Falcon 9 Beats Shuttle’s Reflight Record, But Still Has A Long Way To Go

Put simply, the goal of any reusable booster is to reduce the cost of getting a payload into space. The comparison is often made to commercial aviation: if you had to throw away the airliner after every flight, nobody could afford the tickets. The fact that the plane can be refueled and flown again and again allows operators to amortize its high upfront cost.

In theory, the same should hold true for orbital rockets. With enough flight experience, you can figure out which parts of the vehicle will need replacement or repair, and how often. Assuming the fuel is cheap enough and the cost of refurbishment doesn’t exceed that of building a new one, eventually the booster will pay for itself. You just need a steady stream of paying customers, which is hardly a challenge given how much we rely on our space infrastructure.

But there’s a catch. For the airliner analogy to really work, whatever inspections and repairs the rocket requires between missions must be done as quickly as possible. The cost savings from reuse aren’t nearly as attractive if you can only fly a few times a year. The key to truly making space accessible isn’t just building a reusable rocket, but attaining rapid reusability.

Which is precisely where SpaceX currently finds themselves. Over the years they’ve mastered landing the Falcon 9’s first stage, and they’ve even proven that the recovered boosters can be safely reused for additional flights. But the refurbishment process is still fairly lengthy. While their latest launch officially broke the record for fastest reflight of a space vehicle that had previously been set by Space Shuttle Atlantis, there’s still a lot of work to be done if SpaceX is ever going to fly their rockets like airplanes.

The Space Truck That Wasn’t

When it was being developed, NASA envisioned the Space Shuttle as a sort of “Space Truck” that could fly weekly missions to low Earth orbit. Scientists and engineers wouldn’t need to wait years before they had the opportunity to fly their payloads, just strap it down in the back of the next scheduled Shuttle flight. The cavernous cargo bay ensured there would always be plenty of room for whoever wanted to come along for the ride. NASA hoped to skip the airliner analogy completely, and compared their reusable spacecraft with over-the-road shipping.

Flight preparation as imagined by early Shuttle concept art.

But of course, that never happened. The Space Shuttle was an extremely complex vehicle, arguably the most intricate piece of machinery humanity has ever produced; and it’s been said that even by the end of the 30 year program, engineers were still learning something new about it after each flight. With so many components in the system, it was difficult to predict what kind of repairs each orbiter would need when it got back to Earth, making the refurbishment process between flights far more expensive and time consuming than NASA had anticipated.

In the end, it usually took months to prepare each Shuttle for its next mission. There was an expectation that the time between flights would have been reduced as ground teams gained experience on the vehicle, but in actuality, it was the opposite. The fastest turn around between two flights of the same orbiter was in 1985, between the first and second flights of the newly constructed Atlantis, with 54 days elapsing between missions STS-51-J and STS-61-B.

As the program went on, and especially after the tragic loss of Challenger in 1986, the changing culture of safety at NASA demanded ever more stringent inspections on the nation’s only human-rated spacecraft. Inherent flaws in the winged orbiter were partly to blame, as the chances of crew survival in a number of failure modes were considered unlikely at best. Failure was truly not an option with some of the Shuttle’s systems; either they worked perfectly, or seven astronauts would almost certainly die. So they took their time.

Back to Basics

Despite being one of the most modern rockets in the world, the Falcon 9 is a very simple design compared to the Space Shuttle. There’s no winged orbiter, no solid-propellant boosters. While it does have nine engines on the first stage, they’re much smaller and less complex than the three colossal RS-25 engines used on the Shuttle. The Falcon 9, with the exception of the hardware added to facilitate its propulsive landings, could be thought of as a return to “classic” rocket designs of the Space Age but with the benefit of modern construction techniques.

Still, the roughly nine-minute suborbital trip to space and back that the first stage of the Falcon 9 makes is no pleasure cruise. Even with such a comparatively simple booster, it takes a lot of work to ensure it’s ready for another flight. While SpaceX has been unusually tight-lipped about how much money and effort goes into returning a previously launched first stage to flight, we do know that on average it takes them months to complete the work.

A recovered Falcon 9 being prepared for its next flight.

That being said, the pace is certainly picking up. It took just shy of a year to examine, refurbish, and validate the first reused booster back in in 2017. In comparison, the Falcon 9 that recently launched South Korea’s ANASIS-II satellite into orbit was the same rocket that returned the United States to human spaceflight just 51 days prior. After holding the record for 35 years, the Space Shuttle has officially lost the title as the most rapidly reusable space vehicle in history.

