Aerodynamicist and Other Opinions

I would like comments to the following:

On my modified Twister III I am planning on using small tip plates at the trailing edge of the wing similar to but somewhat larger in proportion than those I see on the tips of United Airlines AB 320's.
The goal is to at least alter the tip vortices so that instead of heavy rolling in wake the vortices are reduced to a more gentle wing rocking should they be encountered. The second main feature to the Twister III mod is to increase the stab/wing moment arm by 1.5" and then use a stab/elevator combination that is around 30% of the wing area. The stab/elevator thickness will be increased to 1/4" and incorporate aerodaynamic balance like found on the Van Loo Chipmunks. The use of the aerodynamic balance is due in no small part to Dave Fitzgeralds' Dee Tails article in Stunt News. I considered using a carbon fiber landing gear but passed on that idea so that every beginner could do the modifications easily at minimal costs. Some other features are adjustable tip weight, adjustable line rake and a removable tail wheel that can be lengthend as needed. The wing will be extended 1" by using 1/2" balsa tip plates on each wingtip. The 1/8" lite plywood tip ends will be used to have an attachment point for the trailing edge tip plates. They are of course experimental and can be changed or removed if no improvement is gained. In order to make the stab stiff enough, I will cover it with two layers of .02 CF veil using waterbased polycrylic for application. The landing gear will have a stainless steel connecting shaft through the fuselage to help strengthen it. Also the fuselage will have a seperate 1/16" birch plywood nose tripeler on the outboard side and the inborad side will have a light 1/4" balsa nose tripeler. The tank will be a plastic uniflow mounted similar to the method that Mike Pratt uses on his profiles - Thanks Mikey! Of course there are better profile airplanes out there but the good old Twister is still one of the cheapest and that's usually what a beginner wants is something that won't break the bank if it's busted up in the learing process. O.K. all, comments on these changes to a good old venerable C/L airplane anyone?? Or in other terms let the .... oh well, whatever comes!

>On my modified Twister III I am planning on using small tip plates at the trailing edge of the
>wing similar to but somewhat larger in proportion than those I see on the tips of United
>Airlines AB 320's.

>The goal is to at least alter the tip vortices so that instead of heavy rolling in wake the
>vortices are reduced to a more gentle wing rocking should they be encountered.

Jim-

I don't know whether I'm thinking of the same airliner tips, but I'm guessing that they are a sort of Whitcomb winglet, as on the Rutan Vari-EZ - a canted, near vertical tapered fin with inboard camber and set back along the tip chord?? If so, are you planning a symmetrical treatment (one above and one below)? (Edit: Now that I think of it, that was a silly question for me to ask about cambered winglets! Please disregard.) As I recall, these were meant to "unwind" the vortices and regain some of their energy.

Off hand, this sounds like some complexity. Although I suppose a flat plate might interrupt the vortex some, I'm wondering whether you might better achieve your goal with extended tip treatment, either raking the tips back, with a tip span of around half the chord or greater, or separating the vortices with a couple in-line tips (more structural problems). The latter not only break the vortex into smaller ones, but separate them vertically, when operated at some aoa, possibly causing them to interfer with each other. Edit: You might find the Stanford aeronautical site's paper on non-planar wings interesting, since it has quantitative changes for everything from plates to C-wings to bipes.

Just food for thought...

>I would like comments to the following:
>
>On my modified Twister III I am planning on using small tip
>plates at the trailing edge of the wing similar to but
>somewhat larger in proportion than those I see on the tips
>of United Airlines AB 320's.
>The goal is to at least alter the tip vortices so that
>instead of heavy rolling in wake the vortices are reduced to
>a more gentle wing rocking should they be encountered.

Hi Jim The purpose of tiplets on airliners is not primarily aimed at reducing tip vortex. Cambered tiplets are angled to the vortex cross flow in order to gain some forward thrust from it. That however the angle is critical and only works at one specific trim (crusing speed) to reduce fuel cost. This obviously will not work for all the filght conditions on a CL. Flat tip plates and in fact any specially shaped tip structure have only minimal effects on drag. That is because the vortex is a very large structure. Both vortices together are twice the span across so any effective tip structure must be exceedingly large to have more than a minor effect. However, you may get some improved, stable tip flows from some shapes so worth experimenting.

>The
>second main feature to the Twister III mod is to increase
>the stab/wing moment arm by 1.5" and then use a
>stab/elevator combination that is around 30% of the wing
>area.

Increasing control movement gets to a point where the drag increases more than control moment. Anything above 30 deg is getting dubious, The old 45 deg rule is really the outside limit I think. Increasing stab area is going to increase stability as well as pitch moment. Both these things can benifit manouverability up to a point. If you want to tighten up the loops, you might be better trying to improve lift over mass. Reduce weight, increase wing area, use a heigher lifting section, flaps, even turbulators can help.

