Symmetrical Craft Spinning Out

[Magicide]

Hello!

I have been having fun with KSP for a few months now and have sent many kerbalnauts to their doom in the name of pushing boundaries. However since .22 I've noticed a problem and I don't know if it's the game or me.

I have had several ships lately that should have flown straight and true instead spin wildly. Attached is a photo of my prototype concept for a super heavy lander. It is built in 6x symmetry and the center of mass and lift are both dead center on the middle of the vessel. If I reattach the girders in 4x symmetry it will sometimes launch straight but other times it will spin as well. The only time I have a 100% success rate regarding spinning is if there is only one central engine.

I have noticed that my heavy launcher which is a central orange container and three additional layers of orange containers in 6x symmetry also spins every single launch while the same design used to be perfectly stable. I have my suspicions that something more is going on but maybe it's a simple fix. I am not running any mods. Hopefully someone can help me out.

Thanks!

http://imgur.com/S9q0V01

Edited November 1, 2013 by Magicide

[TaranisElsu]

Try turning off the gimbals on the outer engines (or using the LV-T30 which does not gimbal). Only the center engine should be allowed to gimbal its engine. Also try adding RCS, Reaction Wheels, or fins to counteract the spin.

I have noticed a small tendency to spin on some of my rockets, but it is usually very slow at first and easily countered.

Edit: I just noticed all the fuel lines. Make sure that the fuel is flowing evenly. Just a little bit off balance makes a big difference.

Edited November 1, 2013 by TaranisElsu

[BigFatStupidHead]

Sometimes, part clipping causes phantom forces that cause spin. Small girders are known to be a particularly bad offender for this. I couldn't see anything wrong in your design, but you might want to check for this as well.

Edit: Actually, how are those tanks attached to the sides of the girders? That might be enough.

[Dwight_js]

If the fuel line placements aren't all symmetrical then I think that they can cause spinning too.

[yongedevil]

Joints in Kerbal are not perfectly rigid and will flex some in flight. For engines placed out from the centre of mass this flexing will usually lead to some asymmetric thrust vectoring resulting in spinning. At least that's the best explanation I have for the phenomenon.

I've found the best solution to be to stiffen the rocket so there is less movement and adding controls to counter what is left. 1-2 struts between each booster and the core will usually be enough, with in very bad cases, an extra 1-2 struts linking each booster to its neighbours.

[Magicide]

I attached the girders directly to the tank and then the outer tanks directly to the other end of the girders.

As an experiment I just tried a stripped down craft consisting of only a fuel tank, probe core, girder and small tank/engine attached to the girder. Anything from 2x to 8x symmetry leads to the ship rotating at an increasing rate. I experimented with the other girders, decouplers and pylons as well. They all exhibit rotation right from launch if any sized tank + engine is attached to them. I also tried several different engines and they all rotate. I tried stiffening the assembly with struts but it doesn't appear to make any difference.

What does work is disabling the engine gimbal so it's the most likely culprit. This wasn't an issue prior to .22 so I'm thinking something got buggered up in the last patch. This definitely throws a wrench in my plans as the interplanetary vessel I've been assembling in orbit relies on nuclear engines attached to girders... now it looks like I get to deorbit it and start from scratch.

[Magicide]

Actually I just tried running struts from each of the outer tanks to the central core and then between each of the outer tanks. It appears to fix the issue but my poor ship goes from advanced rocketry to looking like the wright flyer.

Joints in Kerbal are not perfectly rigid and will flex some in flight. For engines placed out from the centre of mass this flexing will usually lead to some asymmetric thrust vectoring resulting in spinning. At least that's the best explanation I have for the phenomenon.

I've found the best solution to be to stiffen the rocket so there is less movement and adding controls to counter what is left. 1-2 struts between each booster and the core will usually be enough, with in very bad cases, an extra 1-2 struts linking each booster to its neighbours.

[Zephram Kerman]

It seems to me the outer tanks are attached to the trusses slightly off-center. So when those engines fire, they will cause the truss connections to twist a little bit and spin your craft.

[Tex_NL]

Since your craft has a relatively low profile it wouldn't surprise me if, due to fuel drain, the center of gravity ends up behind the cente5r of thrust. Aerodynamic control surfaces reverse their behavior if this happens, engines do not. As a result your engines will steer right when the should steer left and left when the should steer right making it nearly impossible (or at least very hard) to control.