Lofty Goals

The Falcon 9 might have beaten the Space Shuttle at its own game, but SpaceX still has to pick up the pace considerably if they’re ever going to meet Elon Musk’s stated 2019 goal of relaunching a first stage booster within 24 hours of landing.

We intend to demonstrate two orbital launches of the same Falcon 9 vehicle within 24 hours no later than next year. That will be, I think, truly remarkable to launch the same orbit-class rocket twice in one day.

We’re now halfway through 2020, and SpaceX is still nowhere near that sort of launch cadence. The next manifested launch of a reused Falcon 9 might actually break the 51 day record they just set, but only by a few days. Just as NASA underestimated the difficulty in flying weekly Space Shuttle missions, it’s entirely possible, perhaps even probable, that SpaceX will never hit their goal of launching twice in 24 hours. The argument could even be made that there’s not sufficient industry demand for daily flights to low Earth orbit.

The first recovered Falcon 9 in 2015.

But that doesn’t mean there isn’t room left for improvement. Within just a few weeks of the first successful landing of a Falcon 9 in 2015, SpaceX made a thorough examination of the flown booster, returned it back to the launch pad, and fired up all nine engines to test their performance.

There was no attempt to actually launch the booster, but that was never the point. Before pinning the future of the company on the concept of reusability, they wanted to answer a simple question: was it actually possible to refuel and reignite a rocket that had just returned from putting a payload into orbit.

Clearly, they were happy with the results of that test. It was an important step towards proving the Falcon 9 could be reused safely, and lay the groundwork for the post-landing examination and refurbishment process that SpaceX is still refining. The challenge now is streamlining the process so they can get the turnaround time as low as it was on that first attempt. It would mean once again succeeding where NASA failed, but SpaceX seems to be getting pretty good at that these days.

77 thoughts on “Falcon 9 Beats Shuttle’s Reflight Record, But Still Has A Long Way To Go

  1. This is a bit unfair to the Shuttle. I’m not saying that the Shuttle was a great idea or design or anything (I’m also not saying it wasn’t, so don’t take it either way) but the Falcon 9’s payload to LEO was like 15 metric tons when reused. The Shuttle’s was 27.5 metric tons, and that’s treating the Shuttle like it’s a glorified rocket and ignoring the humans it was transporting and the significant weight cost that implies.

    I’m really in awe of how reusable the Falcon 9 program has been (especially with the double-fairing catch recently!) but this is a bit apples-and-oranges. I’ll be far more impressed when Starship’s reusability passes the Shuttle’s (which I fully expect to happen, although at this point I’m just waiting for Starship to stop exploding).

    1. Are you aware that Falcon Heavy is > 50 tons to LEO reusable (and 64 tons expendable if necessary)?

      And Space Shuttle had very serious (and this is not a hyperbole) design bugs regarding human safety like not having any kind of launch escape system or abort solutions for many parts of flight.

      After those solid fuel candles were lit there was no escape, no contingency for anything going wrong until they burnout.
      Just the very first flight had ejections seats for pilot and copilot.

      1. “Are you aware that Falcon Heavy is > 50 tons to LEO reusable (and 64 tons expendable if necessary)?”

        Yup. I’m also aware that Falcon Heavy’s only launched 3 times. And this article isn’t about a Falcon Heavy breaking the Shuttle’s relaunch rate.

        I mean, I guess you could claim that the side boosters have probably outdone the SRBs? Maybe? I don’t actually know how long it took the SRBs to be refurbished and reflown, but the Falcon 9s pretty much have had to beat them, I guess. None of the actual “Falcon Heavy”-specific parts (the center core) have ever reflown.

        “And Space Shuttle had very serious (and this is not a hyperbole) design bugs”

        Please note my second sentence. I’m not trying to either tout or disparage the Shuttle, I’m just trying to point out that comparing the turnaround time on a vehicle like the Falcon 9 to the Shuttle’s turnaround is a bit silly. They’re not remotely similar vehicles in terms of lift capabilities.

