Perfect Orbit?

[WeirdCulture]

I am fiddeling with some orbit and docking maneuvers.. and throu this I try to get the perfect orbit... but I can't archive the same apoapsis and periapsis . I just get about 99% never 100%. Is it able to get the full 100%? so that apoapsis and periapsis are the same?

[painking]

Getting an orbit to be -exactly- circular is damn near impossible, even mechjeb can't do it. Getting them to within tens of meters is usually the most anyone will bother with, even with geostationary orbits since they'll still have their orbital period to measure by. You could try with xenon engines, but even then, the orbit will probably have some tiny tiny amount of eccentricity.

[krenshala]

Theoretically, yes, you can achieve a perfectly circular orbit.

In practice, once you get the difference it down to tens of meters (instead of hundreds or thousands) then it normally isn't worth the time and effort needed to refine it further.

[zarakon]

you can, but it will change slightly as soon as you timewarp or even if you rotate your craft. Probably something to do with the center of mass not quite moving exactly right

[Boffinbox]

you can get very very close. i'm under the impression that you cannot get 0 eccentricity, but if you get close enough to be within rounding errors like 100,000.4 and 99,999.6 for example, that would loook like it's perfect

[Frederf]

You can get exactly right if you text edit the save file. It looks horrible to get too close as your nodes will go mad. It's better to not. If you need an exact synchronization you want to look at period aka semi-major axis.

[HeadHunter67]

The numbers don't have to be the same - eccentricity is what matters. If you have an orbit that's 100km altitude, that's a circle that's 1400km across (Kerbal has a radius of 600km, according to the KSP wiki). Even if the periapsis is 98.6km and the apoapsis is 101.4 km, that's still essentially a perfect circle for orbital purposes, as each is only one tenth of a percent off nominal.

Yes, you can get it closer - an exceptional pilot or MJ user could do so. But there's not much reason to seek a "perfect" orbit. Even real spacecraft seldom do so.

[AncientAstronaut]

I have done it. I don't have the screenshot anymore because it was like .16 or earlier, but it was pretty exciting. The moment I snapped the picture, the orbit was to the meter.

[WeirdCulture]

You mean that a 99.76% orbit is close enough for a randevouz?

[Superfluous J]

You don't need to make the orbits the same to rendezvous. All you need to do is set the target, get your closest approach down to a KM or so, and then when you get to that closest approach burn so that your speed relative to the target is almost 0. Boom, you'll be in almost the exact same orbit and then you can thrust your way slowly to the target.

[blizzy78]

For rendezvous, you might want to check out my tutorial (see signature.)

[SuperWeegee4000]

I wonder if Scott Manley could do it.

[blizzy78]

He sure can, but I'd guess he doesn't care. What's the advantage of having a perfectly circular orbit anyway?

[DaRocketCat]

It's possible to get it within one meter if you use a single ant engine, and some tiny bursts.

[Lheim]

It's possible to get it within one meter if you use a single ant engine, and some tiny bursts.

Or fine-controlled RCS. But that's beside the point, surely - isn't every orbit mathematically suppose to be an ellipse?

[Pugspaceprogram]

You can do it and it is very easy, if you are quite patient. All you need is a few xenon tanks and 1ion engine, and lots of energy, perhaps 4 Radioisotope Thermoelectric Generators should do it, alongside 2 big-ass solar panels

[Javster]

You mean that a 99.76% orbit is close enough for a randevouz?

It's spelled rendezvous. From French.

Anyway, yes, it is possible, but not worth the bother.

[RoboRay]

there's not much reason to seek a "perfect" orbit. Even real spacecraft seldom do so.

They never do so. Even satellites in GSO have a "box" they drift back and forth in, rather than remaining absolutely stationary in respect to the surface from having the impossible-to-attain perfectly circular orbit.

