The Evolution of the Canard
(or at least, the evolution of THIS canard)
I originally built two of these C.A.T. Canard AST Tailless planes from plans #1073 in the August 1990 RC Modeler. It was designed to be a twin, push-pull .40 size plane. Aircraft #1 was just that - it had two .40 engines and it flew just fine. Aircraft #2 had the exact same planform, but I put a .60 on the rear and balanced it out with nose weight. It also flew just fine with that setup, but I sold plane #2 and kept flying plane #1.
|August 1990. This plane had some interesting flight characteristics. I originally had the landing gear too far back to allow good rotation for take-off. With both engines running at full speed, the plane would roar down the runway at Mach 3 and never rotate. It would finally hit some bump in the runway that allowed for just enough rotation, and then it would headed straight up at Mach 3. I fixed that problem by moving the landing gear forward. On the second flight, I stalled the plane to see what would happen. It turned into a VERY flat spin and would not recover. The plane came down like a helicopter. It broke the landing gear through the wing, but no other damage was done. I have never stalled it since.|
|Summer 1994. I decided a few years later to move the .40's to another plane and put a .61 on the rear. I extended the nose and balanced out the rear engine weight with nose weight. I also put wing extensions on to give a little more lift. The plane flew about the same after the mods.|
|August 2000. At this point, I was really starting to get creative. I moved the original wing to a Four-Star body I recovered from the trash at the flying field, and I built this forward swept wing to replace it. The plane never flew in this configuration, but it looked neat.|
|June 2002. I decided that the original canard needed some
augmentation, so I changed it for this canard, the remains of another
crash. This plane was ready to fly, but I never took the time to compute where
the CG needed to be to get stable flight. This configuration never flew either.
|May 2003. This configuration has flown. After
calculating the center of aerodynamic pressure for each lifting surface (the AC for the
main wing is actually 4.2 inches in front of the wing center leading edge), the projected
CG came out far forward of where I had it in the previous configuration. This
required a major rework of the whole plane. I moved the nose cone (minus some of the
weights) to the tail. The main landing gear were reconfigured for the new CG. I
moved the engine and fuel tank to the nose. I also changed the rubber-banded wing to
a bolt-on wing, after deciding they would probably twist quite a bit due to wind
loads. I moved the servo and radio compartment behind the canard so the fuel tank
could fit in the nose. The canard surfaces are now just ailerons for roll control,
since they are within a few inches of the CG, and won't have much effect on pitch.
The main wing surfaces are elevons, so they control all the pitch, along with roll. The
nose steering and rudder are normal.
The CG calculation was done using Ron Van Putte's canard calculations available on-line at: http://www.palosrc.com/instructors_corner.htm using my own Excel spreadsheet that I put together that allows easy use of these formulas.
|April 2005. The previous configuration flew, but did not
have enough elevator authority. The large canard moved the elevators themselves too
close to the CG to get enough pitch control, so all the elevator control was done by
mixing elevon control into the ailerons. I have swapped back in the original canard
but farther forward than before. The new configuration keeps the center of pressure about
the same, but the elevator now has better pitch control authority.
Previously the horizontal elevator had to be reflexed for level flight, so I also took the
fixed positive pitch out of the canard and it now sits at 0 degree incidence.
To get the CG just right, I have used a couple CG calculators. My Excel spreadsheet from Ron Van Putte's canard equations and this canard web page are in close agreement, but not precisely the same. To verify completely, I built a small scale (.292 of real) model of the full size plane and trimmed it for best glide. Best glide should be almost on the neutral point. The experimental neutral point from the scale model agreed exactly with the equations at the canard web page.