Maneuverability And Stability

[RocketTurtle]

So, I've been building planes for a while, have a few functional SSTO's, lot of atmo planes, VTOL's, they're all quite stable and nice to fly, but one thing they lack is maneuverability. I can either make an insanely maneuverable plane the flips out at the touch of a button and slams into the ground try to do its own aerobatic maneuvers, or it has a turn radius about the size of Jool.

So yeah, how can I make a maneuverable and stable plane?

Thanks, by the way, the sooner I get help, the less deaths Jeb experiences...

[xxjetterxx]

Hey RocketTurtle. Personally, I find that having Ferram Aerospace Research installed helps a lot, because the wings actually work like... well... wings!

Anyways, give it a shot, this is mostly due to KSP's bad aerodynamic physics

[RocketTurtle]

Well, yeah, and hey. But I like to make stock planes so I can share them with people. But yes, I agree, FAR is worlds better.

[Zephram Kerman]

Well, two things make a big difference:

• lots of lift; maybe you can rotate, but it's lift that makes you actually change direction of travel

• CoM just slightly ahead of CoL; it must be ahead, but the distance between these two has a huge affect on stability/maneuverability.

Oh, and #3, canards are cool, but I avoid them on small planes. If it's really necessary, and if you have FAR, tilt it up 5° so that it stalls before the control surfaces in back.

[ferram4]

Maneuverability and stability are always at odds with one another; if you want more maneuverability the stability of the plane will suffer, and if you want more stability the maneuverability of the plane will suffer. It seems like you've seen the results of running to either extreme. What you want to do is to take a stable plane, and make it less stable; put the control surfaces further forward than they are. Then, if it's still not maneuverable enough, move the fins even further forward. If it's not, then shift the fins back a little bit.

[numerobis]

As ferram says, instability comes from having the center of lift ahead of the center of mass, so that when you start to turn, you generate more lift, which on average is applied ahead of your center of mass, which makes you turn harder, and suddenly you flip out.

Putting control surfaces further behind the center of mass will improve maneuverability, but in their rest configuration they'll return you to point prograde. So you'll get more maneuverability when you touch the controls, up to a maximum, but you'll be stable when you don't touch the controls.

Gyros and thrust vectoring will improve maneuverability without affecting stability.

[RocketTurtle]

Alright... I've just seen a few videos with very maneuverable and *relatively* stable.

[RocketTurtle]

Also, putting canards on the plane makes it very maneuverable, but no matter the placement, made the plane very unstable.

[allmhuran]

KSP, being a game does not follow all real life constraints. One sure-fire way of increasing agility without sacrificing stability is simply adding more control surfaces while keeping the CoL in the same spot.

1) Build a stable plane with CoL just behind CoM.

2) Add control surfaces both in front of and behind the CoL so as to not shift its position.

3) GOTO 2.

[Murph]

It's worth noting that real world fighter-type jets are actually designed to be inherently unstable, to the point of being impossible to fly without computer assistance. This is what gives them the ultimate manoeuvrability required for the type of plane, and the pilot gets to pray to the software gods that the avionics combined with their own skills manage to keep them on the survival side of the ragged edge. When it all goes wrong, they get a free Martin Baker tie, if they survive to receive it.

[allmhuran]

The funny thing about KSP is that most planes are very unstable in that they will easily initiate a pitch, yaw, or very violent roll. The big difficulty is trying to maintain a high turn rate during a normal turn. Pull back on the stick all you like, the plane will only turn so hard and then the nose will start flopping back down, and begin oscillating. Try to do this in a real plane and you'll stall... but KSP doesn't have stalling, hence this odd behaviour.

Having all of your mass centralized (minimizing your moment of inertia) and just spamming control surfaces front and back can go a long way to overcoming this. It is possible to build craft that can pull 180 degree turns in 2 or 3 seconds (with a bit of throttle modulation).

[allmhuran]

(duplicate removed, damn network)

Edited December 4, 2013 by allmhuran

[el_coyoto]

I agree with all of the above, especially the placement of CoL in relation with CoM and the need for aerodynamic surfaces (canards help a lot).

On top of being an awesome 101 on plane building, the following thread is wonderfully illustrated and has a nice "state of the art" bonus at the end of the opening post :

Basic Aircraft Design Explained Simply With Pictures.

It is so well written that it helped completely grok the anhedral/dihedral wings setup, how and why they work, while this I was stuck on this for years...

[RocketTurtle]

Yeah, I've read that.

But I've found that setting the wings low in a delta-canard configuration gives nice maneuverability and stability.