Flying wing and elevon deflection

[Slow Riser]

I have been experimenting with flying wings, "mini-zagi", and to get the wing to fly level, my elevons have to be up to fly level (please see attached graphic). With this amount of deflection, I'm under the belief that the wing is not flying as efficiently as possible. If the elevons were in line with the wing "mid-point", then the wing would dive...which it does.

Question:

1. Is this same elevon setting required in other flying wings like the Zagi?
2. The wing profiles are generated in Profili. The dotted line is the line the bisects the "upper" and "lower" airfoil. Should the plane fly at this dotted line?
3. If I wanted the elevons to be inline with the line in question #2. How would I modify the airfoil?

Now with all this said, I am very happy with the performance of the wing thus far...my only concern is the wing is not flying as efficiently as possible.

The following is a link to the wing:

http://www.rcgroups.com/forums/showthread.php?s=&threadid=57130

Thanks...Hank

[Yogi]

Hank,

The phenomemon you are referring to is called reflex, and it is totally necessary for a flying wing to be able to fly. All flying wings need some reflex to fly level, although some need more than others. From what I understand, this is one reason that flying wings are not as efficient as their fused flying bretheren. From your picture, the amount of relex you are showing looks sortof extreme. It should line up with the curve on the bottom of the airfoil. My two Zagis only need what looks like about 1/8 to 1/4" of relex. If you want your wing to go faster, look into thinner airfoils, and the Wedgie specifically. This plane totally rocks with a plain ole speed 400. The first time I saw one I thought for sure it was a brushless, but it wasn't. My big fat Zagi needs a brushless to go as fast as a Wedgie.

~~fred
Atlanta

[Sparky Paul]

What Yogi said.
The reflex (yours does look like too much) substitutes for the lack of a horizontal.
This does diminish the effectiveness of the wing, which is why all-wing planes aren't seen that often. They have to be larger to get the same lifting capacity as a normal wing-tail airplane.
Larger is more bux.
.
Sparky Paul
http://www.angelfire.com/indie/aerostuff
PJB's Seriously Aeronautical Stuff
http://home.earthlink.net/~pjburke1/aindex.html

[rparigoris]

What yogi and sparkey say.

Your diagram does seem extreme though.

Foward CG will do it.

Here is a good way to adjust CG:

Go up to altitude, turn off power and set up a nice slow glide, trim for hands off. Now dive at a 45 degree angle and let go of stick. If pitch up you have foward CG, if nose down then CG is too far aft. If too aft it will get to be a real handfull.

Once CG is set pretty good, now do the same, set up a nice power off glide and add full power, if nose pitches way up, or it begins to dive with proper CG then the motor angle is incorrect.

Desirable for me is a slight nose up power off CG test, and a slight nose up with power on test but only when the airspeed increases alot, and a slight climb will begin.

Good Luck
Ron Parigoris

[Slow Riser]

Thanks for the info, not having flown a wing before, I was worried I had some setup problems.

I guess I should have said the diagram was not to scale . The "reflex" I have is about 1/4".

Nice to know its common on wings to have this as I was trying to trim when the elevon was level and it kept nosing into the ground on hand tosses.

I'll try adjusting the CG and see if its set correctly. On my first wing, I could do loops from level and with the new one, it kind of stalls out at the top so I figure the CG is to far forward.

Would adjusting the thrust angle up have any effect? My thinking is the wing would try to pitch up a bit more and I would have to use less reflex.

In any event, I'm happy with the results so far in my flight tests.

I didn't realize wings are so much fun!

[Slow Riser]

I did a search with reflex, now that I know what it is, and wings and someone mentioned airfoils that have a built in reflex such as the MH and Zagi airfoils.

Is there any added efficiencies to using these foils with built-in reflex vs airfoils with no built-in reflex?

My second model uses a modified Clark Y and when I compare it to an MH45, its almost identical with the exception of the MH where it is thinner at the trailing edge.

I know the Wedgie uses the MH45 but I would imagine the trailing edge is prone to damage since it so thin.

Hank

[Sparky Paul]

For a messing-around-the-sky airplane like a ZAGI, the airfoil isn't that important. Practically anything will work, with reflex.
If you are trying to get real performance, especially endurance, then the specially designed profiles like the MH's, when built properly will outperform a "normal" airfoil with reflex added as an afterthought.

[rpm3]

Slow Riseer
I have 3 Wedgies....and have noticed the same thing. It looks like the elevons are angled to hi, but thats the only way they all trim out. At first I thought the CG may be off, and the deflection was to correct it.
I dont think so now.....all of my Wedgies are setup about the same....and Yogi is right on.....the Wedgie ROCKS!!!!....i use the speed 400 and it is ahead of my skills by 6 months or so...and the best news is that although it is not nearly as pretty.....it still flys great.....
One benchmark I see used is to lay a ruler on the bottom side of the wing, and set the elevons there for neutral.....just a benchmark, but it has worked for me......and if I error, I add justg a bit more.....not an expert by any means......just got some experience with the Wedgies!

Get one.....you'll love it!

[Slow Riser]

This will certainly aid in my wing designs.

Sparky Paul...anyone tell you you look like Tom Skerrit?

Cheers!

:)

[Sparky Paul]

I think Tom Skerrit looks like me!
And he hasn't handled advancing age as gracefully! :D
And he's probably younger, to boot!

[MrThermal]

Hi guys....learned alot reading this thread. My 34" red herring w/MH45 airfoil is just about ready for the electronics to go in.

It has a 12" root chord and I plan on using 1" elevons. Now I know where to set the elevons for neutral trim.(about 1/8-1/4" up)

Question: should I use pianowire or threaded steel for my pushrods? I want to use z-bends at servo end and nylon clevis"s at horns. Using HS 55's and 1/2A horns.

Also, should I put any up thrust into my GWS EDP-50XC powerplant? I know it may be marginal in the power dept. but not looking for alot of speed just cruise and loops, maybe a slow roll now and then. Main thing is to use my Qcoms and get some duration. thanx for any input

[njones]

First - as everyone has already said, the airfoil isn't critical to a well flying plane, but if you want to chase that last 10% of performance or just kill some time, read on...

The issue isn't really reflex - it's actually about pitching moment - reflex is the characteristic of an airfoil that affects pitching moment, control surfaces also modify pitching moment.

If you look at a set of plots for an airfoil, three curves are plotted against angle of attack (AOA) for one or more reynolds numbers (reynolds number = airfoil chord X airspeed). The three curves are 1) coefficient of lift 2) coefficient of drag and 3) pitching moment.

The first two are pretty well understood by most modellers, the last refers to the twisting force (around the main spar) produced by an airfoil. A "typical" airfoil has a negative (nose down) pitching moment - this requires a tail surface producing an equal down force to counteract it. In your case the counteracting force is produced by your raised elevons - and apparently your choice of airfoil has a pretty strong negative pitching moment.

Airfoils designed specifically for flying wings have a pitching moment close to zero at a particular AOA (usually the intended "cruise" AOA). A wing flying at this AOA requires no counteracting force from it's control surfaces to balance it.

Deflected control surfaces always increase drag so aiming for zero control surface deflection at "cruise" maximizes efficiency. This doesn't mean that flying wing airfoils have less drag - they just have a zero pitching moment without control surface deflection. A flying wing is not necessarily more efficient than a classic two surface design - it all depends on the total drag for a given design.

The Horten brothers (Germany) were early pioneers in flying wing design in the 30's and 40's. I have used their airfoil successfully in a flying wing design with similar planform to the Zagi - but it still needed slight reflex in the elevons, I presume due to the much lower reynolds number my model flys at (smaller size, lower speed) and AOA (>0) my model requires for level flight.

Flying wing research was also done at Northrop which built a flying wing bomber prototype, the YB-49, which was flown in 1947.

There are numerous web resources available - now you have the keywords to search for...

Run this airfoil through an airfoil analysis program (like Javafoil) and see what I mean...

Horten HIII Airfoil (I understand they used this aiirfoil on many of their powered and glider models)

1.00000000 0.00000000
0.95000000 0.00488000
0.90000000 0.01014000
0.80000000 0.02236000
0.70000000 0.03660000
0.60000000 0.05294000
0.50000000 0.06910000
0.40000000 0.08099000
0.30000000 0.08639000
0.25000000 0.08522000
0.20000000 0.08223000
0.15000000 0.07527000
0.10000000 0.06331000
0.07500000 0.05533000
0.05000000 0.04544000
0.02500000 0.03198000
0.01250000 0.02145000
0.00000000 0.00000000
0.01250000 -0.01417000
0.02500000 -0.01924000
0.05000000 -0.02685000
0.07500000 -0.03229000
0.10000000 -0.03601000
0.15000000 -0.04011000
0.20000000 -0.04232000
0.25000000 -0.04323000
0.30000000 -0.04362000
0.40000000 -0.04251000
0.50000000 -0.03907000
0.60000000 -0.03390000
0.70000000 -0.02704000
0.80000000 -0.01898000
0.90000000 -0.00962000
0.95000000 -0.00475000
1.00000000 0.00000000

Hope this is helpful.

