SpaceX Drops The Ball On Catching Fairings

You don’t have to look very hard to find another rousing success by SpaceX. It’s a company defined by big and bold moves, and when something goes right, they make sure you know about it. From launching a Tesla into deep space to the captivating test flights of their next-generation Starship spacecraft, the private company has turned high-stakes aerospace research and development into a public event. A cult of personality has developed around SpaceX’s outlandish CEO Elon Musk, and so long as he’s at the helm, we can expect bigger and brighter spectacles as he directs the company towards its ultimate goal of putting humans on Mars.

Of course, things don’t always go right for SpaceX. While setbacks are inevitable in aerospace, the company has had a few particularly embarrassing failures that could be directly attributed to their rapid development pace or even operational inexperience. A perfect example is the loss of the Israeli AMOS-6 satellite during a static fire of the Falcon 9’s engines on the launch pad in 2016, as industry experts questioned why the spacecraft had even been mounted to the rocket before it had passed its pre-flight checks. Since that costly mistake, the company has waited until all engine tests have been completed before attaching the customer’s payload.

SpaceX’s concept art for propulsive landing

But sometimes the failure isn’t so much a technical problem as an inability for the company to achieve their own lofty goals. Occasionally one of Musk’s grand ideas ends up being too complex, dangerous, or expensive to put into practice. For instance, despite spending several years and untold amounts of money perfecting the technology involved, propulsive landings for the Crew Dragon were nixed before the idea could ever fully be tested. NASA was reportedly uncomfortable with what they saw as an unnecessary risk compared to the more traditional ocean splashdown under parachutes; it would have been an impressive sight to be sure, but it didn’t offer a substantive benefit over the simpler approach.

A similar fate recently befell SpaceX’s twin fairing recovery ships Ms. Tree and Ms. Chief, which were quietly retired in April. These vessels were designed to catch the Falcon’s school bus sized payload fairings as they drifted down back to Earth using massive nets suspended over their decks, but in the end, the process turned out to be more difficult than expected. More importantly, it apparently wasn’t even necessary in the first place.

Deadliest Catch

Credit where credit is due, both Ms. Tree and Ms. Chief did successfully catch fairings during their tenure with SpaceX. The ships proved the concept was viable, and on two missions in 2020, they even managed to capture both fairing halves. But taken as a whole, their success rate was quite poor. According to a tally from SpaceXFleet.com, of the 37 missions on which one or both ships attempted to recover the fairings, they only managed a total of nine catches. That’s already a pretty bad average, but when you realize each mission actually has two fairings that needed to be caught, it’s abysmal.

The low success rate is bad enough, but even when the ships actually nabbed one of the fairings in mid-air, it didn’t always end well. During the October 18th, 2020 Starlink mission, the live video feed from Ms. Tree briefly showed a fairing ripping through the net and smashing down onto the deck. With each fairing half estimated to weigh approximately 950 kilograms (2094 pounds), having one break lose presents a clear danger to the crew and equipment aboard the recovery vessel, to say nothing of the fairing itself. After all, the goal is to recover them intact so they can be used on a subsequent flight.

Only a few seconds of low-resolution video were shown before SpaceX cut the stream.

It might seem odd that SpaceX had so much trouble catching these relatively large and docile objects as they drifted down to the surface under their parafoils, especially when compared to the fire and fury of the Falcon 9’s first stage landing. Over the last three years SpaceX has managed to maintain a success rate of around 90% for booster recoveries at sea, and at least on the surface, it would seem both procedures are more alike than they are different.

But ultimately, it’s a question of command authority. The active grid fins and thrust vectoring capabilities of the Falcon 9 make it far more maneuverable on descent than the steerable parafoils used by the fairings. Even with the recovery ship actively communicating with the fairing’s own avionics and attempting to plot an intercept point, a strong gust of wind at the wrong moment was all it took to knock them off course.

Making a Splash

From the beginning, SpaceX believed that they’d need to catch the fairings with a net because allowing them to come into direct contact with salt water would damage them beyond the point of economical repair. While specific details are hard to come by publicly, it’s widely believed that the concern stemmed not so much from the electronics onboard, which could presumably be waterproofed, but the unique construction of the fairings themselves. Made from an aluminum honeycomb structure sandwiched between layers of composite material, water intrusion could be a serious problem; as once salt water got inside the structure of the fairing itself, getting it back out quickly and economically might not be possible.

But in the face of a recovery program that seemed to be going nowhere, the engineers at SpaceX have apparently figured out a way to make it work. Closeup photographs of recently constructed fairings show that various vents and openings have been relocated so they’ll be higher from the surface of the water, and rumor has it that the internal sound dampening panels are now considered a consumable and discarded after each mission rather than trying to dry them out. What, if any, steps were taken to prevent water from seeping into the  fairing’s aluminum/composite construction is currently unknown.

