Launching Starlink with Starship

[TheRadicalModerate]

That's not really an either/or.

No, but it's probably an 80/20.

[GregA]

It doesn’t matter so much if they lose some as long as no humans are involved and there’s a new generation that fixes the problem coming along in a month or two anyway. Not that they are being careless but they’re designing a system that isn’t going to shut down for a year and be obsessed about what went wrong if there is a failure. There’s no assumption any generation of Starship won’t be quickly scrapped.

Will that really work?
I mean... I agree Musk would be happy to learn with periodic RUDs (on launch or return), but if they shutdown missions for 6-12months to investigate that would be bad.

Can they adjust to expecting periodic explosions and continuing.

[GregA]

Whereas the average launch of 25 Starships in a year of ~1 per month from a probable each of 2 pads (LC 39A and Boca Chica) is easily achieved. Where 18 SS are manufacture per year and 6 SH are manufactured per year. That is currently Boca Chica's build rate capacity not requiring additional equipment/tooling other than that currently being installed.

I think the numbers are too low, but I may be misunderstanding you.

Shotwell said "We are aiming to be able to drop Starship on the lunar surface in 2022" which will require multiple refuelling missions. Before then they'll need to do several practice refuelling missions... though perhaps these will BE the practice refuelling (makes sense, they like to learn as they go).

Plus to send 2 cargo starships to Mars at the end of 2022 will require about 10-12 launches (to refuel each 4-5 times).

If they're only planning 25 launches a year there'll be little room for any Starlink launches in 2022.

edit: I don't mean to say they'll achieve this, after all it may be the first year SS/SH flies at all. Just to say their aspirational goal is for a LOT of launches in 2022 even before Starlink enters the equation.

[Robotbeat]

Even if Starlink is barely profitable. It will fund the launch of dozens of Starships a year.

Starship will likely progress rapidly to multiple reuse of SH and SS.

SH reuse is likely after about 6 moths to a year of orbital flights. (2022) After that the number of flights achieved per SH will rapidly climb to as many as a value of 4 in one year.

Reuse for SS may take 2 years to achieve. But the flight numbers after that likely will rapidly increase such  that by 2024 with 8 flights per SH and 4 for SS. This could represent a per launch cost as low as $25M.

Which has more bang for the buck:  using Starlink profits to fund relatively more Starship launches, or using them to fund building more Starlinks?  Starlink is exactly the kind of business where you'd expect massive returns on scale.  Launching stuff is not.

I'm not really sure what you mean about launching not having massive returns to scale. Can you explain?

Do you mean the market for space launch is pretty limited? Because that's true. Do you mean there's not much efficiency to be gained from launching larger or more often? Because that's definitely not true.

Gotta realize that Starship isn't even ultimately necessarily about profit. It's a way to get to Mars. So Starship's greatest customer could be Elon. That's why there's a large return for launching more often and with larger payloads: because it means Elon can pay for the same platform to launch orders of magnitude more stuff to Mars for the same upfront cost.

[TheRadicalModerate]

I'm not really sure what you mean about launching not having massive returns to scale. Can you explain?

"Return on scale" is a phenomenon where productivity (and therefore profit margin) increases as the amount of your product that you provide increases.  It's why car companies are big.  It's why Amazon plowed everything back into the business until they had close to a monopoly position.

Vanilla-flavored access ISPs scale OK, because all of their OAM&P systems and personnel can cover however many access POPs they deploy, but the POPs themselves don't scale particularly well:  each one needs roughly the same amount of CPE, the same amount of cable and fiber plant, head-ends, or DSLAMs, and some kind of presence in an IXP or central office to get access to the rest of the internet.  All of this adds up to an environment where the number of subscribers per investment doesn't increase very much.

Starlink has the same CPE (the "UFO" ground station) as an ISP, but the coverage areas for the POPs are larger (especially when you can thin out the ground stations by doing two hops to the lightly-subscribed ones), and the "last mile" is air and vacuum.  But even more important, the average number of subscribers per bird explodes as you can have a bird service different groups of subscribers along more of its ground track.  That's the principal reason why Starlink has a good chance of being a (literal) world-beating business and mass cash cow.

Do you mean the market for space launch is pretty limited? Because that's true. Do you mean there's not much efficiency to be gained from launching larger or more often? Because that's definitely not true.

You have to get to high demand scale before the "efficiency gained from larger and more often" kicks in.  In a market where there's nothing but bigsats, there's no efficiency to be gained at all.  Likewise for small constellations of smallsats.  It's only when you get to big constellations or big things--which will happen over time, but not in the next 5 years--that the theoretical specific cost reductions possible with Starship have any relationship to the actual specific costs.

