Canard Forum: Show,Discuss, Learn - Page 114 - RC Groups
Thread Tools
Feb 15, 2008, 10:21 PM
Registered User
Don Stackhouse's Avatar
OK, let's try to straighten out this whole incidence-stability-C/G-trim issue.

First of all, some of you have the tail literally wagging the dog here. Stability is NOT the result of a particular incidence, and C/G required is also not the result of incidence.

1. Required C/G location is the result of the relative sizes of the wing and canard, and the required stability.

2. Required incidence is the result of (among other things) the C/G location.

A little review:

If you could add together all the little lift and drag forces all over the wing, and find a single place where you could replace those little forces with one big single force, the point that force would act through is the "Aerodynamic Center", or "AC" of that flying surface. We can also find a point in space that is where the combined forces of a group of flying surfaces act through.

The AC of an individual flying surface is typically located at or near 25% of the way back from the leading edge of the "Mean Aerodynamic Chord", or "MAC". The MAC is NOT the same as the average chord. On a trapezoidal (straight-tapered) wing, it's at the location along the span where exactly half the area of the panel is inboard of there, and the other half is outboard. There's an article in the "Ask Joe and Don" section of our website that describes a graphical method to find the AC of a straight-tapered wing panel. If you have a more complicated shape, you can break it up into a series of trapezoids that approximate its shape, then use the weighted average based on their individual areas to find the AC of the total. Yup, it's a lot of work. Fortunately there are some programs that let you punch in the numbers for the panel break locations and chords, and give you the coordinates of the MAC and AC of the complete panel.

All moment arms are measured from the AC's of the flying surfaces involved, typically parallel to the longitudinal axis of the airplane (normally the fuselage datum line).

The AC of the entire aircraft is found by using the weighted average of the areas of the individual flying surfaces.

For example, if your canard is 30% of the area of the wing, then the total area is 1.3 times the wing area.

If you plot the AC's of the wing and tail (or canard) on the top view, and a line for the tail (or canard) moment arm, the AC for the combination will be on that line, at a distance from the wing AC equal to the tail (or canard) area, divided by the total area, times the length of the moment arm line.

For example, if our 30% canard's AC is 20 inches ahead of the wing's AC, then the AC of the combination of the wing+canard is:

(0.3 / 1.3) * 20 = 4.615 inches ahead of the wing's AC.

Now, there are some complicating factors, but if we can neglect those, the Neutral Point ("NP") is going to be located approximately at the AC of the entire aircraft, that 4.615" location in the case of our example. If you put the C/G at that location, the plane will have neutral static pitch stability. In other words, the plane will go wherever you point it. Pull the nose up to a new pitch attitude and let go of the stick, and it will stay there with no tendency to go back to the original pitch attitude. Push the nose down with elevator and let go, and it will stay down.

To have positive static stability (i.e.: the plane wants to come back to the original angle of attack when you let go), you have to put the C/G ahead of the Neutral Point. How far ahead is measured by what we call "Static Margin", which is typically expressed as a percent of the wing's MAC. Typical numbers are around 5 to 10 % of the wing's MAC, so if your wing's MAC is 7", then the Static Margin would typically be 0.35" to 0.70" ahead of the aircraft's AC.

One of the earlier posts in this thread talked about making the canard larger, maybe going to 30% of the wing area instead of 25%. Note that this changes the ratio between the canard area and the total area.

In our example with the 20" moment arm, the dimension for the AC using a 25% canard was .25/1.25 * 20" = 4.000" ahead of the wing AC. If we increase the canard area to 30%, then the aircraft's AC (and its Neutral Point) is now at .3/1.3 * 20" = 4.615" ahead of the wing's AC. In order to keep the static pitch stability the same, you would need to move the C/G forward that additional 0.615". If you didn't, and the plane's original static margin was aft of that point, then making the canard bigger could make the plane statically unstable in pitch.

So, like I outlined at the beginning, the amount of static stability the plane has (i.e.: its desire to return to the original angle of attack if something disturbs it) depends on the relative sizes of the flying surfaces, and the location of the C/G.

Note also, if the fuselage has things like large chines on it (such as the SR-71 or the F/A-18 Hornet), or is unusually big (like on the Super Guppy cargo planes), that fuselage can count as a flying surface, and may need to be included in C/G calculations.

