ESA in discussion with SpaceX on launch requirements

[Robotbeat]

Counterpoint: Each megaconstellation can support around 50 launches per year on average, and therefore at least one RLV. There are 2 active megaconstellations and another one coming. That's 3 RLVs. There may well be more megaconstellations in the future, plus launch demand from NASA (this is not insignificant as NASA leans harder and harder on commercial launch...). Artemis 3 will involve roughly 1000mt IMLEO from SpaceX, about 10 launches of Starship (actually double that as SpaceX will need to demo it). NASA requires continued launches to ISS (crew and cargo), to Gateway (currently just cargo), and for HLS (first Starship, then adding another provider) plus the CLIPS missions. In addition to the occasional Earth Observation, Astrophysics, and Robotic planetary science missions.  All told, NASA's launch demand could count as another RLV.

More precisely:
Kuiper masses about 2000mt for its initial configuration (~400mt annually?). Starlink about 3500mt in initial configuration and around 40,000mt in its final configuration (~8000mt annual?). OneWeb is 200mt initially and about 500mt for the second rung (~100mt annual?).
Artemis is about 1000mt per year for the HLS, 100mt per year for ISS servicing. 50-100mt IMLEO per year for Gateway. Maybe 150-250mt per year once CLIPS is operational and another 50mt for random non-HSF launches for a total of about 1500mt per year of launch demand from NASA. Altogether, those are about 10,000mt per year in launch demand. Split equally among 20t launchers, that's a healthy 500 launchers, enough for 10 RLVs, 5 RLVs with a healthy 100-per-year launchrate.

That's altogether pretty healthy, and I didn't cover other commsats or military stuff or space tourism. Space Tourism could be easily just as much, starting with 50mt per year for occasion LEO flights, getting much higher with Starship (Polaris Project) and then flying around the Moon (Dear Moon) to say an additional 200mt per year... Once that becomes a thing, you might get 10 tourist flights per year to the Moon. That'd equal all other demand above.

A wild card would be SBSP. The stuff the European study asks for is about 100,000t per year. I get the skepticism, but this could be conceivable at this scale.

It won't happen if we don't try.

[su27k]

You're double counting, the 8,000t for Starlink and 1,000t for HLS is only there because of Starship, it wouldn't go to 20t launchers. Take that out and your market shrinks by 90%...

[Asteroza]

A wild card would be SBSP. The stuff the European study asks for is about 100,000t per year. I get the skepticism, but this could be conceivable at this scale.

ESA EHLL is targeting 10,000t/year lift target to support SPS deployment, looks like you added an extra zero there. Your Starlink annual mass of 8000t suggests EU Starship EHLL can cover it's own Starlink class constellation, which might intermingle with any OneWeb expansion attempts.

Plus there's always the dark horse constellations in the corner that is space 5G systems like Lynk.Global and Spacemobile, if they ever get direct support from the likes of Apple or Google to seal their global smartphone hegemony.

Additionally, the numbers here are informative with respect to the chinese Guowang constellation, which would be facing similar tonnage concerns over sustainability of chinese commercial heavy RLV's in their ostensibly closed domestic market.


But we are starting to veer away from the original topic. With the announcement of EHLL, which appears to be a direct Starship competitor, how long would ESA tolerate sending payloads to F9 and Starship to avoid undercutting EHLL wooing bulk customers to drive that kind of tonnage? Ostensibly EHLL needs to support SPS deployment in 2035, which implies first flight before then. That puts perhaps a healthy 6-8 years of uncontested Starship availability temping ESA payloads, while F9 being close to Ariane 6 means payloads in lower weight classes (but above Vega) will waffle (departure window constrained payloads may consider escaping on F9).

[M.E.T.]

Counterpoint: Each megaconstellation can support around 50 launches per year on average, and therefore at least one RLV. There are 2 active megaconstellations and another one coming. That's 3 RLVs. There may well be more megaconstellations in the future, plus launch demand from NASA (this is not insignificant as NASA leans harder and harder on commercial launch...). Artemis 3 will involve roughly 1000mt IMLEO from SpaceX, about 10 launches of Starship (actually double that as SpaceX will need to demo it). NASA requires continued launches to ISS (crew and cargo), to Gateway (currently just cargo), and for HLS (first Starship, then adding another provider) plus the CLIPS missions. In addition to the occasional Earth Observation, Astrophysics, and Robotic planetary science missions.  All told, NASA's launch demand could count as another RLV.

More precisely:
Kuiper masses about 2000mt for its initial configuration (~400mt annually?). Starlink about 3500mt in initial configuration and around 40,000mt in its final configuration (~8000mt annual?). OneWeb is 200mt initially and about 500mt for the second rung (~100mt annual?).
Artemis is about 1000mt per year for the HLS, 100mt per year for ISS servicing. 50-100mt IMLEO per year for Gateway. Maybe 150-250mt per year once CLIPS is operational and another 50mt for random non-HSF launches for a total of about 1500mt per year of launch demand from NASA. Altogether, those are about 10,000mt per year in launch demand. Split equally among 20t launchers, that's a healthy 500 launchers, enough for 10 RLVs, 5 RLVs with a healthy 100-per-year launchrate.

That's altogether pretty healthy, and I didn't cover other commsats or military stuff or space tourism. Space Tourism could be easily just as much, starting with 50mt per year for occasion LEO flights, getting much higher with Starship (Polaris Project) and then flying around the Moon (Dear Moon) to say an additional 200mt per year... Once that becomes a thing, you might get 10 tourist flights per year to the Moon. That'd equal all other demand above.

A wild card would be SBSP. The stuff the European study asks for is about 100,000t per year. I get the skepticism, but this could be conceivable at this scale.

It won't happen if we don't try.

Kuiper 2000mt in total? That’s 20 Starship launches in total. 400mt per year? That’s 4 Starship launches. $40M launch revenue per year at $10M per Starship launch. Yet you state that each mega constellation could support a RLV?

Using Kuiper as the example, why develop an entirely new RLV for something that could be satisfied by 4 Starship launches per year? For $40M?

[Robotbeat]

You're double counting, the 8,000t for Starlink and 1,000t for HLS is only there because of Starship, it wouldn't go to 20t launchers. Take that out and your market shrinks by 90%...

sure, but the stuff Starship enables is stuff others will also want. People want an alternative to SpaceX as an option. Kuiper especially. But Congress and NASA would prefer a second HLS. I know some folks developing space stations would really like not just SpaceX but a similar competitor (like Terran-R).

Bezos wants to do similar stuff to what Musk is doing, just different destinations.

And note that with new technology and competition, launch mass can grow by a couple orders of magnitude even if revenue is basically flat. Operating an RLV isn’t much more than an ELV, even at lower launch rates. But the marginal cost of launch is FAR less.

I think 5 RLVs is likely too much. But 3 or 4 is feasible.