[efarmer]

>>the airfoil isn't critical to a well flying plane

Someone must have misinformed you in the aerodynamics design department... the airfoil is one of the primary performance and handling determining aspects of a wing/airframe. To a typical power flyer, there are three airfoils: flat bottom, semi-symmetrical, and fully symmetrical. To anyone who actually has half a clue about airfoil and wing design, there is an infinite number of airfoils that exist, each and every one with it's own unique handling and performance.

As a gross generalization, the airfoil determines flying speed, stall characteristics, amount of lift developed, amount of drag induced, structural limitations, etc etc. Yes, there are other factors that come into play with these issues (fuselage/tail design, build quality/precision, aeroelastic issues, parasitic structures, etc), but the airfoil is one of the primary, if not THE primary determinate.

>>the twisting force (around the main spar) produced by an airfoil

Around the spar? What if the isn't a spar? The axis of pitching is the aerodynamic center of the airfoil (usually somewhere near the high point of the airfoil)

>>If you look at a set of plots for an airfoil

These are called "polars".

>>A flying wing is not necessarily more efficient than a classic two surface design - it all depends on the total drag for a given design.

It also depends on the total drag for a given design... and more importantly, the l/d (ratio of lift produced to drag produced)

>>Airfoils designed specifically for flying wings have a pitching moment close to zero at a particular AOA (usually the intended "cruise" AOA). A wing flying at this AOA requires no counteracting force from it's control surfaces to balance it.

I assume you mean that airfoils that have a Cmo "close to zero" have only a slightly negative Cmo. Airfoils can also have positive pitching moments. Check out the Selig 5010 if you want to see one.

>>Deflected control surfaces always increase drag

Not necessarily. If the surface (lets say an aileron, for example) is deflected in such a way which creates a "new" airfoil with a lower Cd, the drag is decreased. On most modern glider designs, if you want to go faster (ie: less drag AND less lift) you raise the TE (reducing the effective camber). This is a control surface deflection that does NOT increase drag.

>>but it still needed slight reflex in the elevons, I presume due to the much lower reynolds number my model flys at (smaller size, lower speed) and AOA (>0) my model requires for level flight.

You say you used the Horten airfoils, but you changed the planform, probably dihedral, CG, and twist (I assume). Maybe it's not the size difference (although this does have a small effect, but not much). Airfoils like the horten sections are designed to work properly on a single COMPLETE design. The airfoil design (especially on a flying wing) takes planform, twist, CG, etc all into consideration. If you suddenly change one, or more, of these attributes, the airfoil is no longer going to work the way it was designed.

I also see many guys referring to a control surface as if it is separate of the wing/airfoil. It is not. It is a part of the airfoil, and should not be thought of something that is in addition to a particular section. A control surface essentially CHANGES the airfoil in a manner which creates leverage to pitch/roll/yaw the model in the desired direction (or to control speed/efficiency, etc).

Hope this helps.

-Eric

[Slow Riser]

is there any advantage to having the tapered elevons like the Zagi (elevon thickness at tips wider than at root)? There are some other "high performance" wings like the Wedgie that have constant chord elevons.

[berniep1]

Hi Slow Riser!

The reason why you see wings that have highly tapered elevons (inside small, outside wide) is because the effectivness of the elevons is almost nonexistent in the inner regions of the wing -- in fact they just create more drag. I personally think that having the elevons just on the outer part of the wing is sufficient -- the only problem then is that you have to relocate your servos (unless you use torque rods) to an area of the wing where you don't have a lot of "meat"...and you need long wires to boot.

Bernie

[efarmer]

Slow Riser-

Ideally, your elevon size should be a constant (or very near constant) percentage of the chord. This is generally the most efficient way to doing things. However... the most efficient is not always the most effective and/or desired. Guys like responsiveness, and the larger chord elevons give this.

There are definate down falls to the larger-at-the-tip elevons. 1) Flutter. At high speed, this type of surface is almost sure to flutter. Especially when the control horn is located at the extreme inner portion of the surface. 2) Handling. Personally, I believe the larger elevons induce bad handling characteristics. Primarily, tip stall. Make your elevons taper slightly (as a percentage of the chord. IE: 20% of the root chord at the inboard end, 22% of the tip chord at the outboard end).