OK, so much for stability and C/G. Now, what about incidence?

As I outlined in an earlier post in this thread, the incidence of the wing relative to the fuselage determines what the angle of attack of the fuselage is when the plane is flying at some desired airspeed and G-load. The incidence of the canard relative to the fuselage, or the decalage (the incidence angle between the wing and the canard) determine what angle of attack (and what airspeed in level flight) the plane flies at for a given set of other operating parameters, including (but not limited to) the C/G location.

There are a bunch of different nose-up and nose-down forces and moments acting on the airplane in flight. At least one of those (the typically nose-down force of the plane's weight acting through the C/G) is essentially constant regardless of airspeed.

Conversely, the aerodynamic forces tend to get stronger with increasing airspeed. Some of those forces, such as the aerodynamic pitching moments created by the camber in the non-reflexed airfoils of a typical wing and canard, as well as the lift of the wing acting through the short moment arm between it and the C/G, want to pitch the nose down. However, the lift of the canard, with its higher lift coefficient acting through its longer lever arm ahead of the C/G, wants to pull the nose up.

There are other forces as well, such as the drag on the landing gear acting horizontally below the C/G, trying to pull the nose down. The force of the propeller's thrust could try to push the nose down or push it up, depending on whether the thrust line passes above or below the C/G. There are in-plane forces acting on the propeller disk that try to pitch the plane up or down. There are lift forces on the fuselage that could try to push the nose up or down, depending on the fuselage's angle of attack and its shape and location relative to the C/G.

In a statically stable aircraft, the nose-up forces become more dominant as the plane flies faster. If you shove the nose down, the plane starts going downhill, increases airspeed, the aero forces on the plane increase because of this, but because the plane has positive static stability, the nose up forces dominate, and try to pull the nose back up and bring the airplane back to its original angle of attack. As the plane starts going uphill, it begins to slow down, causing the nose-up aero forces to become less dominant.

At the "trimmed airspeed", all of these various forces and moments add up to exactly zero.

If you add "up" elevator, you're changing the effective chord line, incidence and camber of that flying surface. This alters the balance between all of these various forces, whcih then changes the airspeed where the forces and moments are all back in balance.

If you have a plane with a C/G that's in the "stable" range (ahead of the Neutral Point), but not enough canard incidence, then the plane's natural trimmed airspeed will be faster, perhaps so fast that the plane has to be in an almost vertical dive in order to reach that speed.

It's also possible to have too much canard incidence, enough that the canard is stalled. This could also cause the plane to dive steeply into the ground on launch, because the stalled canard can't make enough lift to hold the nose up. There is only so much angle of attack a surface can handle, and only so much lift it can make at a given airspeed. If you demand that it make more than that, it's likely to go on strike and quit flying for you.

As far as whether your plane can climb with some given canard incidence, that's a matter of what the canard is capable of doing, and what you do with the elevator. If you want to fly slower and climb (assuming you have enough power to do so), then add some up elevator. If you want to fly faster, add a touch of down elevator. In both cases that changes the effective incidence and camber of the canard, and therefore changes the trimmed airspeed. These are not free-flight models folks! We have the power to change things in flight, at least within reasonable limits. You are the pilot, so FLY THE MODEL!

Yes, flying around with some elevator deflection does add a little bit of drag. However, the drag involved is typically extremely small, certainly nothing to get all obsessive-compulsive about. Still, we want to set up the plane to be as efficient as possible.

With that in mind, figure out from your mission profile (you did take the time to figure out just exactly what you wanted your plane to do, didn't you?) what airspeed and power setting is most important (probably cruise in most cases). Set your wing incidence so the fuselage is level at that airspeed, and set your canard incidence so that your required elevator deflection is zero at that airspeed.

