high school. Oh well!

One problem I have with aerodynamics--and the plots--- is the Reynolds number---usually because it is introduced and then the comment about picking the value as a function of the size of the design element you are looking at. Sort of makes it hard to understand exactly since I always wonder what size are they talking about---thickness, chord length,... (of course they are usually within each other by a factor of 10.

Serge_Krauss · « **Reply #25 on:** November 04, 2009, 12:34:25 PM »

I've always had similar feelings about RN, because it's such an imprecise concept - as you hinted. But it's pretty necessary in handling our inability to scale the air to suit our models, because it makes such a big difference. For instance, it reverses the choices of airfoils from from 12% for full-sized aircraft up to 18-24% for our stunt models. The 12% airfoil's maximum-lift curve crosses and dips below those for the 15% and 18% sections as RN decreases; so the Ringmaster must be lighter or fly faster than modern stunters to turn as well. Shapes change too with RN - like leading edges and relative thicknesses for small stabs verses larger stabs or wings. That's what's so nice about graphs; you can look at the lines representing those thicknesses and just follow them to see their relative heights and positions and where they cross. If they jump around or disappear in any range, you know that something unusual or unsteady is happening with the air for that section at that RN. I suspect that just what you said about thickness vs. length affects some of this; in addition to easing contour from l.e. to high-point, the thicker sections cause the air's speed to change as it passes, and this really amounts to a greater local RN for the thicker wing - just a guess.

Another of these reports (at higher RN) covers the effects of turbulent air. When you put these together, you get a hint that my old Skyray would benefit from a turbulator or sharp l.e. on its narrow, thin stab, similarly to something that I think Igor concluded from his CFD work (look way to the left on the graphs). These seemed to help the Skyray and another of my smallish models. I think that you can read some of this in the graphs' shapes, while getting reasonably good numerical data - like the profile drag coefficients of the sections of different thickness. A little arithmetic shows that the drag (based on area) is not a penalty for the thicker wing down where that 10" chord wing operates. Actually, these rare graphs tell us a lot, not just about why things have evolved the way they have in our hobby, but how our own particular models would best fit in. I suppose CFD will completely eclipse things like these graphs, but they contain information that Mr. Jacobs and Mr. Sherman actually found out about our model flying surfaces, something that no one else bothered to do for us.

Well the graphs do bear some study and familiarization. The values can be read easily enough, but there's more yet in the pictures themselves. It would take a lot of words to describe all the information in these graphs, which both picture and quantify it.

SK

Igor Burger · « **Reply #26 on:** November 04, 2009, 12:45:04 PM »

Good point Serge, so it is now clear why my indoors have completaly flat wings and brakes on TE to compensate airfoil drag

Shultzie · « **Reply #27 on:** November 04, 2009, 01:43:41 PM »

IGOR...WOW! What an interesting design photo!!

Since the first day I saw that indoor video of your's on Flying Lines....I have aways been astounded at your ability to not only create very interesting concept models but also equally astounded at your ability to design-build-n fly them with such expertise.
Do you have any more photos or info on this project of yours...or even maybe a 3 view?

Howard Rush · « **Reply #28 on:** November 04, 2009, 06:20:19 PM »

Quote from: Alan Hahn on November 04, 2009, 08:38:39 AM

One problem I have with aerodynamics--and the plots--- is the Reynolds number---usually because it is introduced and then the comment about picking the value as a function of the size of the design element you are looking at. Sort of makes it hard to understand exactly since I always wonder what size are they talking about---thickness, chord length,... (of course they are usually within each other by a factor of 10.

It's another nondimensionalizing convention. For airfoils, chord is the convention. For other stuff, the person citing the Reynolds number should specify the length upon which the Reynolds number is based.

Igor Burger · « **Reply #29 on:** November 05, 2009, 03:00:40 AM »

Thanx

There are many tricks used on that model, I know I should write more ... but time is my enemy ... may be "once"

At least I will post some pictures, everything is visible, it just needs "good eye". Brakes are not on pictures from building. I found I need them later. Plan is in corel draw if anyone wants (well - it is not plan, just main parts)

Span is 800mm, weight 160g lines 5m prop 10x4.7 at 4300rpm and lap time ~5s, leadouts are ~20 degrees back from CG, thrust line 2-3 degrees down.

