How do I evenly space out my communications satellites?

[godefroi]

After (much) trial and (even more) error, I've managed to follow the RemoteTech tutorial for a four-satellite network. At the end of the tutorial, it suggests that you might want to edit the save file to align the semi-major axes in order to get perfect orbital periods for them.

Me, being OCD, I decided that I'd just hyperedit them into absolute perfection. I zeroed out everything I could zero out, set the SMA to a nice round number, and now, the only problem is, how do I calculate the necessary mean anomaly at epoch in order to space them out evenly along the orbit?

I'm OK if someone just happens to know this number and doesn't want to bother with the math. It's interesting, but I'm trying to calm my disorder here. I think it's somewhere in the vicinity of 1.45.

[Taki117]

Your first sat should be at 0, the next one depends on how many SATs you have. For a constellation of 4 they should be 90 degrees apart. Also make sure that the Epoch is the same for all the SATs.

[Nathair]

I find it almost trivial to throw up a set now. To place three satellites into three hour orbits (1585.2km) in a nice triangular formation I just put all three onto one launch vehicle. From launch get your apoapsis to the target 1585.2km and then raise your periapsis to 550km, that puts you into a nice two hour orbit. Then every time you hit apoapsis, drop a satellite and circularize it to a three hour orbit. The two to three ratio puts them into a nice formation and as long as your orbital periods are correct you don't need to sweat getting the apsides exact. Minimal skull sweat.

[godefroi]

Your first sat should be at 0, the next one depends on how many SATs you have. For a constellation of 4 they should be 90 degrees apart. Also make sure that the Epoch is the same for all the SATs.

Uh, yeah, I can divide a circle into four pieces, but the "mean anomaly at epoch" isn't in degress.

Hm. Reading the wikipedia entry again, it looks like it goes from 0 to 2pi each orbit. That means it's closer to 1.57.

Putting that in, however, it's clear that's still not right. Grr.

Edited March 9, 2015 by godefroi

[AlexinTokyo]

If the mean anomaly at epoch is not in degrees then it's most likely in radians.

This being the case the angle is exactly À/2 or approximately 1.5707963267948966.

[Taki117]

Uh, yeah, I can divide a circle into four pieces, but the "mean anomaly at epoch" isn't in degress.

Hm. Reading the wikipedia entry again, it looks like it goes from 0 to 2pi each orbit. That means it's closer to 1.57.

I might have mislabeled things. It's the Longitude of Periapsis I was referring with a Mean Anomaly of 0. This way your satellites will start at the same position in their orbit (Periapsis I believe) but will be spaced relative to each other.

I point you to an awesome tutorial type thing I made a while back.

Edited March 9, 2015 by Taki117

[superm18]

The Ap/Pe is not important, it's the orbital period you wanna get right.

This works for any orbital height, just make sure your orbital period is the same and the sat's will never drift apart. (tip: at KEO (2868 km) the orbital period is 6 h)

[ilo]

You might want to take a deep look (and breath) to the visual remotetech planner tool. As mentioned by @supermun18, you don't really need to match ap/pe that perfectly, just make sure the orbital period (in msecs) is as close as possible: they sats will get closer and far during the orbit, but will keep their average distance stable. So I recommend you to use the smallest engine available and also reduce the thrust while adjusting the orbit.

You must also take care of how much battery do you have in your sats. The bigger the orbit, the MORE time in the shade because of body eclipses, the BIGGER battery required. Additionally, the higher the orbit (the bigger the orbital period) the less deviation you will get because of different satellites speeds: if the difference is one second between two of the satellites, they will completely get out of their triangular shape at 1sec every orbit.. so if 1 hour (3600 secs) is 360 degrees, that means that 1 second of difference is 0.1 degrees: In 10 orbits (10 secs) the satellites will be 1 degree de-synched. Try to find somewhere in between that gives you some pace, you will have to balance.

For kerbin 'ground' control, you just need 3 sats with a communotron 16 antenna at 200km orbit, but that will be too hard to maintain because of this same issue..

Btw, have you ever put a satellite in geosynchronus orbit in the mun?.. I did!! (i tis not in a geosynch orbit as it is not possible in the mun, but the mun relay sat stays in the same position relative to the mun during its whole orbit, and is orbiting kerbin).

There is a satellite in the same orbit as the mun separated by 2.600km, enough to reach the mun sats using a communotron 16 antena..

These are my sats in the mun, they don't need a directional antenna because this is the purpose of the satellite in the mun orbit.

I can have full coverage in the mun surface for a deflectron DP-10 antena.. with only C-16 antennas.

Javascript is disabled. View full album

And yes, only used two geosynchronous satellites in kerbin pointing to the mun, enough to reach the space center 24h.

Edited March 9, 2015 by ilo
error in the battery description

[mudkest]

http://www.ksp.interrutt.com/?OrbitalBody=4&Orbit=700&Ratio=2%2F3

in this case you can have a ship with 3 satalites, put it in a 700ap/84.1pe orbit and release 1 satalite(and circulize its orbit) at the AP,

or(bit less acurate) have 1 satalite in 700km orbit, get the 2nd one in 84.1 then make a node to intercept the satalite as close as you can, then circulize its orbit then 2nd time you're at pe(first time being next to the other satalite)

[godefroi]

If the mean anomaly at epoch is not in degrees then it's most likely in radians.

This being the case the angle is exactly À/2 or approximately 1.5707963267948966.

You'd think, but that's not right either.

[StevenLawyer]

I've created an Excel spreadsheet that will calculate even spacing for satellites.  You can find it here:  Satellite Spacing Calculator