>The stab/elevator thickness will be increased to 1/4"
>and incorporate aerodaynamic balance like found on the Van
>Loo Chipmunks. The use of the aerodynamic balance is due in
>no small part to Dave Fitzgeralds' Dee Tails article in
>Stunt News.

Mass ballance can aid smooth control too.

>Some other features
>are adjustable tip weight, adjustable line rake and a
>removable tail wheel that can be lengthend as needed.

Tip weight a good idea. I just finished a spanwise center of lift routine. If you tell me the span and line length to geometric centre of the wing, Ill run it for you. I'm interested to see what you think of the results.

>The wing will be extended 1" by using 1/2" balsa tip plates on
>each wingtip. The 1/8" lite plywood tip ends will be used
>to have an attachment point for the trailing edge tip
>plates. They are of course experimental and can be changed
>or removed if no improvement is gained.

Good idea for experimentation.

>The landing gear will have a stainless steel connecting
>shaft through the fuselage to help strengthen it.

Not sure about the weight might be lighter and stronger to streangthen the outer surface of the fuselage like a tube structure.

Serge, Ken,

Thank you for your replies. Just two replies to this?? Gee, I thought that I put my deoderant on?? Well, I know there are lots more out there thinking about a reply - I hope anyway!

Let's see now - Ron, Bill, Keith, the other Keith, Tom (McClain), Rollin, Curtis, Randy, the other Randy, Ted, Brett, Soil, Paul, Alpha, Al and many others. I know your computers aren't broken. I do see other posts on here!

Jim,

Not an Aerodynamicist, so I'd definately be an other opinion. And my expert opinion is: beats me.

My computer WAS broken. Actually, it was the SecureID, which made the computer worthless for dial-in.

The tip plates sound cool, maybe they will work. I often have trouble translating big plane stuff to our smaller Reynolds numbers. The concept sounds okay. I'm picturing mirror Hoerner tips in my mind and would extend them to cover the flap tips as well. Make up a batch of differing sizes and let's try them out. Even if they don't work, the "coolness" should net a few impression points in competition.

I think double layers of CF on 1/4" balsa on the stab is overkill. One layer each side should be plenty and will save some money. Your front end sounds just like mine, plus I would fiberglass the fuse (1 1/2 oz. glass on the nose, 0.5 oz. on the rest).

Sounds like fun.

Just one quick note - longer tail will need more effective elevator - little more deflection or elevator to stab area.

Why?

Jim Here is a picture of a plane I built around 1985.Notice the tip plates on the flaps at both ends and the middle of the flaps.At this time Bill Rich and I flew together at least twice a week. When watching his foam wing planes fly in the wind I saw the nose of his plane was wondering in and out in the wing over and level flight. This movement was very small and he did not notice it.But watching him fly flight after flight I could see it, and in a lot of other foam wing planes doing it also.I didn't see this in ribbed wing planes.I built this plane to see if it would stop this action and increase the flap use in windy weather.This plane did not have swept back wingtips as many do today.This plane was called the Nostome Wind Machine and did fly very well in the wind.It had a short life as a solid leadout broke and it went in and I gave it away as not wanting to repair it. It was about this time that many started to use the swept bask wingtip and that seemed to correct the problem.I think a lot of these projects we take on give us satifaction even if it does not help the plane. AS Bill Werwage has said "the plane we are flying at the present allways seems the best even if it's not"
Ed

The doodads on the ends of the flaps are very cool. Too bad you didn't get to try the airplane with and without them.

A Northwest local yokel built an Impact with the last couple of inches of flap fixed. It suffered from aerodynamic perversions (I forget just what), presumably because of the gap that opened between the inner and outer flap when the flaps were deflected. He gave up and connected the outer flap to the inner per the original design. I wondered if sealing the crack the way you did would have helped.

>>Just one quick note - longer tail will need more effective elevator - little more deflection or
>>elevator to stab area.

>Why?

My thoughts:

The longer the tail,...

1) the more positive the neutral stab/elevator angle of attack in a turn (that is, against the turn, a la Zaic).

2) the greater the polar moment of inertia (despite having a greater tail moment, it's an "R^2" thing, right?)

3) the greater the acceleration needed to maintain quickness over a greater distance (see #2).

Did I miss anything?

Sounds right, given a constant cg. #1 is Cmq, I presume.

SORRY!

#1 was worded wrong in post #9 - just corrected. The relative motion of an undeflected tail to the air in a turn is such that it fights the turn - i.e. it's stabilizing. For an inside loop, the stab/elevator in its neutral position would meet the air at a positive aoa. So it would need some deflection (trailing edge up) just to get back to neutral effect (sustaining turn), before it could increase the turn rate. The longer the tail, the greater this aoa difference between tail and wing. 'hope that didn't muddy things up too much.