For combat, bigger at the tip is fine. For anything else, I would venture to say that the smaller, bigger-at-the-root elevons work better... significantly.

As a side note... if you're flying a flying wing with lots of sweep (Zagi, Boomerang, Wedgie, etc etc), try putting about 2 deg of anhedral (tips down) in it. The rearward sweep of the wing gives a net effect of dihedral. The anhedral will cancel out much (but not all) of the dihedral effect, and typically improves roll response, overall handling ability, and pitch response.

-Eric

[rparigoris]

Hey efarmer

Stated like a Pro!

High speed and heavy with wider at the tips on a foamie, unless you have it glassed or sheeted is likely to twist the wing at the tip and get you control reversal or severe osalations.

Constant cord is king in that arena. It is also more efficient to move a large surface a little, as compared to moving a little surface alot.

If you are going to make a wider at the center elevon, do not just make it wider and hang it off the stock trail edge. Cut the trail edge down so you have the same area of wing, but just make more of it move.

For a pretty rigid and strong elevon try making out of bass. For half way stronger and more rigid than balsa, laminate bass and balsa (bass on bottom) with spray 77. Also makes sanding easier.

Last, more efficient to hinge elevons on bottom, and less drag to put horns on bottom. HL sells gap seals i use for bottom gap. Also use some fairings to clean up and protect horns. So far in 3 or 4 dozen flights, no problem with horns on bottom.

ron Parigoris

[berniep1]

@efarmer

The anhedral will cancel out much (but not all) of the dihedral effect, and typically improves roll response, overall handling ability, and pitch response.

Roll response? How? :confused:

And adding anhedral will yield a completely unstable plane if the negative moment of the airfoil is too high....

Or am I missing something?

Bernie

[efarmer]

Bernie,

>>Roll response? How?

Think of a conventional stabbed airframe with tons of dihedral, and ailerons (this represents a typical swept flying wing, with a flat wing, or dihedral). The roll control is typically not very crisp, and they almost always have horrible dutch roll.

Now, take the same airplane, with very little or no dihedral (this example represents the swept flying wing with anhedral to cancel the net dihedral affect of the sweep), and tell me how the ailerons respond. They will be much more crisp, and the dutch roll will be hugely reduced.

The addition of anhedral does NOT mean the airplane sees the affect of anhedral, all it does is cancel out some of the dihedral affect which is induced by the rearward sweep. If you get way out of hand with the anhedral, you'll run into problems. But around 2deg is just enough to reduce the net dihedral effect.

>>And adding anhedral will yield a completely unstable plane if the negative moment of the airfoil is too high

That is true for a conventional stabbed and unswept (or little sweep) airframe... but we aren't talking about sections with a negative Cmo. We're talking flying wings. Flying wings MUST have a positive Cmo to fly. If they didn't have a positive Cmo, they would have a stab... and would therefore no longer be a flying wing.

The anhedral simply reduces the net dihedral affect which is induced by the sweep of the planform. It DOES NOT give a net anhedral affect (unless of course, you really over-do it).

-Eric

Message edited for dumb finger typos...

[njones]

Thanks for the corrections efarmer, I learned a few things. I was trying to keep it simple, but as you pointed out I actually simplified it to the point of inaccuracy.

Zagi not only has increased elevon chord at the tips, but since the elevons aren't part of the airfoil (they are extensions) and the wing chord decreases at the tips, this actually means that the Zagi elevons are a pretty poor example of aerodynamic design - right?

So If I understand the previous messages correctly; I should try an elevon that is a constant % of chord which will actually make them narrower at the tips - right?.

I have not experienced flutter at all, but I suspect my rolls are not a crisp as they can get due to the elevons twisting under stress

But if I want stronger roll response, increased chord elevons are in order, but I should still follow the constant % chord rule? And for minimum drag the elevons should be part of the airfoil and not extensions?

[Sparky Paul]

The extended elevons on the ZAGI would be a poor example of a design.... if .... the ZAGI were a full-scale airplane.
As it is, it's adequate for the task, which does not include aerodynamic efficiency!
And yes, the elevons ARE part of the airfoil! That's why they need to be reflexed! They provide the nose-up pitching moment the forward c.g. works against.
As for increased roll response, that comes with experience.. and surface deflection. There is such as thing as too much.. more drag than lift created for instance.
The optimum setup for a ZAGI is most likely seperate elevator and ailerons. The elevon close to the "body" contributes little roll effect, while an elevator in its place would enhance the pitch response. And outboard ailerons increase the roll moment because the effective area moves outboard. At the expense of adding a servo..
I made this change on my Evan's Simitar and improved everything.. pitch response improved, rolls were better, and the "unspinnable" design now spins nicely!

