RemoteTech - How many sats are really needed for interplanetary connections?

[Cairol]

My question is in the title.

How many sats are really needed for interplanetary connections with full coverage?

This guy here says, you should place 3 polar orbital sats for interplanetary connections.

For example to cover a Duna transfer from Kerbin.

In my Imagination I think additionally to the 3 sats in aequatorial orbit around Kerbin,

just 2 more medium/long range sats in polar orbit at 180 degrees to another will suffice for full coverage of another planet like Duna.

Because even in worst case when Kerbin is blocking LoS from another planet to one of the polar orbit sats,

the second sat will be visible as Kerbin only blocks a small angle from such a large distance.

Is this correct, or am I missing some use case here?

[hendrack]

I only have only one geostationary satellite with long range dishes and it works fine. I used to send satellites to my destinations too but now I build a command ship with 6 kerbals and make sure it has LoS. Though I used hyperedit to make it properly geostationary after putting them into an approximate orbit.

Javascript is disabled. View full album

Edited July 4, 2015 by hendrack

[Alshain]

Technically speaking, if you want Full Uninterrupted Coverage you need quite a few. 6 per planet minimum, and then a solar network in a polar orbit to ensure the sun and other planets do not block line of sight to Kerbin. Realistically though, 3 per planet should be enough to get most areas of the planet and the sun and other planets do not block LoS for very long (relatively speaking) so that network can wait till late in the game.

[maltesh]

Technically speaking, if you want Full Uninterrupted Coverage you need quite a few. 6 per planet minimum, and then a solar network in a polar orbit to ensure the sun and other planets do not block line of sight to Kerbin. Realistically though, 3 per planet should be enough to get most areas of the planet and the sun and other planets do not block LoS for very long (relatively speaking) so that network can wait till late in the game.

Actually, the Draim Tetrahedral Configuration can do it with four satellites in specific positions in inclined, elliptical orbits.

All four of the satellites maintain line of sight to each of others at all times, as well as providing full surface coverage.

I wound up writing up a challenge to set up a minimum-size Draim Tetrahedron over Kerbin a few years ago, and spent some time writing a computer program to simulate the surface coverage of one earlier this year. As it turns out, for the minimal Draim, you'll need communications links of about 11,000 km to maintain connections between the satellites.

The navigational tools KSP provides, however, are not very good for setting up your own Draim Tetrahedron, and the three-polar, three-equatorial satellite solution is definitely a lot easier to fly in KSP.

Edited July 5, 2015 by maltesh

[Cairol]

Short interruptions due to planets / moons / sun in the way of LoS are acceptable for me right now.

What I want is to not lose connections due to planetary rotation (or current satellite angle when not in synchronous orbit).

And therefore I guess two polar orbital sats will be enough?

As Alshain said, the total coverage is more for the late game.

I've just started out with RT and I am currently setting up my first interplanetary connection.

So it does not have to be perfect. Just decently reliable to work with.

I just wanted to know in advance how many satellites I have to bring for connecting a neigboring planet like Duna appropriately.

[RainDreamer]

Actually, the Draim Tetrahedral Configuration can do it with four satellites in specific positions in inclined, elliptical orbits.

The paper is behind a paywall so I can't read the details. Do you have a tutorial for replicating the configuration in KSP?

[John FX]

You could put two on highly elliptical polar orbits (100kmx40Mm) so they take days to orbit kerbin then one keostationary one feeding them. Mun and Minmus would not get in the way when they are so high.

The problem I find is that often the burn you need to do to capture at the other end needs to be on the far side of the planet...

[maltesh]

The paper is behind a paywall so I can't read the details. Do you have a tutorial for replicating the configuration in KSP?

To be honest, I've never actually done it without guideposts. When I fly to place a Draim configuration, my typical tactic is to edit the persistence file to place targets at the desired points, and then I fly my spacecraft to rendezvous with them. In the Draim configuration, all four of the satellites are in orbits with inclination 33°, and eccentricity 0.28, and a semimajor axis that gives a periapse high enough for the satellite to do its part in covering its section of the planet; Draim's parameters for Earth were these:

The minmum 27-hour requirement over Earth is how he specified the semimajor axis for his orbit. This works out so that, for any particular world, the minimum semimajor axis for the Draim configuration is about 7.25 times the radius of the world. As such, these were the parameters I picked for the challenge, which are a little larger than the minimal Draim configuration over Kerbin.