PerttiMe · « **Reply #30 on:** November 05, 2009, 03:57:36 AM »

Your foamie reminded me of these:
http://shop.donuts-models.com/boutique/fiche_produit.cfm?ref=01DM-GEBEER3INDOOR&type=13&code_lg=lg_fr&num=14
http://www.rcgroups.com/forums/attachment.php?attachmentid=2811438&d=1255308105

Is there a leadout guide somewhere on the wing? I cannot see it.

Igor Burger · « **Reply #31 on:** November 05, 2009, 04:04:07 AM »

yes and yes :-)

the shape comes from gb r3 but it was too thinn, I needed more side area so I modified it

LO is visible on the last picture under the wing

Alan Hahn · « **Reply #32 on:** November 05, 2009, 07:26:32 AM »

Igor,
We just need to program your controller to apply the dive brakes in the descending leg of the maneuver!

I should get a copy of Don Hutchinson's SBD Dauntless profile plans!

Shultzie · « **Reply #33 on:** November 05, 2009, 10:18:40 AM »

Igor...
THANKS...GOOD BUD!!! GREAT PHOTOS!!!
(What an interesting full sized full blown stunter project this little electric model of yours might make?

Igor Burger · « **Reply #34 on:** November 05, 2009, 10:21:08 AM »

well ... I am thinking about fuselage like that ... but it will need composite technology ... we will see :-)

PerttiMe · « **Reply #35 on:** November 05, 2009, 10:35:25 AM »

There's an Axi 4130/16 / .61 2-stroke / .91 4-stroke R/C ARF
http://shop.donuts-models.com/boutique/fiche_produit.cfm?ref=01DM-GEEBEER3&type=16&code_lg=lg_fr&num=14

Might take a bit of modding, though.

(The ones I posted are from Mirco Pecorari's concept drawings. I'd give him a shout if I wanted to sell anything based on his work.)

Igor Burger · « **Reply #36 on:** November 05, 2009, 10:39:03 AM »

But that wing is really too thin for such a C/L model

Air Ministry . · « **Reply #37 on:** November 05, 2009, 04:23:20 PM »

Quote from: Erik Janssen on October 26, 2009, 08:09:25 AM

A wing profile has a drag coefficient, a typical value for an aerobatic profile we use at a lap speed of 5,2 sec is Cd = 0,014
All calculations in metric system so we can easily compare drag * speed = Watts and Volts * Amps = Watts too so you can easily compare.

The drag of a wing is a function of air density, speed (quadruple), frontal area and this Cd factor.
For an aerobatic wing like the Stiletto of Les McDonald this is:
F = ½ density * speed^2 * Cd * Frontal area
Density of air is 1,2 kg/m^3
Speed is 25 m/s
Cd = 0,014
Frontal area is 1,5 m span * 0,054 m thickness
F = ½ 1,2 25^2 0,014 1,5 0,054 = 0,42 Newton

-------------------------------------------------

Aspect Ratio is somewhat relevant, particularly during manouvres and glide.

A (say ) 23 % thick wing of a set area will have the SAME frontal area at any A/R ,Though drag will be less
(at our speeds) on a high A/R (hopefully) This is assumeing similar torsional ridgidity,or we have an ornitopia.

Igor Burger · « **Reply #38 on:** November 05, 2009, 05:27:10 PM »

Matthew, several notes:

1/ It was caculation for wing at 0 lift. Means induced drag is 0 and thus does not make induced drag. Because:

induced drag = something * (lift coefficient ^2) / Aspec ratio

this is on top of arfoil drag, so you are right that drag at low speed (gliding) or maneuverig at higher AoA will be higher at low AR

I note again that the formula used by Erik is incompatible with that Cd=0.014 !!! proper formula uses wing area
It can be calculated by frontal area, but ina that case the drag coefficient vill be different (much higher) number

2/ BUT ... as Serge already wrote, we must calculate with RE number at our speeds and if you will take NACA 0018 as anexample for two wings of the same area (also the same frontal area), you will see that half span will lead to twice RE number and thus lower drag coefficient, so in some cases you can see just opposite result

phil c · « **Reply #39 on:** November 06, 2009, 07:59:50 AM »

Quote from: Matthew Spencer on November 03, 2009, 12:01:15 AM

apparently 30 % thick would have good lift, but'd look a bit odd .

