stunthanger.com
Design => Stunt design => Topic started by: Frank Wadle on February 06, 2018, 06:43:16 AM
-
Hello,
Just recently I came across some pictures a russian F2B pilot posted on facebook. He is currently building prototypes of a new F2B contest airplane, fully take-apart, all build in molds.
On the pictures you can see that he is using a rather unconventional airfoil. It is a Wortmann FX-71-L-150/K25 and the flaps are integrated into the airfoil.
This started a discussion between some friends of mine and myself if this is a good choice.
And this discussion rose the question what exactly is the "best" airfoil for Stunt airplanes?
What exactly makes an airfoil suitable (or unsuitable) for stunt?
I'm not an expert but know about drag coefficient, lift coefficient and moment coefficient curves, but how do these curves have to look like in order to make it a good stunt airfoil?
And how should these curves look like to make it "the ideal" airfoil?
Best regards from Germany
Frank Wadle
-
That looks like a very good airfoil to me.(But I'm NO expert) HB~>
Of course the side of my shoe (Shoeman) looks good too! LL~ LL~ LL~
Jerry
PS: I like those apple cheek cowls!
PPS: AFAIC, there is NO "Best" airfoil. All airfoils are good in one respect and not so good in another, i. e., it'll go fast and not turn or it'll turn and not go fast. That's my experience in glider racing. (Slope Soaring)
-
Dr. Wortmann published a paper in 1973, "Symmetrical Airfoils Optimized for Small Flap Deflection", which included the 71-L-150/25. I don't know the difference, if any, between that and the 71-L-150/K25. I presume that K is for Klappe (flap). He designed these airfoils for surfaces such as vertical and horizontal stabilizers to have minimum drag when the control surface was deflected as for trim. I don't think this is applicable to stunt, but Phil Cartier and I looked into using this airfoil for combat planes. I found my hand plot of the 71-L-150/25 this morning. I think I went so far as to make a flap, but I didn't finish the airplane. As I recall, to break even for induced drag from the added weight, we would have had to build the flap and mechanism for less than 50 grams. That seemed impractical.
The "Symmetrical Airfoils Optimized for Small Flap Deflection" (note American spelling) paper heading had the logo of OSTIV, and at the bottom was "Aero-Revue 3/1973". The 71-L-150/25 and similar airfoils are in Stuttgarter Profilkatalog I, by D. Althaus, Institut für Aerodynamik und Gasdynamik der Universität Stuttgart, 1972.
I think the primary consideration for a stunt airfoil is maximum lift coefficient. That allows wing area to be minimized, so turbulence response is minimized for a given cornering capability. Other considerations are thickness, ease of building, and linearity of the lift-vs.-angle-of-attack curve. The thicker the airfoil, the lighter the wing can be built, and airfoils with high max lift coefficient tend to be thick. Airfoils that are thick and convex just forward of the flap hinge are easier to build and finish than airfoils such as the 71-L-150/25. Igor mentioned linearity of the lift-vs.-angle-of-attack curve and wrote about airfoils with a hump in the curve. I had not noticed such a hump, but looking at the 71-L-150/25 lift-vs.-angle-of-attack curve today, I see that it has that hump at large flap deflections.
I don't think drag is a big consideration for stunt airfoils. Airfoils with the highest maximum lift coefficient tend to have relatively low drag in maneuvers (which is a virtue) and slightly higher drag in level flight than other airfoils. Because we fly the same path every flight, I think we could calculate the energy required for a flight as a function of airplane weight, airfoil section drag, and span loading, and find out for sure how much airfoil section drag matters. I have been too lazy to do that.
Using XFOIL, which I probably don't know how to do, the best stunt airfoil I've found is the Impact's.
Frank's next question will be about the optimal progression of rib spacing from wing root to tip.
-
Optimal rib spacing prigression is out of question a function to the power of 2 ;-)
I agree that the overall drag is not so much an issue.
My theory goes as such (but please dear experts take it easy on me. I'm not an aerodynamics engineer)
I think induced drag has to be minimized (difference of drag in lever to corner flying)
I also think lift should be nicely steadily increasing as you pull on the handle, no "bumps" in the lift curve.
I also think shift of centre of lift should be minimized.
But all this is not really based on textbook knowledge or empirical tests, it's just my intuition.
So I'm calling all the experts out there (hello Igor, Bob, Ted, Dave,....). Please enlighten me. Please tell me that I'm wrong, please tell us what to look for when we look at airfoil diagrams.
-
So I'm calling all the experts out there (hello Igor, Bob, Ted, Dave,....). Please enlighten me.
I did, several years before you asked that question :- )))
You can reread my SN article, it is reposted somewhere here on SH or on FB in my profile or on FB max bee page, you certainly saw it :-P ... now it is time to READ it, I mentioned there why that airfoil is not the best way. (and I know several guys proofed that fact :- P )
-
Oh well, I certainly didn't come across it YET!!!
I will dig for it, but the internet is vastly big place.
Can you give us a clue where to exactly find it? A link maybe?
-
He designed these airfoils for surfaces such as vertical and horizontal stabilizers to have minimum drag when the control surface was deflected as for trim. I don't think this is applicable to stunt
Exactly, that airfoil is for tails, not for wings, the main difference is AoA of those airfoils, while wing airfoil operates on positive AoA (regarding produced lift ) tail operates at NEGATIVE AoA ... means its lift is created by deflected flap, not by AoA of fixed part. So quick look to lift curve of that polar shows all problems, while lift curve on left side of 0 AoA axis (negative angles) is nice straight, right side has bumps and other strange excesses.
Result is clear - model is difficult to trim, having very pure tracking, and I have seen model which was not able to fly round loops, it did usually something like hexagons, whatever pilot did :- ))
The other problems of sharp LE were unexpected stalls.
-
Oh well, I certainly didn't come across it YET!!!
I will dig for it, but the internet is vastly big place.
Can you give us a clue where to exactly find it? A link maybe?
Simplest way is this I think:
http://www.maxbee.net/download/MaxBee_1.pdf
http://www.maxbee.net/download/MaxBee_2.pdf
Hope it still works
-
Jees , looks like a P-51 airfoil , He must be trying to get to Berlin. And Back .
(http://airfoiltools.com/images/airfoil/fx711525-il_l.png)
Hardly suit a scale stunt Hawker of Supermarine . >:( cept maybe attacker / sea Fang . :(
-
Igor,
I read your article now. Thank you very much. That explains it very well.
And it is nice to see that your mathematic proof comes to pretty much the same conclusion as my intuition.
I really like your explanations about what happens in a corner vs. level flight.
I wonder what happens in the transition.
At first there is deflection of flap and rudder, then the plain starts turning and goes into the corner.
Then the deflection is rather abruptly gone and as a consequence the plane stops turning.
Have you had a look at that yet?
Frank
-
I wonder what happens in the transition.
Not too much. Hand only hardly makes so quick movement that we can call it "immediate" in comparison to corner time. Corner is not radius, it is more hyperbolic (becaise of that limited deflection speed), and tail (on modern stunters with relatively long tails and large areas) usually follow deflection easily. However still, transcient and especially limted speed of handle movement can make some kind of transcient like you describe. The tail of my max and wing load are designed with respect of 2 things:
1/ Aerodynamic transcient - the efect you described. The point is, that the tail with immediate deflection should not get stalled before the corner - that is easy to analyze, air flow is straight and we have tools for that. The second problem is immediate move of elevator back to neutral while model still flies radius. Since we know AoA of stab in radius, it is also relatively easy. My tail is designed with both on the edge of stall. And since I know I have limited speed of controlls movement, I know it is on safe side. And if pilot does it too strong, it can even do some negative feedback and make turns more repetitive (I mean stall on elevator because of too strong input - when it forms separtion bubble at LE of stab) ... at least I hope. VD~ I feel something like that when I do stressy turn, looks like model does what it always did when I do "normal" corner.
2/ there is also mechanical transcied. This one is important how to get model fly clean for human eye. This is some kind of cheating, and we spoke about it already in past. Since we know AoA of model coming from airfoil properties, wing load, flaps deflection etc, then if I know AoA in corner is say 10 degrees, then model will get proper angle for fly off 10 degrees before it does 90 degrees turn. Means absolutely clean fly off will need controll movement taking time 10 degrees of corner radius. Means not immediate movement. Something similar happens on begin of corner. I do not know if it is clear what I say, may be it will need some pictures. I think I posted them somewhere on stuka years ago. I want to say that properly designed model will give chance pilot to make relaxed handle movement while judge can see very quick corner. It has something to do also with that logaritmic unit on flaps (limited flap deflection). This is something we know when model comes too light, model tends to make end of turs somewhat difficult to handle, has tendency to make waves after corner etc. and for surprise some balast in CG can make it fly much better.
And that will probably (here I am not 100% sure with this) define aslso optimal tail arm length - too long and tail will stall in turns and at stop, too short and it will not make enough torque, plus thanks to short distance from wing, the air stream will be paralel with stab and it loose self controll (that stabilizing buble at LE of stab) and probably leads to too "jumpy" model and will need heavier nose (CG at 15 to 18% MAC). ... but here I am only guessing, I did not have time and chance to test.
... well yes I know little bit chaotic answer, but this is certainly not subject for short forum reply ;D
-
... I think induced drag has to be minimized (difference of drag in lever to corner flying) ...
Other's have thought this, too. Because induced drag is almost purely a function of aspect ratio and wing tip design, they went out and built really high aspect ratio stunt planes. The results cornered almost too well (although that may be a training issue) and didn't handle wind nearly as well as the 5:1 to 5.5:1 aspect ratios that you usually see in the winner's circle.
-
Amount of induced drag is almost function of "span load" - means weght to span ratio (well +/- something) ... we also spoke about it some year ago
-
Amount of induced drag is almost function of "span load" - means weght to span ratio (well +/- something) ... we also spoke about it some year ago
Sorry, yes -- for a given weight, speed and wing area, it's all down to aspect ratio, but it's really driven by speed and span loading. It still means that all else being equal, if you want to reduce induced drag you need to increase aspect ratio. And while I wasn't around for the high aspect ratio stunter era, I've certainly heard stories about the consequences.
-
Here it is, the Wortmann FX-71-L-150/K25 airfoil. One can find its data here on page 43:
http://www.bilimuygula.com/FileUpload/bs774731/File/deneme_modellflug_profilesammlung.compressed.pdf
K25 was designed for 25% chord flap.
-
Huh,
It is sorrow to see this beautiful fuselage, designed for wings with FX-71-L150. Why?
To choose this airfoil for C/L stunt, is an absolute, elementary mistake.
This so-called "laminar" airfoil is designed for fullscale sailplanes' vertical stabilizers, for Re at least 1.000.000, for minimum drag and high efficiency.
(We used to fly sometimes no more than 350.000, pretty below the critical Re value of FX-150)
You can see: the thickest point is very-very aft, at 38% of chord. (Ours is around 25%).
If you modify this airfoil to bring forwards the thickest point, (it is not a big task with "Profili-II" software), the result can be usable. That will be the well-known "pollywog" airfoil. I use similar, with good results for 30 years, with 30% to 20% moving part (flap) and 16,7% to 15% thickness. It never moves more than 20°by me.
One remark: it is hard task to build correct, warp-free concave flaps by conventional technology. (This question does not touch me since I build of bluefoam, covered by thin glass+epoxy).
-
And this discussion rose the question what exactly is the "best" airfoil for Stunt airplanes?
Al Rabe used to claim that he designed his airfoils by tracing around his shoes. If I could walk a mile in that man's shoes -- I'd trace them out first.
-
>>>Amount of induced drag is almost function of "span load" - means weght to span ratio (well +/- something) ... we also spoke about it some year ago<<<
Sorry, yes -- for a given weight, speed and wing area, it's all down to aspect ratio, but it's really driven by speed and span loading. It still means that all else being equal, if you want to reduce induced drag you need to increase aspect ratio. And while I wasn't around for the high aspect ratio stunter era, I've certainly heard stories about the consequences.
The point of that mine statement is that the aspect ratio itself is not whole answer. The induced drag is function of both lift coeffcient and and aspect ratio (linear to both). So if you make shorter chord, you will get proportionally smaller area and thus higher lift coefficient (having all other flight parameters equivalent) and you will get proportionally better aspect ratio. Unfortunatelly one compensate the oyther, so induced drag will not change as expected. Means there is only span which can change induced drag, not aspect ratio itself. Because elongating the span, we will get larger area and thus smaller lift coefficient AND better aspect ratio. So it makes it power of 2.
It means 2 things for induced drag:
1/ you can change aspec ratio withouthout changing span - nothing happens
2/ you can change span without changing aspect ratio a induced drag will change
That means the span defines induced drag (for the model of given weight), not aspect ratio itself. Or better written - weight to span ratio.
-
First I would like to thank everyone for the great input.
This has definitely shed some light on this very interesting topic. I still have a few questions, but I will come to that later.
Meanwhile I would like to clarify what I meant by induced drag.
I believe we have to think in percentages. I think it is important that the drag increases by as little percentages as possible.
This can also be achieved by selection of airfoil.
Many years ago when I designed my Big Kahuna airplane I used the software profili to investigate the influence of different trailing edge shapes.
For structural reasons I wanted to use a “thick square” TE and not a sharp one. And I found out that the thick one has a higher drag overall. And the change of drag in percentages was smaller than with a razor sharp TE.
So there is definitely an influence of the airfoil on this subject.
Let me be the devil’s advocate for a moment…
Why do we want as little increase in drag as possible?
Because we want as constant as possible speed.
This could also be achieved by dragging a parachute behind the plane.
The drag of the parachute is constant, no matter if you fly level, if you fly up, down or if you fly a corner. And that would even out the speed of the plane.
Wouldn’t that mean we have to increase the drag of such parts that don’t create lift as much as possible?
High drag fuselage…
Yak 55…
:-)
-
If your model has in-flight drag 10N and drag variation in corner is +5N, then you will balance that in-flight drag by 10N prop thrust. So speed variation in 0,2s corner will change by:
0,2s * (110N - 110N + 5N) / mass of model = 0,2s * 5N / mass of model
If you add a parachute with 100N, the total in-flight drag will be 110N and drag variation in corner is +5N, then you will balance that in-flight drag by 110N prop thrust. So speed variation in 0,2s corner will change by:
0,2s * ( 10N - 10N + 5N) / mass of model = 0,2s * 5N / mass of model
does it answer what we have to minimize to limit speed variations in corner enough? H^^
-
I understand.
I thought it goes as such:
The power needed to fly level is
P=F * V
And the power of the motor is always the same for the reason of a simpler math :-)
That means in level and in a corner you have the same power (kW) to fly the plane.
Low drag:
F(level) * V(level) = P = F(corner) * V(corner)
V(corner) = F(level) / F(corner) *V(level)
V(corner) = 10N / 15N * 25m/s = 16,666m/s
High drag:
V(corner) = 100N / 105N * 25m/s = 23,8m/s
I must have looked at the problem from the wrong perspective. I'm sure your explanation is correct and mine is wrong. HB~>
-
I'v made a quick analysis with Javafoil of FX71-L-150/K20 airfoil with flaps at 27 degrees. I think a result talk by itself...
-
I understand.
I thought it goes as such:
The power needed to fly level is
P=F * V
And the power of the motor is always the same for the reason of a simpler math :-)
That means in level and in a corner you have the same power (kW) to fly the plane.
Low drag:
F(level) * V(level) = P = F(corner) * V(corner)
V(corner) = F(level) / F(corner) *V(level)
V(corner) = 10N / 15N * 25m/s = 16,666m/s
High drag:
V(corner) = 100N / 105N * 25m/s = 23,8m/s
I must have looked at the problem from the wrong perspective. I'm sure your explanation is correct and mine is wrong. HB~>
I agree ... at least with the last sentence :- ))))))))))
if you expect that the power in both cases the same, you must also expect that the speed is the same, If speed changes, also the power changes.
-
Correct. If speed drops then power output of Propeller drops.
So even More drastic effect.
The difference is that my formular applies to equilibrium only. So I guess it does not so much apply to corner. But maybe more for wingover shortly before we go back to level flight.
-
I'v made a quick analysis with Javafoil of FX71-L-150/K20 airfoil with flaps at 27 degrees. I think a result talk by itself...
Yes, that exactly shows what I wrote above.
1/ The lift curve shows that deflected controlls push that airfoil to regime when it abruptly loses lift as it get to higher AoA = just opposite what pilot expect and VERY hard to fly wanted shape path.
2/ The moment curve shows that it has little bit positive feedback in pitchig effect, it means also pitch stability is so low, that it will be hard to conroll at all.
This is just opposite what we need from aitfoil. And it also shows that such airfoil is for AoA less than 0, you can see it is very nice linear airfoir in AoA left of the 0 AoA axis. = great airfoil for tail, but certainly not for wing.
-
Correct. If speed drops then power output of Propeller drops.
So even More drastic effect.
Try my props, they pull more, not less when model slows down :- ))))))))))))))
-
Dear Vitalis,
What do you want with FX-71-L150 at Re=400.000? (instead of 1.000.000)
Dr. Felix Wortmann DID know what he designed, and what is it for...
Fullscale sailplane builders DO know why they use this airfoil...
C/L stunters using this airfoil simply DO NOT know what they do.
It is so simple.
-
Try my props, they pull more, not less when model slows down :- ))))))))))))))
They ALL do that (within usable pitch ranges). What yours does better (for your purposes) is also fall into the drag bucket when slowed down. That's why we liked similar prop airfoils on the 46VF and the PA51 (which run very smoothly with little reaction to drag torque) and had to go WAY off the reservation to get it to work on the PA40/61. In fact we found the PAs worked better with flat-backed airfoils because the engine itself reacted plenty enough to not have to count on the drag going down.
Of course you have solved that dilemma by having your feedback system, which senses the slowdown directly, instead of indirectly as it is with an engine, and use the drag bucket issue to partially mitigate the finite bandwidth of the controller. The fact that only a few people even see a glimmer of what you are doing makes it pretty hard to explain, of course.
Brett
-
Correct. If speed drops then power output of Propeller drops.
So even More drastic effect.
The shaft power may drop but the thrust goes up drastically, and the system absolutely counts on that. The increase in induced drag from creating more thrust is what makes the RPM/power drop in the first place. Unless you have an undercambered prop with a conveniently-placed drag bucket like an Eather or Igor prop, in which case, you get more thrust and no RPM drop, or at least much less RPM/power reduction.
Brett
-
Dear Vitalis,
What do you want with FX-71-L150 at Re=400.000? (instead of 1.000.000)
C/L stunters using this airfoil simply DO NOT know what they do.
It is so simple.
It's too late, bro, to apologize - molds have been already made and a first ship is almost ready to fly. Let's wait for a flight review.
-
They ALL do that (within usable pitch ranges).
Of course :- )))
I just little bit provocated Frank, he will certainly understand : -)))
BTW, regarding flat back and underchambered props, it is still here also with electric, for example I like flat back props, while some guys especially with Yatsenko models (for example Orestes) like underchambered (I mean both with my regulation). May be it has something to do with model type, but I am not 100% sure.
-
Let's say shaft power is constant. Then the prop output power will decrees with the decreasing velocity of the model.
At stand still (model on ground with prop spinning) the efficiency of the prop is 0%
I've once built a propeller test stand in a wind tunnel that proved this.
Anyway... We are drifting away.
I admire the courage this Russian builder has. He has put a lot of work into making moulds. I keep my fingers crossed that his efforts will be rewarded with a well flying airplane.
In our recent F2B history there was another successful design using a rather unconventional airfoil.... Beringer used something like the CAP21 airfoil for his designs and was very successful with them.
How does this airfoil tie in to Igor's theories?
-
Let's say shaft power is constant. Then the prop output power will decrees with the decreasing velocity of the model.
At stand still (model on ground with prop spinning) the efficiency of the prop is 0%
I've once built a propeller test stand in a wind tunnel that proved this.
Well, fact that prop has 0 efficiency at still, is clear from physics, you not need stand to proof it, if velocity is 0 and trust whatever, then power comming from force * speed must be also 0 since the speed is 0
... just the same like when prop does not make any thrust at pitch speed - the efficiency is also 0 because whatever the speed is, multipied by 0 force you have 0 power ... before that point power INCREASES when model slows down.
... so you can see that there is also speed range where that theory does not fit reality :-))
(https://www.mh-aerotools.de/airfoils/images/f3dprop3.gif)
However it is false expectation that power on shaft is constant. Much better expectation is that RPM is constant.
-
In our recent F2B history there was another successful design using a rather unconventional airfoil.... Beringer used something like the CAP21 airfoil for his designs and was very successful with them.
How does this airfoil tie in to Igor's theories?
That was actually MUCH better airfoil compared with FX 71. The problem of those models were too small flaps and therefore small lift (may be Keith R. can repost some analyzed triangle which we saw on some meeting) and thus it needed extremly low wing load and thus making troubles in turbullent air.
-
In our recent F2B history there was another successful design using a rather unconventional airfoil.... Beringer used something like the CAP21 airfoil for his designs and was very successful with them.
How does this airfoil tie in to Igor's theories?
It's been discussed here (one of my favorit topics btw):
https://stunthanger.com/smf/engineering-board/ice-cream-cone-airfoils/
-
Basic problem was, what is better?
A gain in drag of 10% or a gain in drag of 20% due to induced drag?
And I know there are several ways to achieve the lower increase percentage.
-
Basic problem was, what is better?
A gain in drag of 10% or a gain in drag of 20% due to induced drag?
And I know there are several ways to achieve the lower increase percentage.
I think you're dwelling too much on one thing. And Igor's explanation early in this thread about the importance of not having any sudden changes in lift or drag makes a lot of sense to me. We have gobs of available power, from our motors or engines -- we don't need super-efficient low-drag machines. What we need are super-predictable airplanes that go where you point them when you point them. I do not believe that drag even approaches being the biggest factor in that equation.
-
Of course :- )))
I just little bit provocated Frank, he will certainly understand : -)))
BTW, regarding flat back and underchambered props, it is still here also with electric, for example I like flat back props, while some guys especially with Yatsenko models (for example Orestes) like underchambered (I mean both with my regulation). May be it has something to do with model type, but I am not 100% sure.
Depends on how much "boost" you can tolerate, or want. I have been very surprised at how much boost/brake people have been willing to use, even when it was very clearly hurting their overall performance. I am not sure why that is, mostly, I think it is that people tend to line up with a particular guru, and just do what the guru says, and aren't even willing to try anything that might incur the guru's disapproval. Or just general unwillingness or inability to make sense of experimental systems.
People can get to be pretty good by getting their system just good enough and then flogging it with 2000 flights a year. I think it has gotten a lot harder to win that way, but not impossible.
Brett
-
Basic problem was, what is better?
A gain in drag of 10% or a gain in drag of 20% due to induced drag?
And I know there are several ways to achieve the lower increase percentage.
I think the the result from speach is clear, we need minimize CHANGES of drag, not average value of drag, and since we have drag of airfoil which gets higher at higher lift and plus induced drag, we must optimize airfoil for high lift - means with deflected flap - means keep is as small as possible ... while extra drag from inefficient airfoil shape with straight flap is welcome. That is vhy we use "added" flaps while integrated flaps are suboptimal. Just like I show in my article :- ))
But it is slill only about drag, I see TIm wrote already that there are other aspects of airfoil making them good or bad.
-
Depends on how much "boost" you can tolerate, or want. I have been very surprised at how much boost/brake people have been willing to use, even when it was very clearly hurting their overall performance. I am not sure why that is, mostly, I think it is that people tend to line up with a particular guru, and just do what the guru says, and aren't even willing to try anything that might incur the guru's disapproval. Or just general unwillingness or inability to make sense of experimental systems.
Probably also a reason, people want it best of the best, strongest boost, largest prop etc ...
So happy as it pulls on top of loop, but prop cannot brake so well as it can pull, so it overspeeds on bottom especially in wind ... so they compensate with underchambered prop which will dump such wild reactions :- ))
-
Here's a side question. Is there a point when these airfoil design arguments become somewhat futile, when all of the "best airfoils" are now, or have been seen sporting the addition of either zig-zag turbulator strips or vortex generators to augment lift?
Here's a few notable examples:
-
At what point do these airfoil design arguments become futile, when all of the "best airfoils" are often seen sporting the addition of either zig-zag turbulator strips or vortex generators?
Here's a few notable examples:
Those guys? Meh.
What it really tells you is that to first approximation, unless you go out of your way to do something silly, it's just not that critical. Maybe, just possibly, a very subtle reshaping of the airfoil might let any of those airplanes fly as well without VGs, but probably not. You have airfoils that are nearly 3" thick (ARF Strega) that have less capability than airfoils maybe 1 3/8' (Diva), so it matters some, but (in decreasing order of silliness):
don't make the LE pointy (known and documented in the early 70s)
don't make the aft portion a straight line/flat to the hinge line
don't make the aft portion concave
don't insist on fairing every LE shape fair into a 1/4 square LE set on edge
and you will probably be OK.