Note that since the required canard incidence and elevator position for a given airspeed and power setting depends on a whole shopping list of other factors, the required wing incidence and canard incidence is going to depend on the design details of each particular aircraft involved, as well as the airspeed where you want it to fly. For this reason, rules of thumb such as "the correct incidence for the canard is two degrees", or "four degrees", or "3.5 degrees" are pretty much useless. There is no "correct" incidence, other than whatever incidence makes your airplane fly with zero elevator deflection at the airspeed that is most important to you.
Sign up now
to remove ads between posts
Feb 16, 2008, 09:17 AM
Registered User
Don, Thanks for that super informative discussion. I read it carefully and enjoyed every bit of it! I especially enjoyed the discussion of lift and drag forces about the CG to obtain equilibrium. The use of up or down thrust to to accomplish this is a subject that I would like to hear discussed when applied to high or low wing planes. It seems that a low winger would require some UP thrust with a forward static margin but some argue that down thrust is needed in the design for no good reason. I feel that you agree that speed changes are in order with changes in decalage. Charles
Last edited by canard addict; Feb 17, 2008 at 07:07 AM.
Feb 16, 2008, 09:21 AM
Registered User
Don Stackhouse's Avatar
As you can see from the above, the whole problem of calculating the wing and tail/canard incidences required for a given aircraft and flight condition gets quite involved. A spreadsheet can be a huge help in solving a problem like this.

There was a series of articles in "Sport Aviation" magazine (the Experimental Aircraft Association's member magazine) running from Feb. '90 to Feb. '91 called "Designing Your Homebuilt", by John Roncz. The series is probably the finest text on preliminary aircraft design I've ever read, thorough, scientifically robust, but in plain, readable, and even entertaining English.

There are a series of Excel spreadsheets included in the articles that take a lot of the tedium out of the required calculations. The one for calculating incidences is constructed section-by-section over several articles, with the final result in Article 9. However, to fully understand and use it, you need to read the articles leading up to it.

Check with your local library, they may have these issues available, or be able to get them through an inter-library loan.

The spreadsheets are written for full-sale aircraft, but apply surprisingly well to models, especially with a few modifications.

Now, regarding some recent comments on this thread:

...Since canards in general tend to be more efficient than conventional setups...
Generally this is not true. It may be possible to come close, but in general canards end up being less efficient in actual practice. For a given amount of stability, the basic detriment to induced drag for a canard vs. a conventional tail ends up about equal. However, the canard (because it is smaller than the wing) is a less-efficient lift-maker than the wing, and on a canard aircraft you have to make the canard do more than its fair share of the lift-making. This reduces the overall efficiency of the combination.

In addition, because the wing has to be kept from stalling in any flight condition, it ends up being larger than the wing of an equivalent aft-tailed layout, resulting in a greater total whetted area and resulting skin friction for the complete aircraft, which hurts the high speed performance.

These factors tend to suggest a smaller canard to maximize efficiency, the smaller the better. The end result of that school of thought is a canard with an area of zero, which is of course a flying wing. Flying wings can indeed be designed that are more efficient than either an equivalent canard or aft-tailed design. However, it generally takes a lot more effort to do this, and it's usually possible to achieve this for only a narrow range of flight conditions, usually with a net detriment at the other flight conditions. No free lunch.

...The lower the surface area or lift of the canard versus the main wing, the more positive incidence might be needed to compensate....
No. The "wing loading" and lift coefficient of the canard needs to be greater than that of the wing, but that doesn't necessarily mean the incidence is higher. There are a whole bunch of other factors that enter into that. Also, for stability reasons the canard's lift coefficient generally needs to be greater than that of the wing, but making the canard smaller does not mean the lift coefficient needs to be even greater (which is what increasing the canard incidence would tend to do). If you make the canard smaller, then the AC of the complete aircraft moves aft, pulling the C/G aft with it, which takes load away from the canard and gives it to the wing. The lift made by the canard decreases, but the lift coefficient of the canard (and therefore its required incidence) tends to stay about the same.

OTOH, making the canard smaller does tend to reduce stability, just as making the stab does for an aft-mounted tail. However, if you increase the moment arm of the stab/canard by the same proportion as the change in area, the static pitch stability of the plane remains the same, while the dynamic stability (the abilty to damp out oscillations) actually increases.

...I've read that when a canard is under 5" at the root it makes little difference if it has an airfoil or is just flat foam...
That's a common belief in the model community for both canards and horizontal stabilizers, but in actual practice it's not really valid. It's mostly a way to rationalize not taking the extra effort to design and build a proper airfoil for that flying surface. It's easier to get away with on an aft-mounted stab than on a canard, and it can be made to work, but I would not call it "better", or even "just as good". On a canard in particular, the stall characteristics and angle of attack for stall are critical parameters, and a flat plate airfoil tends to be quirky in those regards.

