Reusing Boosters: Launch, Land, and Re-Launch

[Cosmacelf]

And we did!

Again, Elon/SX have been interested in satellite internet since just about SX was founded. It’s probably always been an element of the master plan.

Are you channeling ChatGPT and hallucinating? Link? Specific references? Anything?

 

[bxr140]

Are you channeling ChatGPT and hallucinating? Link? Specific references? Anything?

Sigh.

• Very early on Elon hired Larry Wiliams , a former Teledesic and ICO guy. Yes, THAT Teledesic, the one that wanted to put 1000 big ass satellites in LEO for broadband services. (ICO had similar if not slightly less grand plans). While Larry did a lot of things, you don't hire a guy that was critical to the concept development of by far the most audacious data constellation ever to do everything but figure out how to come up with a more better audacious constellation...
• In 2005, SpaceX invested in Surrey who, at the time, were just about the only real player in the "not expensive" satellite space. While this timeline drops us in the post 90's environment where Globalstar and Iridium had fizzled out and Teledesic was clearly too ambitious (and Dotcom didn't help), Surrey was astute enough to realize those exercises indicated that there was a glimmer of there there in internet-from-space. Being well positioned as the "not expensive" satellite manufacturer, Surrey was quite keen to lay their foundation in anticipation of a hopeful resurgence of internet-from-space. This company direction was well aligned with SX who, at the time, knew they wanted to get into the internet-from-space game AND BIG (remember the guy that wanted to put 1000 satellites in the 90's)...but didn't know how or when from the perspective of closing a business case...and they certainly weren't a satellite company at the time (all of their fledgling resources were focused on their rockets.)
• A few years later if was clear to SX that Surrey wasn't going to crack the code on technology that would enable satellites to be both cheap enough and powerful enough to return the right amount of black. (Among other things, phased arrays were just becoming a thing terrestrially, but the things like the cost of solar power, Li-ion batteries, etc. were just a bit out of reach). Surrey knew this as well and were pivoting into both a) the earth observation realm (they were early players in what became the early teens EO frenzy) and also b) getting that sweet sweet state funding from participating in Galileo. Both were more lucrative than waiting for technology to come good for internet-from-space, but neither of those interested SX because they weren't core to the top level charter of Mars. So...SX sold their stake.
• In this time, and especially toward the end of the Surrey period, SX really noses to the grindstone on driving toward a reusable Falcon (which of course wouldn't actually be realized for another decade), based on Rocket =Horse, Constellation = Cart logic. By this point it was already quite clear that any megaconstellation was going to require significant lift and frequency capability, which without a cheap rocket (that he was developing) was essentially untenable. And the corollary, it was quite clear that developing a cheap rocket that could do significant lifting was not going to happen financially without an untapped source of ROI--something well beyond what the space industry at large could return. And so, it was quite clear that the only way to create enough demand to satisfy the development cost was to do so internally, with a megaconstellation. Elon bet big--as he did with Tesla--that he could dump a ton of money (a ton of money to SX anyway) into making F9 reusable (which, again, wasn't realized for years after it started flying) and that the spending would come good once the constellation got off its feet.
• Later, but before what we know as Starlink was a thing, the infamous Musk-Wyler bromance conspired to huck many hundreds of big ass satellites into MEO. By this point Elon was pretty confident he had the rocket side solved, but the satellite side was still not solved (and Wyler wanted RF ISLs where Elon thought Optical ISLs were The Bees) and the bromance flame blew out.
• Finally, Starlink comes into play, once Elon was confident that he had both the rocket and satellite side solved (by doing both himself).

 

[Cosmacelf]

Sigh.

• Very early on Elon hired Larry Wiliams , a former Teledesic and ICO guy. Yes, THAT Teledesic, the one that wanted to put 1000 big ass satellites in LEO for broadband services. (ICO had similar if not slightly less grand plans). While Larry did a lot of things, you don't hire a guy that was critical to the concept development of by far the most audacious data constellation ever to do everything but figure out how to come up with a more better audacious constellation...

Teledesic was a complete flop and never went anywhere. And yes you DO hire a guy like that to do something else. His linked in bio says he was doing something else entirely for SpaceX. Nothing about Starlink:

”Member of the initial executive team at SpaceX responsible for international business development and government relations. Wore many hats and helped expand the company's revenue base from around ~$4 million to almost $4 billion.