Brett
-
But Youve Left Half of Them OFF . S?P
(http://www.thunder-and-lightnings.co.uk/javelin/full/cov99513.jpg)
those airbrakes look rather usefull , too .
(http://controlline.org.uk/phpBB2/files/stiching_elevators_437.jpg)
not the best illustration , usualy use 1/2 W x 3/8 deep ( 3/4 O.A. ) Double Upholsterers Thread . if you dont countersink it ,
one would presume it must have some ' readhearance ' effect on the airflow . Despite looing centuries old .
The Olde Sailmakers palm , Binding of spars and rigging etc , would convince of the immense strength and coheshion of members .
-
You have airfoils that are nearly 3" thick (ARF Strega) that have less capability than...
Does anybody know why Brett is so sensitive about Strega, Patternmaster and all the derivatives of Big Jim's legacy? Cmon, Brett, lay down your pain finally b1
But let's talk about Strega's airfoil particularly. I'v been very curious myself about that since I became an owner of two Strega ARF kits, thinking if they do worth my endeavour as my future project. Now when I'v learned to use Javafoil a little bit, I can think this airfoil is not really that bad. It definitely has a potential, but ofcourse only after drastic LE modification (blunting).
Firstly even just in visual comparison to the classic and proven PW Impact's airfoil, a modified Strega's airfoil doesn't look so much exotic by the means of it's thickness.
And secondly, as Javafoil analysis has shown, up until flap deflection to 22 degrees, it has a very smooth Cl curve, a very reasonable Cm and quite a low Cd up until Cm=2.0.
This and all the flight reports of Stregas with blunted airfoil, convinced me to begin a work on my current project. Still I decided to use only a wing and fuse from a kit, all other parts have been redesigned and built from a scratch.
Vitalis
-
Does someone have an airfoil analysis of the Yatsenko airfoil as used on a Classic or a Yak?
-
as Javafoil analysis has shown, up until flap deflection to 22 degrees, it has a very smooth Cl curve, a very reasonable Cm and quite a low Cd up until Cm=2.0.
22 is too little deflection for corners, go over, to 30 or so and you will see forming those bumps on lift curve and abrupt changes of momoment.
-
Okay Igor. I have a question.
Why do you assume we need 30° in a corner?
I have seen in flight pictures of models in a corner and it was visible that the model had far less than 30° flap deflection in the corner.
I have even seen models that weren't capable of 30° flap deflection, there was simply not enough travel in the controls. And they still flew quite nice corners.
Did you take pictures in corners of your plane and analysed them? Or what makes you say that you need 30° ?
-
Here's a side question. Is there a point when these airfoil design arguments become somewhat futile, when all of the "best airfoils" are now, or have been seen sporting the addition of either zig-zag turbulator strips or vortex generators to augment lift?
Here's a few notable examples:
Any VG or turbulator will degrade parameters of airfoil (lift /drag) on airfoil over critical RE and we ARE over. At the same time it will improve controllability if it makes problems mentioned earlier. I never got success with VGs, but I had to use turbulator on wing because of not so perfect surface on my wing happened in one heavy rain on a contest (visible on first picture under the "S"), so I rather installed them on wing and it cleaned some imperfections. Since then I use composite wings and during trimming I always test also turbulators but I always removed them, so I did not have them in Perth and also not on my newest model.
Completely different situation is on tail (as described in article by Dave F. in Stunt news), if you take stab itself, it can be considered on edge of critical RE number. And analyze will show that there can happen very strange effects in boundary layer when transition point can abruptly jump between LE and hingeline what will certainly make some pitching effects like hunting), I also wrote it some years ago, so I not only use those mentioned sharp edged LE as Brett wrote (iy pushes critical number down, because it is source of turbulation), but I also install turbulator close to LE - if it is necessary or not. It will not hurt anything, I did not find any ill effects yet.
-
Okay Igor. I have a question.
Why do you assume we need 30° in a corner?
I have seen in flight pictures of models in a corner and it was visible that the model had far less than 30° flap deflection in the corner.
I have even seen models that weren't capable of 30° flap deflection, there was simply not enough travel in the controls. And they still flew quite nice corners.
Did you take pictures in corners of your plane and analysed them? Or what makes you say that you need 30° ?
Take those numbers from polar and make the math, you will see it is not enugh :- ))
... and the next question which you are going to write - yes I did that math, so I do not assume, :- ))))))))))
You go from totally opposite side of the problem ... you have to go down with area to load the airfoil to its maximal numbers, and it is 30 degrees on some, 22 on another and 15 on even other, but that wich will work well at 30 will give certainly better model, so you can use also that 15 you can even fly without flaps, it is only your choice, however which is better is clear :- ))
-
22 is too little deflection for corners, go over, to 30 or so and you will see forming those bumps on lift curve and abrupt changes of momoment.
Dear Igor, of course your airfoil is perfect in terms it doesn't have any bumps on ascending part of Cl curve, even with flap at 55 degrees (and I'm not kidding here). And Cl with flaps at 30 is phenomenal.
But hey, all those numbers are just numbers, although you really made me doubt ~^
As we all know even on your own model flaps are restricted to 27 degrees. At this setup Cl of your airfoil can theoretically reach almost 2.35 at 8 degrees of AoA.
Now.. if another (clearly not so noble) airfoil can smoothly reach Cl=2.1 at the same 8 degrees AoA and stay there up until 20, does that really matters to what angle flaps are deflected 27 or 22 ???
Yes it starts to form a bump with flaps deflected over 22 degrees and probably a model with this airfoil will not be able to perform extremely sharp corners, but I believe it's going to be at least a competitive one.
-
Now.. if another (clearly not so noble) airfoil can smoothly reach Cl=2.1 at the same 8 degrees AoA and stay there up until 20, does that really matters to what angle flaps are deflected 27 or 22 ???
Yes, it is mechanically limited, but I explained in my article the air flow hits the wing so, that the real deflection is higher, it can be 30 degrees.
And yes, you can make it travel less and ger lift say 10% less, it will need wing those 10% larger. ... plus you must tune flap to elevator ratio.
-
Does anybody know why Brett is so sensitive about Strega, Patternmaster and all the derivatives of Big Jim's legacy? Cmon, Brett, lay down your pain finally b1
The most important observation is above - at least the ARF Strega version has a pretty serious defect. It is NOT a "patternmaster" and the differences are significant and make a huge difference. Just looking at the results, I would suggest Phil Granderson knows a lot more about stunt airfoils than whoever drew the ARF Strega plans (which is not that surprising, frankly). In fact, the Brodak website contains this gem:
http://www.brodak.com/files/file/Strega_Building_Wing_Instructions.pdf
telling you to round of the LE, which probably originated from an SSW post by Gerry Arana from a conversation with Ted and I in 2007.
http://www.clstunt.com/htdocs/dc/dcboard.php?az=show_topic&forum=103&topic_id=280959&mesg_id=280959&listing_type=search
This fixes it, good on everyone involved, there was an issue, and it was addressed.
But once you realize that the LE is the problem, you can see how they came up with the other "features" - gigantically thick section and gigantic flaps. The Strega is far from the worst case, the MIG and others from the era had 40% flaps (which is derived from Abbott and von Denhoff, Figure 115). This in attempt to "get more lift" since it didn't work otherwise.
Of course, fix the LE, and you can run much smaller flaps and a much thinner airfoil, which then permits you to trim the airplane conventionally, instead of trying to manufacture tremendous amounts of line tension and run the CG forward to "get more flap deflection". That was the fatal flaw with Windy's system, and NO ONE competitive ever watched one of those airplanes fly, and then ran off saying "boy I better get the plans for that because I am going to lose otherwise".
People tell me the Strega plans actually show the LE shaped the same way, which I cannot confirm, since I have never seen the Strega plans. I do know that the Red/Silver Strega that Kent Tysor was flying didn't have that shape to it, and in fact, could not be achieved using the construction of the ARF. It seems to fly just fine and has none of these problems. I don't know about the original Patternmaster or BJ plans from Big Jim, himself, because I never did more than glance at them over someone else's shoulder one time.
The annoying part is that it has been repeatedly claimed that my buddies copied it or our airplanes are "based on it", in particular, that the Infinity, Trivial Pursuit, and even the Impact were "patternmaster derivatives". This is absurd from any perspective, but the most annoying is that it seems like we would *bother* copying something that already wasn't competitive. Why rip off Windy when I can rip off Paul/Ted/Billy (whose designs are astronomically more successful)? And in any case, I make absolutely no claims otherwise, freely admitting where I ripped off things and from whom, and where I came up with it myself. I did rip off Big Jim when it came to solid fuselage construction, which I also freely admit. I probably stole more stuff from Paul than anyone, and anything I might have gotten from Ted and David is kind of hard to pin down, since it was a free collaboration. I would never admit I stole anything from David, just on principle, of course.
But I also get all sorts of ideas and useful information from these threads, even the ones from beginners and neophytes.
For a long time, perhaps still, Windy was claiming that the Vector 40 was "based on the proven Profile Cardinal aerodynamics", which I am sure Randy greatly appreciated and is one of the most absurdly and demonstrably false bits of ad copy I have seen in this business.
It's that kind of stuff that drove everybody crazy about Windy. It was insulting and false, and served to take credit for other people's success. No one gave a flying tinker's damn about what he did on the field and I never had any issue with what he did as a competitor (although he was a fascinating case study). He was/probably still is an exceptional stunt competitor, with a record beyond 99.999% of everyone who has ever flown this event. Most people would be absolutely thrilled to be in even *one* Top 5 flyoff, much less dozens.
The off-field stuff, we "resolved" in private and we each know where each other stands.
Big Jim, I talked to him for a few minutes one time 34 years ago, seemed like a good guy, and he had some very good observations that advanced the state of the art, in engine operation at least. His design experiments were moderately successful, maybe it didn't lead too far - but no one knew that when they started and ultimately there's only one way to find out for sure. I have seen several pretty good-flying Patternmasters - but an ARF Strega is not a Patternmaster and only vaguely related, as far as I can tell.
I have said the same thing to people in public and private for years. Guys like you come around trying to spin up a controversy every so often, I don't know what you get out of it, but it is something I have come to expect. And besides, I am not the one trying to build a stunt wing with a conformally-mapped Ringmaster airfoil. We could all be wrong about that, too - but I doubt it.
Brett
-
Gosh .
Fair Enough Too .
This thing is Prety Much the P M 60 ,
(http://library.modelaviation.com/system/files/styles/zoom916/private/ma/ma199610/ma199610_047.jpg?itok=EO8buX8g)
(https://stunthanger.com/smf/index.php?action=dlattach;topic=16163.0;attach=58427;image)
As you can see , it cant be totally hopeless .
' The Story ' is its supposed to voek in da vind , But WHAT KIND of WIND . The Blowy Bumpy rough stuff , We Presume . As In it Dosnt get Blown into the Ground .
But Did That CHIPMUNK with the Thick wing Predeceed it ? . ( Yes ! ) Wot appens if you blow up the Chipmunk rib to the 9 & 11 in Chord ? How Thick . ?
This is the SIG Super Chipmonk that Rib Sets were availiable For , more than one plan in F M useing that rib set . RIGHT .
Sarpoulous , Sheeks , . . . .
(https://stunthanger.com/smf/index.php?action=dlattach;topic=33783.0;attach=139428;image)
Effective Incidance ,
------------------------
If we allow theres a ' seperation point ' at the L. E. , and the Departure is at the Aft Flap Edge , wot we have as a Line Through . Effective Incidance . ( Flap Deflected ) .
plus allowing for the tangental radius bit , as per Igor .
SO . . . . . . . . . . . . IF the FLAP is TWICE the Chord , Half the Deflection gets the Same Effective Incidance ( Thereabouts - leaving out a bit of algebra etc )
or a extreme example would be , Flap 3/4 of Total Chord . . . almost a variable incidace Wing ! .
if we put the wing at 5 or 10 5 O. A. Chord and the FLAP at 95 or 90 % , you can see where we're getting .
Less Extreme but no less , the Wide Chord L J Flap - Assume Rear Edge Deflection - Comparable to Std. deflection of common narrow chord flap -
and you start to see the picture .
But IS the moment / force AGAINST the Turn Comparable ??
Presumably with the flap deflected the C/L moves aft , approx % of O A Chord - as % of Wing Only Chord , Non Integral airfoil ( FLAT ) Flap . :-\ Thereabouts .
Thus The Dungers Shown below , Dont Approach the 1:1 Flap / elevator deflection ratio .
