So I'm excited to see this tech develop but I wonder how much of a market there really is for super-heavy lifters. I can't wait to see a future version where they land the various stages rather than just dumping them into the sea. The first Falcon Heavy launch was super impressive.
SpaceX already has the Falcon Heavy and there have only been a handful of launches, primarily military.
I guess the argument is it'll open up new opportunities but will this really replace the Falcon 9 workhorse, which at this point is I believe the most successful launch system in history?
Won't someone make a fully reusable smaller launch vehicle that'll suit commercial needs?
FH had the problem that it had a comparatively small fairing compared with an upper stage that's "only" okay-ish for deep space insertions, so you can neither put really huge LEO payloads on it, nor can you give a deep space probe a really big kick stage to make up for the deficits of the upper stage.
Starship solves all these issues: The upper stage is more fuel efficient, and it has more room for really big payloads and/or kickstages.
> Won't someone make a fully reusable smaller launch vehicle that'll suit commercial needs?
Half of the people tried went bankrupt already due to F9: It is already too big for most payloads, so it does a lot of rideshare missions that pool multiple smaller launches together. It's very hard to compete with that.
So even if, for some reason, commercial customers don't really want to exploit the capabilites of Starship (ignoring the fact that multiple did already), SpaceX can again offer ride shares at a larger scale for F9-class payloads.
Starship might honestly have a similar payload issue with the weird door design, the way it hinges up means you need a more complex release plan than most which just pop straight off the front of the booster.
During the third test flight they also tested their weird side eject design for Starlink (or other flat pack style satellites) and the video looks like the door completely ripped itself apart.
The door design isn't final yet, there's no point in whining about it. They need tankers and landers for NASA contracts short term (neither of which require payload deployment), anything else is a nice to have that can be tinkered with on the side until it works.
I'm just getting really irritated by the amount of concern trolling surrounding SpaceX. Everything they do "must" have a gotcha, because clearly they cannot be as far ahead of the competition as they daily prove to be.
Yeah, I agree with you. Healthy skepticism is generally a good thing but now SpaceX has clearly demonstrated an unprecedented ability to solve a large number of insanely difficult problems. At some point, it becomes unreasonable to "yeah, but..." less difficult things like cargo doors.
> Healthy skepticism is generally a good thing but now SpaceX has clearly demonstrated an unprecedented ability to solve a large number of insanely difficult problems.
I'd add "again" into your sentence. They already did it before. Now they proved that they hadn't lost that ability yet.
The door is a major issue to using super heavy to deliver other payloads which is a goal long term and the need for a heat shield on the bottom makes it hard to make it fully open towards the front. Kind of need to have this fairly well sorted from the beginning because new designs mean new testing and certification which are expensive.
Here's where things get really counter-intuitive. If Starship lives up to even a fraction of its potential, it's likely Falcon 9 will be completely retired, because Starship will cost less to launch! The entire point of the Starship is complete and instantaneous reuse. The idea is to have it launching something up, landing right back into its launch pad slot, and then going again. The ridiculous cost saving potential is what makes all of this so much more revolutionary than most realize.
This isn't just a new big rocket. This is the most powerful rocket ever built, with the goal of launching it for less than the cheapest rockets cost. The current goal is to aim for $10 million within a few years, and then keep pushing it lower. For contrast, a Falcon 9 currently costs about $67 million to send 18 tons to orbit. Rocket Lab's Electron micro-rocket costs $7.5 million to send 0.3 tons to orbit. Starship can deliver 150 tons to orbit, a number that is planned to increase substantially.
The thing about space is that the potential is infinite, but it only becomes possible to start doing stuff once you get launch costs really low. Falcon 9 has brought launch costs down by orders of magnitude, but most people don't even realize this because unless you're a giant telecoms company or something, then $2000/kg doesn't sound that different than $50,000/kg --- wayyyyy too expensive for anything. But now imagine a world where you could launch things for $10/kg. Suddenly the entire universe opens up to expansion and exploitation, and life as we know it would basically change overnight.
Calculating the cost per kilogram for LEO with Starship gives me a new startup idea: small business (or even personal) interplanetary postal service.
