Encounter and orbit insert in one go

[Mythos]

I thought moving straight into orbit wasn't possible. Don't mind the polar orbit I tried to match, which didn't work out. So I tried to get closer by some gravity assist. After some fumbling I ended in this trajectory. That's obviously a patched conics only feature.

[Dangerous_Beans]

i've never heard of it, but i suppose it's possible. a gravity assist uses the gravity of a body to accelerate you in some direction, so you must have just lined everything up to have it pull you into a stable orbit.

impressive!

did it actually happen, or did enough rounding errors creep in to mess it up?

[regex]

It's an error, the Mun's gravity isn't enough to do that.

[Mythos]

did it actually happen, or did enough rounding errors creep in to mess it up?

I kept the plan like it was shown. Other missions will happen in the meantime, I will wait and see.

- - - Updated - - -

I backed up my save and tested. It was stable until and after the maneuver.

But mid course it changed drastically.

[Eric S]

But mid course it changed drastically.

Not surprising, my first instinct was that it was a flaw in the patched conics prediction.

Basically, the reason it shouldn't work (and won't, except in cases where the simulation isn't as accurate as the math underlying the simulation), is because of the basics of orbital mechanics. Baring some outside force, you will arrive at the same altitude as you started at, and with an upward velocity equal to the downward velocity you started at.

[Superfluous J]

While mathematically this is impossible, KSP iterates trough time so it becomes possible in the game. It was far more possible before 1.0 when time warping through an SOI would totally obliterate your path compared to the predicted one.

[samstarman5]

Should have left it at the first and kept the possibility real! Now you ruined everything!

[Eric S]

Hang on fellas... If your periapsis just makes it to the edge of the Mun's SOI, ahead of the Mun's position, couldn't the resulting gravity drag pull you into a stable orbit? Also, didn't I hear somewhere that planets and stars can capture rogue bodies in this way IRL?

Not really. With patched conics, if something enters an SoI with a given speed, it's going to exit the SoI with the same speed if it's only affected by gravity. Gravity slingshots don't change the orbital energy of the slingshottee within the SoI of the slingshotting celestial body, just the parent.

Planets and stars capture rogue bodies mostly due to interactions with other bodies (N-body physics, gravitational braking using other celestial bodies orbiting the parent body, etc.) which won't happen in this situation. An asteroid getting captured around Kerbin due to interaction with the Mun or Minmus could happen, however.

[FancyMouse]

Hang on fellas... If your periapsis just makes it to the edge of the Mun's SOI, ahead of the Mun's position, couldn't the resulting gravity drag pull you into a stable orbit? Also, didn't I hear somewhere that planets and stars can capture rogue bodies in this way IRL?

Nah not possible - KSP has neither n-body nor oblation, which implies that your orbit will be solid stable around a given body. If you enter a body, without thrust you'll for sure leave, except for numeric errors (from simulation and/or floating point arithmetics)

Gravity drag/assist is not relevant - e.g. if you flyby Mun, you're using the gravity of Mun to change your velocity w.r.t. Kerbin. If the topic is about orbiting Mun, then there's no such thing as gravity drag. It's just simple 2-body orbit mechanics here.

[Reddeyfish]

Hang on fellas... If your periapsis just makes it to the edge of the Mun's SOI, ahead of the Mun's position, couldn't the resulting gravity drag pull you into a stable orbit? Also, didn't I hear somewhere that planets and stars can capture rogue bodies in this way IRL?

With a SOI system, it is completely impossible.

When inside a sphere of influence, only the main body of that sphere matters. All other bodies are completely ignored. Therefore, you can think of a sphere of influence as if it was a single body floating on it's own, so gravity acts as a conservative force for the spacecraft with respect to the body. If you have 100 m/s at 100,000 km on the way in, you have 100 m/s at 100,000 km on the way out.

Now, KSP doesn't simulate SOI exactly, because of things like rounding errors (as mentioned above), so it is possible (but rare) in KSP for it to happen.

EDIT: and ninja'd

Edited July 8, 2015 by Reddeyfish

[Merinsan]

Not really. With patched conics, if something enters an SoI with a given speed, it's going to exit the SoI with the same speed if it's only affected by gravity. Gravity slingshots don't change the orbital energy of the slingshottee within the SoI of the slingshotting celestial body, just the parent.

Planets and stars capture rogue bodies mostly due to interactions with other bodies (N-body physics, gravitational braking using other celestial bodies orbiting the parent body, etc.) which won't happen in this situation. An asteroid getting captured around Kerbin due to interaction with the Mun or Minmus could happen, however.