NACA did further testing with airfoils up to 24%(if I remember right). It appeared that the lift advantages disappeared or went down over 21-22% thick(we're talking non-flapped airfoils here). It is really hard to beat the NACA 0018 for flap-free models in our size ranges. You can graft on the leading edge from a 5 digit section so the wing doesn't look so fat. I've tried that and it seemed to work just fine with no noticeable problems. Making a baseball bat leading edge(just the opposite) also helps improve lift a little, the combat guys have been doing this for eons. Put the high point at 18% or so of the airfoil. Even the big guys are doing it. take a look at the Extra 300: http://www.extraaircraft.co.uk/img/Extra300S_HA-RED.jpg

Gary James · « **Reply #40 on:** November 07, 2009, 07:31:55 AM »

Another thing that helps improve Cl max on single-element airfoil sections, although counter-intuitive, is to have a BLUNT trailing edge. While the "linear aero types" might insist that a sharp trailing edge is essential, in reality, Cl max is mostly a function of how long you can keep the airflow attached to the surface. A blunt or squared off shape causes a small vortex to form just aft of the trailing edge. This vortex acts like a "pump" and re-energizes the airflow a bit which keeps it attached to higher angles of attack. The effect can be dramatic. The blunt t.e. also increases the section drag coefficient, but on a stunt plane, that's not too important as it tends to cause the plane to fly at a more constant speed. Not a good idea for a racer, however. A good choice... 18%-20% thick, max thickness at 18-20% chord, a fairly blunt l.e., and a blunt t.e. that is 2-3% of the chord in thickness. Just like the other guys have been recommending.

Gary James · « **Reply #41 on:** November 07, 2009, 07:36:36 AM »

Quote from: Howard Rush on November 03, 2009, 05:42:41 PM

I found TR 586 in 1962. That first chart won me a lot of combat matches.

Erik Janssen · « **Reply #42 on:** November 08, 2009, 02:36:00 AM »

The Impact I am building has a blunt trailing edge on the plan from 1994. I wondered what the effect would be. Thanks for the explanation.

My current plane has sharp training edges, maybe I will sand the TE down before I retire the model to see what it does.

I had a chance to see the jatsenko models flying, in a square they seem to hit the handbrake, make the turn, come to a dead stop turning and start flying again. A truly impressive flying style. The models are jig built and very light. With a 13cc they are only 1600 gram. Taking into consideration that they are fully detachable this is really impressive.

I found these drawings, on the classic airplane the flaps are part of the airfoil with a sharp trailing edge.

It seems that there are multiple ways to buld a good airplane

Igor Burger · « **Reply #43 on:** November 08, 2009, 03:57:51 PM »

Quote from: Erik Janssen on November 08, 2009, 02:36:00 AM

flying, in a square they seem to hit the handbrake, make the turn, come to a dead stop turning and start flying again

A beginner reading this thread must come the conclusion that the best airfoil for c/l stunter has low drag in level flight and high drad (or even brake) in corner, but it is just opposite, proper airfoil is optimized to have low drag at high lift and if possible high drag in level.

Erik Janssen · « **Reply #44 on:** November 09, 2009, 03:26:16 AM »

In this post I refer to the package not just the profile, I am convinced that the 13cc motor and the low weight of the model have at least or even bigger influence on the characteristics as the wing profile.

From the desription of the profile you seem to be looking for the CAP21 airfiol like Beringer is using. Big engine and very light model too.

Two totally different models with good flying abilities.

