How can you achieve functional elevator control on canards? What are the tips for beginning to test flights?

Whether it is a canard or anything that flies the longitudinal aerodynamics of the airplane have to end up to 0 at the center of gravity. With the canard planform it is still a matter of positioning the CG such that the lift from the canard times its moment arm to the CG is equal to the lift of the wing times it moment arm to the CG. Once the balance of moments is made you have an airplane that will fly. If the Neutral Point of the airplane (basically the center of the aerodynamic forces) is aft of the CG then the airplane will be stable. Put the canard at a higher angle relative to the fuselage than the wing is set at and you will have an airplane that will trim out at some airspeed to give level flight.

Then the control question comes in. A conventional moveable elevator on the rear of the canard surface will allow changes in lift to be made. Elevator trailing edge down gives increased up lift on the canard. This results in a nose up pitch rate and just reverse the control movements for nose down.

Along the same lines an up flap on the wing will give a nose up on the airplane but at the loss of a lot of lift. Normally on an airplane when you want the nose of the airplane to go up it is to make the airplane to go up also. Loss of lift in that situation is not a good thing!! So the movable surface is put on the front surface.

It is a neat thing the way things work out.

A neat side-effect (for some people...) to having the elevator control on te Canard surface... you give up-elevator control, the canard surface AOA effectively increases. If the plane is near stall speed. the Canard surface stalls... nose drops main wing NEVER stalls. (the nose just goes into a gentle rise and fall sequence if you hold FULL up elevator just at stall speed.)

A stable Canard design is somewhat difficult to force into a condition that stalls the main wing...

Which is more effective to prevent nose up tendency in canards; adding more weight to front site or decreasing the canard stab area?

There are probably places (I don't know where however) on the net where the appropriate formula can be found for canard type airpalne design. They generally would suggest an area ratio of canard area to wing area that is also based on the distance from the wing to the canard.

If you have such a layout then shifting the CG is done for stability.

Once stability is achieved then canard incidence angle is set to fly level.

Larger canard area allows the CG to be moved more forward.

A nose up tendency on a stable canard airplane can be cured by making the incidence angle of the canard lower or giving nose down canard elevator commands.

The statement about the canard stalling first and making a canards main wing stall being hard is correct, only if your CG is correct.

A rearward CG will stall the main wing (called a deep stall). This is usually unrecoverable and has resulted in many full size fatalities.

The trouble with canards is they have not really been studied enough for the rules to be known so it's all about trial and error. Set your CG using the Area/Moment rule as Ben said (using the centre of lift as the moment). It is all trial and error from there, but be careful of that deep stall. Maybe fit a stall recovery chute is you are building something large and expensive. I plan starting a 1/3 scale LongEZ soon (built from the full size plans, with required elevator size changes etc).

I have a St Croix LongEZ (62") and the plans for the fullsize LongEZ, Vari-Eze, Vari-Viggen and Solitaire (all from Burt Rutan). I hope one day to build a full size 4 seat Aerocanard (www.aerocad.com)

.
Locating the c.g. is very important for a canard.
The safest way to do that is build a smaller glider, and toss it with the c.g. where you think it should be, then make corrections if needed.
Risk is small.. just some balsa sticks.
Then for the real one, use the c.g. you've found for the first trial flights.
Limit your control surface motions. Too much will make a bad situation worse if you get into trouble.
If it's powered and gets into trouble, the FIRST thing to do is cut the power OFF! The unpowered plane will not hit the ground as hard.. and just might self-recover.

May I second the use of small models to determine settings and physical layout of potential designs. I use 1/8 inch foam from trays or whatever to make small electric RC models as a way of evaluating airplanes. They are not perfect for this application but are a step up from the basic glider (still a good starting place).

Look at

http://142.26.194.131/aerodynamics1/...ty/Page11.html

for an nice interactive idea of stability and control of a canard.

To get some more information of how the stabillity of a canard will effect the loadings on the forward surface look at

http://www.desktopaero.com/appliedaero/appliedaero.html

Section 14.4.4

There are no equations in this page, very basic assumptions - No fuselage, no wing section pitching moment, just basic stuff. The graphics could be more exciting though.

The definitions of the inputs in the figure are below

sm = static margin, the measure of stability based on the reference chord. Distance between the CG and the Neutral Point.

Wing AR and Tail AR = changes in the picture as you change the numbers. What you see is what you get

St/SW = the ratio of tail area to wing area, easy stuff.