(https://stunthanger.com/smf/as-time-goes-bye/bob-baron-photos/?action=dlattach;attach=58419;image)
( Just Laminated Doublers for a Std ish Narrow Flap Blunt LE thing , simultaeneously ready to go on a wide flap sharp clean airfoiled ' wind ship ' ,
Both Fly Completely Differantly - but horses for courses . First bucks & Buffets in wind but good in calm , second WORKS Well with air moving, dull in calm air .)
( But do I build the BIG layout of the wind ship with ST 76 , or the STD one with the HP 40 , found Id done the mid sized trundler when I checked . :P Just as Well )
-
...and in fact, could not be achieved using the construction of the ARF.
Thank you for your honesty, Brett - it's a very interesting story.
But actually it's not so difficult to modify a wing of the ARF, one just needs a long metal ruler and a sharp snap-off blade.
-
Igor,
I understand your explanations, your formulas and thoughts. It all makes sense to me and I agree with most if not all of it.
What I don’t understand is how a Yatsenko airfoil ties in with this.
- They have a blunt leading edge but with a slight sharp edge. Almost like the STREGA ARF but not as extreme.
- They have almost no curvature behind the thickest point.
- They have an almost fully integrated flap, so at around 5° Flap deflection the “hingeline-kink” occurs that seems to be so fatal.
It seems they did everything wrong. Yet these airplanes fly very well, very predictable, very maneuverable, and very stable. Overall they are a very good package both with IC and electric powerplant.
Can you explain why?
Could it be that they make use of the “saw tooth” in the Cl diagram? Assuming that there is one at Flap deflections higher than 5°.
Could it be that this saw tooth comes so early that the entire pattern happens beyond that point?
I have not seen an airfoil diagram of a Yatsenko airfoil so I’m just guessing here.
-
Dear friends, dear colleagues,
I would like to have a little request of you, as follows:
I would like to make some statistics of experiments. Please, fly one half inside loop, as tight as you can, and a half outside loop, similarly.
Please, write me/us the achieved minimal heights. It can be estimated very precisely, and shows far more than analyzing any videos on square /triangle loops' corners' radii.
Will you help me? (and all of us...)
Thanks: Istvan
PS: explanations of the reason later...
-
Igor,
I understand your explanations, your formulas and thoughts. It all makes sense to me and I agree with most if not all of it.
What I don’t understand is how a Yatsenko airfoil ties in with this.
- They have a blunt leading edge but with a slight sharp edge. Almost like the STREGA ARF but not as extreme.
- They have almost no curvature behind the thickest point.
??? The few Yatsenko planes I have looked at closely have nearly a perfect ellipse as the LE shape, with even more curvature than the Infinity in the area about 1/2" behind the LE. There was nothing like the 45 degree pointy wedge. Take a try-square and put it on the LE of the ARF Strega, it fits perfectly and from the LE to about 1" along the surface, it's a DEAD STRAIGHT LINE at 45 degrees to the chord line.
Again, I would note that this IS NOT the same as the Patternmasters or the (not-ARF) Strega like Kent Tysors, at least those I have seen. You cannot make the ARF LE look like Kent's airplane with a wood rasp, if you tried you would wind up with 3/4" less chord and no structure to speak of. You would have to completely remove the ARF LE wood and make a new LE "plate". But you don't need to, the modification shown on the Brodak website seems to be sufficient, at least to resolve the worst of the problems.
Similarly, the Yatsenko airfoils I have seen are nearly flat around the high point, and have decent curvature which is required to make it down to the hinge line. The first 2.5-3" from the high point have a noticable hump. It's not like Igor's or Al Rabe's, with a lot of curve towards the TE, but it's certainly not like the Frankenstunt or other examples I have seen that are drawn with a ruler from right behind the high point to the TE. There's no problem with the ARF Strega or Patternmaster, or any other common airfoil, in this regard. That's why you can fix the ARF by rounding off the LE, even with the skinny 1/4" LE wood.
Note that the flat aft section TE is intentional, it's made that way so you can build the wing with the TE on the building board and lay the straight aft section flat on the building board. But every airplane I have flown with that airfoil has had ASTRONOMICAL control loading, almost no matter what you did with the rest of the airplane.
So, I don't see what you are getting at all, the Yatsenko Shark airfoils (and other similar Iron Curtain airplanes from others) I have see seem perfectly fine to me and I actually like the LE shape, it's very similar to the examples I came up with by plotting ellipses and then tacking a reasonable-looking aft section (using my shoe or any available curved smooth edge). Say what you want about the Yatsenko planes ruining stunt, but they *do* fly pretty well even by current standards, and at least the wing seems fine to me.
Brett
-
What I don’t understand is how a Yatsenko airfoil ties in with this.
Here is a quick analysis of Yatsenko Shark Ellipse airfoil with flaps at 30 degrees. The airfoil can be found in drawings here:
http://discovery-aeromodels.com/en/andrey-yatsenko-shark-ellipse-2-control-line-f2b-model.html (http://discovery-aeromodels.com/en/andrey-yatsenko-shark-ellipse-2-control-line-f2b-model.html)
Note: this is NOT an official analysis and I have not been provided with airfoil data from the manufacturer. It is very sketchy, as it is based on a low resolution drawings. You should not make any evaluation or conclusion from these images about true performance of real Yatsenko Shark stunt models.
-
Here are some drawings and pictures of the Yatsenko Shark airfoil for study.
The drawing was supplied directly by Yatsenko, according to PipeMakerMike. PDF and JPG are attached.
http://www.clstunt.com/htdocs/dc/dcboard.php?az=show_topic&forum=103&topic_id=311411&mesg_id=311411&page=9&topic_page=2
-
Here are some drawings and pictures of the Yatsenko Shark airfoil for study.
The drawing was supplied directly by Yatsenko, according to PipeMakerMike. PDF and JPG are attached.
http://www.clstunt.com/htdocs/dc/dcboard.php?az=show_topic&forum=103&topic_id=311411&mesg_id=311411&page=9&topic_page=2
Right, there's nothing like a point on that, and the aft section has a good curve to it. And it's negligibly different from a NACA 0020. The small differences shown here make NO difference in the results.
Brett
-
Andrey’s airfoil (Shark) has the rear surface more flat than Yuriy’s. Andrey told that he took the airfoil from some real aerobatic plane, I forgot which one, propably Sukhoi.
I know this because we had some issues with Shark wing skin resonating and breaking the wing ribs, a non-existent problem in Yuriy’s designs.
Also, Yuriy went back to a little sharper l.e. because some issues in landing glide (FAI 1 lap from 1,5m altitude), a blunter foil did not penetrate well enough. L
-
Andrey told that he took the airfoil from some real aerobatic plane, I forgot which one, propably Sukhoi.
Im not sure that a good rate of roll is neccesary for F2B Ship . Hence a less blunt Airfoil may be a better proposition .
-
not a clear one of Berringer Airfoil . ( used a Sharper Derivative on My Yak 1 / 11 , light , lively, responsive , bigger ( 200% ) Flaps . Light W/L ) .
(http://s018.radikal.ru/i526/1611/26/29b1f4c254d8t.jpg)
(http://www.clstunt.com/htdocs/dc/user_files/20236.jpg)
The Flat Flank Airfoils ( Flaps inboard 50% Span only, seem to ' air brake ', 'G' Wise - and hold in turn , under pressure / variable winds )
Igors Airfoil seems Very Similar to me to Al Rabes , ( 3rd & 5th down .)
(http://www.airplanesandrockets.com/airplanes/images2/sea-fury-al-rabe-aam-march-1973-12.jpg)
Id used his Mustang 5 Airfoil on a 17.5 / Sq Ft. Martin Baker MB3 . The Clean Corner - no bobble - apparently the no L?D Bump curve .
Good Penetration ( at that weight VD~ with G 51 Going Hard. In V Wind & V Rough Air , latter was going to down pitch to 4 in ,
which helped smooth the bumps- Torpedo 40 8011 with perry pump - still got the on/off 2-4 bit , but the fuel kept up , not ' out 'G'd .
Was a hair stand on end , razors edge job at that loading - but accurate & almost flew itself exact . Pilot just hung on and hit the handle
at the appropriate moments . ;D LL~
Not Exactly Relaxing - equivilant of Dodge 440 precision driving test on wet skidpan with the throttle stuck . But very Rewarding .
The Windships underway , with the bearers & doublers in . Laminate up a good spar worthy of a Windjammer round the Horn , Next .
S?P S?P H^^
P.S. ,
The ' BUMP ' in the Lift Drag Curve , would be what gave the olde ' Nailed & Turned ' Squares , aka Hunt etc , with the 4-2 run .
In FACT I BELIEVE THAT WAS WHY THE 4-2 was so popular - overcome the sudden increse in drag , IN the TIGHt Corners . :-X
Do the Hunt / Werwage Airfoils have the sudden bump in the L/D curve ?? H^^
*
Al's 25/22 AIRFOIL IS AT THE root only , Progresses to 22/22 at the deheadral break/ gear leg position , with simetrical out from there .
-
Andrey’s airfoil (Shark) has the rear surface more flat than Yuriy’s. Andrey told that he took the airfoil from some real aerobatic plane, I forgot which one, propably Sukhoi.
That is a common thing on full-scale aerobatic airplanes, but it seem to cause extreme control pressure on CL planes - probably for the same reason it improves the aileron effectiveness. Note that the full-scale airplanes also usually have spades or other control-pressure-reducing counterbalances. Just a little curvature seems to change it completely. I would add that the cambered version (i,e, slice the same thing in about half so it is flat-bottomed) seems to work great for propellors, and for HLG wings.
The airfoils that I am talking about (exhibiting all the bad characteristics) look like diamond-shaped flat plates with the edges slightly knocked off. A 5th bad characteristic is lots of curvature near the high point. That is necessary if you start with a diamond and knock off the corner formed at the high point. That is literally how I used to modify props, flat facets, then round off the sharp edges, and how you carve/sand HLG wings.
The ARF Strega only has the diamond shape in the front 1.5" or so (the rear half seems to be fine), presumably explaining why knocking the point off to a still-pretty-sharp 3/6 or 1/4" radius seems to fix it/ The Frankenstunt and many full-scale aerobatic airfoils have all of them. For those we have usually ended up using turbulators taped near the high point to get the air to go "around the corner" that the high point represents. It turned Frankenstunt (which is an Uncle Jimby airplane) from a disaster to merely extremely difficult to fly.
Bobby Hunt watched Jim fly the Frankenstunt at the NATs, and told him "you have a 200 mph airplane and 100 mph skills".
Brett
-
DIAMONDS ? yea, airfoilss are mathmatical ?? ( :-X) derivatives or ellipses & teardops / raindrops . I Should Think . :o
(https://ars.els-cdn.com/content/image/1-s2.0-S1290072914001823-gr1.jpg)
A bit of pushing & pulling , and youve got most of them . %^@ S?P
-
(https://stunthanger.com/smf/index.php?action=dlattach;topic=16163.0;attach=58427;image)
As you can see , it cant be totally hopeless .
' The Story ' is its supposed to voek in da vind , But WHAT KIND of WIND . The Blowy Bumpy rough stuff , We Presume . As In it Dosnt get Blown into the Ground .
Matt, be advised that I was present for that entire week, and watched every single official flight Bob did very closely. I was also one of the few people still willing to talk to him - briefly.
I saw him do a practice flight on Monday and it was immediately obvious that he was the favorite - just like the 81 Team Trials, if he was really "on", he was nearly untouchable. He wasn't always "on", and unfortunately for him (and the rest of us) he was unable to tell the difference or accept that other people could tell the difference.
As it happened, he just blew through the week as expected, and through the first flyoff flight. Then, it got a bit windy, and he started having problems. But so did Ted, with first his engine just fading for reasons we never figured out, then the infamous "bubble gum" flight. Then things got very close, and clearly Ted was flying better. But at the time, the scoring range in the flyoff was very narrow, and once Bob had posted his first flight score (560), it would have taken a big shift of something to catch him. It ended up pretty close but he still won.
Interestingly he was calling Mike Rogers between flights to try to get some coaching and suggestions over the phone. Mike's Patternmaster was far and away the best example I have seen.
BTW, I also note that you can see, even in the picture, that the LE is rounded off to the same degree or more than we suggested for the ARF Strega. And to give credit where credit is due, Al Rabe made the same point about pointy LEs in his American Aircraft Modeler column in about 1974, and when we did it more-or-less like Al recommended, it solved the problem.
As a completely unrelated aside, of all the really great fliers I have been around, including those who are much better in general, I probably try to emulate the way Bob Baron flew at his best (81 TT and 96 NATs) more than anyone - "clean" and "simple".