Even if you do get your orbit circular to the meter, you are still eccentric by centimeters. If you get it down to the centimeter, you are still eccentric by millimeters. You will not attain a perfectly circular orbit, and there is no reason to attempt to get one. Even NASA calls it circular if the Ap and Pe are within a couple of kilometers of each other.

Save yourself a lot of headaches and wasted time and don't worry about trying to get a "perfect" orbit.

[Tiron]

They never do so. Even satellites in GSO have a "box" they drift back and forth in, rather than remaining absolutely stationary in respect to the surface from having the impossible-to-attain perfectly circular orbit.

Even if you do get your orbit circular to the meter, you are still eccentric by centimeters. If you get it down to the centimeter, you are still eccentric by millimeters. You will not attain a perfectly circular orbit, and there is no reason to attempt to get one. Even NASA calls it circular if the Ap and Pe are within a couple of kilometers of each other.

Save yourself a lot of headaches and wasted time and don't worry about trying to get a "perfect" orbit.

NASA also deals with a lot larger orbits, given that Earth has about 10 times the radius of Kerbin to start with, and the atmosphere goes higher too, so a 'couple of KM' is a lot less of a difference in the real world than in KSP.

The problem is that achieving a perfectly circular orbit is impossible in KSP, not just because of the difficulty of the task, but because of the same rounding errors in PhysX 2 that caused the Space Kraken. There's small-scale instability in orbital parameters, which is mostly noticeable when going in or out of time warp...or when you've got a nearly perfectly circular orbit. Your AP and PE get very unstable when you've got extremely low eccentricity, so much so that ANY control input will alter them. And even without input they'll shift around a bit. The closer you get to perfectly circular, the more unstable it gets.

[TheTriniFlyer]

Also in Real Life even if you have an orbit - circular to the millimeter, it will eventually become perturbed over time as:

1a. The Universe has n-body physics (refer to 1b)

1b. Over time gravitational influences from the Sun, Moon, Jupiter... will perturb the orbit.

2. Even at 500> km there is a incredibly sparse but ( talking on the millimeter scale ) non-negligible atmosphere which will decay your orbit by millimeters then centimeters then meters...etc (you get the point)

3. Solar Particles like photons will cause an (extremely) small perturbations by radiation pressure...

I'm done talking now as you get the point.

*no scientific belittling was intended by this post*

Edited August 6, 2013 by TheTriniFlyer
Spelling

[Kimberly]

Adding to TheTriniFlyer's point: the current official definition of a planet requires that it has "cleared the neighborhood" of minor objects, i.e. that there is no such thing as a stable orbit around the planet, at least on an astronomical time scale.

[HeadHunter67]

Also, gravity isn't a constant at any altitude around any celestial body. Varying planetary density in different regions results in a slight change in gravity - meaning the orbital speed increases or decreases as it passes over these areas. So you can see that this will eventually, albeit slowly, influence an orbit.

[Tripzter]

Its a waste of time. To make it worse a tiny little shake in the vessel will throw around your ap and pe around the orbit like mad. IMO you want ~1k difference so you dont have to deal with the ap/pe shake as much.

That said having a 1km difference between the ap/pe is less than a .1m/s as far as docking goes.

Edited August 6, 2013 by Tripzter

[strongest_2hu]

If you actually want perfect orbits for your GPS constellation orbit or whatever, orbits that won't ever get out of synch and cause you to lose coverage in places, your best bet is to edit your save file and then never load those craft from the space center or get within 2.25 km of them ever again, as they'll get thrown out of their perfect orbits when the physics loads.

[maltesh]

Adding to TheTriniFlyer's point: the current official definition of a planet requires that it has "cleared the neighborhood" of minor objects, i.e. that there is no such thing as a stable orbit around the planet, at least on an astronomical time scale.

That's not what "cleared the neighborhood of its orbit" requires. It means that the object is, by a wide margin, the most massive object in the region of its orbit around the sun. It does not prevent the object from having long-term satellites; if it did, Earth would not be considered a planet.