Plane control surface question (elevator/canard)

[phemark]

Hi,

I am trying to get better at building and flying planes, so I have built this plane. However, although it looks good (to me), it barely lifts from the ground at the end of the runway, and then it almost doesnt respond to elevator controls:

However, if I just add two canards in front of CoM, it flies SO MUCH MUCH better:

So I wanted to ask why is this?

1) I can see that CoL changes a bit in the second picture, however is this small change the biggest culprit?

2) Or is it because front canard is giving all the help?

3) Or is it because in pic1 my two wing elevators are almost with CoL?

I would like to learn to build a plane without using canards.

P.S. 4) How to put bigger elevators on the end vertical wing? I cannot do that...

Thanks

Edited November 2, 2014 by phemark

[Master Tao]

There are a few important changes:

• The canards shift the aerodynamic center forward, making the plane less stable.
  
• They are farther from the center of mass than the rear elevons, so they can provide more leverage.
  
• They add more control surfaces, giving more pitch authority.
  

Your vertical stabilizer has a built-in rudder – there's no need to add another control surface to it. Other parts, like the small delta wing, can have control surfaces attached to them.

[phemark]

Thanks for the answer, a bit clearer now.

5) Am I correct in thinking, that canards in the front, given everything even (distance of CoL and CoM, distance of canard/elevator) will be better than elevator in the back? If yes, why?

6) And could you please explain me why this elevator (control surface lift rating 0.7)

is worse than this elevator (control surface lift rating 0.25)

?

Thanks

[Wanderfound]

Basically, your plane is drastically undersupplied with elevators, and adding the canards partially corrects that. Canards are essentially just forward-mounted elevators.

Tail mounted elevators should be built off a horizontal stabiliser. Like so:

[Starman4308]

Control surfaces act like levers. In the case of the elevons on your wings, they are so close to CoM that they provide little pitch authority*. The canards, however, are much more distant, and so allow for much greater pitch control. You could also accomplish this with a more traditional empennage (tail section). Given the size of your spaceplane, it probably would be a better idea to increase the rudder size by replacing the winglet with a full wing (with elevons), which would additionally let you place elevators in a T-configuration.

*They will do well for roll authority, because when you change perspective to roll, your wing elevons are distant from CoM.

[Starman4308]

Thanks for the answer, a bit clearer now.

5) Am I correct in thinking, that canards in the front, given everything even (distance of CoL and CoM, distance of canard/elevator) will be better than elevator in the back? If yes, why?

All that matters is how far it is from CoM. There are other reasons to have a tail empennage vs. canards, but for pitch authority, just think of control surfaces as levers.

6) And could you please explain me why this elevator (control surface lift rating 0.7)

is worse than this elevator (control surface lift rating 0.25)\s

Lift rating isn't the same as control capacity: the first is a winglet with a small built-in control surface, whereas the second is all control surface.

[Wanderfound]

In terms of control surface authority, canards and elevators at equal distance from CoM are identical: they're both pitch controllers, and their strength is purely a product of how much air is deflected and how much leverage does it have on the CoM. However, canards tend to promote manouevrability over stability for two reasons:

1) They pull the CoL forwards. The more the CoL is behind CoM, the more stable and less manouevrable the plane will be.

2) Control surfaces react in a stall in opposite ways depending on if they're in front or behind CoM. Rearset surfaces tend to return to straight flight, forwardset surfaces tend to exaggerate the diversion from straight.

Note, however, that all of this applies to FAR/NEAR realistic-ish aerodynamics. In stock aero, all bets are off; there are some places where sensible aero design principles apply, but also many where they don't.

[phemark]

All that matters is how far it is from CoM. There are other reasons to have a tail empennage vs. canards, but for pitch authority, just think of control surfaces as levers.

Thanks,

And would you mind mentioning other reasons to have tail empennage?

[phemark]

In terms of control surface authority, canards and elevators at equal distance from CoM are identical: they're both pitch controllers, and their strength is purely a product of how much air is deflected and how much leverage does it have on the CoM. However, canards tend to promote manouevrability over stability for two reasons:

1) They pull the CoL forwards. The more the CoL is behind CoM, the more stable and less manouevrable the plane will be.

2) Control surfaces react in a stall in opposite ways depending on if they're in front or behind CoM. Rearset surfaces tend to return to straight flight, forwardset surfaces tend to exaggerate the diversion from straight.

Note, however, that all of this applies to FAR/NEAR realistic-ish aerodynamics. In stock aero, all bets are off; there are some places where sensible aero design principles apply, but also many where they don't.

Thanks for explanation, i understand it better now.

[Starman4308]

Thanks,

And would you mind mentioning other reasons to have tail empennage?

Going to defer to Wanderfound here, who has vastly more experience with spaceplanes.