        You could definitely make an argument for the Falcon Heavy, although I’d say Starship is the one that would really surpass it. If you treat the orbiter *plus* its payload as the Shuttle’s lift capacity, you’re talking about 100 metric tons to LEO, which is Starship-land. It’s not *entirely* fair to include the orbiter, but it’s not entirely fair to exclude it either, given that the orbiter and its capabilities were actually used.

        1. and all of this is without taking into consideration that the orbiter was, and remains, the only craft that can both launch and recover an orbital object in an enclosed bay and subsequently return it to the surface intact, or bring it to the enclosed bay for maintenance using lighter-duty Vacsuits than the full EVA rigs (was this ever done? I’m not up on the flights enough to know)

        2. I think payload capacity is 95% orthogonal to the point of this article. It’s the airliner vs. rocket comparison.

          I grew up watching the shuttle. It’s definitely got a place in my heart… for a long time that’s what US spaceflight looked like. That’s what I thought of when I thought about humans in space.

          I also watched the first shuttle explode almost live (on a class conference call with NASA at the time) … I had parts of the second rain down less than 2 miles from where I was at the time.

          As much as I want to like the shuttle, I have to say I totally believe it was a serious mistake. It was so costly and limited (or maybe treated too gingerly) that it restricted our growth into space for many decades. We were running to LEO and back, over and over again. We should have an established moon base by now.

          I was very sad when they stopped flying the shuttle. I understood the basic reason, but it definitely was a loss. It’s ridiculous we were out of human spaceflight that long. I just think if step outside my emotions, the thing was a bad idea.

      2. “After those solid fuel candles were lit there was no escape”
        not quite, they have these hatch-looking things at the bottom, their purpose is to be blown off should they ever need to cut thrust on the SRBs.
        Of course the fuel will still happily burn, but the engine will have next to no pressure and thus next to no thrust.

        The problem is that you still have to separate everything, get the Shuttle into an attitude that allows egress and then evacuate. All of this takes a lot of time, so chances of success in a real scenario are not very high, especially if they don’t have the altitude.

        1. They only had the bail-out option after the Challenger disaster. Before STS-51L, the shuttle was a certain “Loss of Crew and Vehicle” scenario until roughly 350 seconds into flight (way past even SRB separation). Basically, if it didn’t make it to orbit, the crew died. Past STS-51L, they invented the bail-out option and some other abort modes, but the chances of actually making those work were considered slim to hail-mary by most of those involved.

          1. There was also the duffel bag transfer system.
            Lesser crew members zipped themselves up in a large “duffel bag” with a breathing system, and waited for a crew member in an EVA suit to carry them to another shuttle in orbit that came to rescue them.

      3. What people these days seem to forget is that you can´t make an omelette without breaking a few eggs. Being a pioneer has always been fraught with danger, it comes with the territory. The shuttle was designed with that in mind. Attitude changes is what killed it off.

    2. You do realize the Shuttle was designed in the late 60’s as a stop gap untll NASA could design a better shuttle?

      The problem is our aerospace companies are brain dead and couldn’t do and thus forced NASA to rely on the shuttle mere than they wanted to,

      1. “our aerospace companies are brain dead and couldn’t do and thus forced NASA to rely on the shuttle mere than they wanted to”

        Not having a better shuttle wasn’t the fault of the aerospace companies. It was political. NASA didn’t have the budget to develop anything else and politics didn’t give them the budget to do so. NASA came close several times, but each time the budget was cut and projects ended just before they started bearing fruit.

        1. Yup. Nearly every failure of a defense contract can be traced to scope creep that occurred after contract award. The Shuttle compromises that were made to support certain military applications were no exception. (I forget the exact details, but the Shuttle was required to support a military mission that was something along the line of launch, deploy payload, and return to origin in only one orbit. Achieving this fundamentally compromised the rest of the design.)

          1. The military compromises with the Shuttle started before funding ever happened: the Shuttle’s military use is pretty much what allowed it to be funded in the first place.

            That’s why I don’t really talk about the Shuttle in terms of “bad” or “good.” NASA did what they had to do to get it funded, and in the end, it actually did do amazing science, so I’m not sure I can exactly criticize them. In an ideal world of course you’d make different decisions, but in the real world, I’m not sure there actually were better options.

  2. “You just need a steady stream of paying customers, which is hardly a challenge given how much we rely on our space infrastructure.”

    Wouldn’t that be space ultrastructure?
    e.x. infrared v. ultra violet
    B^)

    1. Lol very true.

      Musk is a blow hard. By his claims we should already be traveling by Hyperloop (which isn’t feasible), should have conducted a manned mission to Mars, and should have submersible cars all by now.