While the refurbishment process for these “wet” fairings is undoubtedly more costly and time consuming than if they had been caught in the net, the difference is evidently not enough to justify the continued operation of Ms. Tree and Ms. Chief. Instead, SpaceX has chartered the much larger Shelia Bordelon to take over as primary fairing recovery vessel. Intended for underwater research, the 78 m (256 ft) long ship features wide open decks, a built-in crane, and a Triton XLX remotely operated vehicle (ROV) that can dive down to 3,000 m (9,840 ft).

Shelia Bordelon unloading a fairing half. Photo by Kyle Montgomery.

The integrated crane makes it easier to pull fairings out of the water and drop them on the dock, but otherwise, this vessel doesn’t seem particularly well suited to the task at hand. For one thing, its underwater capabilities are being completely squandered. But more importantly, the rapid launch cadence demanded by Starlink missions means that the recovery vessel should ideally be able to hold four fairing halves before returning to port. So either the Shelia Bordelon is going to be getting some modifications of its own soon, or SpaceX is only using it temporarily until they can come up with a long-term solution.

91 thoughts on “SpaceX Drops The Ball On Catching Fairings

  1. Failure is the fastest method to learn about a mistake. It can be costly and even dangerous but if time is your primary concern then failure is often the best path to success.

    1. Which actually goes well with the saying, gaining popularity, that you go alone to go fast but go together to go further.
      At one level, it’s about a sense of responsibility. “Move fast and break things” is fine when you hack something in your own workshop. Not as much fun when you play with people’s livelihoods. Which is part of the reason private organisations have more leeway than public institutions (e.g. NASA).

      Sometimes, we need to move exceedingly fast. Other times, it’s puzzling how hurry comes from mere hubris. Narcissism has consequences.

      1. Elon Musk’s mode of operation is rather “fail slow”. In other words, he takes on tasks and goals where the failure/success criteria is sufficiently ambiguous and hard to prove that he can run the show for a couple decades before admitting that it doesn’t work and all the investor money was wasted.

        For example:
        >Over the last three years SpaceX has managed to maintain a success rate of around 90% for booster recoveries at sea

        If the economic break-even point for a re-usable booster is let’s say 10 flights, then the probability of any one booster being cheaper than a conventional launch, at 90% recovery rate, is just 35%. Two thirds of the boosters will fail to break even and cost more than the equivalent regular rockets. The rocket that fails on flight #1 will cost ten times its worth, whereas the rocket that survives flight #20 has only saved you half the launch cost. You do the math – the rockets that fail early outweigh the ones that survive past break-even and the whole business makes no sense. You need to catch them almost perfectly every time, and also launch them almost without failure, or else it just does not offer any advantage. They will have to beat the industry standard launch failure rate with a system that has many more points to fail, and they simply haven’t done so.

        The point is, it takes a lot of launches and many years to prove the point. All the while Mr. Musk can argue that they’re improving, they’re learning, this is just teething problems etc. and collecting money and making profit off of a product that fundamentally does not work – simply by promising something fantastic. Same thing with the Tesla Autopilot, Hyperloop, etc. promising a pie in the sky that looks plausible on the face of it, but cannot actually be reached.

        1. And on top of that, he continuously shifts the goalposts for himself anyways, and the fanboys go along with it. Anyone still remember the original three step plan with Tesla? Anyone still remember how Musk promised the Model S would cost half as much as it actually did?

          1. … okay?

            Remind me, how’s your private spaceflight and electric car business doing? If you think you can do it better than Musk, well… here’s your chance. I’ll wait.

          2. Wasn’t his plan,
            Create a low volume, high priced car.
            Create a medium volume, medium priced car.
            Create a high volume, low priced car.
            Which part has he failed on? Volume is going up and price is going down.

        2. >If the economic break-even point for a re-usable booster is let’s say 10 flights,
          Which orifice did you pull that number out of? For that to be the case, the increase in build cost for a booster would have to be ten times the net per-flight cost benefit. With a new expendable booster running about $20 million and recovery and refurbishment costs in the single-digit millions, that’s pretty unrealistic. While nobody outside of SpaceX has access to SpaceX’s exact numbers, it’s more likely that the extra-cost features of a reusable booster (grid fins, legs, beefed-up heat shielding) are paid for after a single reflight.

          1. That’s just an example. Elon Musk himself has said “about 12”.

            The increase in build cost is not only in the booster, but for the preparations and running of the whole operation. In order to have re-usable boosters, someone has to go out there and catch them, someone has to fix them, someone has to inspect them…

          2. Also remember that the comparison isn’t between a single-use Falcon 9 but a much smaller and cheaper rocket that has an equivalent payload capacity /without/ the extra size and mass for the extra stuff.