It only costs $150 per kg in the near future to send objects into space with Starship; so I could, for example, send a Raspberry Pi (47 grams) into LEO for ~7 dollars (as long as I also had 149 tons of other objects from other people to send). A more useful use case would sending fully automated manufacturing facilities (probably either for semiconductors (https://www.nasa.gov/general/the-benefits-of-semiconductor-m...) or crystals (https://uofuhealth.utah.edu/newsroom/news/2017/07/proteinxl))
I'm not fond of speaking in specifics. The reason is that everything I am saying is, to the best of my knowledge, correct but the results sounds just so difficult to imagine. It's something like the opposite of cognitive dissonance. I know what I'm saying is true and it also leads to other things that I believe to be true, but the result is so difficult to imagine, that I have difficulty even believing myself. So it's quite odd!
But taking things in the abstract, single digit prices per kg to space would transform the Moon (or space in general) to just another place to build stuff and expand out into. Obviously there are technological hurdles but nothing particularly challenging, especially in contexts where we could have regular traffic flow to/from Earth. The possibilities of this rapidly enter into the sci-fi domain. But there seems no real reason for that "fi" suffix to really exist in these scenarios. So again, it's all just quite weird.
If things work out as they seem increasingly likely to, then are all living through what will be the most critical segue in humanity's history to date, and perhaps ever. So far as we know space really is the final frontier, and once we can start expanding outward into that, humanity's existence will be radically shifted, and our perpetuation as a species will also be all but completely guaranteed.
Large orbital habitats, like Elysium. Asteroid, moon, and Mars mining. Colonization of the aforementioned, space trade and shipping between Mankind's Interstellar Empire.
Demand will rise for energy production and terraforming, all kinds of things will boom. All of a sudden we'll have 100,000x more resources per man than we have now.
For one, Starship+Superheavy will enable launching of large objects like space telescopes without forcing object in question to be engineered with expensive, delicate, failure-prone folding mechanisms (like the James Webb Space Telescope was). Just build the thing as big as it needs to be and launch it in its final form (aside from minor folding bits like solar panels).
It could have similar impact on other scientific missions like rovers and probes. The ceiling for what’s possible is much higher when you’re not having to question the worth of every gram and square millimeter.
So JWST has (IIRC) a 6.5 meter mirror once deployed and yes, it was a challenge to develop that tech. Plus it added risk of failure. The Starship Super-heavy seems to have a max payload dimension of 9 meters. I imagine some buffer is required (ie it won't just allow a static 9 meter mirror) but I could be wrong.
So that's larger but not that much larger. Remember the JWST was a huge step up from Hubble's 2.4 meter mirror.
I expect NASA/ESA will take the opportunity to deploy even larger mirror by using the folding tech they've developed.
But here's the main point: these kinds of flagship missions don't support and sustain a commercial launch system. There are only so many JWST 2.0s that you can and will build, launch and deploy. Your bread and butter is going to be commercial communications satellites and other than deploying large constellations like Starlink, I'm not sure what the market is here.
I agree that it you can fold it up, sure, go for a bigger mirror. But widening the mass constraints will also be a game changer.
As for the commercial market, I think it opens up a new branch of materials science for high-end manufacturing by making zero G financially accessible. E.g. growing crystals or perfectly spherical things. If you or I already knew exactly how that would be valuable, we'd already looking for seed investment ;)
Also, let's conservatively estimate its payload to be 50 tons and triple the target launch cost to $30 million. That's $6000 per 10 kgs.
If there's a regularly scheduled flight to Korea/Japan, how many manufacturing and mining operations that are offline for want of a critical part would be willing to pay upwards of $6000 to take delivery of a 10 kilo part in <24 hours when they're losing a few thousand dollars per minute?
Since for telescope what matters is the surface size of the mirror (light collected is proportional to surface which is proportional to r^2), the 9m telescope is ~1.9x better than 6.5m one. This is a big difference. Not even counting the loses from a non perfect alignment of a folding mirror compared to a fixed one.
> There are only so many JWST 2.0s that you can and will build, launch and deploy.
I remember reading about building a virtual mirror out of hundreds telescopes, scattered around Solar system. It will allow getting kilometer-resolution images of Alfa Centaur system planets.
Building and launching that with Superheavy can become viable in the next 30-50 years.
Yes, there won't be as many customers purchasing 150-200 tons of lift, but that's the point of "rideshares". All that really matters with space launches is the cost per kg and if it's capable of lifting multiple payloads into multiple orbits, it'll have 10-15 customers per lift, not one. The current model has a kind of pez-dispenser but for chucking out multiple payloads.
There are purchasers for the full lift capacity too, like ISS modules and major telescopes.
If you think about this, it doesn't make a lot of sense because different satellites are going to sit in very different orbits.