      He throws crap at a wall.

      1. Yup. Anyone who has been following Musk shouldn’t be surprised that they’re nowhere close to something he said he’d be doing in only a year. Just like the snake oil he keeps on peddling with FSD. That whole house of cards is falling down – even the most ardent Tesla cultists now are telling other people not to bother with FSD because it’s overpriced and underdelivering, so Tesla’s funding stream to do further FSD is drying up, leading to Elon jacking up the price of FSD along with delusional claims that it’s worth over $100k per vehicle, leading to even fewer people purchasing the option.

        1. There is so much room for criticism of Musk, but you cannot, in good faith, deny that SpaceX have completely revolutionized the launch industry and have essentially lapped the competition. He’s made some pie-in-the-sky claims about Starship/Superheavy, yet they have already developed a very promising engine for it that is pretty cutting edge. It remains to be seen how much work is left on the engine, but it definitely works long enough for a short lateral flight.

          The point is that, in the midst of all the hyperbole, it’s still difficult to overstate how wildly successful and disruptive SpaceX have been. They could scrap SS/SH, rest on their laurels, and this would still be true.

      2. A lot of that crap sticks though. It could also be argued that humanity has been dragging its feet and there should indeed already have been manned missions to Mars. As for the hyperloop, it is indeed at least impractical and at most impossible to do on earth. But everything Musk does, literally everything must be seen with Mars in mind.

        With SpaceX and Tesla it is undeniable that 2 of hist companies are severely disrupting 2 major industries. So does lighting often strike twice, or are you just hating for other reasons?

        Odd of there being a functional hyperloop system somewhere in the solar system is an infinitely better bet than betting on you doing only half of what Musk has every accomplished.

  3. >But there’s a catch.

    There are far more catches than just the recycling time. The time and cost to re-launch is of course a major factor, but far from the only one.

    For one, the re-usable airplane doesn’t lose payload capacity (significantly), whereas the rocket does by virtue of having to spend extra fuel for the return maneuvers.The SpaceX first stage imparts 60% less kinetic energy on the second stage to save fuel for return, so it can only lift small payloads or reach low orbits. If the rocket is only going to be used a handful of times, then the famous rocket equation dictates that a rocket that was designed and built for that exact payload and orbit would be exponentially smaller and thereby cheaper to construct.

    So the question is, how many times they can recycle the same big rocket that performs the same as a small rocket? There comes two questions: launch reliability, and customer profile. SpaceX itself has estimated that the re-usable rockets start to pay after 10-12 launches, but the industry average failure rate is around 2% so there’s about a 19% risk that the same booster (assuming it was brand new every time) would fail before it pays off. About one in five boosters would not make ends meet, and that puts additional requirements on the remaining boosters.

    The other is the flight profile, because you can’t assume every customer or group of customers wants to put a small satellite to LEO. When a higher load or orbit demand comes, SpaceX expends the booster because they don’t have the fuel to return. Ideally you’d have a bunch of old boosters lined up that you can send for a final journey, but remember that each booster hast to operate MORE than 10-12 times – much more – so only a small portion of your customers can demand anything but a light launch. Otherwise you have to start expending boosters that haven’t broken even yet, and that’s a losing proposition.

    So, SpaceX has built a very big rocket that performs like a small rocket, and only breaks even financially if it is used like a small rocket, and even then it probably requires it to reach an unprecedented reliability score that rivals the Russians. It’s hard to see how this is going to work in the end – other than being a big smoke and mirrors show that is kept up because NASA is paying.

    1. >”SpaceX itself has estimated that the re-usable rockets start to pay after 10-12 launches”

      Care to give a citation for that statement?

      I haven’t heard about them making that statement.

      1. Replying to myself:
        I decided to check some number, and I personally don’t trust these numbers, but I will cite them:

        “The boosters, which SpaceX has often landed, make up about 60% of the total cost of the rocket, or about $37 million.”
        https://www.cnbc.com/2020/04/16/elon-musk-spacex-falcon-9-rocket-over-a-million-dollars-less-to-insure.html

        “Cost per launch: : New: $62M (2020), Reused: approximately $50M (2019);”
        https://en.wikipedia.org/wiki/Falcon_9

        I’m going to assume that the 10 flights is with the 12 million dollar discount

        So, if it takes 10 launches to pay for the 37 million dollars, that’s recoup at only 3.7 million per flight. Plus the 12 million discount puts the financial difference at 15.7 million dollars, per flight.

