Sun-Synchronous Orbits in KSP- are they possible? How?

[Northstar1989]

I've thinking about deploying a satellite to a satellite to a sun-synchronous orbit around Kerbin (there are NOT to be confused with heliocentric orbits- the orbit still lies within Kerbin's Sphere of Influence)

http://en.wikipedia.org/wiki/Sun-synchronous_orbit

My reason why is irrelevant to the discussion, though it's to deploy a KSP-Interstellar Microwave Beamed Power Transmitter operating off solar power (moderators, please DO NOT move this thread to the 'Add-Ons' section- this is a discussion about orbits, not mods...)

What I am curious is, is it even possible to deploy a satellite to a Kerbin orbit such that it is NEVER eclipsed by Kerbin?

Some preliminary calculations of mine indicate that a satellite placed at the absolute limit of Kerbin's Sphere of Influence (84 159 286 m from its center, or 83 559 286 m above sea-level) would move East relative to the surface at such a speed when crossing the equatorial plane if its inclination were 73.709 degrees that its orbital period would be equal to Kerbin's sidereal orbital period (9 203 545 s, or 106 d 12 h 32 m 24.6 s- a Kerbin year) if it were able to orbit on the equatorial plane at that speed- however this is only at the equatorial plane, and the component of velocity in the direction of Kerbin's rotation increases to a maximum at the greatest height above/below the equatorial plane of such an inclined orbit- so that's not all that helpful here...

I suspect that such a sun-synchronous orbit as I am looking for would lie outside Kerbin's Sphere of Influence, however I would like somebody to confirm this if possible...

Regards,

Northstar

P.S. I am aware that there is no precession (rotation of orbits relative to the inertial frame) in KSP. I am trying to figure out whether it is possible to have an orbit that always stays at the same position relative to the Terminator (the line separating day/night) with all motion relative to the surface constrained to being in a North/South direction...

Edited June 9, 2014 by Northstar1989

[Kasuha]

You can put it in an orbit that is in resonance with Kerbin's year. Something like that it will circle Kerbin exactly five or six times a year.

Then, you can adjust its inclination so that when it passes the equator, it is never in straight line between Kerbin and Sun.

[AlexinTokyo]

P.S. I am aware that there is no precession (rotation of orbits relative to the inertial frame) in KSP. I am trying to figure out whether it is possible to have an orbit that always stays at the same position relative to the Terminator (the line separating day/night) with all motion relative to the surface constrained to being in a North/South direction...

What you describe here is an inclined geosynchronous orbit, which is possible in KSP.

A sun-synchronous orbit, where (for example) the satellite ascends across the equator 12 times each sidereal day at the same local time each pass may be possible for Kerbin, but I haven't done the maths yet. Of course, as you say, there is no precession in KSP, so the orbit would fall out of local time sync over multiple orbits.

Edited June 9, 2014 by AlexinTokyo
Because I wasn't thinking straight.

[GoSlash27]

The only way I see that as a possibility is if it's out of Kerbin's SOI. It would have to be in the exact same kerboocentric orbit as Kerbin itself and just ahead or behind to minimize the distance. It won't get eclipsed by Kerbin, but it will get blocked by the Mun once a month.(edit) Unless, of course, you have 2 of them on opposite sides...(/edit)

Beyond that... maintaining zero angular velocity with respect to Kerbin means that it's angular velocity with respect to Kerbol is exactly the same as Kerbin's; same orbit.

Best,

-Slashy

Edited June 9, 2014 by GoSlash27
(addendum)

[Tommygun]

I've done this by accident several times while launching a 90 degree polar orbit.

I think I was launching about a half hour or so before sunset.

It never seemed to last more than a week or two before slowly going behind Kerbin.

I don't think you could get the precision down to do it permanently.

Even real satellites in Sun-Synchronous orbits also need to adjust their orbits from time to time to stay in the sun.

[cantab]

You could simulate the effect by making periodic stationkeeping burns over the poles to rotate your orbit around. With an ultra-high-Isp thruster this could be very practical.

[michaelhester07]

I don't think this is possible for KSP as its a 2 object orbit system instead of a multi-object system. The only way I would think to achieve this is to have a small constellation of satellites: An equatorial ring of 6 or so in geostationary orbit (which does exist in ksp) and two in polar orbit. If you could do small burns like the above suggestion to rotate it then that would keep it from being eclipsed. The 2 in polar orbit setup would be placed in perpendicular polar orbits so that one of them is always in sunlight.

[Northstar1989]

I don't think this is possible for KSP as its a 2 object orbit system instead of a multi-object system. The only way I would think to achieve this is to have a small constellation of satellites: An equatorial ring of 6 or so in geostationary orbit (which does exist in ksp) and two in polar orbit. If you could do small burns like the above suggestion to rotate it then that would keep it from being eclipsed. The 2 in polar orbit setup would be placed in perpendicular polar orbits so that one of them is always in sunlight.