In particular, flat plate airfoils tend to be prone to hysteresis, where local separation bubbles form on different places on the airfoil as you approach a particular angle of attack ("AOA" or "alpha") from one direction vs approaching that same alpha from the opposite direction. This means that if you reach a given alpha by pushing the nose down from a higher angle, the resulting lift coefficient and lift from the canard would not be the same as if you reached that same alpha by pulling the nose up to it from a lower alpha. The net result can be an airplane that refuses to fly level. Pulling the nose up even a tiny bit results in a climb, but pushing the nose down an equal amount results in a shallow dive, with no elevator position that results in level flight at that airspeed. It tends to make the plane rather frustrating to fly.

It is true that at lower Reynolds numbers (thinner air, smaller chords, and/or lower airspeeds), the required airfoils tend to be thinner. However, in my experience the required distribution of thickness and camber becomes more critical. Designing a really good low-Re airfoil with good performance and handling characteristics is one of the more challenging tasks in the field of aerodynamics.

Generally speaking, a "flat plate" airfoil is not going to make as much lift coefficient as a properly airfoiled section for the same set of operating conditions. For a canard, this does tend to help make sure that the canard stalls before the wing. However, it tends to make sure that the canard stalls a LONG WAY before the wing. This means you get to use even less of the wing's total lift-making ability, which then requires even more wing and canard area to accomplish the same mission profile, and a net loss of perormance because of the increased whetted area.
Feb 16, 2008, 09:47 AM
Registered User
Don Stackhouse's Avatar
Originally Posted by canard addict
... I especially enjoyed the discussion of lift and drag forces about the CG to obtain equilibrium. The use of up or down thrust to to accomplish this is a subject that I would like to hear discussed when applied to high or low wing planes. It seems that a low winger would require some UP thrust with a forward static margin but some argue that down thrust is needed in the design for no good reason.
That's really not a good way to deal with the issue.

Aero force equilibriums need to be dealt with separately from thrust issues. Thrust varies all over the map depending on flight condition. A given alpha and airspeed will result in a certain amount of drag, lift and pitching moment from the wing, but the amount of thrust existing during that situation could be anywhere from negative (windmilling drag) to maximum, full-throttle thrust. Trying to balance one type of force with the other is generally going to give you an inconsistent, poor-handling airplane.

Step 1: Design and adjust the flying surfaces so that they keep each other in the desired balance at the desired flight conditions, with or without power.

Step 2: AFTER you've achieved that, then pick a thrust line angle (generally the angle that results in the thrust line passing through the C/G in both the horizontal and vertical sense) that does not mess up your pitch trim when you change throttle setting.

I feel that you agree that speed changes are in order with changes in decalage.
Anything that changes the effective incidence of the wing and/or canard will change the pitch trim, and therefore the one-G airspeed at which all the forces are back in balance. In that regard, a change in the incidence of the wing or canard, or a deflection of the elevators or of elevons on the wing all have the same effect.

However, the efficiency with which they do this does vary. In particular, deflecting elevons on the wing "up" to raise the nose and slow the airplane down also reduces the effective camber of the wing, and therefore its ability to make lift, just at the time when you're trying to fly slower, and therefore need more lift. This is counterproductive.

Changing decalage does change the trimmed airspeed. However, depending on whether you change the decalage by changing the canard incidence or thw wing incidence, the angle the fuselage flies at could also be changed, which would then change the fuselage drag. If you aren't planning for this deliberately, and just guessing at what to change and by how much, odds are that the fuselage drag is probably going to change for the worse!

It is possible to guess at the design of an airplane and then get lucky. However, deciding each of the parameters and changes for a specific, thought-out reason is generally a faster and much more sure way to end up with a successful design.

As an added benefit, doing it this way also increases the chances of understanding how you got to that end result, and the odds of being able to accomplish that again on your next design.
Feb 16, 2008, 10:24 AM
Registered User
critterhunter's Avatar
Good info. Printed it out to read over a smoke so don't have much to say at the moment. However, with the quick skimming I just did I don't think I saw any mention of percentage of canard surface area to the main wing's? From the limited reading I've done early on in this thread (first three pages) I saw somebody saying that the canard should be roughly 30% of the main wing's surface area. At a minimum it should be 20% but the magic number seems to be 30? What I read was that if the canard surface area is too small it will cause problems at lower landing/take off speeds I think. Can't remember the exact details (have it printed out for reference) but it made a lot of sense to me at the time. So, should I be shooting for that?