The one exception to that is obviously Starlink.  My objection to what you said further up-thread was that you asserted that it made sense to emphasize "more often" at the expense of "larger".  In the absence of other paying customers, it's always better to be launching the minimum number of Starships needed to keep up with satellite production, with the maximum possible number of satellites.

Gotta realize that Starship isn't even ultimately necessarily about profit. It's a way to get to Mars. So Starship's greatest customer could be Elon. That's why there's a large return for launching more often and with larger payloads: because it means Elon can pay for the same platform to launch orders of magnitude more stuff to Mars for the same upfront cost.

SpaceX is in business to make money.  Elon, as SpaceX's leader, has given the company aspirational goals, but those goals can only be achieved by making money.  Whether that money is reinvested directly by SpaceX into Mars operations, or whether Elon takes his SpaceX profits and hires his own company to do the things he wants is a pretty blurry distinction, but it's all driven by making the money in the first place.  For any of these aspirational goals to be realized, SpaceX needs a cash cow.  Starlink is it.  Optimizing for anything other than getting Starlink deployed and scaled ASAP is bad business, and bad business is bad for developing Mars.

[Robotbeat]

Nope, you've got it backwards. SpaceX is not in business to make money. It's in business to get to Mars, to make humanity multi-planetary. Making money happens to be an efficient way to do that.

As Musk said during his 2016 presentation, there's really no amount of money that would get you to Mars right now using existing techniques, certainly not a self-sustaining civilization. So being a trillionaire doesn't help unless you use that to invent the technology to make it feasible. That's what Starship is.

SpaceX was started as basically a purely money-losing venture to send a greenhouse to Mars.


EDIT: Also, Starlink isn't the only cashcow. Tesla is another. Starship as a Mach 20 commercial airliner (as unlikely as that is) potentially is another.

[GregA]

SpaceX is not in business to make money. It's in business to get to Mars, to make humanity multi-planetary. Making money happens to be an efficient way to do that.

True, so anything which furthers getting to Mars is good and doesn't need to have profit at its core. But anything that is supposed to help fund Mars needs to make money..

...so Starlink is in business to make money Except for starlink satellites sent to Mars.

(And any method that increases "potential" profitability for any venture will be good for its goal too).

[Robotbeat]

SpaceX is not in business to make money. It's in business to get to Mars, to make humanity multi-planetary. Making money happens to be an efficient way to do that.

True, so anything which furthers getting to Mars is good and doesn't need to have profit at its core. But anything that is supposed to help fund Mars needs to make money..

...so Starlink is in business to make money Except for starlink satellites sent to Mars.

(And any method that increases "potential" profitability for any venture will be good for its goal too).

If Starlink merely pays for and justifies Starship's existence but otherwise just breaks even, then it will have be EASILY worth it.

[GregA]

If Starlink merely pays for and justifies Starship's existence but otherwise just breaks even, then it will have be EASILY worth it.

Absolutely.
But this is all pretty well the same thing right?

Spending too much on Mars and Starship is quite okay as it's the goal.
Only make Starlink if it helps Mars and Starship.

[TheRadicalModerate]

Nope, you've got it backwards. SpaceX is not in business to make money. It's in business to get to Mars, to make humanity multi-planetary. Making money happens to be an efficient way to do that.

As Musk said during his 2016 presentation, there's really no amount of money that would get you to Mars right now using existing techniques, certainly not a self-sustaining civilization. So being a trillionaire doesn't help unless you use that to invent the technology to make it feasible. That's what Starship is.

SpaceX was started as basically a purely money-losing venture to send a greenhouse to Mars.


EDIT: Also, Starlink isn't the only cashcow. Tesla is another. Starship as a Mach 20 commercial airliner (as unlikely as that is) potentially is another.

I'm willing to stipulate that SpaceX's motives are much loftier than making a profit, at least for purposes of this argument.  But, for those purposes, it doesn't matter a whit.

Remember, we're talking about how to match cadence and payload capacity for Starlinks once Starship is already working, at least for ascent.  It's not like we're taking money away from Starship development; we're just not spending it foolishly.  And if you want to maximize the capital available for doing Mars stuff later--that pesky "making money" thing again--then rolling out Starlink as quickly as possible is the best way to do it.

