Lagrange points?

[azazel1024]

Anyone know if the L points exist in KSP and if so what the location/math for them is based on scale and gravity? Oh, and especially how you'd calculate them based on the impact of Minimums (which, pardon the pun, I assume would be minimal impact due to its much greater orbital diameter and smaller size/gravity compared to the Mun).

Just thinking about in the future trying to setup a station and/or relay satellites in a few of the L points

Thanks!

[Cykyrios]

L points are not simulated in KSP, since the way gravity is simulated uses only 2 bodies: the spacecraft and the planet/moon/star. Maybe a special way to implement them exists, but it is not planned as of now (nor is switching to n-body, AFAIK).

You can still find their theoretical position using the formula of gravity: F = G m/r² where m is the mass of the body you are orbiting, and r is the distance to the center of gravity (altitude + radius).

[karolus10]

Lagrangian points aren't existing in KSP, because of patched conics system specification (and for making Lagrangian points simulation possible we need major changes in game, not friendly for game performance anyway),

Also Lagrangian points aren't too useful for KSP, because orbits around Lagrangian points are generally unstable (they only places when gravity forces of two bodies are equal, so they counteracting each other) ,Even small Deviation from L points center made orbit unstable, so you need surgical precision ,even for putting small satellites .

[maltesh]

Lagrange points aren't places where the gravity forces between the two bodies are equal hey're places in which the gravity of both the larger and the smaller object provive gravitation in the proper direction to make at least limited stable orbits possible in 1:1 resonance.

L1 (a point between the two objects), L2 (a point farther out from the central body than the satellite), and L3 (a point on the opposite side of the sattelite's orbit of the central body, a little farther out than the satellite) are only stable in limited directions, so objects tend not to stay there.

L4 (60 degrees ahead of the satellite) and L5 (60 degrees behind) provide stability in all directions, so those can wind up occupied naturally. There are over five thousand known asteroids occupying the L4 and L5 points of Jupiter's orbit, and two of Saturn's moons have smaller moons occupying the L4 and L5 points in their orbits (Tethys has Telesto and Calypso, Dione has Helene and Polydeuces).

Edited August 29, 2012 by maltesh

[illectro]

Lagrangian points aren't existing in KSP, because of patched conics system specification (and for making Lagrangian points simulation possible we need major changes in game, not friendly for game performance anyway),

Also Lagrangian points aren't too useful for KSP, because orbits around Lagrangian points are generally unstable (they only places when gravity forces of two bodies are equal, so they counteracting each other) ,Even small Deviation from L points center made orbit unstable, so you need surgical precision ,even for putting small satellites .

Please stop spreading the myth that proper orbital mechanics are somehow hard to simulate because of performance issues! The extra number crunching required is relatively trivial compared to things like the graphics and simulation or inter-part interactions of a rocket.

It is simply that it would take developer time and there are many more things that developers could be doing that would affect a lot more people. The number of players who legitimately have interesting things to do with lagrange points is tiny, from a simple cost-benefit analysis devs are better to look elsewhere.

Anyway, to clarify the nature of Lagrange points - they are places where the restricted 2 body system produces exact solutions to the equations of motion which can be solved analytically, these are places where the gravitation forces exactly balance the rotational psuedo forces and allow massless particles to follow orbital solutions that are exact.

The main uses that we find in reality aren't that interesting. The comsat people would probably love to put a satellite in a halo orbit around the point on the far side of the mun, but that's not stable. L4/L5 are stable, but not that interesting from a utility purpose. And while you can use them for low energy transfer orbits, such orbits take a really long time and very few people will have the patience and skill to exploit weak stability boundaries (AKA the interplanetary super highway).

[azazel1024]

Well, outside of the boundries of the game, the advantages of using the L points would be to place a satellite at L3 and at L4 or L5 and with 2 satellites you could have LOS communications to all points of the Moon, I mean the Mun, with the exceptions of deep craters in a few spots and a few spots on the poles that might render the satellite below the local horizon also due to terrain features. For the unstable L points, yes they are unstable, but it also doesn't require a significant amount of station keeping. A low powered ion engine would work great for long term positioning on one of those L points.

You'd basically either have LOS directly from the main body to the satellite or you'd have the L4/5 to L3 relay to the far side of the satellite.

Anyway, thanks for answering my question on whether they existed in the game or not.

[maltesh]

Using L4 and L5 as relays for munar communications would double the light-lag time. You're better off having three sats in the Munar orbit high enough above the surface that each one has LOS to the other two.

[azazel1024]

Depends. You would have larger lag, but you'd also have greater cost involved in placing those satellites. In the context of the game (with no current implemented cost/resource structure), at least additional time required to place them in combination with addional objects for the game to have to track and for the Remotetech mod, additional required antennas on the satellites. Current lag, with Remotetech mod at least, is low enough that a doubling of the lag is still on the order of being insignificant.