Igor Burger · « **Reply #45 on:** November 09, 2009, 06:16:10 AM »

1/ Yatsenko fly Retro .60 and it is 10ccm motor, not 13ccm
2/ if it is so good setup, then why he must fly 4.8 lap times ?

And I am certainly not looking for airfoil lile CAP (or Beringers wing design) ... if it is so good airfoil, why we do not see it on good places on results?

It is just opposite what I mean.

Claudio Chacon · « **Reply #46 on:** November 09, 2009, 10:57:54 AM »

Quote from: Igor Burger on November 09, 2009, 06:16:10 AM

1/ Yatsenko fly Retro .60 and it is 10ccm motor, not 13ccm
2/ if it is so good setup, then why he must fly 4.8 lap times ?

And I am certainly not looking for airfoil lile CAP (or Beringers wing design) ... if it is so good airfoil, why we do not see it on good places on results?
It is just opposite what I mean.

Hello Igor,
Interesting observation you've made about Yatsenko's lap times. I got a question: Do you think he flies that fast because he's used to it or just because the plane doesn't perform well at lower speeds, say, 5.1/5.3 laps?

Regarding Beringer's wing designs/airfoils -if memory serves me well- , Remi Beringer won the 2006 World Champs in Valladolid with that design.

Regards,
Claudio.

Erik Janssen · « **Reply #47 on:** November 09, 2009, 12:58:05 PM »

Rob Metkemeijer built his own 13cc MB engine and is flying the shark at 5,3 sec/ lap. He flew this model at the Dutch Championships. Although trimming is still in progress I did like the way it turns.

The Beringer profile was developed by Gilbert Beringer and used by his son Remy who became World Champion in 2006. In the St Etienne area there are a lot of Caudron models flying around. I competed at his competition in St Etienne from 1981 to 1988 and only succeeded to beat Gilbert once.

The French pilots claim that the profile can only work on a very light airplane, 1600 grams should be the maximum. This is very light as he is flying a big four stroke.

Common factor on both airplanes is that they are both very light and have a lot of engine up front. With less power, like my Super Tigre .46 I would prefer the NACA 0018 or Eppler 473.

I want less weight for my next model and more drag to reduce speed in down maneuvers. Maybe increasing the wing area is safer than increasing wing thickness. Althoug wing thickness is a lighter solution than area.

Igor Burger · « **Reply #48 on:** November 09, 2009, 03:05:34 PM »

Quote from: Erik Janssen on November 09, 2009, 12:58:05 PM

The French pilots claim that the profile can only work on a very light airplane, 1600 grams should be the maximum. This is very light as he is flying a big four stroke.

An that is the point, every airfoil will work at low loading, also flat plate

Good arfoil will carry lot of weight and thus will be good weapon to fly in difficult conditions.

Igor Burger · « **Reply #49 on:** November 09, 2009, 03:21:37 PM »

Quote from: Claudio Chacon on November 09, 2009, 10:57:54 AM

Interesting observation you've made about Yatsenko's lap times. I got a question: Do you think he flies that fast because he's used to it or just because the plane doesn't perform well at lower speeds, say, 5.1/5.3 laps?

No I do not think, becuse Andrey fly slower laps with basically same aeronamics. In past I wrote about airfoils, I do not know if you saw it. One problem is lift and drag of the airfoil and the second is how "well" it makes the lift.

The amount of lift is important for high wing load as I already wrote in previouse post (look to latest WC and EC what are winning wing loads and airfoils), but very important for good properties of stunter is also how "well" it makes the lift. We see model which will do what we expect every time in every condition same way. Some flapped airfoils does not have linear response lift to AoA because of unstable separating on upper side of airfoil. If it does not happens "repetitively" the model does not have good properties. Aifoiled flaps like used on that modes are such example. So if we want to trim such model, we must push it out of that unstable speed/lift somewhere to stable airflow. And it always means some compromising. I would say that his model just fly well at that speed. I am sure it can fly slower, but I do not thing it will fly that well.

He fly well but if you will look longer, you will see that also after hours and hours of practice, he do not have consistent bottoms. He will do 20 perfect turns and then one bad. And that is sign of that problem.