IF
I’m pretty sure with the thermal-tile system they have that Starship is already reusable with minimal refurbishment. It’ll be interesting to see the first time a reused Starship is re-entering. But before that, witnessing a 2nd stage tower catch attempt will be nuts. I predict they’ll attempt this in the next month. Probably could have done it sooner but the risk of flying over populated areas is not taken lightly.
I had the luxury of watching the entire Starship 14 flight. Seeing all 26 satellites deploy and then watching each stage land. It’s beyond amazing that this company started in 2002 and is now the premier launch system in the world. When Russia suggested they would abandon the Space station Musk stepped in and said Space X would handle orbital corrections. And then went on to do it. I still have trouble contemplating the simultaneous landing of 2 boosters. It’s hard to believe a vehicle can decelerate from 17,500 mph and stick a landing site that’s measured in inches of centerline. 700 successful flights of the Falcon 9. 11,131 satellites put in space. The newer Starlink V3 satellites are 10 times more powerful than their predecessors.
This is already beyond science fiction.
The “killer app” of space, the one people looked for for decades to support a robust commercial launch sector, turned out to be large constellations of satellites with multiples sharing an orbital plane, so that it made economic sense to build a reusable launcher big enough for that many satellites at a time.
I’m not sure it’s clear yet that Starlink is a killer app. It’s certainly a revenue stream, but the books are so opaque on Musk’s companies that it’s hard to be sure which one is subsidizing the other, or if the combination of the two actually brings in net profit, or what role the government contracts play.
Undeniable of course, but still I think that what makes the difference is a use (how good a one may be debated) for constellations of satellites sharing an orbital plane. One launch putting 20 satellites in orbit at a time is a lot more economical than twenty launches with twenty separate payloads. I can’t think of anything pre-Starlink that could avail itself of that economy of scale.
Payloads for which it’s specifically desirable for many satellites to share the same orbit are rare. But there are a lot of different payloads where what’s needed is just for them to be in orbit, and what orbit is largely immaterial. You could bundle a bunch of those together.
Of course, Starlink satellites all being identical probably also makes it easier to bundle them, but that’s a fairly small engineering problem.
GPS & the various non-US equivalents all use the same-plane approach to orbiting.
The numbers per plane are much smaller, and since the satellites were each hand-built the Old Space way there was never a critical mass to launch a bunch at once.
But I could imagine a Son-of-GPS system that involved large numbers of much smaller satellites. Harder to destroy a functionally significant percentage, harder to jam from above, much more likely to be able to get a fix in difficult locations, etc.
Unrelated to the above
I could imagine a spysat constellation that was similar. Large numbers of planes and large numbers per plane. Giving near continuous coverage of vast swathes of Earth. Add in some synthetic aperture signal processing and they may be able to get by with much less fancy (read “insanely expensive”) optics than traditional reconnaisance sats.
There would seem to be a saturation factor there, though?
Once you have the system in place, would there be a reason to keep launching new satellites at a rapid cadence?
Observation satellites benefit from being in fairly low orbits. Anything less than 700 km high is going to be subject to residual air drag, requiring reboosting or replacement.
True. But the same can be said of Starlink itself.
Net of course of replacing malfunctions, fuel exhaustions, and orbital decays. And of course generational upgrades with improved capabilities sliding into existing slots in existing orbits.
Starlink or similar competitors have the potential that demand for their bandwidth just keeps going up and up and up forever. And that maybe the cheap way to satisfy that is to launch ever more satellites, not 1-for-1 replace them with ever-improved satellites.
You could well be right. Bloody hell, orbit space is going to get awfully crowded if we’re not careful.
Kessler and all that…
Starlink satellites are designed to be deorbited at the end of their lifecycle or as they are replaced by later models. I believe they purposely deorbited 200+ V1 models last year.
Far from space being the wide-open frontier, LEO may become the most heavily traffic-controlled space in the Solar System.
Somewhere around here I recently (last year or two) wrote a Feynman analysis of trying to Kessler attack the Starlink constellation. Be damned if I can search it up though.
Punchline relevant to this thread is that even if we simplify the actual orbits a bit to further concentrate the constellation, and assume it’s fully built out, each refrigerator-sized target has a 2D area of space the size of West Virginia all to itself. And in 3D terms it’s a thin sheet-like volume the lateral size of West Virginia and about 5 feet thick. Which is gossamer thin when scaled down to human scales.
Space is big. No, bigger than that. Starlink in toto is small. Very very small.
Late add: The prior analysis was: