Target centre of mass giving me incorrect information.

[Vereox]

So I've received a contract to redirect a class D asteroid to Gilly.

I successfully redirected it around Kerbin to some inclined orbit (I simply just captured it) 200 km Pe and 70 Mm Ap. And then I changed the orbit to a 150 km orbit around Kerbin.

I had no problems with redirecting it. The centre of mass was targeted perfectly fine and the onboard RCS and reaction wheels allowed for minor errors in targeting the centre of mass.

However, once I assembled my Gilly transfer vehicle and attached it to the asteroid facing the centre of mass, it spins out when I activate the engine.

In the album below, there are screenshots showing the alignment to the target of centre of mass before and after I connect to the asteroid. After I connect, the vehicle is not pointing through the centre of mass anymore. So I freed the pivot on the grabbing unit and rotated my vehicle to correct the direction its facing. But as you can tell, it looks so wrong. It doesn't look like it's pointing through the centre of mass. And the final screenshot is of when I've activated the engine and the asteroid has begun to spin because of the force not being applied towards the centre of mass.

Javascript is disabled. View full album

I haven't done anything different in terms of targeting the centre of mass, the probe bodies that I use to control the vehicle are in line. So I'm completely stumped as to what is going wrong.

Can anyone help me out with this?

[RainDreamer]

I think the problem is not with not being able to line up your center of mass, but the problem is that you are hooking up big rockets to an even bigger rock with only that tiny claw. It is not structurally stable. The rocket push and the claw would bend and everything get thrown out of balance. You either gotta deal with that, adding addtional anchoring structure using more claws+docking ports, or make a tug instead of a pusher. Things are easier to move when you drag things with you than pushing them.

[Taki117]

I think the issue is that while you are aligned with the center of mass, it's off center from where you grabbed it. I suggest unclawing, moving around the asteroid a bit to line up more inline with the center of mass, and try again.

[Vereox]

I think the problem is not with not being able to line up your center of mass, but the problem is that you are hooking up big rockets to an even bigger rock with only that tiny claw. It is not structurally stable. The rocket push and the claw would bend and everything get thrown out of balance. You either gotta deal with that, adding addtional anchoring structure using more claws+docking ports, or make a tug instead of a pusher. Things are easier to move when you drag things with you than pushing them.

Interesting, I didn't think of pulling it. A question: Does KSP simulate force if an object is behind an engine (directly in line with the engine's exhaust)?

I think the issue is that while you are aligned with the center of mass, it's off center from where you grabbed it. I suggest unclawing, moving around the asteroid a bit to line up more inline with the center of mass, and try again.

I've tried that. It's a very tedious task to do such a thing when the game's engine slows right down when dealing with a craft like this. :S

[Jovus]

One possibility: did you detach the other craft before attaching this one? If not, the center of mass you were targeting was of the asteroid and capture craft, which would put you off when the capture craft detaches.

[RainDreamer]

Interesting, I didn't think of pulling it. A question: Does KSP simulate force if an object is behind an engine (directly in line with the engine's exhaust)?

Yes. It would cancel out if you have the exhaust pointing right at the asteroid as you try to pull it. Gotta leave some space.

[Cirocco]

Yes. It would cancel out if you have the exhaust pointing right at the asteroid as you try to pull it. Gotta leave some space.

yup, blocking the exhaust means damaged structure and no thrust. As RainDreamer pointed out, one solution is to make the pulling vessel either long enough (with engines near the front and grabber on the back) so that the effect is nullified, or wide enough so that your engines fire next to the asteroid. Of course that last solution is kinda hard with E-class asteroids.

A third, and in my opinion quite elegant solution is to tilt your engines outwards. That way they fire next to the asteroid without the need for making a super-wide ship. Note that this does put extra stress on your structure and reduces the efficiency of your engines because part of the thrust goes to forces squeezing the ship together. The loss is related to the angle: the wider the angle, the more loss. For reference: you'll only have 50% efficiency if your engines are angled 45° outwards.

[Kryxal]

No, your engines will have approx. 71% efficiency at a 45 degree angle. It's actually cos A, where A is the angle between the direction of intended travel and the direction of each engine. You may see this sort of thing referred to as cosine losses, along with some similar things.