What I knew offhand, though, is that if you design your canard right*, you can get it to stall before the main wing; that'll cause your nose to tip downwards and return you to stable flight.

*And if you do it wrong, it stalls later​, making the problem worse.

[phemark]

Thanks for the quick and really good info guys.

As for my plane, with current set up, I had to move my engines right to the front, if I want to use back elevator:

I think I'll need to redesign my plane, if I want to put engines in the back, as it was my original intention, but at least I'll know what to do.

Thanks again!

[Starman4308]

I think I'll need to redesign my plane, if I want to put engines in the back, as it was my original intention, but at least I'll know what to do.

Thanks again!

Suggestion: way bigger tail section, with full wing sections and separate elevons. The elevators can help pull the CoL backwards, letting you drag the engines back as well. You might also ditch the tail connector: it's not really doing you that much good.

One final demon to look out for is CoM shifts as fuel burns: I use RCSBuildAid to find where my dry CoM is. I've had an egregiously expensive (fortunately unmanned) spaceplane lawndart on re-entry because the CoM shifted far backwards.

[Wanderfound]

phemark, it sounds like you might benefit from looking at this thread: http://forum.kerbalspaceprogram.com/threads/52080-Basic-Aircraft-Design-Explained-Simply-With-Pictures

As for the empennage thing...horizontal stabilisers are good for the same reason that vertical stabilisers are good: they add a lot to stability.

Basically, in each axis you need stability and the ability to disrupt that stability. This allows the plane to fly straight most of the time, but change direction when you want it to.

Breaking is easier than making. To get stability, you use a big airfoil (e.g. tailfins and tailplanes). To disrupt stability, you deflect a small airfoil (the control surface).

[phemark]

I read that thread a couple of times, and it was my guide. I just didnt expect that slight change to CoL/CoM location makes a such big difference. And also, that thread doesnt mention elevator positioning (far from CoM), as you guys explained to me here.

And thanks, I will try to look out for dry mass CoM

[Wanderfound]

I read that thread a couple of times, and it was my guide. I just didnt expect that slight change to CoL/CoM location makes a such big difference. And also, that thread doesnt mention elevator positioning (far from CoM), as you guys explained to me here.

And thanks, I will try to look out for dry mass CoM

As Starman mentions, RCS Build Aid is your friend there; it adds an extra marker in the SPH for dCoM and gives you the CoM/dCoM offset in metres. Try to keep it < 1m.

[Wanderfound]

Thanks for the quick and really good info guys.

As for my plane, with current set up, I had to move my engines right to the front, if I want to use back elevator:

http://i.imgur.com/zg5i2Syl.jpg?1

I think I'll need to redesign my plane, if I want to put engines in the back, as it was my original intention, but at least I'll know what to do.

Thanks again!

On the present version:

* The amount of pitch authority you have is tolerable, but at the lower end of what will work. Consider a tailplane built from wing sections with multiple control surfaces. This will also pull your CoL right back, allowing you to shift the engines rearward.

* CoL doesn't need to be on top of CoM, just behind it will do.

* Make sure you use the tweakables to set the control surfaces to only influence the appropriate axes: yaw for the rudder, pitch for elevators, roll for ailerons.

* The wing control surfaces are for roll authority only. The further from CoM the better; move them to the end of the wing.

* Your current build will have huge tailstrike problems. Losing the tail boom will help with this: so will raising the landing gear on hardpoints.

* If all of those fuselage sections are full of fuel, you've got enough juice to circumnavigate Kerbin a dozen times over. Empty out all but one of them.

[phemark]

Awesome tips, thanks!

One more thing - how can I attach elevators to vertical wing?:

They seem to interfere with each other.

P.S. With tweakables, do you mean right click on control surface, and disable two not needed axes?

[Starman4308]

Awesome tips, thanks!

One more thing - how can I attach elevators to vertical wing?:

http://i.imgur.com/5UsI9PWl.jpg?1

They seem to interfere with each other.

P.S. With tweakables, do you mean right click on control surface, and disable two not needed axes?

Slap a couple wings on with bilateral symmetry, and attach the control surfaces to those. An example is in Wanderfound's post.

[Wanderfound]

Awesome tips, thanks!

One more thing - how can I attach elevators to vertical wing?:

http://i.imgur.com/5UsI9PWl.jpg?1

They seem to interfere with each other.

P.S. With tweakables, do you mean right click on control surface, and disable two not needed axes?

Elevators to a vertical stabiliser: don't. Put some horizontal wings at the height you want the elevators, and attach the control surfaces to that. Or hook a pair of horizontal all-moving winglets (AV-R8's, canards) to the tailfin. Basically, build it as if it were a real plane. Control surfaces go along the trailing edge of wings, not anywhere else.

Have a poke through the Kerbodyne thread and look at the screenshots. There's enough diversity of design there that it should give you clues on how to construct most things.

Re: tweakables, yes.