Geosynchronous satellites are an obvious case where satellites will collect into a limited number of orbits but they vary on what point of the Earth they sit over. Also getting to geostationary orbit takes a lot more fuel so the rocket has less room for payload than, say, low EArth orbit. I'm not sure one rocket can launch a geostationary satellite above the Americas and above Europe in the same mission.
But you can't really launch a satellite in a polar orbit and an equatorial orbit in the same mission, for example. Likewise, how economic is it to deploy one at 150km and another at 250km?
Starlink is a special case because it's a related constellation of satellites where a number of satellites are in the same orbit.
The (unproven) target cost per kg of a re-usable starship, from even the most conservative source I could find, was under $300/kg[2]. The next cheapest, the Falcon Heavy, is around $2.3k/kg[1]. The cost difference is astronomical, and so low that it becomes viable send less payload and more orbital adjustment fuel, not to mention its (again, unproven) designed to be refueled in orbit. At that price, you could fly multiple refueling flights and still be under the cost of any other life provider.
> I'm not sure one rocket can launch a geostationary satellite above the Americas and above Europe in the same mission.
It can. Geostationary satellites are a certain distance above the equator. If they adjust their orbit a tiny bit lower than that they start to drift east, if they adjust their orbit a tiny bit higher they start to drift west. This process is called "repositioning".
Generally there is a tradeoff between how much fuel you spend on it and how fast the repositioning is done. So you can do it quick and then your sat will have less fuel for position keeping. Or you do it "slow" and then you preserved more fuel potentially extending the lifetime of your satellite.
But these are all done with tiny bits of fuel (compared to the fuel needed to put the satellite up there in the first place) because the delta-v involved is very small.
> I'm not sure one rocket can launch a geostationary satellite above the Americas and above Europe in the same mission.
Easily. Moving within an orbit is a matter of fine adjustment. For example, any stationkeeping that expands the orbit slightly will cause the satellite to "fall back" over time. Geostationary satellites are the best orbit for this, since every satellite in such an orbit essentially shares it with all others, differing only in position along the orbit.
I'm not sure this is true. if it were, there wouldn't be launch windows because any correction within a given orbit would be, as you call it, "a fine adjustment" yet we clearly do have launch windows.
Also if you're in a geostationary orbit to deliver one payload you have to leave that orbit to get to another geostationary orbit because there are other satellites in your way.
It is a fine positioning, which takes time. The launch windows get you to the right spot right away. Someone paying for a dedicated launch doesn't necessarily want to wait around to get their satellite operational. Someone launching for cheap on a rideshare might be willing to.
The vision is that the cost per unit of mass to orbit will come down massively with Starship, once it's launching like the Falcon. That will open up hitherto unimaginable missions and markets. And customers. It's all about the the cost!
It's not only about cost per kg but also maximum payload mass. If you can build bigger satellites then you don't need to optimize for weight as hard and can use cheaper components/standardize. Which means both launch cost and sat costs will come down.
Or entirely new capabilities get developed. Look how long it took for the F9 Heavy to get any business because fitting payloads really only got planned and developed after it demonstrated its abilities.
With the Starship, there will be single payloads of 100t or more - Elon is even talking about 200t in future versions. That is a total game changer. A station like the ISS could be set up much quicker. You could start designing real spaceships with e.g. ion drives. And a 100t payload might even cost less than currently a single F9 flight.
Like? What industry really needs things floating in space that are only constrained by cost to launch? I can see lots of science mission perhaps, but even that seems somewhat limited.
There have been tests of producing fiber optic cables (iirc) made in zeroG. There are other things as well that are way too cost prohibitive now, but might become viable opportunities with this type of capability.
Nobody has mentioned space-based solar power yet (https://www.nasa.gov/wp-content/uploads/2024/01/otps-sbsp-re...: "Launch is the largest cost driver..."), which would the cheapest (and currently only technologically feasible) route to turn humanity into a Kardashev Type 1 (or 2 if we construct a Dyson swarm) civilization (without really cheap fusion reactors).
There’s lots of things that are just too big and heavy and need launch vehicles like that.
It might be overkill for satellites, but space stations and habitats need the payload capacity of something like this to become anything resembling economical.
A bit like asking how many 30 story buildings are there when we first started building modern steel and concrete buildings. How many cathedrals could we possibly need?