There is no reason to wring every last penny out of Starship launch costs until you have an application that's crucially dependent on those low launch costs.  If you think that that's going to be sending colonists to Mars at scale, or even P2P suborbital travel, any time remotely before the Starlink system has been deployed to the point where adding more has hit the point of diminishing returns, you're crazy.  The more money Starlink makes, the faster those follow-on applications can be funded.

As for Tesla as a way for Elon to make money to plow it into Mars development, it's irrelevant for purposes of this discussion.  This is only about optimally rolling out Starlink.

[Robotbeat]

I think that of course SpaceX will try to make Starlink very profitable to fund Mars. But even if it just breaks even and provides something to pay for Starship, then that will still advance Mars by a LOT.

It's a simple point I'm making, here. Starship needs demand. It needs more demand than the entire global launch rate (tons IMLEO per year) combined. Starlink can provide that, even if it just is breakeven.


And Tesla is not irrelevant to that point. Because Elon has other sources of funding for Mars, not just Starlink as you imply. So if all Starlink does is push Starship far down the experience curve and pay for its development, it'll help Mars tremendously as Elon has other funding sources. See the thread about Musk's asset accrual.

[TheRadicalModerate]

I think that of course SpaceX will try to make Starlink very profitable to fund Mars. But even if it just breaks even and provides something to pay for Starship...

First, define "break even".  Are you talking about the Starlink making enough revenue to cover its costs?  If so, then all you're doing is providing Starship a steady stream of launches, not funding any other part of its development.

Are you talking about it making enough profit (revenue minus costs) to cover Starship development?  If so, Starship development to what level?  To orbit?  That'll already be done by the time that Starship is deploying Starlinks.  To EDL and refueling capabilities?  That can be done with whatever cadence is dictated by launching max-payload batches of Starlinks.  Or are you talking about developing Starship to the point where its launch costs are within a few percent of their minimum?  But that only makes sense if you assign a time by which that minimum cost is needed.  I assert that it's so far in the future that increasing cadence by reducing Starlink payloads (i.e., raising the specific launch costs of Starlink) is unnecessary and counter-productive.

Remember that Starlink isn't a one-and-done operation, where launches stop after full deployment.  Roughly 20% of them are disposed of every year, requiring 20-30 launches a year just to maintain the constellation if it stabilizes at roughly 50,000 birds.  That alone is enough cadence to drive costs down, even if no other customers materialize.

... then that will still advance Mars by a LOT.

No, it really won't, because Starship development costs are only a tiny fraction of what would be necessary to enable a Mars colony.  During the time that Starlink is being built out to its maximum size, those costs will be dominated by the development and manufacturing costs for the various pieces of surface habitation and ISRU equipment.  Launches for Mars at high scale won't begin until Starlink has been fully built-out, and that will have required the hundreds of launches that are necessary to have Starship launch costs near their minimum.

It's a simple point I'm making, here. Starship needs demand. It needs more demand than the entire global launch rate (tons IMLEO per year) combined. Starlink can provide that, even if it just is breakeven.

I understood your simple point, but it's wrong as you have it stated.  Starship needs enough demand to drive it down the cost curve before other applications need it at its minimized launch cost.  The timeframe is crucial.  Starlink can provide enough demand to hit that timeframe with maximum payloads and only adequate cadence.  Intentionally driving up Starlink specific cost just to increase cadence is silly. 

You're further wrong about the breakeven:  Either Mars or P2P will require massive amounts of capital, far in excess of what it took to develop and cost-reduce Starship.

And Tesla is not irrelevant to that point. Because Elon has other sources of funding for Mars, not just Starlink as you imply. So if all Starlink does is push Starship far down the experience curve and pay for its development, it'll help Mars tremendously as Elon has other funding sources. See the thread about Musk's asset accrual.

Tesla is a public company, with fiduciary duties and oversight that prevent it from destroying itself to satisfy Elon's need for Mars capital.  If he wants to maintain control, he's severely limited in the amount of stock he can sell, and it's not going to be throwing off any appreciable dividends for decades if it intends to compete with the big boys.  Cars are yet another business where returns on scale are essential.

One way or another, SpaceX operations will fund whatever ambitions Elon has for Mars.

[Zed_Noir]

<snip>
Tesla is a public company, with fiduciary duties and oversight that prevent it from destroying itself to satisfy Elon's need for Mars capital.  If he wants to maintain control, he's severely limited in the amount of stock he can sell, and it's not going to be throwing off any appreciable dividends for decades if it intends to compete with the big boys.  Cars are yet another business where returns on scale are essential.
<snip>

You do realize that Musk only gets pay in stock options? He gets more stock as Tesla market capitalization increases.

