I thought this was impossible...

[Spaceisbeautifulul]

Err well, uhm. Lets let the pictures speak for themselves shall we?

Is this a bug, or did I just prove its possible, or what?

Unfortunately... I got there, and it didn't get me an orbit like the map view said.

Only this:

So... whats up with that?

Thanks in advance.

[Vanamonde]

It's impossible in reality, but there is a tiny degree of imprecision in the game's simulation, and the farther away you are, the more room there is for small inaccuracies to have large results.

[Mr Shifty]

It's impossible in reality, but there is a tiny degree of imprecision in the game's simulation, and the farther away you are, the more room there is for small inaccuracies to have large results.

I think it might be more accurate to say that it's possible in reality, but not in the game because of patched conics. Simply put, once you enter a body's SOI in the game, you're guaranteed to leave it again because of how orbits work. (Specifically, you can't orbit around and return to your entry point/velocity without leaving the SOI.) IRL with a continuous gravity gradient, you could be captured into an elliptical orbit.

[nhnifong]

It is actually possible to enter an orbit like that in KSP without thrusting. If you see the maneuver nodes predict it, pass through the SOI change at a relatively high time warp (to maximize floating point errors) and if you are lucky you will end up in orbit. But to be clear, it's most definitely a bug, although probably not an easy one to fix.

[stupid_chris]

Actually, in real life, if you have no other influence than a planet's SOI, it's impossible to get capture in orbit, you will always be ejected. Satellites can be captured by a planet (most often the gas giants) when they pass by one of the already present's moon SOI and get slowed down by their gravity. A collision can also do the job. But it's impossible to have less than escape velocity when entering an SOI.

Another thing that could work is having a binary system. When it enters a new SOI, then because of gravitational disturbance, one body aborbs a large part of the energy of it's binary partner and get's ejected while the energy abosre allows the other body to stay in orbit around the newly entered SOI. But no matter what, there needs to be a way to expel energy if you want to be captured.

[MR4Y]

Actually, in real life, if you have no other influence than a planet's SOI, it's impossible to get capture in orbit, you will always be ejected. Satellites can be captured by a planet (most often the gas giants) when they pass by one of the already present's moon SOI and get slowed down by their gravity. A collision can also do the job. But it's impossible to have less than escape velocity when entering an SOI.

Another thing that could work is having a binary system. When it enters a new SOI, then because of gravitational disturbance, one body aborbs a large part of the energy of it's binary partner and get's ejected while the energy abosre allows the other body to stay in orbit around the newly entered SOI. But no matter what, there needs to be a way to expel energy if you want to be captured.

I thought that was possible because of Lagrange points?

[stupid_chris]

I thought that was possible because of Lagrange points?

Under restricted conditions, yes, but I started my statement by saying that there should be no other influence than the planet's SOI. A Larangian point automatically means there is a new SOI that interacts with the system.

There would be two possibilities then:

If we talk about the parent to the planet you are flying by, then it's possible only if the Larange point is within the planet's SOI. For example, with Earth this wouldn't be possible because the interesting Larange points here, L1 and L2, are both at 1.5million km from Earth while it's SOI only extends to about 925000km. But given the planet was dense enough, then it would be possible.

If we're talking about a Larange point with a moon, then it directly implies another body assisting the capture again. But that would be possible that by passing through a Larange point you could slow down enough relatively to a planet to be captured.

[nhnifong]

Suppose two objects enter the SOI of a system from opposite sides. They're moving at just slightly over the escape velocity of the system, but now that these two objects have been added, the mass of the system has increased just enough that the escape velocity has increased above the velocity of the entering objects. They both end up in orbit right?

[stupid_chris]

Hmmm I'm not sure adding a satellite changes the escape velocity of a system. That would need to be checked I'm not 100% sure, but it makes sense I guess. And if so, to significantly change the escape velocity, those objects would have to be really massive, and probably that the gravitational waves they induce with the planet would have a much bigger effect than the escape velocity change.

However, again this requires a third extra influence. Natural captures are possible, it's just hard to get the required conditions. And for example, a natural capture around Venus or Mercury would be near impossible because they don't have any current moon. The most likely way it would happen would be that a binary ateroid/comet would enter the SOI and leave on in orbit while the other one is expelled at high speeds. Or by a lucky influence with the Sun.