- In 2005, SpaceX invested in Surrey who, at the time, were just about the only real player in the "not expensive" satellite space. While this timeline drops us in the post 90's environment where Globalstar and Iridium had fizzled out and Teledesic was clearly too ambitious (and Dotcom didn't help), Surrey was astute enough to realize those exercises indicated that there was a glimmer of there there in internet-from-space. Being well positioned as the "not expensive" satellite manufacturer, Surrey was quite keen to lay their foundation in anticipation of a hopeful resurgence of internet-from-space. This company direction was well aligned with SX who, at the time, knew they wanted to get into the internet-from-space game AND BIG (remember the guy that wanted to put 1000 satellites in the 90's)...but didn't know how or when from the perspective of closing a business case...and they certainly weren't a satellite company at the time (all of their fledgling resources were focused on their rockets.)

From your own linked article:

“They could have accepted investments from just about anyone I think,” Musk said. “SSTL is a high-quality company that is probably the world leader in small satellites. We look at this as more a case of similar corporate cultures getting together. For now there is nothing firm about working together. But when you have two companies that see the world in the same way, they can usually figure out things to do together.”

ie. They had no plans to work together. And Surrey never did any work on telecom sats. And reading Eric Berger’s and various other books about SpaceX, Elon was not thinking about satellites in 2005!

• A few years later if was clear to SX that Surrey wasn't going to crack the code on technology that would enable satellites to be both cheap enough and powerful enough to return the right amount of black. (Among other things, phased arrays were just becoming a thing terrestrially, but the things like the cost of solar power, Li-ion batteries, etc. were just a bit out of reach). Surrey knew this as well and were pivoting into both a) the earth observation realm (they were early players in what became the early teens EO frenzy) and also b) getting that sweet sweet state funding from participating in Galileo. Both were more lucrative than waiting for technology to come good for internet-from-space, but neither of those interested SX because they weren't core to the top level charter of Mars. So...SX sold their stake.
• In this time, and especially toward the end of the Surrey period, SX really noses to the grindstone on driving toward a reusable Falcon (which of course wouldn't actually be realized for another decade), based on Rocket =Horse, Constellation = Cart logic. By this point it was already quite clear that any megaconstellation was going to require significant lift and frequency capability, which without a cheap rocket (that he was developing) was essentially untenable. And the corollary, it was quite clear that developing a cheap rocket that could do significant lifting was not going to happen financially without an untapped source of ROI--something well beyond what the space industry at large could return. And so, it was quite clear that the only way to create enough demand to satisfy the development cost was to do so internally, with a megaconstellation. Elon bet big--as he did with Tesla--that he could dump a ton of money (a ton of money to SX anyway) into making F9 reusable (which, again, wasn't realized for years after it started flying) and that the spending would come good once the constellation got off its feet.

F9 reusable didn’t need any extra outside investment or other rationale other than it being a cheaper way to do business. Yes, it was also a component of getting to mars since Elon realized you couldn’t build a city of Mars without reusability, but that had nothing to do with Starlink. This is the section where you are waving your hands since you don’t even have wrong sources to back up your assertions.

• Later, but before what we know as Starlink was a thing, the infamous Musk-Wyler bromance conspired to huck many hundreds of big ass satellites into MEO. By this point Elon was pretty confident he had the rocket side solved, but the satellite side was still not solved (and Wyler wanted RF ISLs where Elon thought Optical ISLs were The Bees) and the bromance flame blew out.

2014 is when SpaceX started work on their ITS (precursor to Starship) and is indeed when SpaceX started thinking about how to pay for it. So, yes, around 2014 is when SpaceX started thinking about Starlink. And the only reason for Starlink was as a method to pay for Starship development.

Look, until someone asks Elon these questions, there just isn’t documented evidence of Elon’s thinking in various times. Maybe Eric Berger will tackle this when his Falcon 9 book comes out next year.

• Finally, Starlink comes into play, once Elon was confident that he had both the rocket and satellite side solved (by doing both himself).

 

[scaesare]

Sigh.