If we're to build a Moon base, we're going to have at least this frequency of flights - really, I'd prefer to have a great margin on top of that, because Moon is much harder than LEO, and we might need more resiliency to safely explore.
Each flight to the Moon will likely need to involve 10-20 Starship flights (rough number) to LEO. So even if we're flying twice a year - and 6 month stay on the Moon right now looks like a pretty serious expedition - we need to have a Starship flight every ~10-15 days.
So even for a robust Moon exploration program we need as many Starships per year as the whole world was launching rockets per year just some ~20 years ago.
Mars launch window is every two years. It is very inefficient to launch at other times.
As for moon, I'm surprised with the estimate you have provided. Apollo needed just one launch for each mission. Even if SpaceX will do orbital re-fueling, it's just two-three launches, why would you need more?
BTW, the idea of getting heavy Starship to the moon and back is interesting, but at the end flying the vehicle optimized for re-entry far away and back is suboptimal. My prediction that they quickly will go to specialized LEO-LMO vehicles with LEO re-fueling.
> Even if SpaceX will do orbital re-fueling, it's just two-three launches, why would you need more?
Wikipedia says Starship weights 120 ton empty and 1320 ton fueled, plus 100 ton payload (approximate numbers). That means fuel weights 1200 ton. So to carry fuel to LEO to fuel up a Starship you need 1200 / 100 = 12 flights. You can change this number maybe 2 times into both directions, but I doubt you'll fuel Starship with just 2 or 3 flights of tankers. Would be glad to err here.
But you don’t have to fuel it fully to go to the moon and back.
It’s a bit hard to compare to Apollo since Apollo dropped stages at every step of the process, but it seems they used 70 tons of fuel in the third state of Saturn V for original trans lunar injection of 45 ton Apollo. Apollo itself was 2/3 fuel. So it’s ratio of 15 tons to 90 tons. I.e. 1/6.
> But you don’t have to fuel it fully to go to the moon and back.
Numbers: from LEO (low Earth orbit) to TLI (translunar injection) - approx. 3.1 km/s; from Moon hyperbolic to LMO (low Moon orbit) - approx. 0.8 km/s; from LMO to the Moon surface - 1.6 km/s, back to LMO - 1.6 km/s, total - 7.1 km.s . Numbers are optimistic, no errors. You absolutely, positively have to add delta-V for landing, Apollo LEMs, for example, had ~0.6 km/s, and also for docking - about the same, so total is already 8.3 km/s . This is more than Starship needs to get to LEO after SuperHeavy boosted it from the Earth.
So... no, if you're planning to get to the Moon anything near the Starship regular payload - 100 ton - you absolutely have to fuel it up fully.
> My prediction that they quickly will go to specialized LEO-LMO vehicles with LEO re-fueling.
Two comments here. First, we assume now SpaceX is going to have Starship HLS - human landing system - which doesn't go back to Earth, doesn't have flaps or heat shield, and is going to be used between low Moon orbit (LMO) and Moon surface - maybe one roundtrip, maybe more. Yes, for each following roundtrip HLS needs to be refueled.
Second, Musk mentioned "Moon base Alpha" in his talk. Having a serious Moon base makes it possible to produce some of propellants there. Oxygen is plentiful in the form of oxides on the Moon, and by mass it's 2/3 - 3/4 of the propellant load of the Starship, so it might be useful to produce it on the Moon.
5 years from concept to prototype, another 5 years to operational and then another 5 years to full capacity.
Starlink was super quick, but it's design started in 2014.
Iterations on existing concepts like telecom or imaging will be quicker, but truly new fields like mining or tourism are at least a decade out before they're using substantial lift capacity.
I'm skeptical because satellites, like pretty much any technology, tend to get smaller over time. I remember reading about how it was profitable for someone to buy up 4 geostationary slots and replace 4 satellites with 1 that was probably smaller than any of the 4 (because geostationary slots can be incredibly valuable).
There are large bespoke payloads (eg JWST) but these are inherently so expensive anyway the launch vehicle costs almost don't matter.
I'm not yet convinced there's a huge demand for super heavy payloads.
They’re expensive (and often delayed and over budget) in part due to the ridiculous demands of fitting everything in a small faring and reducing weight e.g. needing it to fold up and using expensive high strength low weight materials. Lessen those constraints and things get cheaper and easier to build with standard methods and materials.
> There are large bespoke payloads (eg JWST) but these are inherently so expensive anyway the launch vehicle costs almost don't matter.
