Achieve accurate orbit with Xenon drives

[Tibrogargan]

I'm trying to build a comms array with small relay satelites with 3 satellites spaced evenly around the equator of a moon.  Getting the satellites into their initial position is easy enough (followed the tutorial), but the burn time is so long using xenon drives that the orbits end up being quite off center.  I could do this with multiple small Hohmann transfers every time I hit AP - but it requires quite a few burns on multiple satellites - it ends up being very tedious.

Is there a way I can plan my burns so I don't have to do quite so many?

 

[bewing]

Pick one of the sats as your reference.

Decide whether you want to move the other sats forward or backward in their orbits.

Adjust their orbital periods, so that they either gain or lose a few seconds (or a fraction of a second) every orbit on the reference sat.

Wait many orbits.

 

[Spricigo]

There is many ways to setup a relay network, we need to know better how you are doing it to offer better advice.

@bewing's advice rely on the fact that for many satellites formations getting the orbital period right is the key factor and the system will work just fine even if other parameter are a bit of (e.g. in a triangular formation two satellites end 125º apart instead of the 'ideal' 120º00'00") .Unless your satellites are near the limit of their range or it's a very complex formation (e.g. Draim Tetrahedron)  you probably don't need to worry about the precision of the other orbital parameter.

Personally I don't consider the Dawn as an option for relays. Packing all the necessary deltaV  in a chemical alternative tend to be more convenient (cheaper, better TWR, requires less EC,...). A higher deltaV budget of the ion alternative can be an advantage for future corrections but I'd rather solve that kind of issue with redundancy (MOAR relays).

 

 

[Starman4308]

Once you have the relative inclination down to where you want, take a look at the relative phase angle. If your satellite is too far ahead, burn a little bit prograde, wait until the phase angle is where you want, then burn retrograde at periapsis to get orbital periods to match. If it's behind, burn a little retrograde, etc.

Creating a communications network is a little bit of a shadow rendezvous, a fair bit of patience, and then tuning orbital periods very finely.

Also, if you're sending a batch of relays on the same launch, one trick I'll often do is separate one at a time, and put them into resonant orbits with the main satellite. If I'm going for a triangle, I might put one into a 5/6 orbit, wait two orbits, and it'll be 120 degrees ahead of the launch bus, and put the other into a 7/6 orbit, wait two orbits, and it'll be 120 degrees behind. MechJeb can set up these maneuver nodes for you, otherwise you can hand-calculate periods and semi-major axes and Hohmann transfers.

[foamyesque]

If you're willing to throw dV away, you can do these all in one burn, but a better choice is probably going to be grabbing a small chemical engine and tank -- Ant, Spider, maybe Twitch or Spark depending on the size of your satellites -- and use their higher TWR. With small objects and low dVs, which it sounds like is true in this application, they can wind up being more mass efficient than ions are since you can seriously cut down on your electrical infrastructure and they have better tank MRs and engine TWRs.

Edited January 19, 2018 by foamyesque