        But, if the booster costs 37 million dollars, that would mean that refurb costs more than 20 million dollars (more than half the cost to build new).

        Does it really cost them 20 million dollars to refurb a booster?

        Will it continue to cost them 20 million dollars?

        What if they can reduce the Refurb cost to 17 million? Then it’s only 5 flights, which seems pretty reasonable to plan to cut costs by 15%.

          1. That’s not the whole story. The Block 5 engines themselves are designed to run for 10 flights without overhaul or major refurbishment, which is what they’re counting there: $1-2 million to cycle them back to service. The whole booster still needs a whole lot of work, plus the cost of recovering and transporting it.

          1. Your statement is wrong. The total refurb cost is $1-2 million. This encludes the airframe, engines, avionics, transportation, labor, storage, infrastructure, and other odds and ends.

          2. It is 24 pages on my phone and it is not some random forum on the internet. Nasaspaceflight. Com is one of the premier space related web sites and resource. A large number of people who are actually in the industry go there to share ideas and discuss space topics. You do not need to read the whole thread as the latest information is in the last few pages. Yes they have gotten the cost of refurbishment down to $1-2 million, but nothing I say will sway your anti Elon stance. You can either set your biases aside or not, but I done arguing with a wall.

          3. >Yes they have gotten the cost of refurbishment down to $1-2 million,

            Look, if that was actually the case, you could point to a direct quote. I’m not taking a bunch of speculation from a bunch of random people on a forum, no matter how “premier” you think it is.

        1. Correction: Block 5 engines are designed for 10 launches.

          Obviously, if the booster itself tips over and gets damaged on recovery, or gets hit by lightning on the way, or has a structural problem due to metal fatigue, it costs a whole lot to re-build.

          1. Again your statement is correct but incomplete. The entire first stage is designed for a minimum of ten uses. During the refurbishment, very little is replaced. It is mostly inspections.

          2. >During the refurbishment, very little is replaced. It is mostly inspections.

            IF there is no damage to the booster stage or the engines. If there is even a beginning of a crack in the aluminum structure, it has to be replaced because the next launch will tear it apart.

            $1-2 million could be the minimum cost of doing nothing, but that’s not the average cost to recycle the average booster.

          3. Yes it is the average cost of refurbishment. You seem to forget that they only need to rely once to save money over building new boosters for every launch.

    2. Check your numbers. Refurb cost of the first stage is about $1-2 million. They are saving money over a new stage on the 2nd flight. Also, if the rocket carries twice what you need but cost half as much as the next guy, who would you choose? It is kinda like using a 18 wheeler to deliver a pallet. As long as as the customer pays enough to cover the cost and some profit, who cares if the truck is very under utilized.

      1. That’s not the sole cost. You have to maintain the infrastructure to catch these rockets out on the sea and bring them back, as well as the entire infrastructure to refurbish them. The direct cost to rebuild one isn’t the whole cost.

          1. I can’t believe we still can’t edit posts… That should have said:

            That would be true if the van company was not charging twice as much as the 18 wheeler provider…

          2. >That would be true if the van company was not charging twice as much

            That’s true for US corporations – but check out what the Indians are doing.

          3. The F9R costs $28 million to lunch. That includes the first stage refurb, new second stage and fairings, integration, facilities, personal, etc. Add whatever profit the market will bear. What cost are the Indians launching for?

            https://www.google.com/url?sa=t&source=web&rct=j&url=https://www.cnbc.com/amp/2020/04/16/elon-musk-spacex-falcon-9-rocket-over-a-million-dollars-less-to-insure.html&ved=2ahUKEwjAzsKotu7qAhWxc98KHVkEBHoQFjATegQIBhAB&usg=AOvVaw2FRq_WBDNgX649e-Hojpff&ampcf=1

          4. The article you link doesn’t support your claims. It says it costs $62 million per launch. The article is written in a contradictory fashion, but it makes a direct quote that the company can “bring launches down to below $30 million per launch.” which is not an achieved price but a promise. The booster according to your article is quoted at a price of $37 million and 60% of the launch cost.

            https://www.engadget.com/2017-04-06-spacex-is-saving-a-ton-of-money-by-re-using-falcon-9-rockets.html
            >”The historic SES-10 mission rocket’s first stage cost half as much as a new one.”