My FW42 is for the most part scale, but the main wing is slightly wider at the root and the canard is slightly smaller in width and span. When I ran the numbers through the canard COG calculator it gave me this:

At a static margin of 10%:
MAC: 7.21
Sweep Distance at MAC: 1.29
From Root Cord to MAC: 10.53
From Canard Root LE to AC: 21.59
From Canard Root LE to NP: 18.38
From Canard Root LE to CG: 17.3
Wing Area: 320.6
Canard Area: 48.9
Wing Aspect Ratio: 6.32
Foreplane Volume Ratio, Vbar: .42
Wing Area: 320.6
AUW: 32 (this is a guess at this time and probably will weigh more)
Max Life Coefficient: 1
Wing Loading: 14.37
Cubic Loading: 9.63
Stall Speed: 18.8 mph

A few things concern me here. First, my canard is only 6.55 percent the surface area of the main wing. I'll have to run the numbers on the canard's size scaled up to the 4 foot wingspan of the main wing because I know for sure it's smaller than scale right now. Still, I wonder if the scale size will still be much lower than even 20% of the main wing's surface area. What are your thoughts on this? Should I scale it up to 30% or just stay scale, and what will be the pros and cons? Luckily the canard just bolts on so it'll be real easy to change it as well as shim incidence in.

The next thing that concerns me is the wing loading. Right now it's already getting hefty due to the real dense EPS foam I'm using, but I can always drill some holes into the fuse belly to hollow it out some and reduce weight. The stall speed at only 32 ounces says 18.8 mph. That doesn't sound too good. What would be a good number to shoot for? Final question...As I increase the canard's surface area I would figure this should reduce the total wing loading? Come to think of it, I'm not sure if the wing loading calulator I used was taking a canard into considering. Should the input for wing loading be the wing surface with the canard's added to it?

Thanks for all the help. Canards are very new to me and thus tricky to figure out.
Feb 16, 2008, 01:43 PM
Registered User
G_eronimo's Avatar
Originally Posted by critterhunter
Think I've got the numbers figured out with the canard COG calculator. The COG at 10% puts my COG roughly right at the main wing's leading edge to about an inch in front of it. Glad I ran those numbers because somebody was telling me it probably should be a good inch or more behind the leading edge of the main wing, making it a tail heavy disaster waiting to happen.

Now that I'm sure of the proper window for COG I can start finalizing component placement, tape and cover where needed, and then use the battery to find out where it's chamber needs to be. I'm shooting for between the leading edge of the wing and the cockpit, as that's the easiest place to access and also the fatest portion of the fuse.

My rudder and V-stab size is also scaled up a bit to insure decent control. Good thing is that the COG is further forward than I was figuring, so this should translate into better rudder control. I was almost talked out of using a rudder but I think it will work well. Even the little FW42 found earlier in this thread had really good rudder response.

If you do a little reading on the FW42, the initial wind tunner models had the canard at the top of the fuse. It was dropped to give the pilot and forward gunner better vision. Later models reduced the four v-stabs to two. After that it was found that one v-stab worked almost as good as two and was settled on for ease in production.
I have the original Fw-42 article from Luftfahrt International 16/1976.
One thing to mention is, that the elevator was build up like a "double wing" (like some Junkers wings: Ju-52, Ju-87).

The fuselage of the Fw-42 has a very large and long (side view) area in front of any CoG. To compensate for that you need a large stab behind the CoG and/or it has to be farther away from the CoG.
I tested the original stab size and position first. It was way to small at that position. Without help from the ailerons there was only a small tendency to turn. Not enough for safe handling with rudder alone. So I increased the size to about 150% and moved it back several inches. If you fly with ailerons it should work, but I would consider it unstable.
As they didn't build a real plane, it is hard to say if the original size of the vert. stab. wouldn't have been increased due to results from flight testing. They added fins to the Fw F19, after the first plane crashed. And the vertical stab. is already very large in comparison to the one on the Fw-42.

I found the CoG markings on the fuselage. It was around 1.5 inches in front of the wing l.e. So I would say too, that any CoG behind the wing l.e. is not good (for this plane).

