What is making my rocket unstable?

[T.C.]

I'm a new player experimenting with basic rocket designs, and I'm having a hard time figuring out what contributes to instability.

I've created a rocket capable of putting a payload into orbit. With no payload, it gets into orbit easily. When the payload is a small rocket weighing 27.22 tons, it gets into orbit easily. But when the payload is a space station module weighing a mere 12.13 tons, the rocket becomes impossible to control during the gravity turn, and I end up doing cartwheels. SAS modules don't seem to help.

I don't think the explanation has anything to do with flexibility because I've used strut connectors to make sure the station module is rigidly connected; I certainly don't see any flexing during flight.

What other explanation can there be?

-TC

P.S. I'd like to post screenshots, but I can't figure out how to do that...

Edited May 25, 2014 by T.C.

[Othuyeg]

Do you have the steam version of the game?

If so, press f12 (may need to be pressed a few times) while in the game, then when you have finished taking screenshots, you go to your game library in steam and scroll down on the KSP page and there is the screenshot viewer. You click that and click show on disc and then you go to for example imgur.com, press upload pictures and you drag and drop the images there. You can then upload them and post a link to them here.

[T.C.]

Here are some screenshots.

This is stable.

When I put a rocket on top, it is still stable:

But when I put a lighter, lower center-of-mass station module on top, it becomes impossible to control during the gravity turn.

I've tried lots of things -- adding SAS modules, adding control fins, disabling reaction wheels in the cupola modules, capping the station module with a nose cone, ... Nothing seems to help.

-TC

[ArcFurnace]

Looks like you're probably getting flexing at that small decoupler/docking port joint. Decouplers and docking ports have always been notoriously wobbly. You obviously know that, since you've put some struts across the joint. Try adding more struts, say from the wide bottom of that 3.75m-2.5m adapter to the top of the station module, and see if that works any better.

[SlimeCrusher]

Are you using any mods? Or rather, are you using FAR?

[Jouni]

Your problem is caused by the cupola modules. The rocket is basically too small to lift two of them. The cupola modules have twice the normal drag coefficient, meaning that drag will pull them behind the center of mass, if you point away from prograde when flying close to terminal velocity. Adding more reaction wheels may help, and you may also try flying slower or climbing higher before starting the turn.

[T.C.]

Thank you for the replies. The drag explanation makes sense and matches what I'm observing. I'll try slowing down the rocket and see if that helps.

-TC

[Taki117]

Thank you for the replies. The drag explanation makes sense and matches what I'm observing. I'll try slowing down the rocket and see if that helps.

-TC

If you are not using FAR try to keep it below terminal velocity, you can Also get a mod called Procedural Fairings to make your rocket more aerodynamic (Almost a requirement with FAR)

[T.C.]

I'm not using FAR. Is it recommended?

-TC

[CaelumEtAstra]

I'm not using FAR. Is it recommended?

-TC

Not really. All it does is implement a more realistic drag model.

[Sirrobert]

If you are not using FAR try to keep it below terminal velocity, you can Also get a mod called Procedural Fairings to make your rocket more aerodynamic (Almost a requirement with FAR)

I'm pritty sure fairings don't do anything for aerodynamics in stock.

[LethalDose]

In stability of the rockets during launch may not be the only problems you'll have with what you're working on. I don't see any batteries on those rockets, and maybe a Gigantor solar array on the last one (the station hub with 2 cupolas?).

As an alternate recommendation, you may want to consider starting in Career mode to learn the basics rocket science from the ground up. Career mode starts you with small, simple parts, and unlocks more as you gain science. The science mechanics aren't very intuitive, but they're fine after you get the hang of them.

[Sirrobert]

In stability of the rockets during launch may not be the only problems you'll have with what you're working on. I don't see any batteries on those rockets, and maybe a Gigantor solar array on the last one (the station hub with 2 cupolas?).

As an alternate recommendation, you may want to consider starting in Career mode to learn the basics rocket science from the ground up. Career mode starts you with small, simple parts, and unlocks more as you gain science. The science mechanics aren't very intuitive, but they're fine after you get the hang of them.

There is a battery on the adeptor. More than enough to power the rocket to orbit, where the gigantor can be deployed

[cantab]

Try being cautious with the gravity turn. Make it slowly keeping your heading on or near the edge of the prograde circle, instead of beaning it over to 45 degrees. If that doesn't work, try also making it later than normal, reducing the drag forces.

[Taki117]

I'm not using FAR. Is it recommended?

-TC

Recommended yes, but it adds a unique challenge to the game that most players aren't ready for. By completely changing the Drag model that KSP uses making spaceplanes and rockets more difficult to fly if they are not aerodynamic.

[FleshJeb]

I think there's some good solid advice in this thread.

I can only add that you should always turn off thrust vectoring on your side boosters. They can really magnify latent instabilities.

Likewise, the location of your command unit makes a big difference. If you have a slightly wobbly upper stage, and you're trying to command from there, the stabilizers will try to correct the flight based on that wobble rather than the rocket as a whole. I always put a probe on my second-to-last stage and fly from there. It's typically more stable, and it also allows you to deorbit that stage.

Lastly, the lighter payload is going to result in a much higher TWR. As you near burnout of any particular stage, you're going to have a lot of thrust at the end of a long empty tube, and a heavy weight on top. Lots of turning torque + very low moment of inertia = welcome to flippy land.

[michaelhester07]

I would suggest a better strut technique. I picked this up in another thread here: Engineers love triangles. Strut to make triangles:

Boosters to the main body: 3 struts in zig-zag down each side of the booster

Main body to playload: 8 struts (4x placement) to make zigzag pattern between decoupler and payload. To reduce on-orbit part count make sure you strut from the launch system to the payload. The struts will unload when your launch system is out of range and you'll have fewer parts total in the end.