• Very early on Elon hired Larry Wiliams , a former Teledesic and ICO guy. Yes, THAT Teledesic, the one that wanted to put 1000 big ass satellites in LEO for broadband services. (ICO had similar if not slightly less grand plans). While Larry did a lot of things, you don't hire a guy that was critical to the concept development of by far the most audacious data constellation ever to do everything but figure out how to come up with a more better audacious constellation...
• In 2005, SpaceX invested in Surrey who, at the time, were just about the only real player in the "not expensive" satellite space. While this timeline drops us in the post 90's environment where Globalstar and Iridium had fizzled out and Teledesic was clearly too ambitious (and Dotcom didn't help), Surrey was astute enough to realize those exercises indicated that there was a glimmer of there there in internet-from-space. Being well positioned as the "not expensive" satellite manufacturer, Surrey was quite keen to lay their foundation in anticipation of a hopeful resurgence of internet-from-space. This company direction was well aligned with SX who, at the time, knew they wanted to get into the internet-from-space game AND BIG (remember the guy that wanted to put 1000 satellites in the 90's)...but didn't know how or when from the perspective of closing a business case...and they certainly weren't a satellite company at the time (all of their fledgling resources were focused on their rockets.)
• A few years later if was clear to SX that Surrey wasn't going to crack the code on technology that would enable satellites to be both cheap enough and powerful enough to return the right amount of black. (Among other things, phased arrays were just becoming a thing terrestrially, but the things like the cost of solar power, Li-ion batteries, etc. were just a bit out of reach). Surrey knew this as well and were pivoting into both a) the earth observation realm (they were early players in what became the early teens EO frenzy) and also b) getting that sweet sweet state funding from participating in Galileo. Both were more lucrative than waiting for technology to come good for internet-from-space, but neither of those interested SX because they weren't core to the top level charter of Mars. So...SX sold their stake.
• In this time, and especially toward the end of the Surrey period, SX really noses to the grindstone on driving toward a reusable Falcon (which of course wouldn't actually be realized for another decade), based on Rocket =Horse, Constellation = Cart logic. By this point it was already quite clear that any megaconstellation was going to require significant lift and frequency capability, which without a cheap rocket (that he was developing) was essentially untenable. And the corollary, it was quite clear that developing a cheap rocket that could do significant lifting was not going to happen financially without an untapped source of ROI--something well beyond what the space industry at large could return. And so, it was quite clear that the only way to create enough demand to satisfy the development cost was to do so internally, with a megaconstellation. Elon bet big--as he did with Tesla--that he could dump a ton of money (a ton of money to SX anyway) into making F9 reusable (which, again, wasn't realized for years after it started flying) and that the spending would come good once the constellation got off its feet.
• Later, but before what we know as Starlink was a thing, the infamous Musk-Wyler bromance conspired to huck many hundreds of big ass satellites into MEO. By this point Elon was pretty confident he had the rocket side solved, but the satellite side was still not solved (and Wyler wanted RF ISLs where Elon thought Optical ISLs were The Bees) and the bromance flame blew out.
• Finally, Starlink comes into play, once Elon was confident that he had both the rocket and satellite side solved (by doing both himself).

Thanks for the references/background... it provides some context regarding whether Starlink (as at least a concept) existed early enough for it to have been material in helping fund/justify the development of F9 , the discussion of which was prompted by the Ariane 6 development debacle.

I agree that it's likely that Elon was considering getting in to the satellite game earlier than the 2015 announcement. Heck once you announce something, you've had to have been planning it for some time... especially something that big.

What I think we differ on are the goals for (what became) Starlink and what impact it had on F9.

Elon has repeated several times that he sees Starlink funding Starhip/Mars:

Space.com: "We think this is a key steppingstone towards establishing a self-sustaining city on Mars and a base on the moon," Musk said of Starlink. "We believe we can use the revenue from Starlink to fund Starship."

cnbc.com (in 2019 “We see this as a way for SpaceX to generate revenue that can be used to develop more and more advanced rockets and spaceships,” Musk said.

spacenews.com: Musk also briefly discussed Starlink, SpaceX’s planned broadband satellite constellation that will provide connectivity to underserved areas and a low-cost alternative in more urbanized areas. Starlink, he added, will also provide revenue needed for SpaceX’s Mars ambitions.