I didn't run a CoG calculation when I build the plane. There was nothing available (try and error ). Right now I would (and will) use Dieter Schall's ENTEX program:
There used to be a free download for testing, but I could not find it anymore.
It is an Excel program and it is in German only.
When I have some time, I will run the data of the Fw-42 with this program. The results are usually most of the time comparable to the results from the canard cog locator program.

Feb 16, 2008, 06:39 PM
Registered User
Critter, I got your canard wing as 15 % of the main. I feel that the canard area should be increased. This will move the CG forward and if you have a tail heavy problem, probably depending on the weight of the motors and batteries, you could reduce it to what seems correct to you. When using the calculator to figure loading, you add the areas together because both wings carry the weight. Must go now but will always be happy to help and I hope that Don, John and others will also pitch in. Charles
Last edited by canard addict; Feb 17, 2008 at 06:51 AM. Reason: Thoughts reconsidered.
Feb 17, 2008, 12:27 PM
Registered User

Canard Sailplane Fuselage Build

The 1/8" by 1/4" sticks were weighed and ran from 2 to 8 grams. The extremes were in short supply so the 6 gram sticks were selected for the four corners of the frame. The 4 gram ones were chosen for the uprights and angle braces. The total weight of the sticks needed came to about three ounces which was a 1/2 ounce savings over using all 6 gram material. Since about 24 pieces were required to measure 1.5 inches long, it was decided to find a way to measure and cut with the miter box. This method worked surprisingly well and will be shown in the pictures. Charles
Feb 17, 2008, 08:33 PM
Registered User
John235's Avatar
Originally Posted by critterhunter
The next thing that concerns me is the wing loading. Right now it's already getting hefty due to the real dense EPS foam I'm using, but I can always drill some holes into the fuse belly to hollow it out some and reduce weight. The stall speed at only 32 ounces says 18.8 mph. That doesn't sound too good. What would be a good number to shoot for? Final question...As I increase the canard's surface area I would figure this should reduce the total wing loading? Come to think of it, I'm not sure if the wing loading calulator I used was taking a canard into considering. Should the input for wing loading be the wing surface with the canard's added to it?
I think there are probably ways to get canards with smaller forward wings to fly, but if you keep the forward wing 20 - 30% of the main wing area, you will have good control with a conventional elevator setup. I suggest to build an easy flying model for your first canard and stick with the tried and tested design choices that Charles has posted on this thread.

I don't think it is easy to compare the wing loadings of canards to conventional models. The issue is that canards seem to always land faster, so this is something that might influence your choice if you don't have experience with fast models. The good news is that a correctly setup canard model will be more predictable on the landing approach, but that may not help you so much on the maiden flight. I think you are fortunate that the resident Canard Addict has so much experience with canards in the very similar size range to what you are proposing to build. Charles does his wing loading calculations with the total wing area including the canard. Comparisons with conventional models don't mean much, so it is all relative to the type of aircraft anyway. The wing loading of my small 80cm canard based on total wing area is only 10oz per sq'. There is no way I want to have faster stall speed, since hand launching took a bit of practice. Mine is probably fast because of the semi-symmetrical airfoil and the small size, but for a slightly larger scale model, 12oz seems like a sensible limit to me.
Last edited by John235; Feb 17, 2008 at 08:40 PM.
Feb 18, 2008, 06:15 AM
Registered User
John, Thanks very much for the nice remarks! I will have to work very hard to deserve them. I appreciate your valuable contributions to the thread. Helping to make these models fly is where the fun is for me. Charles
Feb 18, 2008, 07:36 AM
Registered User
Flown my duck with flaperons yesterday and today. Just like Don predicted, there is very little effect on slowing down the plane.
With flaps at 30 degrees, the canard nose will pitch down even with full up elevator. As I trimmed down the flaps, I was able to get pitch control when flaps where about 8-9 deg. At this angle, flaps effect was close to zero.
Charles, Don, thank you for all the great advices ! My canard has been a successful project ! I learned what I needed to learn, now it is time to make the next step toward my AP platform !
Best regards,
Aries Bujor
Feb 18, 2008, 07:59 AM
Registered User
In my opinion, flaperons are not worth the effort. There is not enough servo movement available to give enough flap position and still have adequate angular movement for the ailerons. If you want good flap action then add separate flaps and leave the ailerons to do their job as you want them to. I have much more fun with elevons. Bujora, it has been an interesting experience following your canard work. Thanks for sharing it and best wishes for your AP endeavors. Charles
Feb 19, 2008, 11:58 AM
Registered User
Critter, You indicated a canard area of 48.9 and a wing area of 320.6. So 48.9 divided by 320.6 gives a canard area of 15.2%. If you want a 25% canard, it would be 0.25 times 320.6 which is 80 and a 30% canard would have an area of 0.30 times 320.6 or 96. Just adjust the dimensions for the canard in the calculator until the size you want comes up. Maybe 27% would be a happy medium to start or a larger size which could always be trimmed down. Have fun! Charles
Feb 19, 2008, 12:57 PM
Registered User
critterhunter's Avatar
Thanks for all the input. I printed it out to read and reflect over before asking more questions or replying. Thanks again.
Feb 19, 2008, 01:02 PM
Registered User
Don Stackhouse's Avatar
Originally Posted by canard addict
...48.9 divided by 320.6 gives a canard area of 15.2%... Just adjust the dimensions for the canard in the calculator until the size you want comes up...
Since area is proportional to the square of the linear dimensions, then linear dimensions are proportional to the square root of the area.