[wannamoonbase]

SpaceX is not in business to make money. It's in business to get to Mars, to make humanity multi-planetary. Making money happens to be an efficient way to do that.

True, so anything which furthers getting to Mars is good and doesn't need to have profit at its core. But anything that is supposed to help fund Mars needs to make money..

...so Starlink is in business to make money Except for starlink satellites sent to Mars.

(And any method that increases "potential" profitability for any venture will be good for its goal too).

If Starlink merely pays for and justifies Starship's existence but otherwise just breaks even, then it will have be EASILY worth it.

I know what you are trying to say there, but I respectfully disagree.  Starlink needs to do more than pay for it's launch vehicle.

I don't think it will have any trouble doing that though.  Once fully deployed and licensed in different parts of the world I bet Starlink pays for all the SS/SH development quickly.

Starlink could be valued at $150-200B with maybe $5-10B a year in profits (based on 25 million customers globally at $100 a month).  $10B probably pays for all of SS/SH development. 

After year 1 you can start adding it to the Mars pile.

I think that is why SpaceX is raising money as fast as they want.

F9 is working really hard to get Starlink launched, but it seems that we are seeing limitations on quickly they can conduct launches.  Weather that is weather, processing boosters between launches, how quickly the ASDS can be cycled.  They can put resources into this and improve the launch rate, but F9 with 60 satellites at a time is going to be a limitation.  Starship is a must.

[Robotbeat]

The main argument against large, fully reusable rockets is there just isn't anywhere near the launch needed needed to justify them. Even new heavy lift expendables struggle to achieve a high enough launch rate (ask ULA or Ariane), and many industry people were skeptical that F9 would have high enough launch demand to justify even partial reuse. But FULL reuse? of a SUPER heavy lift rocket? the commercial launch demand is about 2 orders of magnitude too small...

If you make a list of things that need a crazy launch rate but that might actually be profitable, the list is incredibly short.

Basically:
1) Space based solar power (100GW requires about 1 megaton at GSO, roughly speaking... a one-time cost, though, every 20-30 years... 30 megatons every 30 years, or about 1 megaton per year could provide the entire globe's electricity requirements). We all know how Musk feels about that.
2) Space-based missile defense. 100,000 "Brilliant Pebbles," each 100 kg each, would mass 10,000 tons. Other concepts may be heavier, on the order of 30,000 to 100,000 tons. Obvious concerns about this idea, plus it requires the government to do it.
3) LEO megaconstellations. 40,000 satellites of 250kg each, would mass about 10,000 tons. Even better, may need to be regularly upgraded. Satellites could grow in mass and number.
4) Point-to-point hypersonic transport. This is about 3 orders of magnitude larger in potential market than 2 or 3 and even larger than #1. maybe 100 million passengers per year, works out to 100,000 flights per year (at high density seating). A flightrate that dwarfs even space-based solar power.

Almost nothing else, except for pure philanthropy, gets close to these.

SpaceX chose the latter two options. #3 is by far the most realistic of the options here. (I suppose there's also space tourism, but...)

So even if you're skeptical about Starlink being a massively profitable cash cow, as long as you think you'll probably be able to break even, then it's a good idea to do it if you care about fully reusable rockets. Because otherwise, you won't have the launchrate to justify a large, fully reusable launch vehicle. There just aren't that many options out there for why anyone would possibly need near the amount of launch capacity that Starship provides.

[Robotbeat]

Look even at Falcon 9's launch manifest this year. 10 of the 16 launches have been Starlink, and it's already September. 6 launches in 9 months is barely enough for even an expendable rocket to be justifiable as you probably need to make around 10 rockets per year for your factory to stay well-oiled. Even Falcon 9's partially reusable economics struggle unless you have the launch demand that Starlink provides.

So profit aside, that's why Starlink would make sense for SpaceX and why it makes sense for it to be so dang huge. SpaceX has to will more launch demand into existence to make fully reusable super heavy lift make sense.

[Nomadd]

Look even at Falcon 9's launch manifest this year. 10 of the 16 launches have been Starlink, and it's already September. 6 launches in 9 months is barely enough for even an expendable rocket to be justifiable as you probably need to make around 10 rockets per year for your factory to stay well-oiled. Even Falcon 9's partially reusable economics struggle unless you have the launch demand that Starlink provides.

So profit aside, that's why Starlink would make sense for SpaceX and why it makes sense for it to be so dang huge. SpaceX has to will more launch demand into existence to make fully reusable super heavy lift make sense.