Brett
-
Vitalis,
If I interpret your diagrams correctly, then this Yatsenko airfoil has a nice Cl diagram without bumps up to 8° or 9° AoA.
It seems that the kink in the profile created by the deflected flap has no influence??
Very interesting!
Maybe it has to do with the hinge construction? I mean the leading edge of the flap itself is a perfect circle. Therefore when deflected there is no sharp corner.
I once talked to Yuriy Yatsenko about the logarithmic device like Igor is using it. He mentioned that his airplanes have something similar…. The hinge gap of the flap. He explained that the hinge gap on his design will change with increasing deflection. I checked this on my planes and indeed, the gap changes a little when you deflect the flap.
That reminds me of an airplane I saw last year. It belongs to a Ukrainian pilot and was designed by Sergey Belko. The hinges were made in such a way that there was almost no gap around 0° deflection. At around 20° deflection the hinge gap opens up to several millimeters. This was made intentionally, not by accident.
PS:
I doubt that a Yatsenk plane will have as much as 30° flap deflection in a corner. They usually have a ratio of 1/1.2 to 1/1.3 flap to elevator. 30° Flap would make 36° to 39° elevator. At least it is like that on my planes.
Here is a quick analysis of Yatsenko Shark Ellipse airfoil with flaps at 30 degrees. The airfoil can be found in drawings here:
http://discovery-aeromodels.com/en/andrey-yatsenko-shark-ellipse-2-control-line-f2b-model.html (http://discovery-aeromodels.com/en/andrey-yatsenko-shark-ellipse-2-control-line-f2b-model.html)
Note: this is NOT an official analysis and I have not been provided with airfoil data from the manufacturer. It is very sketchy, as it is based on a low resolution drawings. You should not make any evaluation or conclusion from these images about true performance of real Yatsenko Shark stunt models.
-
Vitalis,
If I interpret your diagrams correctly, then this Yatsenko airfoil has a nice Cl diagram without bumps up to 8° or 9° AoA.
Indeed. An ascending part of Cl curve is as good as on Igors airfoil and the airfoil can produce a fair amout of lift.
It seems that the kink in the profile created by the deflected flap has no influence??
I think it has. What's so different about Igors airfoil is a top part of Cl curve. It has a very smooth transition from ascending to descending part over high point. That makes Igors airfoil act predictable at AoA higher than optimal.
Maybe it has to do with the hinge construction? I mean the leading edge of the flap itself is a perfect circle. Therefore when deflected there is no sharp corner.
I think you're right. The big radius on LE of the flaps certainly helps to avoid a sharp bump on Cl curve. Is that enough for any possible flight conditions, well I don't know.
I once talked to Yuriy Yatsenko about the logarithmic device like Igor is using it. He mentioned that his airplanes have something similar…. The hinge gap of the flap. He explained that the hinge gap on his design will change with increasing deflection. I checked this on my planes and indeed, the gap changes a little when you deflect the flap.
I think the statement has a sense. There is no doubt that a varying gap does affect flap's performance, the question - exactly how. Unfortunately I have not found a way to reproduce influence of a hinge gap in Javafoil software (still need to learn it) :(
I doubt that a Yatsenk plane will have as much as 30° flap deflection in a corner. They usually have a ratio of 1/1.2 to 1/1.3 flap to elevator. 30° Flap would make 36° to 39° elevator. At least it is like that on my planes.
I have a saved message by Peter Germann about Shark where he claims that "... the flap / elevator deflection ratio is approx. 30° flaps at 45° elevator"
Moreover, derivatives of a Shark made by Leonidov that I'v seen had no restriction in deflection at all.
Regards,
Vitalis
-
I have a saved message by Peter Germann about Shark where he claims that "... the flap / elevator deflection ratio is approx. 30° flaps at 45° elevator"
Moreover, derivatives of a Shark made by Leonidov that I'v seen had no restriction in deflection at all.
I just checked the trimm on my Yuriy Yatsenko planes.
They vary between 1/1,3 to 1/1,4
That would be 39° to 42° elevator for 30° flap.
-
What I don’t understand is how a Yatsenko airfoil ties in with this.
Actually ... perfectly :- ))
Take closer look to pictures posted by Vitalis, if you really look what is happenening at 8 deg AoA, you will clearly see hysteresy problem which I methioned in that article. Step on lift curve making tracking problem and the same on moment curve making positive pitching feedback at that angle.
Means wing load must be lower than with airfoil having it at 10 deg AoA (or not having at all) so they will not need to go close or even exceed that point. And also 8 instead of 10 deg AoA making time for corner transition 20% shorter.
Real life influence is clear and known from real experience, those models fly Ok when light, it means also 20% less resistant in wind and turbulence. They need some extra flight to accomodate for different air density to avoid exceeding those steps, and they lead to oveshooting corners in case of stressy turns.
But I understand you, your model use it, so simply fly it as its is, I happy for that :- )))))
And regarding that gap - well, I am not going to comment its function because I did not analyze, and I will not, because it is making problems with trimming. But what is for sure different, is feedback from hinge moment. My unit limits feedback from hinge moment in corners what allows much easier flying especially in wind. That is second essential property of that unit and I would say most important. That gap cannot do it.
-
They need some extra flight to accomodate for different air density to avoid exceeding those steps, and they lead to oveshooting corners in case of stressy turns.
[
In some models I did seal the gap with some sort of a lip seal and result was exactly that, especially in turbulent wind.
Anoher bad thing is increased friction, especially in landing glide. Kind of a stick’n slip. L
-
They need some extra flight to accomodate for different air density to avoid exceeding those steps, and they lead to oveshooting corners in case of stressy turns.
[
In some models I did seal the gap with some sort of a lip seal and result was exactly that, especially in turbulent wind.
Anoher bad thing is increased friction, especially in landing glide. Kind of a stick’n slip. L
For sure, because sealed flaps make higher lift coeffient and that makes even lower AoA at low weight ... so it perfectly matchs theory and proofs what I wrote :- ))
Friction is clear if you do it on Yatsenko model, because it has pivot point aft of flap LE, means LE is moving up and down. That is actually nice trick to lower hinge moment. But it is very difficult to do sealing on such hinge construction. If you try it on classic hinges where is no moving on sealing tape, only deflecting, you will not see any friction. .. 10000 times tried :- ))
-
For sure, because sealed flaps make higher lift coeffient and that makes even lower AoA at low weight ... so it perfectly matchs theory and proofs what I wrote :- ))
Friction is clear if you do it on Yatsenko model, because it has pivot point aft of flap LE, means LE is moving up and down. That is actually nice trick to lower hinge moment. But it is very difficult to do sealing on such hinge construction.
Once I had the opportunity to look at one of the airplanes closely, I think the attempted construction was to not have the LE moving up or down, but to have it "roll" where the pivot is exactly at the center of the round flap LE. That doesn't really counterbalance anything, it just doesn't increase it as much as it might otherwise. That's more-or-less what Keith Trostle did on his Focke-Wulfs, and Al tested on his car hood (with arguable results). The Yatsenko hinge line construction is effectively identical to Keith's aside from the materials.
Done perfectly, that *at least* solves the "giant discontinuous hinge line" problem you would have with "blended" flaps, and if you make it fit tightly enough, would tend to have a significant effect of reducing the flow through the "gap". Unfortunately, on both the airplanes I saw, the pivot wasn't centered in the flap, so the rounded LE of the flap rose up above the surface when travelling in one direction, and was recessed below the surface in the other direction. This also required the hinge be installed with a much larger gap than would be hypothetically possible to prevent binding, so it was pretty wide open. It was also a much different gap in one direction than the other, once it got towards the end of travel.
Unlike most of the other strange and overblown analysis we sometimes do (and I engage it in too, that is not an accusation...) this sort of thing definitely DOES make a big difference. I have come to mostly the same conclusion of most others back to Al's car hood, you probably do better making the flap and stationary part of the wing thin at the hinge line, and doing everything possible to smooth the flow over the hinge line - putting a bit of curve in the aft portion to better fair in the surface when deflected, preventing flow through the gap, and rounding off all the corners it has to flow over.
In order, the people I ripped off for these tremendous personal insights, in order, Al, Denny Adamisin, and Paul Walker. If you are going to steal, steal from the best, I say.
Brett
-
Just a FWIW comparison, if this old image is clear enough (low res for SSW Forum years ago). I drew up a couple rather "fat" sections: a 23% section with an elliptical l.e. shape to the 30%-chord point and modified NACA aft drawn over an NACA 0023 section (as computed). They are pretty close. The black one is elliptical.
-
Vauge relevance , does have flaps , even if the airfoils ' inappropriate ' .
https://www.youtube.com/watch?v=c_lyV-CBBz4
of couse , all you need is a desert road & a sun roof , preferably in a van , a driver would save steering via string or your toes , while you do the tests . S?P
addn ; Measured from gen u wine PM 60 Patternmaster plan ( the Stilletto look use one ) Airfoil Depth O / A .
2 - 13 / 16 root , 2 - 3 / 32 Tip . L E Dia. , 13 mm root , 12 mm tip . ( checked with circle template ) These Exact measurements .
-
I'm saving this just in case I build a model with wheel pants.
I'll use it for the outline.
CB
-
I'm saving this just in case I build a model with wheel pants.
I'll use it for the outline.
CB
I'm very glad you like it. Actually this photo is distorted by fish-eye effect of a lenses.
Vitalis
-
Now I would like to share with you a Javafoil analysis of another one airfoil. Actually I'v been highly interested in it, since I built an electric version of a Brodak T-Rex last year. So here it is with a flaps at 30 degrees.
As you can see, Cl curve is linear only at negative AoA and does an abrupt dive at AoA 8 degrees. At the very same time Cd curve jumps up, indicating a strong positive pitching feedback.
Another thing is that the airfoil has the same tendency even with flaps at 15 degrees. Obviously this airfoil is good only for flapless designs, in a case with flaps it is necessary to avoid flight conditions where wing AoA can exceed 8 degrees.
Vitalis
-
Saved for ' when they thought its all over '
Id Thought the ' Drag Spike ' combined with the 4-2 run , Gave the ' Nailed & Turned ' Square Corners ;
With the engine break evening the delay & accelerating from the turn . The " CLASSIC " Stunt Run .
Thus horses for courses , the types requireing diferant answers from differing parameters . S?P S?P
-
Now I would like to share with you a Javafoil analysis of another one airfoil. Actually I'v been highly interested in it, since I built an electric version of a Brodak T-Rex last year. So here it is with a flaps at 30 degrees.
As you can see, Cl curve is linear only at negative AoA and does an abrupt dive at AoA 8 degrees. At the very same time Cd curve jumps up, indicating a strong positive pitching feedback.
Another thing is that the airfoil has the same tendency even with flaps at 15 degrees. Obviously this airfoil is good only for flapless designs, in a case with flaps it is necessary to avoid flight conditions where wing AoA can exceed 8 degrees.
Vitalis
Just keep AoA less than 8 degrees. It has plenty of Cl, although not as much as the sealed-flap Shark.
Edited to add some background, although not necessarily applicable to stunt. In bygone years I flew combat planes with an airfoil with an abrupt stall. It was a really good airfoil otherwise. The wings had to be straight, and the controls had to have stops to keep the AoA below stall, but the airplanes worked great.
-
Just keep AoA less than 8 degrees. It has plenty of Cl, although not as much as the sealed-flap Shark.
Actually, if you watch our planes fly in competition, the AoA is kept pretty low, if not negative. When you go through the part of Paul Walker's trim chart where you adjust the plane's elevator-to-flap ratio and bias so that it's level both upright and inverted, you're coming pretty close to adjusting it so that the fuselage flies tangent to a loop, instead of having to point inward as a flapless design would.
-
Now I would like to share with you a Javafoil analysis of another one airfoil. Actually I'v been highly interested in it, since I built an electric version of a Brodak T-Rex last year. So here it is with a flaps at 30 degrees.
As you can see, Cl curve is linear only at negative AoA and does an abrupt dive at AoA 8 degrees. At the very same time Cd curve jumps up, indicating a strong positive pitching feedback.
Another thing is that the airfoil has the same tendency even with flaps at 15 degrees. Obviously this airfoil is good only for flapless designs, in a case with flaps it is necessary to avoid flight conditions where wing AoA can exceed 8 degrees.
Vitalis
Bradley Walker used Bob Hunt's Saturn airfoil when he designed the T-Rex. https://stunthanger.com/smf/stunt-design/wing-thickness/msg135532/#msg135532
Not trying to be contrary, but it's doubtful that anyone heard of Bob Hunt's Saturn having any big troubles turning or falling out of the sky due to AoA.