If launch costs are going to be $250M, you need a budget of that order of magnitude to make a mission viable. At that point, you might was well spend anywhere from $50M to $1B on the payload because that's where your budget is. Or, to put it another way, only payloads with a $50M to $1B budget can afford to exist if the launch costs are of the order of $250M.
However, if launch costs are of the order of $5M, then missions with much smaller budgets suddenly become economically viable. And there are a lot more potential missions out there with $10M budgets than there are missions with $500M budgets.
Satellites get smaller not only because the tech gets smaller, but because launch costs/kg are so expensive, or so limited. Currently it's worth spending $10M to reduce your mass by 10%, if doing so means you can reduce your launch costs by $25M. Or, if doing so means you can double your onboard station-keeping fuel, and double the lifespan of the satellite.
If launch costs are less and available upmass is higher, your budget for engineering to reduce your payload mass is less, and so is the reason to do so.
There are a couple of great examples of this playing out in "reverse" with some missions that, at pre-F9 launch costs could only afford to be on a rideshare or small launcher and thus were expecting to have to deal with all sorts of limits, only to end up being able to afford a dedicated F9.
There was IXPE, which has been the smallest dedicated payload launched by F9, which otherwise would've had to launch on a much smaller, air-launched pegasus rocket to get to the right inclination. I recall that they were able to simplify some aspects of the satellite deployment due to the roomier vehicle.
There was another mission, maybe Psyche? where the original plan would've required the risk of testing a new kind of engine to get to its deep space destination, but being able to get a dedicated ride instead, that risk was eliminated, such that it was going to be able to get there even if the engine tests failed.
That's not accurate. In fact, currently the industry is going in the opposite direction.
The new space revolution start with cube sats and have now grown bigger. Starlink is the best example. Their v1 sats were small, v1.5 are bigger and v2 are even bigger.
There is lots of investment into bigger buses currently. K2 for example.
How is putting solar in space more useful than putting it on Earth? You still have the problem of a capricious atmosphere between the source of the beams and the place where you need the electricity. Sure, you can slightly modulate and do a few things, but the extra energy is extraordinarily unlikely to make up the extra costs even if the transport costs were 0.
Bezos predicts data centers in sun synchronous orbit so they always have solar power. The audio is poor but I consider the below video an excellent listen because Bezos outlines his vision of the future which is very different from Musk's.
Anything that will fit in a 30ft diameter faring weighing less than 150 metric tonnes. I’d love to see commercial space stations that can house large numbers of people in comfortable cabins so it’s more like a cruise ship than a submarine. Gotta power all those amenities somehow without diesel generators. But you could also put things like datacenters in orbit if the cost savings on power production made it worth while. Longer term you need a lot of power for resource extraction and processing and manufacturing. Would also make light sail propulsion of probes or deep space missions possible using lasers or beamed microwave power for ion thrusters so you don’t have to sacrifice mass for nuclear and aren’t constrained by how much wattage you can produce on board.
Is there any realistic way to dissipate the heat from a datacenter in space, even ignoring the transport costs?
Also, the limitations of being in an extreme hostile environment make it completely impossible to get cruise ship levels of comfort in outer space, at least with currently known technologies. The huge amounts of radiation and the bone density loss from being in low-g for any weight of time, coupled with the extreme acceleration on the way up and the way down, will also severely limit space tourism regardless of transport costs.
Ask yourself what is the market for a super heavy lift vehicle that may cost 2M to launch... even if it turns out 20M, thats much cheaper whatever you get today and still order of magnitude cheaper than yesterdays options.
Ever seen the incredible classic Moonraker? Larger satellites, larger rockets, it's about more at a lower cost. Bigger trucks, bigger ships, bigger lifters.
"The human remains aboard the lander won't be the first on the moon, as ashes of Gene Shoemaker, the founder of astrogeology, were buried on the moon in the late 1990s by the Lunar Prospector."
> In addition to the NASA science experiments on board the Peregrine lander are cremated human remains and DNA collected by two private companies, Celestis and Elysium Space.
> People hoping to memorialize their loved ones or colleagues pay the companies thousands to send a few grams of cremated ashes to the moon in metal capsules.
SpaceX already has the Falcon Heavy and there have only been a handful of launches, primarily military.
I guess the argument is it'll open up new opportunities but will this really replace the Falcon 9 workhorse, which at this point is I believe the most successful launch system in history?
Won't someone make a fully reusable smaller launch vehicle that'll suit commercial needs?