[maltesh]

There's no real extra time or cost involved.. You propose two satellites at the distance of the Mun (L4, and L5), and a third satellite farther out than the Mun (which would be L2, not L3), that the game can't currently model.

Easier and quicker to just put three satellites in a circular orbit just inside the Munar SOI, separated by 120 degrees in their orbits, and ignore the lagrange points entirely.

[NovaSilisko]

The single most important reason is this: An n-body system makes it impossible to easily predict what's going to happen ahead of time. With the two-body system in place now, you can set yourself on an interplanetary trajectory, then observe the SoI shifts hundreds of days in advance. With an n-body system, however, you would have to simulate the entire path ahead of time, instead of simply doing a bit of math to predict where you'll be. Also, it renders timewarp basically impossible but that's another story.

As such, it's not going to happen, ever. The advantages of the patched conics system are far too great.

[silent_prtoagonist]

Lagrange points are really outside the scope of KSP. But if you're really ready to start talking about building lagrange point space stations, then you're probably ready to make the jump to Orbiter. It has pseudo-n-body physics, which accurately simulate lagrange points, in addition to allowing other interesting things like sun-synchronous orbits and solar sails. It also, given that it has docking support, is currently a much better simulator for building space stations to begin with (at least until docking makes it into KSP).

There are also numerous full n-body simulators out there that are fun to play around with. Universe Sandbox (which is pretty, but costs money) and Gravity Simulator (which is ugly but free) come to mind...

EDIT: In retrospect that really sounded like an add...I promise I don't work for Martin Schweiger

Edited August 30, 2012 by silent_prtoagonist

[illectro]

Universe sandbox just uses a game physics engine to perform its simulation, it's terribly innacurate and prone to producing unphysical behavior

Proper orbital integrators can simulate an orbit at rates of thousands of orbits per second. So yes nova is right of we wanted a proper n-body system to replace patched conics we'd have to spend 0.001 seconds to figure out if an orbit would intersect another SOI, and the game would have to limit time acceleration in the worst case to x1,000,000

Edit: to get an idea of a good general purpose orbit solver look at 'mercury' which is a hybrid symplectic integrator. In layman's terms this is like patched conics but with perturbations appled Along the orbit, and when it enters a transition state from one soi to another it uses a general purpose algorithm. The real problem to me is that no scientific computer algorithms are written in c many are still in FORTRAN.....-

Edited August 30, 2012 by illectro

[azazel1024]

My proposition was to place a satellite at only one of the L4/5 points and at L3. That would be one less satellite that you'd need than 3 at 120 degrees opposition.

Its not really that big a deal, but just something I had been thinking of. Later with actual planets thrown in, I could also see how the L points could be useful for interplanetary travel. You could have them at, for example, the Murs/Kerbal L4 and L5 points for refueling, so if you were going further outsystem and were not planning on using gravity assist, you could potentially use stations there. The stations could (if added in) use a combination of ion engine station keeping drives to counter act the thrust of a railgun to accelerate fuel tanks with some RCS and stability control systems to outbound flights that would pass nearby for refueling. Murs could refuel and resupply the stations. It would increase, potentially, the time window for missions to the outer planets to enable resupply to them enroute for higher energy burns and more direct travel than having to use gravity assist for the extra velocity.

[maltesh]

One sat at L4 or L5 and one sat at L3 means that a sixth of the Munar surface is hidden from both satellites. You need at least three sats to attempt full coverage of a spherical body. (Six is the minimum to guarantee it.)

Lagrange points aren't good spotsfor fuel depots . If you're planning to head to a specific destination on launch, anything you could do with refueling tanks that you have to rendezvous with at several kilometers per second relative (And they will have a relative velocity to you unless you happen to be running right through the lagrange point, a very very rare occurance) is something you probably should have done with drop tanks at your departure point.

[MrFailSauce]

The main uses that we find in reality aren't that interesting...And while you can use them for low energy transfer orbits, such orbits take a really long time and very few people will have the patience and skill to exploit weak stability boundaries (AKA the interplanetary super highway).

Not sure that I agree. The ITN(Interplanetary Transfer Network) seems like the main reason to include Lagrangian mechanics. The transfer times aren't that much longer than the amount of time that people spend waiting around for hohmann transfer window, and they would allow extremely interesting craft. For example, the genesis mission used ITN to get to L1 and back in 2.5 years(much of which was normal mission operation, not transfer) and propulsion was only 5% of their mass! Imagine the possibilities for interesting missions when your fuel requirements are so low. And yes, the math will be hard for players, but there is no reason to think that people won't develop interesting tools to help with this in the same way that they built the hohmann transfer calculator.

[Tex_NL]

Did you even bother to check the date on this? It is OVER a year old!

[FEichinger]

Okay, guys, I get the whole confusion, what with August not being all that far back, but ... Look at the year please, will ya.

Closed for excessive use of Necromancy. The Mages Council does not approve of Necromancy.

FEichinger