So clearly Elon sees SpaceX being it's own internal customer for mega-constellation launch manifests as an end to a means. It just appears that was never a means to fund development of F9, even if early on Elon might have wanted it to be.


So, your original assertion that led to my delving into this was:

Bottom line, Ariane simply doesn't have the ability to offset today's R&D with future launch cost savings of an internal mega constellation like SpaceX could, and that's the volume that's really needed to push their expendable/reusable inflection point to "duh, reusable".

I just don't see how that's the case. SpaceX did it with Falcon9 years prior to their being a mega-constallation market, even if Elon was thinking about one. They developed Falcon 9 for $300mil. They had "standard" contracts from NASA for ~$400mil. No new mega-market required... just efficient development.

So (a roundabout journey) back to the original issue: If SpaceX could do it in the existing market, why can't Ariane?

 

[bxr140]

Teledesic was a complete flop and never went anywhere. And yes you DO hire a guy like that to do something else.

I love that this crystalizes the pervasive trend here of contriving fault.


Seriously, what you're asserting is that Elon Musk, who was already:

• A successful entrepreneur
• All in on "ok fine, I'll build my own rockets"
• Known for personally grilling interviewees on their past career performance
• Known for ruthlessly demanding a success based work ethic
• Obsessed with the letter X (no relevance, just pointing it out)

Decided to hire a guy that:

• Had spent the last 10 years working in the space industry, explicitly on internet-from-space
• Many of those at Teledesic, the (by that point) industry poster child for ambition over skill
• Some of those at ICO, another company strong on ambition and light on execution
• Had no other space experience

And that hiring was done for the purpose of dude bringing senior leadership value to SX by working on something not related to what he had been doing for the previous decade, but definitely NOT bringing down the fledgling SX as a result of dude's long history as a senior leader at a 'complete flop'.

And you assert that all while definitively 'its unpossible' rejecting the notion that Elon hired the guy to (again, among other things) convert his past ten years of "your job is to find the there there" in internet-from-space into something actionable within Elon's new, agile, space company that was unburdened with the legacy industry overhead--overhead that was a major reason Teledesic actually flopped.

...there just isn’t documented evidence of Elon’s thinking in various times.

Elon was not thinking about satellites in 2005!

<chef's kiss>

If SpaceX could do it in the existing market, why can't Ariane?

Appreciate the effort to come full circle; we're indeed at the same place we started.

1. As described in my initial post (which is in the other thread it seems), it was always unrealistic to assume that an institution like Ariane could do what SX can do even with significantly more funding.
2. While you may not like the explanation, SX did NOT do it in the existing market; a major enabler was knowing that they were going to create their own market. Remember that just now, ~20 years after they were founded, SX starting to turn black quarters. Just now, starlink--which was enabled by reusable F9--is paying back the heavy (for SX) investment put into developing a reusable F9. While SX's financials have always been applied/reported in a non-traditional way, in traditional terms, if Starlink didn't exist F9 would effectively still be a cost center.


Again, Elon is known for playing the long game. Elon is known for making big financial bets that have a long tail to come good. Elon is known for an extreme tolerance relative to operating at the red and operating in the couch cushions. Elon is known for having vision that requires layers and layers of 'coming good'.

(Elon is ALSO known for 'Elon Time', and that's what's leading to conflation of the timelines and motivations here...)


To ensure incorrect assumptions aren't being applied, I have not in any way, in any of my posts in this thread (or anywhere else), asserted that Elon's internet-in-space vision looked like [what we now know as] Starlink. Indeed the most recent non-starlink concept (the one with Wyler) was one of far more traditional and far fewer satellites than Starlink; indeed that constellation concept (and various iterations leading back to the mid aughts) still all demanded unprecedented launch capacity to what had been done to date (basically, Iridium and Globalstar). Knowing he was solving the rocket side of the problem with F9, it took that row with Wyler to really inflect into the right concept to solve the satellite problem. Enter Starlink (As we know it).

 

[scaesare]

Appreciate the effort to come full circle; we're indeed at the same place we started.