Falcon did have some competition. Soyuz did pretty well, if not quite as reliable. Starship will be something else. It will (hopefully) create markets that nobody has come close to seeing. Asking whether Starlink makes Starship possible or the other way around is fun, but I think it misses the big picture.
 That was vague, admitted. I'm drinking Guinness, looking at a very bright Mars over the Starship pad and appreciating the fact that the world hasn't run out of new and wonderful things yet. Sue me.

[GregA]

First, define "break even".  Are you talking about the Starlink making enough revenue to cover its costs?  If so, then all you're doing is providing Starship a steady stream of launches, not funding any other part of its development.
....
One way or another, SpaceX operations will fund whatever ambitions Elon has for Mars.

(some creative naming...)... It's difficult really when Starlink and MarsX are the same thing, but one is made to generate money to make the other one possible. Starlink ends up needing to break even because its profits are channeled into MarsX. Or you do the accounting correctly and say Starlink (in 2030) makes a huge profit, all internally reinvested in MarsX.

At best I think you can say is you hope to get a Starlink ROI in 10(?) years ... profit after that...  and then decide if the entire development of SS/SH should fall under that or not.

[TheRadicalModerate]

<snip>
Tesla is a public company, with fiduciary duties and oversight that prevent it from destroying itself to satisfy Elon's need for Mars capital.  If he wants to maintain control, he's severely limited in the amount of stock he can sell, and it's not going to be throwing off any appreciable dividends for decades if it intends to compete with the big boys.  Cars are yet another business where returns on scale are essential.
<snip>

You do realize that Musk only gets pay in stock options? He gets more stock as Tesla market capitalization increases.

Just dug into this.  Turns out Musk no longer has a majority interest in Tesla, so my argument isn't valid.  It also appears that the 10 tranches of options are set up to be dilutive upon exercise.  So his stake in the company will go up, but not as much as you might think.

Bottom line:  I agree that Musk can finance Mars out of Tesla stock, as long as he's willing to reduce his stake in the company somewhat.

[TheRadicalModerate]

The main argument against large, fully reusable rockets is there just isn't anywhere near the launch needed needed to justify them. Even new heavy lift expendables struggle to achieve a high enough launch rate (ask ULA or Ariane), and many industry people were skeptical that F9 would have high enough launch demand to justify even partial reuse. But FULL reuse? of a SUPER heavy lift rocket? the commercial launch demand is about 2 orders of magnitude too small...

If you make a list of things that need a crazy launch rate but that might actually be profitable, the list is incredibly short.

Basically:
1) Space based solar power (100GW requires about 1 megaton at GSO, roughly speaking... a one-time cost, though, every 20-30 years... 30 megatons every 30 years, or about 1 megaton per year could provide the entire globe's electricity requirements). We all know how Musk feels about that.
2) Space-based missile defense. 100,000 "Brilliant Pebbles," each 100 kg each, would mass 10,000 tons. Other concepts may be heavier, on the order of 30,000 to 100,000 tons. Obvious concerns about this idea, plus it requires the government to do it.
3) LEO megaconstellations. 40,000 satellites of 250kg each, would mass about 10,000 tons. Even better, may need to be regularly upgraded. Satellites could grow in mass and number.
4) Point-to-point hypersonic transport. This is about 3 orders of magnitude larger in potential market than 2 or 3 and even larger than #1. maybe 100 million passengers per year, works out to 100,000 flights per year (at high density seating). A flightrate that dwarfs even space-based solar power.

Almost nothing else, except for pure philanthropy, gets close to these.

SpaceX chose the latter two options. #3 is by far the most realistic of the options here. (I suppose there's also space tourism, but...)

So even if you're skeptical about Starlink being a massively profitable cash cow, as long as you think you'll probably be able to break even, then it's a good idea to do it if you care about fully reusable rockets. Because otherwise, you won't have the launchrate to justify a large, fully reusable launch vehicle. There just aren't that many options out there for why anyone would possibly need near the amount of launch capacity that Starship provides.

You're still equating "fully reusable rockets" with "rockets that have to instantly migrate as far down the cost curve as possible".  Those are two different things. 

You don't need to trade payload for cadence to get better reusability.  Cadence does indeed get you further up the reusability curve (and therefore further down the cost curve) faster, but it doesn't get Starlink deployed faster--nor does it get any of the other things on your list deployed faster or cheaper, because they aren't ready to go in the next ten years.

Just to be clear:  I'm not skeptical about Starlink being a massive cash cow.  Indeed, my argument is that it needs to be as massive a cash cow as possible, and the best way to do that is to use the launchers as efficiently as possible (i.e., at max possible payload).