            Assuming that’s half of $37 million, it’s $18.5 million, not $1-2 million. I very much doubt they’ve suddenly improved from 50% to 97% savings in just three years – that’s just implausible.

            Meanwhile, ISRO was quoting launch prices at $13 million for payloads up to 3,250 kg LEO back in 2016 (latest info I found in a hurry), which when scaled up to Falcon 9 payload capacity is roughly $62.4 million to get the same mass to LEO, which means the Indians were asking the same price with expendable rockets four years ago as SpaceX is doing now with the re-usable rocket. Only, SpaceX has to carry five times more satellites at a time to reach the same economy while the Indians can launch one-by-one as the demand comes, so their customers don’t have to wait. As the satellites get smaller and more efficient, the demand for even smaller launches is rising and SpaceX will be left behind as the mammoth of the industry.

            This goes back to the original point: a single-use rocket designed for the exact payload and orbit is exponentially smaller and cheaper to make.

    3. > then the famous rocket equation dictates that a rocket that was designed and built for that exact payload and orbit would be exponentially smaller and thereby cheaper to construct.

      Wrong. According to the rocket equation a rocket designed for a specific payload would be *linearly* smaller. The exponential growth is relative to the delta-v, independent of the payload or if the rocket is reusable or not. The mass only enters in that equation as a ratio between the initial and final mass, that is, the rocket’s efficiency.

      But theory isn’t everything. In practice there is economy of scale in rocket construction: large rockets are more efficient (large payload/GLOW due to things like avionics, turbo-pumps, minimum gauge issues, etc not scaling linearly) and significantly cheaper per kg to launch than smaller rockets. When you add the cost of designing and certifying all those different sizes of rockets, a single big one that can be launched cheaper with smaller payloads starts to make a lot of sense.

      > SpaceX itself has estimated that the re-usable rockets start to pay after 10-12 launches

      Source? SpaceX has said that even a single re-use is already worth it. The cost of a flight proven first stage is “substantially less than half” the cost of a new first stage according to Shotwell. A first stage half the size of the current one would cost substantially more than half the cost of the current first stage.

      > The other is the flight profile, because you can’t assume every customer or group of customers wants to put a small satellite to LEO.

      SpaceX is capable of first stage landing while putting large satellites (4~7 tons) in GTO. The rare launch larger than that can either go in Falcon Heavy or reuse a flight proven booster for the last time.

      > So, SpaceX has built a very big rocket that performs like a small rocket, and only breaks even financially if it is used like a small rocket

      Half a very big rocket is still pretty big. SpaceX was already posting profits and undercutting the competition before succeeding on re-use. Now they are even under-bidding Pegasus, a small that is kept up because NASA is paying.

      SpaceX has been the lowest bidder in every NASA contract it has earned. Likewise with commercial contracts of course. And they spend heavily on R&D (over $1B among F9 block 5 and Falcon Heavy) with money they could be pocketing as profits. Why would they do that on their own dime if they are just making a smoke and mirrors show?

      1. That’s true if you maintain a constant payload fraction, but the SpaceX rocket isn’t doing that. They have a big rocket which isn’t filled to capacity, except when they do an expendable launch. The rocket is built for twice the payload in case it’s needed, so it’s built bigger and stronger than necessary to support the larger load, and this is just more dead mass they’re dragging along.

        When you scale the rocket down, and lose the landing gear, extra fuel, and all the additional structures required to support all that, your effective “payload” fraction gets smaller and the rocket shrinks more than linearly. Furthermore, the rocket equation is only concerned with delta-V, not air drag, which is also reduced when the rocket is scaled down, so you need even less fuel in the first stage to get to the fun part.

        >and significantly cheaper per kg to launch than smaller rockets

        That’s true for expendable rockets with little to no dead weight to launch, but that is not the case here. ISRO/Antrix is asking roughly equivalent prices per kg to LEO as SpaceX with a rocket that is five times smaller and not re-usable. For their cost breakdown, the rocket itself is actually slightly less than half the cost to launch with the ground crew and infrastructure taking the other half. Since they have a number of rockets already lined up, there’s no turn-around time, and being a smaller rocket the customers don’t have to wait for the bus to get full – they can buy the entire thing and launch whenever there’s a suitable window.