I'm not trying to keep a portion of the satellites in a constellation in sunlight- I'm trying to keep *ALL* my satellites in sunlight. Pretty much any orbit with multiple well-spaced satellites in it will always have some in sunlight- with a ration depending on the Beta-angle. But what I'm trying to do is set up satellites that are *never* eclipsed (to maximize power output- my computer is rather slow, so it's better for me to have a few well-placed satellites rather than a bunch of not as well-placed ones...)

Anyways, I appreciate all the help/suggestions from people about constellations- but I'm already well aware how to do that. What I'm looking to do is to set up a single satellite in an orbit where it is never eclipsed, without going into a sun-centric orbit and without stationkeeping... In real life, it is possible because of precession from the Earth's oblateness, but I'm trying to figure out if it is at all possible in KSP...

Regards,

Northstar

[Taki117]

If you are inside the SoI of the body you are orbiting you will always be eclipsed by something. (This can be minimized but not avoided)

[Kasuha]

If you are inside the SoI of the body you are orbiting you will always be eclipsed by something. (This can be minimized but not avoided)

Of course it can be avoided. I placed my ship on approximately such eclipse avoiding polar orbit. I did not spend too much time fine-tuning its orbital period but I think the result is still decent. Kerbin is small from that distance, it's not hard to avoid its shadow. Here's how its solar oppositions looked like throughout the year:

[KvickFlygarn87]

I don't think that's possible. According to wikipedia, the planet has to be at least somewhat (can't find the real word) non-round.

[MeticulousMitch]

100 Kerbin days was the best I could manage. The SOI boundary seems to be quite unstable if you approach it very slowly.

[Wyrm]

While you can make an eclipse unlikely by choosing an orbit far from Kerbin, it will always appear to precess relative to Kerbin because the terminator itself drifts around Kerbin once every Kerbin year. The sun-synchronous orbit results from perturbation effects that are not simulated.

That said, the orbit solver in KSP is one of the cheapest things in the simulation. I estimate that finding the current position of each object in orbit comes under a hundred flops, even with one numerical root finding for each solution. Unless your computer is positively ancient, it should be able to easily handle a constellation of 20 satellites.

[cantab]

While you can make an eclipse unlikely by choosing an orbit far from Kerbin, it will always appear to precess relative to Kerbin because the terminator itself drifts around Kerbin once every Kerbin year.

As Kasuha mentioned and demonstrated, you can still make an eclipse impossible by making your orbital period a simple fraction of Kerbin's year, such that when the sun is on the satellite's orbital plane the satellite is well away from its sun-relative ascending/descending nodes.

Of course the inevitable small imperfection in orbital period may mess things up eventually, but you're talking many decades if not centuries I expect.

[Taki117]

Straight from the Wiki.

Sun-synchronous orbit

→ See also: Sun-synchronous orbit on Wikipedia

In the real world exists a sun-synchronous orbit, which isn't like a synchronous orbit around the Sun. Instead it describes an orbit around Earth which itself rotates, so it looks like the orbit stays the same relative to the Sun. As it requires an uneven gravitational field it is impossible to simulate in KSP.

[Northstar1989]

Of course it can be avoided. I placed my ship on approximately such eclipse avoiding polar orbit. I did not spend too much time fine-tuning its orbital period but I think the result is still decent. Kerbin is small from that distance, it's not hard to avoid its shadow. Here's how its solar oppositions looked like throughout the year:

http://i.imgur.com/FjZnKsI.png

http://i.imgur.com/8KlfzL2.png

http://i.imgur.com/bxRlCd9.png

http://i.imgur.com/mA6D2gE.png

http://i.imgur.com/xfbDN9R.png

How does it manage to avoid Kerbin's shadow? Is it in a specific resonance with the planet? Because according to my understanding of KSP's physics engine, otherwise it should still *eventually* end up in Kerbin's shadow at some point, at least for a few seconds every hundred years or so... (the orbit crosses directly on the other side of Kerbin from the sun on rare occasions, as I can see).

Regards,

Northstar

[Northstar1989]

I don't think that's possible. According to wikipedia, the planet has to be at least somewhat (can't find the real word) non-round.

The term is "oblate". And that's what I thought. Apparently the same reason sun-synchronous orbits around Venus aren't possible...

Regards,

Northstar

[Northstar1989]

As Kasuha mentioned and demonstrated, you can still make an eclipse impossible by making your orbital period a simple fraction of Kerbin's year, such that when the sun is on the satellite's orbital plane the satellite is well away from its sun-relative ascending/descending nodes.

Of course the inevitable small imperfection in orbital period may mess things up eventually, but you're talking many decades if not centuries I expect.

That might be my best bet. I'm looking for a theoretically-perfect orbit, I understand that slight imprecision may eventually mess it up. And I don't want to spend the fuel to go all the way to a polar orbit on the edge of Kerbin's SOI to effectively minimize eclipses to once ever few decades or so like Kasuha did... Can you demonstrate (with screenshots and math) what this would actually look like?

Regards,

Northstar