I don't remember what you said, but I think it was OK. It's why I like long-tailed combat planes.

Was wondering what type of wingtip treatment was being considered, and not real familiar with AirBus aircraft 'round these parts, tho I did ride on one, United, at that, from Boston to Denver a few years ago. It made awful noises before takeoff. Anyway, I went looking for a picture, cropped out the excess and got a really small picture to post. Dunno if it will show much. Cutting down the turbulance seems like a good idea, but will it be a significant proportion? Steve

Here is another thought on longer tail. There is a useful concept of stability called the neutral point (NP).
Basically :

A = E * T * M /(S - E * T)

Where:

A = Distance from Wing AC to NP
E = Tailplane efficiency (0.6 for low in line stab)
M = Moment Arm (distance between aerodynamic centres of wing and Stab)
S = Wing area
T = Stab area

Take the AC as 25% of mean chord for both wing and stab.

The greater the distance (as a persentage of chord) that the NP is behind the CG, the greater the stability of the plane. So if you lengthen the tail or increase the stab area, you increase stability which for reasons some have mentioned tends to fight manouverability.
However if you use the above to calculate the NP, you can adjust CG to keep the same CG to NP distance (same stability).

Why?

Well the more rearward the CG, the greater distance between centre of lift and CG. As CG is the point through which enertia opposes lift, this increases the moment couple that rotates the plane in a loop
while maintaining the same stability at level trim.

Steve,

Yes, that is the tip plate style I am talking about! It will be
proportionally larger on the model but not be so big as to cause
interference with the ground on Take Offs and Landings. Also I
am going to make the tail wheel longer which will help. The
Idea is for the plate to cover the trailing edge to slightly aft of the end of the flaps. Igor, don't forget that the stab/elevator
will encomapss about 30% of the wing area and be split 55% for the
stab and 45% for the elevators. I think that will give plenty of
power for the square turns. In fact I'm so concerned about flexing,
that is the reason that I figured it probably needed two layers of
the .02 Carbon Fiber.

>1) the more positive the neutral stab/elevator angle of attack in a turn (that is, against the >turn, a la Zaic).
>2) the greater the polar moment of inertia (despite having a greater tail moment, it's an >"R^2" thing, right?)
>3) the greater the acceleration needed to maintain quickness over a greater distance (see >#2).

Yes, that is what I mean and even worse, the same force (lift) on longer tail moment arm makes only linearly higher moment (torque), while #2 gives square relation, but longer tail is also heavier (with appropriate nose weight or length) and thus it is even worse.

On opposite side you have rotating flywheel on nose which eats some amount of force in turn which is easier to deliver by longer tail, but I still think the most important is the difference between incoming air AoA on wing and tail in tight turn.

>>>Why?<<<

That is simple. The tail in maneuver does only little lift (I mean steady conditions, not entering of corner and battling with mass inertia). The lift just necessary to balance the CG and airfoil negative pitching moment and it is not so much. So its deflection is not “to make lift”, it is to keep AoA on wing close its zero lift AoA. If you need 10deg on wing, the tail must have deflected elevator so much that it has 0 lift AoA at those 10 degrees (or little more for balancing those moments). If the tail is longer or shorter, does not matter – but only in straight flight. But once you entered circular path in corner, that very short tail has the same AoA as has the wing, but that long tail makes different incoming air direction on wing and different on tail – those 10 deg on wing could change to 20deg on long tail. So if longer tail has to be able at its maximal deflection keep the same AoA on wing, then longer tail will need little more deflection.

Hi

The Why? was retorical, (I answered my own question)

I agree a shorter tail has better angular effect in square turns, provided the stab area is increased to keep the same Neutral point position. I don't much like the neutral point concept, it doesn't fully explain the physics of stability. It is however a very useful concept for getting stability close to right and comparing different design modifications. The point I was making is that if you can move both NP and CG rearward by equal amounts, Stability is not greatly effected but with GC behind the centre of lift, the resulting moment couple will rotate pitch to a far greater extent than lift and tail moment are capable of.

Neutral point can be shifted rearward by either increased stab area or moment arm (longer tail) but greater stab area will give sharper turns than a longer tail for the R^2 reason already mentioned.

Regards,
Ken Webster

So the tail length that would give the ten degrees of extra (did I get the sign correct?) angle of attack at the tail for a 10-foot-radius loop is 120 inches * tan (10 degrees) = 21 inches = 54 cm. Then there's downwash, which acts in the other direction. Downwash decreases with tail length, but the pitch rate effect increases with tail length, so they both cause required elevator deflection to increase with tail length. Correct? A danger of having too long a tail is that your airplane might not fit in model boxes available for you to borrow.

My transport box (that small European metal type on 4 wheels) allows me to have 60cm tail (AC to AC) on back seat. That is good excuse to fly rather 3.5m corners (140”) instead of rulebook 1,5m.

