Pluto and Charon

[Whirligig Girl]

Eeloo needs a BIG moon. Pluto and Charon are in fact a double-dwarf-planet system, NOT just a planet and moon. Obviously KSP can't actually simulate barycenters very well, or Ike wouldn't orbit Duna, they'd orbit each other. Still, Eeloo very much needs a really big moon.

Maybe, and this brings a bit of a sick stomach to me as well, Ike should be moved to Eeloo.

Maybe Ike should be smallinized so that it's a moon of Duna and not a double planet, and another body entirely given to Eeloo.

Or maybe Ike and Duna and Eeloo and "Charon" should BOTH be sets of double planets.

Thoughts?

[p1t1o]

Well it depends on this: "...can't actually simulate barycenters very well"

[bv1]

Would this kind of situation even be possible with the patched conics approximation?

[Whirligig Girl]

I figured out that it is. You just have to make the two SOIs of Eeloo and "Kosma"(idk, Charon analog name?) touch exactly, and they each orbit around a hidden Barycentre gravity source. The trick is that you can't get close to the Barycentre without being in the SOI of Eeloo or Kosma. This WOULD, however, mean going away from the normal mathematical formula for SOIs, you'd need to fine-tune the SOIs so that they work in this special way. I'm pretty sure the game can handle this, but it would somewhat screw with the perfect implementation of the Patched Conics system.

[bv1]

Something like this? The green dot is the barycenter (KSP would consider it to be a surfaceless "planet"), and the red and blue dots are planets (KSP would consider them to be "moons").

There are two big problems with this hack. First of all, for unequal double planets, the larger and smaller planets must have identical orbital periods about the "barycenter." However, KSP would think that the larger body, being closer to the barycenter, should be moving faster and have a smaller orbital period. Thus, they couldn't have standard orbital elements, and would have to be programmed in to move along special kinds of rails.

Secondly, the barycenter in this situation would be the most stable position in the system, while in reality, it should be unstable.

[Whirligig Girl]

No, the green dot has to be within the SOI of the other two bodies so that you never get close to it.

[Bill Phil]

Well, Eeloo is going to be turned into a moon of GP2, so it doesn't matter.

Although the new Pluto analog could use ghost gravity wells. Same for Lagrange points, you can make ghost gravity wells to orbit round.

[Whirligig Girl]

Well, Eeloo is going to be turned into a moon of GP2, so it doesn't matter.

Although the new Pluto analog could use ghost gravity wells. Same for Lagrange points, you can make ghost gravity wells to orbit round.

I don't think GP2 is coming before 1.0. NovaSilisko has left the team, and at any rate there'd still be another Pluto if Eeloo is changed.

[Sigma88]

Since I know GregorxMun tried (and liked, I hope) my version of the barycenter I'm gonna shamelessly promot it here

(just click into my signature to see it)

Basically I put pluto in orbit around "the barycenter" and charon in orbit around pluto.

This allowed me to recreate what I think is the best feature of that binary system:

The system is also unusual among planetary systems in that each is tidally locked to the other: Charon always presents the same face to Pluto, and Pluto always presents the same face to Charon: from any position on either body, the other is always at the same position in the sky, or always obscured.

Source

click on the photo to see the animated gif:

[tntristan12]

I'm still not convinced the invisible gravity source thing could work, but I do recall Chaos Forge pitching a good idea for barycenters in general. It's linked in his sig.

[Eric S]

Well, Eeloo is going to be turned into a moon of GP2, so it doesn't matter.

Although the new Pluto analog could use ghost gravity wells. Same for Lagrange points, you can make ghost gravity wells to orbit round.

Ghost gravity wells have issues and really don't behave much like Lagrange points. It's been hashed out several times in the past.

Not so sadly, the bug that would best demonstrate what's wrong with ghost gravity wells is a lot less common than it used to be. Long time players probably remember falling through a planet's surface, coming too close to the center of the gravity well, and having floating point rounding errors or other integrator issues basically throwing them out with enough velocity to manage a solar escape trajectory.