[efarmer]

njones,

>>but since the elevons aren't part of the airfoil (they are extensions)

Huh? They ARE part of the airfoil. If they weren't part of the airfoil, they wouldn't be attached, and they would be stand alone surfaces (check out the wing surfaces on a Ju52 if you want a good example). When you put slab balsa elevons on the back of a Zagi type airplane, the airfoil is no longer a MH45, MH60, EH2012, etc etc... it is not a totally new airfoil which shares portions (primarily the fore and mid sections) with your off-the-shelf (MH, EH, etc) airfoil.

>>this actually means that the Zagi elevons are a pretty poor example of aerodynamic design

If you're talking about wanting maximum efficiency, yes. They are a pretty poor example. Quite poor, in fact. HOWEVER, if you are wanting a combat airplane that is easy to build and manufactuer, they are just fine. The "quality" of design is relative to the design's intended use.

>>So If I understand the previous messages correctly; I should try an elevon that is a constant % of chord which will actually make them narrower at the tips - right?.

If you're looking for a very efficient airframe, yes. That is definately leaps and bounds better than the Zagi/Boomerang/etc shaped elevons. And the elevons should be an articulated portion of the wing, not a slab sheet of material taped on the trailing edge of the wing (unless of course, the airfoil is designed with that shape... but I doubt it ;) ). A surface which is actually a tad larger in %age of chord at the tip will probably handle a tad better. It should make up for the lack of Rn at the tips.

>>I have not experienced flutter at all

Then you either 1) haven't gone fast; or 2) built very stiff surfaces and linkages. Most Zagi type airplanes aren't designed to go extremely fast, they are made for combat (slow Vmin speed to recover from a collision). However, if you take a stock Zagi/Boomerang/etc, and try to DS it, I guarentee you'll flutter it. I know... I've tried :)

>>but I suspect my rolls are not a crisp as they can get due to the elevons twisting under stress

Yup. Part of it is from the lack of torsional strength, you're right... other factors that hurt are surfaces that are surely not counter-balanced, mass balanced, as well as probably some bad dutch roll tendancies.

>>But if I want stronger roll response, increased chord elevons are in order, but I should still follow the constant % chord rule?

If it were me, yes. I would stick to a constant, or near constant chord (%age) elevon. I would probably make the tip of the elevon about 2-3% larger (relative to the tip chord) than the root of the elevon. IE: The elevon tapers dimensionally, but widens relative to the wing chord.

>>And for minimum drag the elevons should be part of the airfoil and not extensions?

I wouldn't say "minimum drag"... but rather "maximum efficiency". Again... don't think of the slab balsa type elevons as "extensions" of the airfoil... as soon as they are installed, they become the aft portion of the airfoil.

-Eric

[Slow Riser]

first of all, I have gained a ton of knowledge from all your responses.

We have talked about reflex, elevons, wash-out, airfoils, torsional rigity...now one more twist...and that is engine placement on a wing.

Is it best to mount the engine inline with the wing or many designs have the engine slightly above the wing on a pylon. In the latter, to get the best performance, should the engine be mounted with up/down thrust?

The photo below shows the about half the test wings I have made. The middle photo is the one I am refining and into 2nd generation. The wing now incorporates a bit of wash-out into the cores and has added CF spars for torsional rigidity that was experienced with the thinner airfoils tested. The 2nd generation has servos/esc/rx embedded into the wing.

Next test will involve testing differently shaped elevons.

Flies awesome and glide ratio is excellent with an aspect ratio of 6:1. The wing is stiff and the linkages have no slop. Haven't tested the new wing but 1st generation would cruise around 20-25 mph and high speed passes must be 60+ mph all on an DD IPS motor. 20+ minutes of flight on the Qualcomms.

[berniep1]

Hi efarmer!

Mea culpa (my fault) -- I was fairly tired when I read your post and for some reason (don't ask me why -- not reading thoroughly enough most likely) jumped to the conclusion that you were talking about having the tip airfoil at + 2 degrees relative to the root airfoil....and that's why I doubted the ability of any wing to roll faster (except for the tip stalls of course)
And that's also why I was talking about the (negative) cm0 -- if you take an airfoil with a fairly negative cm0, use that for a wing but twist the tip sections in the wrong direction, you are not going to get a stable airplane. That's what I ment.
Sorry for the confusion..


Bernie