I copied this from a previous post. It seems to apply.

"For what its worth, the Bearcat I and III depict a specific Bearcat which had the Falcon paint job and small wing fences. Because they were on the prototype for the model, I use the wing fences also on the model. The fences may actually have some positive effect on the ability of the 60 powered heavier airplane to turn as well as the lighter airplanes with smaller engines. All the Bearcats were very pleasant to fly and won their share of the hardware."

If you make a small change like the end plates on the flaps, the observed effect is likely to be one more of perception than actuality. The fences on the Bearcat were not put there with any intention of improving performance, but they are big enough they might have had an effect.

I put wing fences like Al's on my current Nats stunter. Geo-Bear. The fence went from the highpoint on the top around the front of the wing and back to the highpoint on the bottom. They had a gradual start from highpoint and were out in front about 3/8". The effects were negative to say the least. It seemed to add the feel of to much tip weight but had roll and yaw issues during hard corners. I flew it like that several times and trimmed on the fence and then I finally I just took it off and all was good again with the plane. It would take a whole new trim setup to work. The current one works so I didnt persue it.

I didn't say they were good. I didn't think much about them as being aerodynamic in the stunt sense. Neither did I expect expected them to be noticable. The two airplanes that used this paint job and tip fences flew fine for whatever reason.

Doug, why did you concentrate all of the area of your fences forward, even to the point of having area ahead of the wing (if I understand your description). Did you ever see anything like that on a real airplane? It would seem that with the airplane flying typically slightly yawed you would have a side force centered near the leading edge of the wing. On typical wing fences like those shown on my Bearcat any side force generated would be more in line with the CG and less likely to cause yawing moments themselves.

The small photo above of the Airbus tips could be an A-320. Brings back memories. That was a cool airplane with digital everything and a pilots side stick control. For a guy who was used to a humongous yoke ahead of me attached to a bunch of heavy cables, the day they replaced my side stick controler and I saw it was attached only to four potentionometers gave me a very insecure feeling about my ability to control the airplane. What the hell. Let's admit it, the computer controlled the airplane and did what I asked it to only if it agreed that my commands were OK. This is an opening round of computers taking physical conrol of our lives.

Hi

Sorry to ask this on this topic but I can't see how to post a new topic. How do I do that?

Also does this board have a resources area for photos, sharinr files etc? If so how do I do that?

Ken: I hope I answered your questions by private message, if you can find it (top left when viewing conferences list, or in "user menu").

Al: According to AirBus' website, that is indeed a 320 wingtip, and it was in United motif, per Jim's spec. If not, they are more screwed up than Boeing. But hey, what was that horrid groaning prior to takeoff? If that was lowering the flaps, oh, MY! It sounded like the rear half of the fuselage was slowly being sheared off. NOT confidence inspiring! There were some things I liked about the 320 vs. the 757, tho.

Jim: I'm not real clear on what you expect from these tip experiments. While smoothing the wake in level flight would be a slight improvement, the real problem is the wake in maneuvers, especially those with hard corners and close to the ground (due to more stagnant air closer to the ground). Seems to me that the best wingtip design could only expect a minimal improvement, because other turbulence sources seems much more significant...flaps, elevator, prop, most especially. However, I appreciate your doing these tests, and hope you report your findings. Steve

>Hi
>
>Sorry to ask this on this topic but I can't see how to post
>a new topic. How do I do that?
>
>Also does this board have a resources area for photos,
>sharinr files etc? If so how do I do that?

Ah. Another nice roving kitty.

Here, read this...

Sigh

Guys,

Thank you kindly for all your replies. I have been in a DOD school for a week so I am just now am getting to read them. I just got a new home computer and am in the process of getting another Juno CD because I lost my other one - ugh! I hate moving! My new home computer will have the capability to process photograps so that I can post them here and I will post some pictures of the plane after I get my CD and get my internet connection going again. I will also post more about the testing of the plates after I get the plane into the air sometime later this spring.

Regards to all!

You moved? When? Where?

Do you have a basement now? How about a flying circle in the yard?
Inquiring minds want to know.

How's my engine doing? Two days after I loaned it to you, I needed it back. Since I'm a nice guy, I called Richard and ordered another one

Ron,

Why did your 4 Stroke Croak??

No, I am in an apartment - still the same one I was in.

No croaking. I'm not sure what's wrong, but the engine just doesn't want to settle down and run smoothly enough for me. I cannot believe it needs more break in, but maybe that's the issue. Anyway - the engine run is far different from my experiences in RC and I need to concentrate on re-learning the CL pattern. So I have pulled it out and have installed a simple Ro-Jett .61 with muffler for the time being.

The experiments will continue. I refuse to let a simple mechanical device get the better of me

Ron,

Could be a resonance problem. Try running it differently set up and see what gives.