        While SpaceX might plausibly save some on the rocket, they have an additional handicap in having to maintain and operate the whole recovery infrastructure, which adds up to the system cost. The more of the rocket they intend to recover, the more elaborate and costly it becomes, while again losing payload fraction to extra stuff and fuel on the rocket.

      2. >When you add the cost of designing and certifying all those different sizes of rockets, a single big one that can be launched cheaper with smaller payloads starts to make a lot of sense.

        The expense of designing and certifying is a one-time cost that is relevant if you only intend to launch a handful of rockets. If your aim is to build rockets for thousands of launches, then making rockets in different sizes is more efficient than trying to operate a one-size-fits-all rocket, because simple disposable rockets are much quicker to make.

        On another note, the fact that you are subjecting the rocket to the vibrations and fatigue stresses of multiple launches mandates that you use thicker everything. For one launch, you don’t care if the structure would break after five minutes, because the launch takes three minutes, and that’s enough of a safety margin. When you try to approach infinite fatigue life (which isn’t reachable with light alloys), you run into diminishing returns where the structure gets heavier and heavier with little improvements in service life. That fact alone makes the re-usable rocket much heavier and less efficient.

  4. I really do wonder about the sustainability of the whole enterprise. On the face of it, the numbers appear worrying.
    SpaceX started 2020 with 8 F9s in the stable. In the 12 launches so far this year they deliberately destroyed one and lost two others, leaving 5 in the active fleet.

    Losing a third of your fleet in just 7 months might sound bad enough, but look at where they went:
    Fully two-thirds of those flights (8) were non-revenue-generating internal program costs (Starlink 2-8 and Dragon abort test).
    Two were ISS (Demo-2 and CRS-20). So (maybe) paid for themselves.
    Only two were real paying satellite customers (discount a few rideshares in the starlinks)

    So, so far this year, they’ve paid for a dozen launches, so roughly $400-500M, plus around the same on the Starlink payloads that were launched. Plus their routine burn rate and the Starship program eating about the same again means they spent $1.5-2B so far this year. Revenue was at most a quarter of that.

    And this after a horrible 2019 saw half the revenue of the previous year.

    No wonder Elon went back to the well for another gigabuck of investor cash earlier this year. If they weren’t actually putting hardware in orbit it would look like a massive con job. As it is it just looks like a very fragile house of cards.

    I really, really hope everything goes well in the remainder of this year for SpaceX, because as sketchy as the whole endeavor looks, I still think it’s the West’s best chance in getting a viable continuing space presence. Elon just has to convince Thiel and the rest of the sillycon mafia to continue to shovel cash in his direction.

    1. Every business is limited by the market they serve.

      I’m pretty sure NASA paid for the Crew Abort (directly or indirectly).

      I’m not sure what the cost per crew flight is, but I’m pretty sure it’s way, way more than what the Rocket costs (I’ve no idea what Crew Dragon costs).

      Starship is a huge-risk/huge-reward project, and may well cripple the company.

      Starlink is their attempt to create their own revenue outside of other payload producers. If they can get paying customers, then it’s an investment (otherwise it’s a waste). But less likely to cripple SpaceX since they could spin it off, sell it, or just scrap it and slowly dig themselves out of the financial hole.

      If SpaceX did nothing but simply launched rockets, then they will be financially afloat forever (though perhaps a bit unstable).

      1. Starlink is doomed to fail because it’s not a good solution technologically or practically. It will run into legal trouble even if simply for the amount of junk that starts raining down from the skies with the inevitable 1-2% of starlink satellites that malfunction and become unresponsive.

        If you have 1,000 satellites and 10 fall down uncontrollably every year over a random point on the earth, the probability of hitting a human-inhabited area is roughly 33%. A starlink satellite will fall on someone every 5.5 years with 90% probability.

        1. >”hitting a human-inhabited area is roughly 33%”

          You mean “hitting dry land”

          The majority of which is not “human-inhabited” or with such low population density that it will not matter.

          You have a higher chance of having a Airplane Ice Poop crashing through your roof than of a starlink sat falling through your roof.