The Figure changes to show the changes, you can click/drag the tail left and right and see how the changes occur.

If the Lt/Lw printed out is positive it indicates a tail up load.
e is the induced drag and is not too interesting.

For a typical canard

sm = .10
Wing AR = 8
Tail AR = 6
St/Sw = .2

Push the compute button.

Use the mouse and move the red horizontal tail left and right and see the load on the horizontal change. If the term Lt/Lw is positive then the load is up.

Notice that the upload on the canard is from 25 to 30 percent of the wings load. Since the canard area is only 20% of the wing it indicates that the canard will probably stall sooner than the wing if the airfoil is the same.

I notice that Rutans canard airfoil as seen in Profili is creating sightly more lift at the same angle of attack as the wing airfoil.

It is a good tool to find out what is happening. Of course if the airfoil has a big moment input or the fuselage, flaps, etc it will change the answer. Sometimes not easy to guess at.

I have done some study on canard airfoils. Even talked to John Ronncz (the deisgner of the Roncz MS1125 used as the replacement canard for the longez).

DO NOT USE the scale airfoils from Rutan. They are great for full size but as with most (all) airfoils dont scale well. The same airfoil at 1/3 scale (or any other scale) will react very differently to the full size.

Any semi symetrical airfoil for the main wing and an almost flat bottom (clark y even) for the canard should work well.

Raiki - What characteristics of the Rutan airfoils make them unsuitable for use at smaller scales? It will be interesting to hear what causes you to use uppercase letters :-)

Granted when you are making small airplanes and desire the most performance out of them there is a chance of pushing something too far. Certainly Rutans canard airplanes are designed around a very narrow set of performance requirement and the canards themselves are working in the drag bucket that the laminar type of airfoils possess giving (hopefully) low drag.

As a result it also keeps the airfoils in a fairly low Cl range. There is speculation that if the airfoils separate that it causes the airplane to take longer to recover than other more robust airfoils. But since it is the airplane that is recovering the comments about the time of the airfoil to reattach is just speculation.

I don't know what the exact separation characteristics of the Rutan airfoil/airplane are. Typically our models have a much lower wing loading than the full scale counterpart. Although Rn effects are not favorable (as with most airfoils) the separation characteristics would tend to become a negligible factor because of the lower wing loading.

From what I have read Roncz apparently changed the airfoil of the canard because the basic Rutan laminar flow airfoil lost lift in the rain. Not a good thing but not unheard of either. In most cases it could be taken care of with available trim. Power to move the canard isn't a problem in model sizes for the most part.

Roncz's new airfoil also had slightly better lift characteristics enabling the span of the canard to be decreased by 10 inches. This is running the design on the hairy edge to fullfill performance expectations isn't it? Again something that is not a problem with models

Comparing the Rutan to a ClarkY is interesting.
Rutan is a laminar flow airfoil, ClarkY isn't.
The Rutan has a drag bucket that the Clark Y doesn't have.
Both have the same Clalpha slope.
The Clark Y has a CLo shift of .3 which is due to camber.
The Clark Y has a more negative Cm which is due to camber.

Again given the difference in wing loading and other things it doesn't seem that the airfoil would be a major factor in how the model airplane would fly.

Ben

When I say do not use the Rutan airfoils I was talking about the canard (forgot to mention that). At these exponentially low renolds numbers I agree that the main wing airfoil wont make much difference between the Rutan (Modified Eppler 1230) and the Clark Y. (I'd use a regular 1230 cos it's easy to the Coords and looks cooler)

I do not know if you have seen the 1145 but it has a very blunt LE and not a great deal of camber. It was created because the original GU25 (the first canard foil) did lose lift with water or bugs (any imperfection) on the canard. (as you said)

The GU25 is almost flat bottom but has a wierd camber that is almost at 50% chord. A funny looking foil. Both of these are very different than any foil I have seen used on models.

There's not much difference between conventional aircraft and canards as far as the importance of locating CG. For either, you need CG to be located so that you have static stability ( unless, for some reason, you want to fly an unstable aircraft ). The main difference is that it is very important on a canard that the foreplane stall first. There are actually fairly well defined design rules to acheive this. Generally, the airfoils of the foreplane and the main wing are selected such that the foreplane stall at a lower angle of attack. Also, the foreplane frequently has a higher aspect ratio, so that it stalls at a smaller angle of attack. The relative camber of the two surfaces can also be used.

banktoturn