Physics question about torque and thrust

[VoidPointer]

I've been wondering about the physics of a certain situation for a while now, but I don't know if the answer I have is right.

So, suppose I have a rocket whose center of mass is not in line with the vector of thrust, due to an unbalanced load or whatever. I know quite well that this creates a torque on the rocket. Now, suppose I slap on enough control surfaces or SAS units to counteract this torque and keep the rocket pointed in one direction while I fire it. My question is this: Will this have any impact on efficiency? I can't visualize the free-body diagram of the forces and torques well enough to figure out if the whole force of thrust will act in the proper direction through the center of mass without losing any energy in the process (electric charge notwithstanding). I think it will work fine, but torque is one of those things I never really understood that well.

Edited September 5, 2014 by VoidPointer

[Kasuha]

No, it will have no impact on efficiency.

[Red Iron Crown]

I think the control surfaces would have a tiny impact on efficiency, they have more drag when not edge-on to direction of airflow. Torque wheels shouldn't, though.

[Scarecrow88]

No, it will have no impact on efficiency.

So you're saying that increasing the mass of his rocket with the parts required to negate the problem off his offset thrust will have no effect on it's efficiency. Okay - I wish I could fly all my rockets in your world.

[Kasuha]

Okay, added mass will of course have (negative) effect on efficiency, be it reaction wheels or control surfaces. I believe that's not what OP was asking about.

Conservation of momentum is the key phrase here. As long as you send your fuel in one direction giving it certain momentum, you will gain the same momentum in the other direction regardless whether that thrust vector goes through your CoM or not.

Edit: just so you believe me I made an experiment. Created a ship with three engines, two on sides and one at the center. I hyperedited it to 100 km orbit, quicksaved, then activated side engine and burnt all fuel prograde. Then I quickloaded, activated the middle engine and burnt all fuel prograde. Resulting orbits were the same within reasonable accuracy - Apoapsis at 1931 vs 1937 km. So if you don't think it's true, go and try it.

Javascript is disabled. View full album

Edited September 5, 2014 by Kasuha

[Kasuha]

It looks like you simply spread out the thrust, leaving the COT through the middle of the ship... OP's question was about putting all the thrust off center (ie: 2 engines on one side, none on the other) - then stopping a ridiculous spin by massive SAS or wings.

Only one engine was running in both cases, you can see it in the picture. The ship was built symmetrical (i.e. with two side engines) to make sure the central engine's thrust aims at CoM.

[undercoveryankee]

In atmosphere, if your method of compensating for offset thrust is through an aerodynamic surface producing a lift force in some direction, that lift should induce additional drag compared to the same aerodynamic surface running at zero lift. I say "should" because stock aero may not model this.

If you have gimbaling engines, the steady-state position will be with the engines gimbaled so the thrust axis passes through the center of mass. Inefficiency may arise in atmo if gimbaling takes you too far from your lowest-drag orientation, or in space if your thrust axis ends up too far from your control axis and you find yourself chasing the node to correct for an off-axis burn.

If you correct using RCS, then you're expending RCS fuel that doesn't contribute to your delta-v (or might, positively or negatively, depending on your thruster placement). If you need that RCS fuel later, or if your RCS shares fuel with engines that you do rely on for delta-v, you might end up with a problem.

Correcting with reaction wheel torque just requires electric charge, which is usually a replaceable resource.

So basically, generating torque to balance out offset thrust doesn't cost efficiency (in the sense of available delta-v from the engine that needs to be balanced) unless it adds air drag or expends reaction mass in some other direction.

[Kryxal]

Of course, depending on just HOW off-center thrust is, you might have trouble holding course ... I had a rocket get just a BIT too low on takeoff from the Mun once, it ended up short a fuel tank and engine, and it was SUCH a pain to execute burns...

[VoidPointer]

I think the control surfaces would have a tiny impact on efficiency' date=' they have more drag when not edge-on to direction of airflow. Torque wheels shouldn't, though.[/quote']

Cool; that's what I thought. Of course, who knows what effect it would have with stock aerodynamics (which I still use because I haven't gotten the hang of FAR yet).

Okay, added mass will of course have (negative) effect on efficiency, be it reaction wheels or control surfaces. I believe that's not what OP was asking about.

Conservation of momentum is the key phrase here. As long as you send your fuel in one direction giving it certain momentum, you will gain the same momentum in the other direction regardless whether that thrust vector goes through your CoM or not.

I really should have thought of conservation of momentum. That makes it make a lot more sense! Thanks! I'll edit the category thing. And yes, I meant purely in terms of the forces involved - I know reaction wheels aren't massless.

[GoSlash27]

Reaction wheels essentially give you "free" thrust other than the added mass of carrying them, so they do not negatively impact efficiency in KSP beyond that. Surprisingly, control surfaces are in the same category. In theory, they *should* cost a drag penalty for correcting a torque moment, but in KSP physics they will actually add free thrust due to the infiniglider effect.

Best,

-Slashy

[rhoark]

With a control surface, you are expending energy to accelerate air particles. A gyro is deferring rather than eliminating the need to expend energy. Thrusting off axis will eventually saturate the gyro, and some other source of torque will have to be applied to bring it back into operation.

[GoSlash27]

With a control surface, you are expending energy to accelerate air particles. A gyro is deferring rather than eliminating the need to expend energy. Thrusting off axis will eventually saturate the gyro, and some other source of torque will have to be applied to bring it back into operation.

These are both true in real life, but they don't apply to KSP.

Best,

-Slashy

[Kasuha]

I can confirm (made an experiment) that control surface that is not oriented directly along the airflow has indeed bigger drag even in KSP. Although the difference is very small. Dropping two SRBs with differently oriented control surfaces on tails (one straight, other inclined by 45 degrees) from 35 km altitude resulted in about 500 m distance between them on impact.

KSP reaction wheels, though, never saturate. That does not have a parallel in real world but it is convenient for game purposes.