The Bradleymobile uses the Saturn airfoil. Which is what I call my "thin standard" airfoil. I believe it is the perfect compromise in the slightly lower aspect ratio T-Rex (I mean Bradleymobile). I was particularly adamant with the factory that the front of the airfoil must perfect match the plans. They did not disappoint.
Viewed in CAD all of these airfoils start looking alike, at least from the Hunt and SV series. Randy's SV airfoils are about 3/32" thicker top and bottom than the Saturn airfoil (if blown up the same chord). The Saturn airfoil was supposedly derived from Billy Werwage. The Saturn airfoil looks suspiciously like a Genesi/Legacy/Buccaneer airfoil. As Brett said, these are all not way off the Imitation airfoil (which is a favorite of mine also).
I would not go thicker than Randy's airfoil. There is no need. Just adds weight, drag, etc.
I would suspect the Saturn airfoil is similar to PT airfoil also.
-
Another thing is that the airfoil has the same tendency even with flaps at 15 degrees. Obviously this airfoil is good only for flapless designs, in a case with flaps it is necessary to avoid flight conditions where wing AoA can exceed 8 degrees.
OK, not to be offensive, but the above suggests you have spent a bit too much time looking at javafoil, and not enough flying model airplanes.
Of course, with flaps, you are far less likely to get to the "critical" 8 degrees than you would without flaps - which suggests (if you believe the simulation, which in this case I don't, or at least not to the degree you are believing it), you would do even less well on a flapless design. The fact that the Saturn, which is known to be one of the better modern stunt designs, uses this airfoil - *very* similar in the important characteristics to others like the Thunderbolt and many other highly successful designs - gives you some important "ground truth" by which you can judge the plausibility of your CFD.
Additionally, a flaw I see with these simulations, particularly with flaps included, is that there seems to be no realistic modeling of the flow around the hinge line, which is known to be absolutely critical. That's not unique to your model, all of those I have seen have the same problem. Meaning it might well be right, or pretty close, with the flaps at neutral, and wildly incorrect (like I suspect) with the flaps deflected.
Brett
-
Javafoil and cfd is useful.... but be careful .... it may lead you to sniffing a brick wall
-
Javafoil and cfd is useful.... but be careful .... it may lead you to sniffing a brick wall
Any simulation has more-or-less the same problem, CFD, space shuttle, satellite, etc. What is relatively new is that people seem to want to believe them with a lot less skepticism than we used to.
It's a variant on the Powerpoint effect. When you had to hand-generate vue-foils, they usually looked crude to one degree or another, even the "production quality" types. With Powerpoint, you can make something that looks so slick and clean, that people tend for forget that the content may be questionable. This is particularly true with management types who have and never will *actually understand* what you are talking about, and judge the plausibility of your arguments based on how well presented it is, how confident/arrogant you seem, and how many hoops you are willing to jump through to prove your point. "Gee he's willing to call the corporate Chief Engineer in Baltimore, this must be really important..."
Same trap is even easier to fall into with simulation data. You can run 10,000 simulations a week, but if you didn't bother to set a particular single bit (out of megabits or gigabits) in any of them - which only one person in the world might grasp the relevance of - you might never find a really fundamental problem that will nonethless completely screw the pooch when you encounter the right conditions.
ANY time you do any simulation, you have to somehow validate the results somehow with reality somehow. The aerospace industry it littered with people who thought that they could dispense with sim results validation and just dot it all on a computer, and usually coming to grief later at some point.
Brett
-
Brett,
Could you please quit your job and US, and move in Switzerland to work for high end/luxury watch industry? L
-
OK, not to be offensive, but the above suggests you have spent a bit too much time looking at javafoil, and not enough flying model airplanes.
I'v had an electric T-Rex for more than a year now, and spent all last year tinkering with it. The weight is 1830g (64,5 oz) RTF. Аll in all I'v flown approx 150 flights. That's not a lot, but still enough to come with some thoughts.
So here I going to speak from my own little experiece.
It is not a bad design, but certainly has its flaws. One of those being a wing. I did not succeed to get a crisp corner out of it, and that was most obvious on triangles - way too big radius. I tried to play with elevator/flap travel, but it did not become any better whatever I did. Just a force on a handle increased tremendously to a level, that a few times I hurt my wrist and literraly I was unable to continue any stunt for the rest of a flight. So I set elevators back anf forgot that.
Another flaw - the plane doesn't have enought stability, at least to my requirements. I'v made a new stab from a scratch with a larger area and sharp LE, also moved it back for a half of an inch. That fixed stability issue quite a lot, but still not to the level of my old stunt plane. But that's not the case here.
Now Javafoil opened my I eyes a little bit, at least I can see thats going on with the airfoil. So I think that was not a bad idea to invest my time in learning it. I am also going to fly my T-Rex this spring (it's my first electric ship also), I will try to make some more adjustments, since I'v had more knowledges through the winter time.
Peace,
Vitalis
-
I really enjoy when Brett Buck adds to conversations. He really gets people wound up. Remember it's for fun!
-
Brett,
Could you please quit your job and US, and move in Switzerland to work for high end/luxury watch industry? L
Brett’s ancestors left Switzerland because the cheese had holes in it, and as Baptists, they refused to guard the Pope.
-
I'v had an electric T-Rex for more than a year now, and spent all last year tinkering with it. The weight is 1830g (64,5 oz) RTF. Аll in all I'v flown approx 150 flights. That's not a lot, but still enough to come with some thoughts.
So here I going to speak from my own little experiece.
It is not a bad design, but certainly has its flaws. One of those being a wing. I did not succeed to get a crisp corner out of it, and that was most obvious on triangles - way too big radius. I tried to play with elevator/flap travel, but it did not become any better whatever I did. Just a force on a handle increased tremendously to a level, that a few times I hurt my wrist and literraly I was unable to continue any stunt for the rest of a flight. So I set elevators back anf forgot that.
Another flaw - the plane doesn't have enought stability, at least to my requirements. I'v made a new stab from a scratch with a larger area and sharp LE, also moved it back for a half of an inch. That fixed stability issue quite a lot, but still not to the level of my old stunt plane. But that's not the case here.
Now Javafoil opened my I eyes a little bit, at least I can see thats going on with the airfoil. So I think that was not a bad idea to invest my time in learning it. I am also going to fly my T-Rex this spring (it's my first electric ship also), I will try to make some more adjustments, since I'v had more knowledges through the winter time.
Don't get me wrong, I am not trying to start anything, but there are about 1000 reasons that your airplane doesn't fly the way you want that I would look into before some obscure hypothetical property of the airfoil. In any case, the airfoil you modeled in Javafoil is not the one you actually have, and it's neither two-dimensional, nor the same when you look at it in 3 dimensions. For instance, you will get potentially relevant change depending on how tight the covering is shrunk, and it is definitely different between ribs compared to where it has ribs. In fact, guessing at what causes your unexpected separation in the math model, billowing or sunk-in covering would tend to DRASTICALLY alter the separation point. Have a few wrinkles in the right place, congratulations, you have turbulators. As far as I can tell, you are not modeling the hinge line flow at all, that, too would potentially have dramatic effects on exactly the sort of hypothetical problems you are having, and how they manifest themselves in real life. That's the problem with CFD modeling without the approrpriate controls or validation, and in this case there are very obvious potential flaws with it that you could easily predict.
I seem to recall the largest airplane manufacturer in the world designing a wing entirely using CFD, then finding the wing had a higher CL with the wing LE slats stowed than with them deployed when they actually flew the airplane. I also seem to recall people trying to do CFD models on stunt planes to prove that turbulators were worthless - even though we managed to get Ted's airplane through a NATs in the rain by taping 0.018 flying lines to the surface. Now, some of the same people have found no theoretical value for vortex generators that nonetheless were last seen with airplanes festooned with them, the bigger the better.
It's entirely within the realm of possibility that you have indeed found something that causes a problem. It's entirely possible that I will get hit by a meteorite on my way to pick up my Powerball winnings with my new girlfriend Cindy Crawford. But after reading back through the thread, I am even more inclined to suspect Frank's dog drawing is in fact on point. I would add that Frank was, among other things, a simulation expert for the Space Shuttle and I ran about 1500 simulations of a highly non-linear and moderately complex system 2 weeks ago, and spent most of the last week validating it with flight data. Mine had no air involved, but the more experience you have, the less you are willing to take simulations as unvarnished truth without validation.
The T-Rex is a decent and reasonably conservative design, if it doesn't fly well, we can probably help you get it to work.
Brett
-
Thank you for your honesty, Brett - it's a very interesting story.
But actually it's not so difficult to modify a wing of the ARF, one just needs a long metal ruler and a sharp snap-off blade.
A method I also suggested here:
https://stunthanger.com/smf/open-forum/strega-arf-trim-video-and-strega-arf-'control-line-world'-written-by-windy/msg456421/#msg456421
Do that, and there's no problem - but one you solve that problem, you also don't need it to be 3" thick any more, nor the large flaps, and by corollary, the nose-heavy trim or to manufacture lots and lots of line tension to deflect all this stuff into the relative wind. Which suggests that you might do better when designing your airplane starting with something that doesn't need to be "fixed".
Brett
-
Brett, I completely agree with you... No simulation should be trusted until it got validated in real life. And I agree that the means of simulation we have (javafoil) can't simulate all aspects of a F2B ship.
Writing this right now I feel rather stupid for starting this thread in the first place. Or maybe not?
Wouldn't a airfoil that is superior to another in theore be superior in real life as well? What I mean is, we make the same "mistakes" when calculating an airfoil for both, the superior and the other airfoil.
I think it is fascinating where this thread went so far.
It is also very fascinating to read the different theories. So many different concepts and they all seem to work. Beringer, Yatsenko, Burger, Windy..... All rather different but all seem to work.
So maybe there is a chance that the Wortmann airfoil in question will work as well!
The designer claims to have built a test plane last year and that seems to work just fine for him.
-
Writing this right now I feel rather stupid for starting this thread in the first place. Or maybe not?
Wouldn't a airfoil that is superior to another in theore be superior in real life as well? What I mean is, we make the same "mistakes" when calculating an airfoil for both, the superior and the other airfoil.
I think it is fascinating where this thread went so far.
It is also very fascinating to read the different theories. So many different concepts and they all seem to work. Beringer, Yatsenko, Burger, Windy..... All rather different but all seem to work.
It wasn't stupid at all, it's a very obvious question that happens to have no definitive answer.
After 70ish years of people cutting and trying, now somewhat enhanced by CFD (with the positives and negatives/pitfalls it brings), we think we have learned at least what NOT to do. And we are still learning, or at least confirming, previous understanding every time someone comes up with something new. In the last few years we have seen what happens when you get the shape, particularly around the LE, correct, you might be able to run 1/3 the thickness we used to use without any lack of lift performance.
Also, your list contains two that I would consider questionable - but nonetheless won a WC and lot of high NATs finishes. I think it might have been in spite of rather than because of the airfoil, but that just shows that if you are willing to thrash long enough and you are in the ballpark of OK, and get other things right, it doesn't make that much difference. It also shows that you can easily get led astray trying to fix an issue by misdiagnosing the problem.
The thing Frank and I were cautioning about was over-interpreting your data. Maybe someone comes up with a theory, tests it with a great-looking CFD run, builds it, and finds that the airplane flies great. That still might prove nearly nothing about the underlying theory because there are a million other things that are varying that *wouldn't* show up in a CFD run. Maybe you still have a crap airfoil or crap theory, but you happened to get everything else righter than you ever had before.
When you can do something (Vortex generators) that could do nothing but degrade the traditional airfoil qualities, and have the airplane fly better, or in some cases, MUCH better, that tells you beyond any doubt that these traditional airfoil qualities are not the determining factor in performance. Same thing with altering the first 1/4" of a 14" chord (rounding off the point at the Strega LE).
The other issue is that people get their egos wrapped up in these "theories" and then defend them to the death, regardless of the objective results. That's hardly unique to airfoil design.
Finally - the best stunt airfoil is the airfoil from the Imitation.
Brett
-
Mr. Wadle, not a bad question at all …. We are all still looking for the best airfoil shape, everyone thinks there might be something better out there. We have learned some things that are good and some things not to do, like building to a ¼ inch square le spar.
One of the biggest PA regrets through the years that I have, is not following up on the wind tunnel tests that were performed in the low speed tunnel at the University of Illinois at Urbana (UIUC) by graduate students of Dr. Michael Selig. Selig was/is a sailplane flyer and was involved in low speed wind tunnel testing during his education. He put out a request for models to test in the tunnel. All you had to do was build a model to the tunnel specs and it would be tested.