1. As described in my initial post (which is in the other thread it seems), it was always unrealistic to assume that an institution like Ariane could do what SX can do even with significantly more funding.

Which is more of indictment of Ariane than it is a discussion of what's possible. And if your original point was "Ariane just isn't built to be able to quickly develop a reusable rocket in the existing market" rather than "It can't be done.", then I probably largely agree...

2. While you may not like the explanation, SX did NOT do it in the existing market; a major enabler was knowing that they were going to create their own market.

Here's where we disagree. SpaceX did develop F9 on the existing market. They spent about $300mil as of V1.0 in 2011 The estimates of costs for both F9 and Dragon were just under $850mil as of 2014 so F9's share probably means they spent $400-500mil on it at that point. F9 first successfully landed shortly thereafter.

They had received almost $400mil from NASA. As you pointed out earlier, they already had launch order for CRS worth $1.6Billion. That's a total of $2 billion, i or about five times F9's development cost. In the existing market. No mega-customer in sight.

Thought experiment: If Elon died in 2015 and Starlink never happened and Starship and plans for Mars ground to a halt, would SpaceX have been able to continue profitably selling F9 launch service? I don't see why not....

Remember that just now, ~20 years after they were founded, SX starting to turn black quarters. Just now, starlink--which was enabled by reusable F9--is paying back the heavy (for SX) investment put into developing a reusable F9. While SX's financials have always been applied/reported in a non-traditional way, in traditional terms, if Starlink didn't exist F9 would effectively still be a cost center.

But that's the point: F9 launching Starlink is paying for Starship/Mars. SpaceX isn't finally "starting to turn black quarters" because F9 could finally launch profitably after all this time. It's because Elon's spending like mad to develop Starship whilst also building that largest constellation that's ever existed.

Look at it this way: SpaceX might finally be in the black despite Starship because of Starlink... whereas without Starlink and Starship, they would have been in the black years ago...

Again, Elon is known for playing the long game. Elon is known for making big financial bets that have a long tail to come good. Elon is known for an extreme tolerance relative to operating at the red and operating in the couch cushions. Elon is known for having vision that requires layers and layers of 'coming good'.

(Elon is ALSO known for 'Elon Time', and that's what's leading to conflation of the timelines and motivations here...)


To ensure incorrect assumptions aren't being applied, I have not in any way, in any of my posts in this thread (or anywhere else), asserted that Elon's internet-in-space vision looked like [what we now know as] Starlink. Indeed the most recent non-starlink concept (the one with Wyler) was one of far more traditional and far fewer satellites than Starlink; indeed that constellation concept (and various iterations leading back to the mid aughts) still all demanded unprecedented launch capacity to what had been done to date (basically, Iridium and Globalstar). Knowing he was solving the rocket side of the problem with F9, it took that row with Wyler to really inflect into the right concept to solve the satellite problem. Enter Starlink (As we know it).

Sure, Elon is always looking ahead... and every indication is that at whatever point he was considering his sat constellation, he was attempting to figure out how to pay for Starship. Falcon9 was already baked and out of the oven and a moneyearning workhorse day one.

No technical reason Ariane couldn't so the same... it's an organizational problem.

 

[Cosmacelf]

The theme that governments (and Ariane is most definitely a government agency in all but name) can spend decades spinning their wheels doing nothing, while private industry tackles the problems and gets it done, is, I would think, well established by now.

I'm reminded of this again when Youtube star Mr. Beast built 100 critically needed water wells in Kenya and Zimbabwe while their governments and assorted NGOs with billions of dollars at their disposal did crap all nothing.

MrBeast faces new criticism, admits his humanitarian efforts might get him ‘canceled’

MrBeast is facing backlash after building wells across Africa in new YouTube video

www.deseret.com

 

[scaesare]

It just keeps getting better and better:


Nate Janzen, manager of launch pad systems and operations for SpaceX at Vandenberg and a 10-year employee of the firm, spoke last week during the 10th annual celebration and Future Forum for the Economic Alliance Foundation, or EconAlliance, at the Santa Maria Country Club:

Quote

Next year, SpaceX will re-evaluate and conduct analysis with an eye toward certifying the first-stage boosters for 25 to 30 flights, he said.

