Plane Spins Out When Turning

[KomradeKerman]

I've been getting back into KSP, and wanted to try my luck at aircraft.  I designed a fairly simple plane, but whenever I try to turn it off of the prograde marker even slightly, it spins out like crazy and I cannot regain control.

https://youtu.be/u6BZjLiumqA

I have several mods installed, but none of them affect physics or aerodynamics, and all parts in this plane are stock.

EDIT: The issue has been resolved.  Simply adding a rudder/vertical stabilizer eliminates this behavior.

Edited October 9, 2016 by KomradeKerman

[ForScience6686]

At minimum, add a vertical stabilizer at the rear for yaw control.  I would also give the wings some incidence to improve lift and pull col closer to com.

[Justicier]

I've been building spaceplanes for a long time and to date, I have not been able to create a 100% reliable "Dreamchaser" style lifting body aircraft with only stubby little 45degree wings/stabilizers. The aerodynamics modeling in KSP works... but in some ways is still pretty hinkey/buggy, and it appears to be that way with alot of lifting body designs as well as flying wings. The angled wings seem to be the primary issue, especially if you are not using a decent vertical stabilizer. Granted I have made some that work well enough at low altitude in-atmo, but when re-entering the atmosphere they tumble like a leaf in the wind. It is something I have noticed for a long time, but haven't been able to work around other than clipping a vertical stabilizer into the body of the plane. It feels cheaty but it does overcome what seems to be a limitation of the aero system in the game.

Edited October 8, 2016 by Justicier

[GoSlash27]

I agree with the others; you need a vertical stabilizer.

Best,
-Slashy

[Karol van Kermin]

Thats a risky plane design, as it only has one wing. I would highly recommend adding a rud and a tail wing to maintain stability.

[KomradeKerman]

6 hours ago, Justicier said:

I've been building spaceplanes for a long time and to date, I have not been able to create a 100% reliable "Dreamchaser" style lifting body aircraft with only stubby little 45degree wings/stabilizers. The aerodynamics modeling in KSP works... but in some ways is still pretty hinkey/buggy, and it appears to be that way with alot of lifting body designs as well as flying wings. The angled wings seem to be the primary issue, especially if you are not using a decent vertical stabilizer. Granted I have made some that work well enough at low altitude in-atmo, but when re-entering the atmosphere they tumble like a leaf in the wind. It is something I have noticed for a long time, but haven't been able to work around other than clipping a vertical stabilizer into the body of the plane. It feels cheaty but it does overcome what seems to be a limitation of the aero system in the game.

The angled wings actually don't really have any major effect.  I had tried a similar design with straight wings and intakes on the underside, but that plane failed in a similar fashion.

I have, however, added a rudder, and the plane now flies without any problems.  Thanks for the help!  I probably should have known to put a tail fin on it, but I thought I had been able to make planes without them before.

[Plusck]

2 hours ago, KomradeKerman said:

The angled wings actually don't really have any major effect.  I had tried a similar design with straight wings and intakes on the underside, but that plane failed in a similar fashion.

I have, however, added a rudder, and the plane now flies without any problems.  Thanks for the help!  I probably should have known to put a tail fin on it, but I thought I had been able to make planes without them before.

When you say you "turn", what do you mean exactly?

You have some torque authority with that cockpit, so you should be able to move the craft around using WASDQE, but some of that movement is not possible using aero forces. It is therefore possible to move your craft into a position which is not aerodynamically possible, if you see what I mean.

In the SPH (or in flight) you should be able to click on the cockpit and make it so that the torque wheels are not used by the "pilot". In that case, it should be virtually impossible to get yourself into a flat spin, since you don't have the physical ability to put your craft in that position.

So yes, you can definitely make craft that fly without a vertical stabiliser. To make it flyable, however, you need to ensure that drag from the body will tend to return the craft to a forward-facing attitude, and you don't have the ability (i.e. torque wheels) to exceed the possibilty for drag to push the craft back to that attitude.

[Rath]

I'm guessing that even though it flies right now, it is incredibly unmenourvable. (This might be my sandbox mode fighter/aerobatic plane builder speaking; I tend to only call something done when it is supermenovorable)

If that is true, the problem is that the col is far off from the com.  That used to be good, but now you want to move it a little bit closer.   I tend to shove the col on the com, but that requires tweaking or jet gimbal to work.

[KomradeKerman]

5 hours ago, Rath said:

I'm guessing that even though it flies right now, it is incredibly unmenourvable. (This might be my sandbox mode fighter/aerobatic plane builder speaking; I tend to only call something done when it is supermenovorable)

If that is true, the problem is that the col is far off from the com.  That used to be good, but now you want to move it a little bit closer.   I tend to shove the col on the com, but that requires tweaking or jet gimbal to work.

It takes about 10 seconds to turn around 180°, but that's really just my personal preference.  I like stability over maneuverability.  In fact, I'd say the plane now turns like a dream: it doesn't jerk about with the keyboard controls, but it isn't sluggish when I try to turn it.

[Rath]

10 seconds seems so long to me...

Do I need an intervention?

[ForScience6686]

Maybe try adding small tail fins in the wing tips angled up and as far back on the wing as you can.  That way be enough, but you may do better with the wings moved further back with this method.  That may mean the need for adding another wing segment up front to keep col forward enough.  Maybe the stake could clip in there and look nice.  Our maybe shift the engines back a bit to bring the com rear.

I like what you got here, I might try a similar design while I wait for the update.

[KomradeKerman]

I've made some optimizations, and the turn speed has been cut in half.

[ForScience6686]

The only things I can recommend are larger control surfaces, or a closer com/col.  But stability will start to suffer again.