I wanted to test a Fancher type airfoil with flaps, a Rabe type airfoil with flaps, a Nobler, etc. , but I never got around to getting my project off the ground. The Selig project resulted in several volumes of data through the years.
http://m-selig.ae.illinois.edu/uiuc_lsat.html
Many sailplane airfoils were tested. The nearest airfoils to a stuntship wing were in some of the later tests of several RC trainer wings. Two of these tests were for the Ultra-Sport 1000 and the Trainer 60 airfoils.
The results of these tests, once again real wind tunnel measurements at Reynolds numbers appropriate to PA aerodynamics, turned out to be fruitful for us. Similar thickness airfoils (UltraSport and Trainer)with different leading edge radius, the two airfoil tests do clearly show an earlier stall of the sharper leading edge section. Although I haven’t done it, it might be interesting to load these two sections into JavaFoil as a comparison of numerical analysis to windtunnel results. Attached are the Selig results for these two airfoils at a Re = 500,000.
-
Please don't call me Mister :-)
-
Frank Wadle,
as Frank Williams suggested, open http://m-selig.ae.illinois.edu/uiuc_lsat.html
On that website, you can download the thesis of Williamson, G.A. Experimental Wind Tunnel Study of Airfoils with Large Flap Deflections at Low Reynolds Numbers.
There you could look at the data of airfoil W1015, with flap deviation up to 45°, at Re=400 000. CL up to 1,9!
Regards,
Wolfgang
-
I built a flapless plane with the Wortmann FX 71-L-150/25 airfoil. It flew just fine. No different than a typical NACA airfoil of the the same thickness. The thin trailing edge does make it heavier to build. I cut it out of foam with a slot for a 1/64in. ply trailing edge. That added enough extra weight to negate any possible performance benefits.
It would be really nice if one of the X-Foil junkies would run the airfoils needed to get a true picture of what actually happens on a stunt plane.- 0 flap at 0 AOA, taking into account that the airfoil with a deflected flap has an angle of attack from the leading edge to the trailing edge of the flap. A sequence of, say, 0 AOA 0 flap, 5 deg airfoil angle 10 deg deflection, 10 deg airfoil angle 20 deg flap, 15 deg airfoil angle 30 deg flap, maybe 20 deg airfoil 40deg flap. Plot the results against the combined airfoil/flap AOA at each deflection. I believe that would give a better idea of what is actually going on.
Igor's idea of contouring the airfoil('ala Al Rabe) to prevent the flap stall is still a good one. It worked for Al, and Igor has refined it quite a bit in other aspects of the plane design.
-
It would be really nice if one of the X-Foil junkies would run the airfoils needed to get a true picture of what actually happens on a stunt plane.- 0 flap at 0 AOA, taking into account that the airfoil with a deflected flap has an angle of attack from the leading edge to the trailing edge of the flap. A sequence of, say, 0 AOA 0 flap, 5 deg airfoil angle 10 deg deflection, 10 deg airfoil angle 20 deg flap, 15 deg airfoil angle 30 deg flap, maybe 20 deg airfoil 40deg flap. Plot the results against the combined airfoil/flap AOA at each deflection. I believe that would give a better idea of what is actually going on.
http://www.clstunt.com/htdocs/dc/dcboard.php?az=show_topic&forum=103&topic_id=371054&sub_topic_id=371056&mesg_id=&page=#371073 Start with post 42. Defining the angle of attack as being from the LE of the wing to the TE of the flap is not very useful and not done by folks who use such data. I did it for Ted's amusement in post 78.
-
It would be really nice if one of the X-Foil junkies would run the airfoils needed to get a true picture of what actually happens on a stunt plane.- 0 flap at 0 AOA, taking into account that the airfoil with a deflected flap has an angle of attack from the leading edge to the trailing edge of the flap. A sequence of, say, 0 AOA 0 flap, 5 deg airfoil angle 10 deg deflection, 10 deg airfoil angle 20 deg flap, 15 deg airfoil angle 30 deg flap, maybe 20 deg airfoil 40deg flap. Plot the results against the combined airfoil/flap AOA at each deflection. I believe that would give a better idea of what is actually going on.
You still have the problem of validation of the results, and the known limitations (like section characteristics VS finite span effects) of the CFD program. What you actually get in flight could be wildly different from the simulation.
Igor's idea of contouring the airfoil('ala Al Rabe) to prevent the flap stall is still a good one. It worked for Al, and Igor has refined it quite a bit in other aspects of the plane design.
That certainly seems to help, and in Al's case, is a good example of "guided cut-and-try" with real-life results. Whether he was trying to solve the right problem is certainly open to debate.
Brett
-
" Quote from: Bradley Walker on November 16, 2009, 06:23:42 PM
The Bradleymobile uses the Saturn airfoil. Which is what I call my "thin standard" airfoil. I believe it is the perfect compromise in the slightly lower aspect ratio T-Rex (I mean Bradleymobile). I was particularly adamant with the factory that the front of the airfoil must perfect match the plans. They did not disappoint.
Viewed in CAD all of these airfoils start looking alike, at least from the Hunt and SV series. Randy's SV airfoils are about 3/32" thicker top and bottom than the Saturn airfoil (if blown up the same chord). The Saturn airfoil was supposedly derived from Billy Werwage. The Saturn airfoil looks suspiciously like a Genesi/Legacy/Buccaneer airfoil. As Brett said, these are all not way off the Imitation airfoil (which is a favorite of mine also).
I would not go thicker than Randy's airfoil. There is no need. Just adds weight, drag, etc. "
There is a huge amount I would like to add to this thread about airfoils, I will if I get time to write it, but I would like to quickly correct the above, To sTart with, not many people know or understand the SV airfoils, The SV airfoils used on many different designs, are DIFFERENT and thinner or thicker, it depends on the size and needs of the stunt ship, in my mind there is NO ONE airfoil that is right for any plane.
The StuntCraft and Stuntcraft Vector , plus the Ryt Flyt airfoil systems are all adjustable, and can be made thinner, thicker, higher or lower A?R , more or less taper etc...
All of the above airfoils have dozens of rib stations to pick from , and you can slide up or down the scale (rib Stations) to design the wing you want, ie the VECTOR is ribe station 5 at the root, and rib 15 at the tip, these can be used to make foam templates , or ALL the ribs in between can be generated by computer
The SV11 based planes I fly are all about 1/8 inch thinner than the plans and kits I sell
The Vector STARIS , SATONA , SHIKE airplanes are smaller thinner SV wings than the KATANA or SV11
The Vector 40 wing is again thinner than the larger VECTOR and STARIS ships
The Elliptical SV VECTRA wing is a SV airfoil but is different than the above, but the curvature is the same
Then there are the Airfoils in the NOVA NOVAR Classic planes that are thinner than the SV large ship airfoils, and have about 5% more chord,
About comparison to others like the Cardinal, my airfoils are as different as night and day, much thinner, high point at a different point, chord, sweep A/R curvature of the airfoils is different , and there is NOTHING you can do with blowing up or shrinking to make them fit
The Hunt airfoils are different than mine, and do not fit either, The SV airfoils did not come from any of these airplanes
I designed the original " GEO BOLT" airfoil/wing that Bill Werwage used in his P-47s , Bill Later tweeked the airfoil thinning it adding a little more sweepback , but that could have been done on my CAD program and replotted, the P47 plans still fit almost perfectly the SV wng, it is just now further down the rib station numbers on the thinner side ie rib station 6 to 17.
This wing has been used by Bob Gieseke , Frank McMillan, Dave Fitzgerald , Bill Rich, the Barrys , Doug Moon , and many dozens all over the world, on Bob Hunts Stunt Flyer video magazine you can hear and see Billy pointing to his P-47 and saying that has a Randy Smith wing in it.
Ted's wing I know from cutting many for people over the past decades is not like mine either, his high point of the airfoil is further forward than mine, I have found that having the high point back, close to where the CG is works better for me. So you pay your money and get your choice ! :-)
So bottom line the SV wing is actually 100 of different wings if needed to be the similar thing is the curve of the airfoil , and I have made them from 500 sq inches up to over 750 sq Inches , and dozens of points in between, This works great for me, and as the have won multiple World Championships, many NATs wins and top 5 placings all over the World , They may not suck too much.
Randy
-
" I would not go thicker than Randy's airfoil. There is no need. Just adds weight, drag, etc. " "
There is NO best airfoil ! and NO fits all stuntships airfoil
Here is a Classic problem with statements like this, that could be true or false, just one example would be Scott Bair, he designed his ships to use very thick airfoils, because he wanted excess parasitic drag , His 700 sq in , 63 in span ships, with huge round bodies, 3 to 3 1/2 inch spinner 20 in tail moments and VERY powerful motors turning up to 16 inch props weighed in a 42 to 47 ounces ! yes I said 42 ounces, he was able to fly these in high winds and every condition , and with his designs he flew 6.2 second lap times, with authority , because of the DRAG he built into the plane.
by the way his thick wings didn't really add much if any weight, only drag
Airfoils are really complicated on stunt ships, and should not be considered as IDEAL for any plane, you need to take the entire airplane in mind, and the airfoils should "fit" what your doing. I have formulas and percentages for most all parts of the planes I use
Randy
-
WARNING; loooong meandering post to follow...
Been watching this thread with interest. My fundamental question is how do you define "best"; Highest lift? Lowest drag? Blazing corner while flying like a trainer in level flight? Only Muhammad Ali could "float like a butterfly AND sting like a bee", so choose wisely! We have seen credible stunt performance with airfoils ranging from 1/4" thick flat plates (the designs from down under) to really thick semi-symetrical airfoils with huge flaps (Rabe Sea Fury) to laminar flow sections to... you name it. While it is possible to design a BAD airfoil, I think there are other elements of the aircraft design that overhelm the contribution of the airfoil. I want to talk about a couple of those details.
GAP SEALS
THANK YOU BRETT for the shout out, presumably about hinge gap seals. My history on seals is this. When a preceeding design fell short of expectations, I started over with the first Eclipse (1980) with a full airfoil flaps. My 7:1 aspect ratio wing used a dead stock NACA 0018 airfoil at the root and a NACA 0021 at the tip modified with the high point moved forward to 25%. Of course the correct way to build flaps into an airfoil is to use a rolling hinge pocket per Keith Trostle and his excellent Ta-152 & Eagle designs. However, I figured the structural details were way out of my league. I resorted to a refined version of a simple hingeline. I limited the bevel on the flap to allow for a max of 30 degrees movement. I assumed a hinge gap and then put the absolute minimum bevel (i.e, less than 30 degrees) on the LE of the flap that would allow 30 degrees deflection The resulting surface gap was actually smaller than many "normal" airplanes I saw on the flight line. I also used a thickish full airfoil on the stab/elevator which ended up about .5" thick at the hingeline, and used the same minimum bevel philosophy there too. The resulting airplane was a complete dog - it would barely manage a 45 degreee round loop. I immediately gave up on the airplane, my brother wanted to mess with it so he took it home. The NEXT day he invited me to fly it again. The performance was now astounding. The difference was taped-over hingelines. The next Eclipse had the seals designed & built in, and I have used them in every airplane I have built since. BTW, I mentioned the hinge gap seals but no one was listening.
It must have been early 1982 Ted Fancher was in town on a flight layover and we visited. Ted had recently crashed his new airplane and he said the only thing worse than crashing the new airplane was he had to go back to an older airplane that he hated. I mentioned my experience with the hinge gap seals - with the usual caveat "what have you got to lose?" Several weeks later Ted won his first NATs with the gap seals on his formerly doggy airplane. Along with the performance gain, Ted mentioned that he had removed other trim elements that he had previously used to get the bird to fly less-bad (it had never really flown well before). In short, the seals gave him a brand new airplane and he started over on trimming. This just in: Ted is an EXCELLENT builder, but even then it seems the gap seals overcame any inconsistencies he may have had in the hinge gaps, prompting the opinion that gap seals should be used all the time for just that reason. Ted mentioned the effect gap seals had, and between his NATs Championship & MA column (where he generously mentioned where he heard the idea) everybody was listening now!
Real world experience seems to range from "didn't notice a change" to game changer (Ted & I). I can rationalize some possible ideas why, but really do not care. I like the way David Fitzgerald once summed it up: gap seals have all the advantages of adding tail weight with none of the disadvantages. I just use the seals and move on to bigger issues.