 

[scaesare]

This paragraph from Berger's Rocket Report this wek is worded oddly, but it sounds like even though they won't re-certify boosters for 25-30 launches (presumably for customers) until next year, they may actually hit 30 re-flights (for themselves I assume), by the end of this year.

Also taking another look at booster longevity ... From one launch four years ago to three the next year and 12 the following year, SpaceX expects about 30 liftoffs by the end of this year. For 2024, the rate could jump to 50, then rocket to 100 in 2025. "Next year, we’ll be launching about once a week, but the plan, in about two years, is about every three to four days," Janzen said. Additionally, next year, SpaceX will re-evaluate and conduct analysis with an eye toward certifying the first-stage boosters for 25 to 30 flights, he said.

I should look and see if a booster has enough launches on it to hit 30 this year...

 

[mongo]

I should look and see if a booster has enough launches on it to hit 30 this year...

Not this year... highest is 18

 

[scaesare]

Not this year... highest is 18

Then I don't understand his numbers... what does he mean by "SpaceX expects about 30 liftoffs by the end of this year"...?

 

[mongo]

Then I don't understand his numbers... what does he mean by "SpaceX expects about 30 liftoffs by the end of this year"...?

Ah, that number is liftoffs from Vandenberg, not a single booster.
The "Also taking another look at booster longevity ..." line is in a weird spot.

 

[scaesare]

Ah, that number is liftoffs from Vandenberg, not a single booster.
The "Also taking another look at booster longevity ..." line is in a weird spot.

Ah ok... that make sense...

Agreed on the booster longevity line, I initially thought maybe he was conflating the planned launch cadence of every 2.5 days in 2025 with the booster re-cert, but those numbers didn't make sense either...

 

[Grendal]

I should look and see if a booster has enough launches on it to hit 30 this year...

I believe that boosters are certified for a maximum of 20 launches at this point. Maybe they are working on getting the certification to 30 this year. Interesting that Elon said that the boosters could keep launching and I thought it was just his optimism talking. Apparently they can go a lot further than I thought before getting retired. Also, SpaceX would retire a booster in one of the more energetic launches and that hasn't happened in a long time.

 

[scaesare]

Is that certification for external customers?

Does SpaceX care if a booster checks out after 20+ flights?

 

[bxr140]

Does SpaceX care if a booster checks out after 20+ flights?

Yes. Rockets not doing what they're supposed to do is a safety hazard. The fleet will be grounded until a root cause is identified and remedies are implemented.

 

[scaesare]

Yes. Rockets not doing what they're supposed to do is a safety hazard. The fleet will be grounded until a root cause is identified and remedies are implemented.

I may have phrased that ambiguously.

Assuming the recertification process is a formal thing for external customer guarantee purposes, would SpaceX itself hesitate to fly a booster a 21st time for say Starlink, even if the "certification" for 20+reflights doesn't exist, as long as the engineers give it a clean bill of health?

 

[JB47394]

Assuming the recertification process is a formal thing for external customer guarantee purposes, would SpaceX itself hesitate to fly a booster a 21st time for say Starlink, even if the "certification" for 20+reflights doesn't exist, as long as the engineers give it a clean bill of health?

The certification process is defined by NASA to establish what sorts of government contracts they can fill. From what I can tell, that doesn't stop SpaceX from flying them as many times as they like with other payloads. But I assume that they want NASA certification so that non-US government customers can be confident in the heavily-reused boosters as well.

 

[Grendal]

Is that certification for external customers?

Does SpaceX care if a booster checks out after 20+ flights?

I may have phrased that ambiguously.

Assuming the recertification process is a formal thing for external customer guarantee purposes, would SpaceX itself hesitate to fly a booster a 21st time for say Starlink, even if the "certification" for 20+reflights doesn't exist, as long as the engineers give it a clean bill of health?

It is definitely a regulatory thing. I'm not sure which agency but I expect it is the FAA. SpaceX will not launch a booster, even for themselves, unless it is certified to do so. Flaunting regulations and certifications would lose them NASA and Military launches for sure. And SpaceX probably has some sort of rigorous testing it does on the "extended life" boosters that go beyond the standard check they do with every booster and reused booster.

 

[scaesare]

Thanks gents... that makes sense,,,