RemoteTech2: KEO constellation or LKO constellation?

[NASAHireMe]

Recently began playing with RemoteTech2 after a new career on 1.0.2. I haven't unlocked enough of the tech tree yet to get to the longest-range omnidirectional antenna, the Communotron 32. So I'm stuck with the Communotron 16 (omnidirectional, range: 2500 km) and the Comms DTS-M1 (directional antenna, range: 50,000 km).

Keosynchronous orbit (KEO) is at 2868 km altitude, beyond the range of the Communotron 16, but well within the range of the DTS-M1.

Question is, is it worth it to build a constellation of 3 KEO satellites at this stage of the tech tree? I lack the communication systems to go interplanetary, so the 3 KEO satellites would have to be replaced or augmented once I unlocked the more powerful directional dishes.

I think it'd be better to build a constellation of 5 or 6 LKO satellites with omnis instead, since most launches and orbital assemblies/rendezvous occur at the 70-150km altitudes, which place them beyond the range of the KEO omnidirectional Communotron 16s.

[Geschosskopf]

I did a tutorial on this: http://forum.kerbalspaceprogram.com/threads/88879-Beginner-s-Notes-on-Remote-Tech-2-in-Career

I recommend holding off on building the network until you've got a dish that can reach outside Kerbin's SOI. Otherwise, you'll just have to do it again later once you have those dishes.

[Eric S]

It can actually be done using three satellites in LKO. I usually aim for an altitude of about 600km. They have enough range until they get up to about 800km, and have line of sight with each other down to about 450km or so. 600km puts it near the middle so that satellites drifting doesn't cause a problem too fast.

[Snark]

My own approach is typically to go for LKO initially. 4 satellites can easily handle it, as long as they're synchronous to each other.

Regarding Geschosskopf's future-proofing suggestion: As long as the LKO satellites have DTS-M1 on them, it's easy to add interplanetary capability later. Just add a single "big relay" satellite in a polar orbit at big radius (say, 30,000 km). It points its DTS-M1 at Kerbin; the LKO's each point a DTS-M1 at the relay. Then the relay does the interplanetary communication. In a polar 30,000km orbit, there's no significant problem with Kerbin occluding the signal; one relay sat is sufficient.

One pragmatic reason why I like this all-the-eggs-in-one-basket approach to interplanetary networks: I find that RemoteTech is crash-prone when there are a whole lot of links present; the game will just suddenly crash when switching between ships, or between a ship and KSC, or even at KSC when switching between the overview and a building. The problem seems to get much worse as the number of links goes up. By sticking to just one big relay per planet, it turns the network into an O(N) system instead of an O(N^2) system and there are a lot fewer links around-- I get fewer crashes.

[42_aerobrake_passes]

The reason geosynchronous orbits are used in real life is because a fixed location on the ground will always be able to point in the same direction to be pointing at a satellite. In KSP+RT, you're never going to be having something like that on Kerbin, making synchronous orbits pointless (outside of RP or "it's cool" reasons - if you're Doing It Wrong because you want to do it wrong, you're Doing It Right). In theory it might be useful if you wanted to use a dish on a ground base on another body and not have to retarget it, though even there I'd prefer an omni with satellites in a low enough orbit to reach it.

[NASAHireMe]

Thanks Geschosskopf, I read that thread. It seems that the RT players and their threads on the forums are few and far between

My goal is to obtain full coverage with as few satellites as possible, without the Comm32 omni (which I haven't unlocked yet). Costs are a secondary matter since funds (even on Hard) are relatively plentiful, so this was more of an intellectual exercise than a practical one.

My main complaint with a KEO system is that crafts in LKO (say, an unmanned emergency rescue pod at 100km) need DTS M-1s to reach the KEO satellites. This means they have to have lots of solar panels (right now, I haven't unlocked anything more advanced than the basic single panel). A second annoyance is that although you can set the DTS M-1 on the KEO satellite to 'active vessel', you can't do the same thing with the LKO craft. As the LKO craft flies around Kerbin, it will eventually lose contact with the KEO satellite that its DTS M-1 is pointing at. This means that you have to either (1): retarget the DTS M-1 every third of an orbit around Kerbin (annoying), or (2): put multiple DTS M-1s on your LKO craft, meaning you have to add even MORE solar panels and batteries.

^This problem could be solved if RT could be set to automatically re-route signals to Mission Control. For example, when the LKO craft's connection to KEO1 is blocked by Kerbin, the antenna automatically retargets KEO2 to reestablish connection with Mission Control.

For those reasons, I was curious about the minimum constellations needed to provide full coverage, with maximum future extensability. I'm sure NASA went through this exercise themselves in the 1960s. It's actually pretty simple geometry, so here goes:

1. Minimum altitude for a 3-satellite system with full coverage of Kerbin, assuming the sats are spaced in an equilateral triangle = 600km. Any system lower than 600km needs four satellites in a square; lower than 248.6km needs five satellites in a pentagon.

2. Maximum altitude for an equilateral 3-satellite system with a maximum distance of 2500km to KSC on the ground = 2145km. Any 3-sat system placed higher than 2145km altitude will occasionally be unable to reach KSC with the Comm16 omni.

3. Minimum altitude for an equilateral 3-sat system with a maximum distance of 2500km to a KEO sat = impossible! With a 3-sat system, it is impossible to maintain a continuous omnidirectional connection with a KEO sat. For approximately 83.4 degrees of orbit (or roughly 23% of each orbit), the connection will be lost.

4. Maximum altitude for an equilateral 3-sat system with a maximum distance of 2500km to the other two satellites in the system: 843.38km. If you system is placed at an altitude higher than 843.38km, at least one of the three satellites in the system will always be connected to KSC, but your link to your other two satellites are broken.

5. Maximum altitude for a second 3-satellite system with a maximum distance of 2500km to one of your lower satellites at the maximum altitude of 843.38km: 2286.8km. If your second satellite system is above this height, it won't have a continuous connection with one of the lower three satellites.

Therefore, the magic range for a 3-sat system with full coverage is 600-843.38km. Within this range, your system will be fully connected both to itself, to KSC, and to any satellites up to 2286.8km altitude with 0% downtime. No need to go KEO anymore!

Best practice would be to launch your three-sat system with docking ports so that new antennas, panels, and batteries can be added as the tech tree is unlocked.

Edited May 14, 2015 by NASAHireMe

[Scientist]

i prefer draim tetrahedron, works perfect, 4 sats, 100% coverage even at poles

i usually do it fairly early (as i start with probes not kerbals) to get kerbin coverage, then later when i get the tech il toss up a station in polar orbit which will upgrade its dishes as needed (usually making modules through EPL+MKS right on the station) to handle long range comms to other bodies

with that setup i havent had issues, but the long range comms arent 100% active, closer to 90% if right orbit, there might be better orbits but i never cared to try, worked enough, though if needed, a 2nd sat with a polar orbit 90 deg diff would raise it to 100%