WING TO FLAP AIRFLOW
Regarding the discussion here on getting a smooth airflow on the top surface when the flaps are deflected. No doubt, my simple flap on the full airfoil without gap seals was nearly the worst scenario possible. Al Rabe's approach certainly looks like the best approach - but what ANGLE of flap deflection do you design for? Logically it seems prudent to design for the best airflow at max deflection. However, please note: a airfoil designed for a smooth airflow path with 45 degrees flap will be SUB-OPTIMIZED for every flap deflection less than 45 degrees. I typically set up my airplanes for 30 degrees or so max deflection, and I never fly up against the stops, so to me designing for a smooth transistion with 15-20 degrees deflection makes the most sense. I also use slow controls, with CG a bit aft (I think) from what most folks like. Brett has a funny story about checking the CG on one of my airplanes...
Another building detail may be more important than the airfoil shape: how should we deal with airflow jumping the hinge gap? It is fashionable to make the front edge of the hingeline (the trailing edge of wing or stabilizer) squared off, and put all the bevel for surface movement on the leading edge of the flaps or elevator. I suggest this approach is incorrect, for the same reason that using a simple hinge on a thick flap is incorrect. The hard edge on the trailing edge enables/promotes airflow separation - that we then want to re-attach the airflow on the top of the deflected flap. I suggest (and my last several airplanes have exhibited) that the trailing edges should be rounded - to enable a smoother transition from the wing, across the gap, and onto the flap. Yup, I am using 1950's philosphy (plus seals) on hingelines! However this approach, combined with Rabe inspired airfoils with the proper design point on flap deflection, and gap sealed flaps, SHOULD result in the smoothest airflow transition. Then do it again on the tail.
A NEW IDEA
My flying buddy Wesley Dick researched flap design and learned that the double slotted flaps should yield more lift with a lower drag penalty than a simple flap. His personal contribution was to design a simple and robust mechanism that would allow the flaps to deflect in either direction. The double slotted approach actually tries to take advantage of the gaps to create more lift. Wes built two of his "Velvet" designs with normal flaps, then 2 with the double slotted flaps. From the outside of the circle the two airplanes with the double slotted flaps seem to fly better than the two with the simple flaps. There are several new Velvets under construction so we will get some more/new opinions on the merits of double slotted flaps. Then SOMEONE is going to try it for the elevator too.
Of course the trailing edges of the flaps and elevators should be as thin and sharp as possible, but that is another discussion..!
-
Wes Dick's Flaps.
-
GAP SEALS
THANK YOU BRETT for the shout out, presumably about hinge gap seals.
That was another one of those ideas that just made me slap my forehead and go D'OH, why didn't I think of that 20 years ago! Everything is obvious - once someone else invents it! Even if the surface were not more effective, it would be worth doing, just because it makes the flow through the gap more consistent (zero is very consistent) compared to having a very small gap that changes by even a few thousandths. It was obvious right from the beginning that the gap was critical, and used to spend hours trying to sand them to the same very small values. My one contribution to the idea is the comment that "no matter how small you make the gap, its still bigger than an air molecule".
Brett
-
I started sealing hinglines in the early 1970s, I do not now who invented it, but Henry Schutte showed me how to do it with Monocoat , or fabric hinges on painted planes. Monocoat / Solarfilm was much easier, so most of my planes were covered with monocoat, and painted on the nose. Monocoat hinges were strong, easy to make, and completely sealed the flap to wing joint, and elevators.
Some people today still use that technique with film, and the cloth hinges that cover the entire gap. I always hated cloth hinges, so i rarely used them.
Randy
-
Thanks Howard and Brett. I totally missed that debate 5 years ago. Other problems were in the forefront. And yes, Brett, CFD is not fool proof, no model is, and careful experimentation such as Igor has done have certainly paid dividends for him and others.
The graph, Howard, is exactly what I was looking for. It shows what I was thinking. In flight the behavior of an active-flap stunt airfoil doesn't have the classic, straight line ratio between lift and drag. It curves to the left of a straight line and makes the top of the curve broader and flatter. Also, the change extra lift/drag falls off dramatically after 10-15 deg. of flap movement. Too bad the flaps 20 and flaps 30 airfoils are already well past stall at 15deg. AOA.
That kinda explains what most of us have seen. A slightly over weight plane, up high into the wind, a gust, and the pilot's frantic attempt to give full control and recover just makes the plane fall almost straight down. Everyone needs to learn that up high, in trouble, more control is the wrong way to go. Back off, whip it if you can, and try to get it back down lower where you may be able to recover. Practice with a clunker that won't wound your ego if it crashes. Push it near it's limits and practice recovering from minor brushes with disaster before you have to recover from a real one.
None of which change the common wisdom. Flaps help, quite a bit, especially with the appearance of the plane during maneuvers and weight carrying. The interplay between all the variables can be tuned to work for most designs. And mother nature has built a wall between precision and tight turns in the planes and in our reflexes.
-
It seems as though R/C Pattern flyers embrace gap sealing more so than stunt flyers. Gap sealing prior to first flights always a good idea.
-
You still have the problem of validation of the results, and the known limitations (like section characteristics VS finite span effects) of the CFD program. What you actually get in flight could be wildly different from the simulation.
That certainly seems to help, and in Al's case, is a good example of "guided cut-and-try" with real-life results. Whether he was trying to solve the right problem is certainly open to debate.
Brett
I know this is an old thread but it's focus is exactly what I am working on today. Many years ago I used to use a "piano" type hinge using different sizes of aluminum tubing. Looked great, not too heavy and it completely sealed the the hinge line but they wore out too fast and started to bind under stress so I abandoned them. The next build of the same design never flew as well and I did not know why. After listening in on some discussions with Al Rabe about recessed flaps, which he had abandoned at the time, I decided to use them on the next one and I have been using them ever since on flaps as thin as 1/4". It takes as long to get the flaps and elevator fit perfect as it does to build the rest of the wing but it really looks nice. What I need to know is whether or not I am actually sealing them. The LE of the flap is never aft of the wing TE and the pivot point does not produce the "Hump" that Al thought may have been a problem in his tests. Bottom line - am I wasting my time?
My brother is a well known test pilot (within test pilot circles) and he would completely and emphatically back up your comments on simulation vs. actually flying.
ken
-
Get your ( wifes ) Vacum Cleaner . ( might pay to foam pad the end . Dont Blame Me , if you dont ! )
three guesses !
Put it on one side and bare fingers the other side of the wing / hinge . Try Suck & blow .
Invite SMOKERS ROUND ( Pref Bob Marleys Cousins ;D) give them each a straw .
Put vacume on suck & blow ( variously ) . And try things .
This is probly easyier than a tube , a bag of smoke , and sticking the wing outa the car window at 60 mph . S?P
-----------------------------------------------------------------------------------------------------------------------------------
If you cut / machined a steel tube in half , lengthwise . For the L E of your flaps .
And maybe machine a radius on a plate , or glue a tube / rod to a thin straight plank ,
youd have TOOLS to fine finish the mating surfaces , wether with fine sandpaper of lapping paste ( Toothpaste ? )
Might save a lot of faffing about .
-
Al Rabe used to claim that he designed his airfoils by tracing around his shoes. If I could walk a mile in that man's shoes -- I'd trace them out first.
Tim,
Didn't that comment originate with Wild Bill Netzeband around the time of his classic "Control line Aerodynamics Made Painless" series back when I was a kid? As I recall it was something like "X airfoil is ".6X%" better than one drawn by tracing around the sole of my dress shoe" or something like that. I think Al was generally more straight forward in developing his with the benefit of a simulated wind-tunnel mounted on a pickup truck and was seldom inclined to speak of such things with the intent to prompt chuckles from his readers.
Ted
Just dug up my old binder full of Netzeband stuff and found the "....Painless" series started in the September/October 1966 issue of American Aircraft Modeler. I was, sort of, an adult by that time; no longer a kid. I have most of his "Round and Round" columns from MAN on file as well but it will take some time to go through all of them to see if his "shoe-foil" comment was in those columns.
-
This thread is heavy on graphs, charts, and arcane theory. That's nice, but I'm wondering how much these finer points really matter in C/L stunt.
I'm an engineer but not much on aero. So I did the next best thing. I used the good-old NACA 0025 with flaps just stuck on the back! Drag is what I wanted, hopefully to prevent speed-up going downward.
To overcome drag is easy; use Horsepower. I tried a Saito 61 4-cycle, and that certainly solved the horsepower question.
How's that for a non-technical response to a highly technical discussion?
-
Just for the heck of it while trying to debate what is and isn't the "best" stunt airfoil I was reminded of a years ago conversation with Big Art Adamisin shortly after his return from down under as a guest of the many stunt builders and fliers who cling to the planet by flexing their toes into the sod. Art had been there as their guest to do clinics and some judging.
One of the things he talked most about with me after his return was how surprised he was to have observed, critiqued and judged one (or perhaps several) Aussies who flew excellent patterns with original designs using flat plate wings (although flapped, IIRC) and were competitive despite what we, in this thread, may have been tempted to describe as perhaps the bottom of the barrel when it comes to describing the "best" stunt airfoil.
Perhaps Denny Adamisin can throw in some additional info based on more extensive discussions on the subject with his Dad. Also, if any of our Oz friends have been following this discussion perhaps they can provide some details on who, what, how well and whether the fliers have continued with their no-french curve involved experiments/experiences. Given the three SH pages of effort and serious debate over maximizing the performance of our stunters through optimization of their primary lifting surface the efficacy of such rudimentary structures would be of great interest.
Ted
-
(http://www.retroplane.net/forum/images/uploads/Philippe26/IMG_4274__Custom__148.jpg)
Philippe Moya .
Ant. Klashnikovs Interpretation . From the Horses Mouth .. ! .
(http://forum.rcdesign.ru/attachment.php?attachmentid=1260506&d=1476269394)
Similar elongated on the 57 in Hurricane was unaffected by Wind , But not flown in bumpy air ,
so couldnt tell you about that , there .
-
Ted
Yep, you remember the story about as well as I do. Dad was in Australia some years back and judged at their Nationals. There was a guy there (no idea what is name is) with a flat plate stunter powered with a 19 (?) diesel. Dad said he watched it, was amazed at how well it flew and later asked the guy fly a pattern for him so he could video tape it (I have never seen the tape). He said the pilot/airplane was never in any danger of winning but he placed. A post by Russell Bond from down under indicated that pilot placed 8th one year and 4th another year. Russell said he was the test pilot for one of the flat wing stunters and he was so amazed that he flew it a few more times - as I recall Russell is an enthusiastic supporter of the flat wing concept for a sport-stunter. I attached an article that I found somewhere (maybe from Russell?) that describes these stunters
Basic layout is a 1/4" flat plate wing with reinforcements at center, some (not all) had full span 1/4" spruce spars embedded in the wing. The "formula" also included diesel engine (holds RPM under load), and 5 degrees engine offset. A lot of them were designed to resemble other full stunters. As you say, it runs counter to basically everything we think we have learned about airfoils, especially the part about LE bluntness. It MUST be generating a lot of lift via camber change & the trailing edge flaps? It certainly is not generating lift based on the airfoil contour...
Based on dad's observations from so long ago, and my own curiousity I am building one just to see for myself... (sorry, no diesel)
-
I have that article and another related one. One contributor to the 1/2-A Forum here posted a picture of a full-sized slab-wing version of Ted's Imitation. When I have time, I'll search for it in my own files. Apparently the extreme thrust offset is important.
SK
-
This thread is heavy on graphs, charts, and arcane theory. That's nice, but I'm wondering how much these finer points really matter in C/L stunt.
I'm an engineer but not much on aero. So I did the next best thing. I used the good-old NACA 0025 with flaps just stuck on the back! Drag is what I wanted, hopefully to prevent speed-up going downward.
To overcome drag is easy; use Horsepower. I tried a Saito 61 4-cycle, and that certainly solved the horsepower question.
How's that for a non-technical response to a highly technical discussion?
I think you may have the answer. We try and fine tune to 1/4" +- 3'
Guilty as charged
Nice looking ship - Ken
-
"snip"
We try and fine tune to 1/4" +- 3'
"snip"
I love it!
That's my maxim and I'm sticking with it!
Ted
-
How's that for a non-technical response to a highly technical discussion?
Maybe everyone can get together a list of topics we are allowed to discuss without someone coming along and mocking the effort? That would be really helpful.
Brett
-
The STUNT - WIND full size plan , is downloadable from Here ( Hippocket ) :
http://www.hippocketaeronautics.com/hpa_plans/categories.php?cat_id=71&page=5
if your logged in there .
(http://www.hippocketaeronautics.com/hpa_plans/data/thumbnails/71/stunt-wind.jpg)