[Sigma88]

Ghost gravity wells have issues and really don't behave much like Lagrange points. It's been hashed out several times in the past.

Not so sadly, the bug that would best demonstrate what's wrong with ghost gravity wells is a lot less common than it used to be. Long time players probably remember falling through a planet's surface, coming too close to the center of the gravity well, and having floating point rounding errors or other integrator issues basically throwing them out with enough velocity to manage a solar escape trajectory.

This problem is avoided if the ghost gravity well is inside another body SoI. This would not work for lagrange points of course.

For binary systems tho, it works quite well since once you enter the SoI of the main body, the game completely ignores the baricenter. I tested this using a big barycenter and while orbiting the planet I just passed through the barycenter without any problems.

I haven't received any complaints about the system yet, but I don't think many people downloaded my files

[Eric S]

This problem is avoided if the ghost gravity well is inside another body SoI. This would not work for lagrange points of course.

Oh, that's what GregroxMun meant. I couldn't figure out how the barycenter could be in both planetary SoIs, I think he just meant it had to be in one of them, and only now do I realize why. Just to be sure, you mean that the barycenter itself, not its SoI needs to be in one of the planetary SoIs with that planetary SoI taking priority.

OK, objection lifted due to missing the point, and nothing else stands out as obviously a problem to me. Out of curiosity, are the SoIs hard coded? I always assumed they were, but I'd never heard definitively.

I haven't received any complaints about the system yet, but I don't think many people downloaded my files

I'll have to try it out.

Edited March 10, 2015 by Eric S
it's is the contraction, not the possesive form.

[Sigma88]

Just to be sure, you mean that the barycenter itself, not its SoI needs to be in one of the planetary SoIs with that planetary SoI taking priority.

Exactly

in my current version of the pluto analogue the barycenter has a huge SoI like IRL

(calculations may be not correct yet, but that's because it's still WorkInProgress, the general idea is ok tho)

the barycenter itself is 1 meter wide, and since it's always between pluto and charon it is always inside Pluto's SoI, making the "ghost body" itself irrelevant

Then I put charon in orbit around pluto in order to let you pass from pluto's SoI to charon SoI

I'll have to try it out.

Hope you like it

[Bill Phil]

Ghost gravity wells have issues and really don't behave much like Lagrange points. It's been hashed out several times in the past.

Not so sadly, the bug that would best demonstrate what's wrong with ghost gravity wells is a lot less common than it used to be. Long time players probably remember falling through a planet's surface, coming too close to the center of the gravity well, and having floating point rounding errors or other integrator issues basically throwing them out with enough velocity to manage a solar escape trajectory.

That's more of a problem with KSP...

Anyhow, lagrange points kind of act like gravity wells. Kind of...

Maybe it could be treated as a completely different kind of object? With no gravity...?

[Renegrade]

Or maybe Ike and Duna and Eeloo and "Charon" should BOTH be sets of double planets.

There's no reason to not have multiple double-planet systems. The Kerbol (or whatever) system is kinda .. sparse .. at the moment. One thing I'd like to say though is that I'd like to see the smaller moons of the Pluto/Charon system modeled as well. However, I'd also like to note that Eeloo is nowhere near Pluto's distance - Pluto's SemiMajorAxis is 39.264 Earth AUs, and Eeloo's SMA is 6.626 Kerbin AUs*. Eeloo is actually closer than Saturn, relatively speaking. That's especially interesting as Duna and Jool are roughly where their real-life analogs are.

As for the 'external barycenter' issue, I have to say I'd rather not see rude hacks (er, any further rude hacks...) become part of stock. I really don't think patched conics can come even remotely close to handling that properly - it seems like it would be an n-body problem to me.

* - using Squad rounding, that is 41 AUs and -NaN AUs respectively.

[Sigma88]

As for the 'external barycenter' issue, I have to say I'd rather not see rude hacks (er, any further rude hacks...) become part of stock.

I think squad put "more planet" on the no ask list, so no reason to "worry"