So, applying that to practice, and taking your present area of 15.2%:

to scale to an area of:_____Multiply linear dimensions by:

25% ____________________1.282 (the square root of 25/15.2)
27% ____________________1.333
30% ____________________1.405

However, the required area depends on a number of factors, pretty much the same ones that apply to the sizing of a horizontal tail for a conventional layout.

Yes, it needs to be big enough that the Reynolds numbers ("Re") of the canard need to be reasonably close to those of the wing, but there's quite a bit of "leeway" there.

However, a bigger issue is moment arm. You can't separate the area ratio from the moment arm. It's the two of them together that control the effectiveness of the tail, and the area of the wing along with the chord of the wing that determines the size of the task the tail or canard has to accomplish.

If you have a lot of moment arm, you can get away with less canard area. In addition, a long moment arm improves dynamic stability, the ability to damp out oscillations. Dynamic stability is linear with tail or canard area, but proportional to the square of the moment arm.

The typical method for looking at this entire issue is tail volume coefficients, as explained in some articles in the "Ask Joe and Don" section of our website Some of my earlier posts to this thread list links to specific articles in AJ&D on this subject, or you can go to the search engine in AJ&D for a list of links.

Looking at your numbers for your airplane, your canard moment arm calculates out to just shy of 25", and that results in a horizontal tail volume coefficient of 0.526 . That's a bit above the middle of the normal range of about 0.45 to 0.55 . That factor suggests that your 15.2% tail is big enough.

However, with a canard layout you also have to consider the canard's max lift and stall characteristics. A small, low-Re canard is going to have a lower max lift, and possibly sharper stall, with a more pronounced nose-drop when that happens.

Going to a thicker airfoil (the typical way to get higher lift and gentler stall on full-scale aircraft) may help stall characteristics but can actually reduce max lift at low Re's! In fact, the way a thick airfoil tends to soften stall characteristics is by causing flow separation on the top (and maybe on both the top and bottom) at lower angles than would happen on a thinner airfoil. Because this reduces the canard's max lift, it increases the stall margin for the wing (good), but reduces the max lift you can squeeze from the wing (BAD).

Going to a larger canard increases it's Re's, which increases the canard's max lift coefficient. This does mean you have to move the C/G further forward to keep the same static margin, but that will probably not offset the increase in max lift due to the Re increase. This means that you can now use more of the wing's max lifting ability (good), but increases the chances that the canard can pull the wing's angle of attack high enough to make the wing stall before the canard (VERY, VERY, VERY BAD!!!!). This problem is not insurmountable, but it means more homework in the design phase and more risk in the flight test phase.

Making the canard larger also means that your canard (which is still smaller than the wing, and therefore a less efficient lift producer than the wing) is now carrying a greater percentage of the aircraft's total weight. This reduces the aircraft's overall efficiency.

OTOH, moving the C/G forward for whatever reason (including due to a larger canard) does have the benefit of increasing the vertical tail's moment arm, improving yaw stability.

As is usually the case in aircraft design, it's all a trade-off.