Any asteroid captured around Kerbin, due to Mun/Minmus interaction, is not in a stable orbit, because it can eventually re-encounter that body and be subsequently ejected. Although, it can be a really easy way to capture an asteroid permanently, since it's a lot easier to rendezvous with, and then adjust it's orbit.

[Superfluous J]

Hang on fellas... If your periapsis just makes it to the edge of the Mun's SOI, ahead of the Mun's position, couldn't the resulting gravity drag pull you into a stable orbit? Also, didn't I hear somewhere that planets and stars can capture rogue bodies in this way IRL?

KSP:Reality::The Ocean:Thursday

Yes, this is how planets are captured (and sometimes ejected) IRL but in KSP the Mun's SOI is a perfect sphere centered on a perfect point-source mass. You will orbit Mun in a perfect ellipse or hyperbola (depending on your velocity). If you enter Mun's SOI you are (by definition) going downward at a certain velocity at the exact distance from Mun that its SOI extends. If you trace that ellipse or Hyperbola to the other side where you're back at the height of Mun's SOI, you are going UP at that exact same velocity. And you escape Mun's SOI.

That this occasionally doesn't happen isn't due to my math being wrong. It's due to the granularity of the simulation on very rare occasions mucking up perfection.

[Xyphos]

That's an error.

I have learned long ago, never ever ever EVER trust the prediction system, just plan your nodes one step at a time and manual burn while keeping a close watch.

[Mythos]

never ever ever EVER trust the prediction system, just plan your nodes one step at a time

I do use multiple maneuver nodes planned ahead, IF it looks reasonable. This one didn't, although it was just one node. As always, you have to know what you're doing and not just trust the instruments.

I totally agree with this beeing physically impossible and the thread went as expected, but it was a funny thing to mention

[Lukaszenko]

It should be impossible because when entering a body's SOI you are, by definition, going at escape velocity or faster. Therefore, you should exit at escape velocity (or faster) as well.

This is also why when an asteroid or something from outer space hits earth, it will ALWAYS hit us at least at 11.2 km/s (25,000 mph).

[Sharpy]

If anything, impossibility of this is a misfeature of patched conics. Planets capture new moons commonly. The Moon captures asteroids from time to time, though they never last as the orbit is destabilized by tidal forces of Earth. It's definitely possible in real life.

[Caelib]

You can typically do it with a body large enough and you have to perform a gravity brake -- this means passing in FRONT of it during your encounter, which means your orbit will be "backwards". More info:

http://wiki.kerbalspaceprogram.com/wiki/Tutorial:_Gravity_Assist#Reverse_Gravity_assist

Edited July 8, 2015 by Caelib

[Red Iron Crown]

It's rounding errors, you can't theoretically use a gravity assist to capture around the body you're using for the gravity assist in a patched conic model. You can only use a child body to slingshot a capture around the parent body. E.g. You can use a Tylo slingshot to capture around Jool but not around Tylo.

[Lukaszenko]

If anything, impossibility of this is a misfeature of patched conics. Planets capture new moons commonly. The Moon captures asteroids from time to time, though they never last as the orbit is destabilized by tidal forces of Earth. It's definitely possible in real life.

Is this true? I heard of planets/ moons being captured, but I understood it was precisely DUE to disturbances from tidal and other (such as a 3rd body) forces.

Edited July 8, 2015 by Lukaszenko

[BurnPrograde]

I had this same orbit a few months ago when burning to encounter Ike. I called it a "lasso orbit"

[Eric S]

Any asteroid captured around Kerbin, due to Mun/Minmus interaction, is not in a stable orbit, because it can eventually re-encounter that body and be subsequently ejected. Although, it can be a really easy way to capture an asteroid permanently, since it's a lot easier to rendezvous with, and then adjust it's orbit.

True, but the same can be said for real life captures as well, to varying degrees. Even things parked in L4/L5 aren't permanent, just stable for millions of years. I tend think of the permanence of a capture as a non-boolean quality of a capture, not a boolean requirement.

[Violent Jeb]

Is this true? I heard of planets/ moons being captured, but I understood it was precisely DUE to disturbances from tidal and other (such as a 3rd body) forces.

[Superfluous J]

Is this true? I heard of planets/ moons being captured, but I understood it was precisely DUE to disturbances from tidal and other (such as a 3rd body) forces.

I cannot stress enough (or often enough, apparently) that we are NOT talking about real life here. We are talking about the game, which mimics real life but does not even approach its complexity.

In real life, yes. Things get captured and ejected all the time. In Kerbal Space Program's patched conics system, it should mathematically never happen but rounding errors and the granularity of the universe (both in time and space) can cause it if you're really lucky, unlucky, or persistent.