          “Half the people in the world cram into just 1 percent of the Earth’s surface”
          https://www.bloomberg.com/news/articles/2016-01-06/half-the-world-lives-on-1-of-its-land-mapped

          1. I don’t agree with THE DUDE, but 33% is probably right for 10 attempts to hit “human inhabited area”. Hitting dry land would be about 33% with just one attempt.

          2. Humans inhabit about 4% of the land area. Calculate from that.

            And Starlink satellites fly denser near the north and south poles, but not over the polar regions, so that reduces the area they may fall over, and makes them denser over places like North America and Europe anyways.

          3. Sorry, about 4% of the earth’s surface, not just land area.

            We farm about 40% of the land area. Whether that counts for “inhabited” is your choice, but don’t be surprised if you find pieces of Starlink satellites in your bag of doritos.

        1. Elon Musk is a conman who sells fantastical promises and drags them on for so long that everyone forgets what the original promise even was. The targets and goalposts shift continuously and the public is paying him through cheap government loans and investments from NASA, DoE, etc. without getting anything in return.

          What. Did you really think YOU would be going to Mars in your lifetime?

  5. Interestingly, I’m better on a bicycle than I was 40 years ago, too, but I’m not kicking sand in the face of my former childhood self. The shuttle worked, despite its imperfections, as the defacto beginning of full-circle reusable U.S. space cargo and deployment, and despite a complete absence of internet billionaire celebrities.

    Exceeding what came before and retired is a baseline, not an achievement.

    1. It’s a miracle the shuttle worked as well as it did really, I’ve got a STS systems reference from back in the day and in that they promised Concorde, what flew was in comparison the TU-144

    1. Exactly. Given that they’re not limited to just one booster, they can pipeline the refurb process. Still, it’s in their interest to make the process as efficient as possible, which generally also reduces the amount of time it takes to do it.

  6. Elon has always projected completion dates that are well ahead of realistic, that is true.
    BUT he almost always delivers. only a few years ago, nasa couldnt launch their probes. no one talked about reusable rockets much let alone ones that could land themselves standing up.
    tesla made ev’s that people actually want that have performance and range. power walls provide needed storage to make the most of reusable energy. the only place this kind of stuff was being talked about before Mr.Musk were tech blogs and popular mechanics 8).
    Yes it took longer and Yes, a lot was funded by taxpayers BUT so is EVERY OTHER PLAYER IN AEROSPACE. At least Musk shows some return on investment. And his stuff works eventually and then improves as it is constantly improve.

    results? nasa can support its own missions without relying on foriegn players. Auto makers are finally putting actual effort and money into Ev’s and people in the farthest corners of the world will have the greatest tool they get to better themselves..knowledge. Mean while, there is a renewed effort worldwide to establish a moon base. And a mission to mars is now more than just a wish list item.

    Is he perfect? No. He could have done something slightly more ridiculous than “Elon Tusk”.

    Still, Of all the tech billys, he seems to be the most focused on solving actual problems whereas bozo, cuck and the rest seem only intent on profits and making their own lives better.

    and cmon! even you critics gotta admit testing the payload capability using a convertable with a “astronaut” at the wheel…
    Smooth…

    1. > no one talked about reusable rockets much let alone ones that could land themselves standing up.

      Ahem…

      https://en.wikipedia.org/wiki/McDonnell_Douglas_DC-X

      Elon Musk was handed down all this technology, and he bought the engineers who were developing the rocket engine technology behind Falcon 9 for JPL/NASA already. He simply took the projects over and combined them into one launch system that the NASA was unable to make because of funding restrictions. By re-dressing the whole thing as a “private” space launch technology, together they were able to secure a new round of funding from the government.

      Elon was given the role of the poster boy for re-hashing old NASA research and pretending that he’s blazing a new trail.

      1. That doesn’t really sound like my experience working there a few years. A large portion of the engineering was and is still done by relatively young but talented people working pragmatically from first principals. No one handed me 40 year old reference materials to copy from. Most NASA meetings boiled down to, “this isn’t the space shuttle in the 90’s, we’re going to use modern technology from at least 2005 and you’re going to like it!” I used the right hand rule a lot the week I had to update some reentry guidance code because the crazy gnc guy crushed his leg scrambling up a mountain over the weekend. Another time I even used calculus once! Turned out it wasn’t necessary though…

        Maybe the economics can be optimised better, but SpaceX seems to be doing fine financially. All I know is that launching a rocket and capsule you helped build and controlling it in orbit from mission control is a unique thrill.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.