[table=width: 800]

[tr]

[td][/td]

[td]Sat 0[/td]

[td]Sat 1[/td]

[td]Sat 2[/td]

[td]Sat 3[/td]

[/tr]

[tr]

[td]Semi-Major Axis (SMA)[/td]

[td]4350000 m[/td]

[td]4350000 m[/td]

[td]4350000 m[/td]

[td]4350000 m[/td]

[/tr]

[tr]

[td]Eccentricity (ECC)[/td]

[td]0.28[/td]

[td]0.28[/td]

[td]0.28[/td]

[td]0.28[/td]

[/tr]

[tr]

[td]Inclination (INC)[/td]

[td]33°[/td]

[td]33°[/td]

[td]33°[/td]

[td]33°[/td]

[/tr]

[tr]

[td]Argument of Periapsis (LPE)[/td]

[td]270°[/td]

[td]90°[/td]

[td]270°[/td]

[td]90°[/td]

[/tr]

[tr]

[td]Longitude of the Ascending Node (LAN)[/td]

[td]0°[/td]

[td]90°[/td]

[td]180°[/td]

[td]270°[/td]

[/tr]

[tr]

[td]Mean Anomaly at Epoch UT=0.0 (MNA)[/td]

[td]0[/td]

[td]-1.57078[/td]

[td]3.14159[/td]

[td]1.57078[/td]

[/tr]

[tr]

[td]Epoch (EPH)[/td]

[td]0.0[/td]

[td]0.0[/td]

[td]0.0[/td]

[td]0.0[/td]

[/tr]

[/table]

Viewed from a point perpendicular to the orbital plane, each of the orbits has this shape:

And the video below has a run of the simulator that I bashed together back in march to show someone that the Draim Configuration does do full-surface coverage.

The satellites were placed using the above parameters. It is probably best viewed on YouTube, in full-screen mode, where you can see the numbers in the table.

In the big upper square, we assume a non-rotating Kerbin for simplicity, and using a plate-carree projection. Each of the thin horizontal and vertical lines represent 30° of longitude or latitude. The secondary vertical lines are at 90° intervals, and the big thick lines represent the equator and Prime meridian for the simulation.

The numbers on the big grid represent the point directly under each of the four satellites on this non-rotating Kerbin. 0 is Cyan, 1 is Orange/Yellow, 2 is Magenta, and 3 is White.

The colored contours around each satellite represent contours of how far above the horizon the satellite appears in increments of 10°. Right below the satellite is the number, the next contour has the satellite appear 80° above the horizon, then 70°, and all the way down to the thick contour, which is the limit on the surface where the satellite can be seen.

Below that is a table. Latitude, longitude, and altitude are the surface coordinates of the satellites.

Horizon is the radius, measured along the surface, of the region of Kerbin each satellite can see.

Surface % is the fraction of Kerbin's total surface that each of the satellites can see at any one time.

Sat0 Link, Sat1 Link, Sat2 Link, and Sat3 link is sort of a split table.

The upper right triangle shows the length of the communications link between each pair of the satellites, with the stipulation that none of the links are allowed to pass closer than 600km from the center of Kerbin (because then Kerbin would be in the way).

The Lower right triangle shows the link path. As it turns out, Kerbin never gets in the way of the link between any pair of the satellites.

Finally the lower three views are animations of the satellites viewed from above the X-Z plane, the X-Y plane, and the Y-Z plane, The reason they don't look elliptical is because all the orbits are at angles relative to the planes of view.

Edited July 5, 2015 by maltesh

[Alshain]

Short interruptions due to planets / moons / sun in the way of LoS are acceptable for me right now.

What I want is to not lose connections due to planetary rotation (or current satellite angle when not in synchronous orbit).

And therefore I guess two polar orbital sats will be enough?

As Alshain said, the total coverage is more for the late game.

I've just started out with RT and I am currently setting up my first interplanetary connection.

So it does not have to be perfect. Just decently reliable to work with.

I just wanted to know in advance how many satellites I have to bring for connecting a neigboring planet like Duna appropriately.

Just be careful with which antenna you choose for which planet. Make sure you understand the cone mechanics. You can't just use the biggest antenna at every planet. Eve and Duna for example are far too close for the really big antennae, as the cone will not cover ideal orbits of Kerbin.

[jbakes]

What does LoS mean?

[Stone Blue]

^^

Line of Sight... A straight line between two points, basically with no solid obstructions between, like planets, moons, or sun...

Edited November 4, 2015 by Stone Blue

[Van Disaster]

Last time I ran a RT game I stuck one sat over KSC, and one enormous relay station in a polar orbit somewhat outside Mun's. The intention was to use two relay stations because obviously there's occasional interruptions, but in the end due to the way the game went that ended up being unnecessary ( not least because even if the GSO relay couldn't see the main relay, I had a network at Mun by then which would relay just fine ).