          1. The 90% figure is given in the article. Probability of a flight surviving n flights is 0.9^n

            Wherever you put the breakeven point (m), the “value” of a rocket that fails on flight #1 is 1/m and the value of a rocket that fails on flight n is n/m. Therefore, we can start counting up the failed rockets (1-0.9^n) and scale up so the value of the rocket at m flights becomes 1, so we get (1-0.9^n) / (1-0.9^m). With m=10 we get a curve that reaches an asymptote at around 1.5 which means for infinite flights, the re-usable rocket is worth about 1.5 conventional rockets.

            And remember, this is still just counting the recovery rate at sea. Other failures also apply, including the fact that whenever they need to launch to a higher orbit or have more payload, they have to expend the rocket. It is unlikely that any booster will reach even 10 flights before it HAS to be expended for other reasons, unless they simply refuse to launch bigger missions.

          2. Why defend the numbers after you’re called on it?

            “Oh, but one of the numbers was real, I only made up half the equation” doesn’t help your argument.

            You don’t have the numbers, and the numbers you made up are ridiculous.

          3. Mind, this calculation applies to an arbitrarily large population of boosters. It’s simply normalized to a scale of 1 booster. Let’s further refine the calculation. Let’s subtract 1

            (1-0.9^x) / (1-0.9^10) -1

            This formula now values a booster that fails before the first launch as a negative 1, because it didn’t even get up in the air. Anything above 10 launches gets a positive value meaning it was worth that much more than the equivalent single-use booster. Now let’s multiply further by the probability of a booster actually surviving x launches:

            ((1-0.9^x) / (1-0.9^10) -1)*0.9^x

            Now we have the expected value distribution for a large population of boosters. We can see that there’s a tiny sliver of positive outcomes for rockets that survive up to 10 – 40 launches and a large wedge of negative outcomes for rockets that fail to reach 10. If you take the sum integral of the curve from 0 to about 50, you get a negative value of -2.23 meaning, instead of saving money you actually spent twice as much.

            The early failures outweigh the lucky survivors and the whole business was not worth it in the end.

          1. There’s a lot wrong with that analysis. The most obvious flaw is the assumption that customers pay a fixed cost per kilogram of payload capacity. That’s not how the great majority of the launch market works — if you’re selling a rocket that’s overpowered compared to what the satellite builders were targeting 5-7 years ago, you just launch with unused capacity and you don’t get extra money for that. There just aren’t many payloads that feel a difference from Falcon 9’s reuse penalty — just look at how few customers are buying expendable F9 launches.

            The second is that it assumes that reuse has no impact on launch cadence, which is critical for amortizing the large fixed costs these companies have. Production rate can hit inelastic constraints, where the factory is maxed out and reacting to increased demand means breaking ground on a new factory. Reducing the production demand per launch eases that and opens up a bigger market.

            If that’s an example of how ULA does their business planning, well, it’s no surprise that they’re getting their asses kicked.

          2. >you just launch with unused capacity and you don’t get extra money for that.

            You’re mixing up price vs cost. Consider that for the cost equation, the rocket could be made much smaller.

        3. The biggest issue rocket builder face is that their “throwaway” rocket are not, they are completely over engineered.
          Having seen some mechanical piece of Ariane rocket, I can tell you that they have enormous cost reduction waiting if they truly want cheap and frequent launch (hint: they don’t).
          But these cost reduction are nothing compared to the pork money and fixed cost of the Arianespace company itself.
          So in the end, they are all dying not because SpaceX reuse make sense, but because they are unable to adapt and Musk is dragging them in oblivion by forcing them to go to the execution ground (ie, forcing them to do re-usable rocket).

          1. These companies have evolved to make use of government pork. Re-structuring would mean cutting off all the little piglets that suckle off of these corporations, which is politically impossible.

            ISRO is competitive with SpaceX on a single use basis.

          2. Dude’s assertion that ISRO is competitive on a single-use basis may be true, but engineering salaries are 3x more in the US, so they must be three time more effective to complete against low-cost labor. The link discussed software developers, but other tech job are roughly parallel in wages.

            https://www.indiatoday.in/technology/news/story/developers-in-us-earn-almost-3-times-more-than-indian-coders-developers-in-pakistan-too-earn-more-1648667-2020-02-21

            The fact is SpaceX is outcompeting in both cost and innovation.

            https://www.aspistrategist.org.au/cheap-reusable-space-launchers-are-still-years-away-for-chinese-military/

        4. You definitely need to check your math there. A 90% success rate does not mean 2/3 of a chance of not making it to 10. It means, on average, every booster will make it to 10. Further that is the 3-year success rate. The 2 year rate is even higher and the 1 year rate even higher still. The process is improving greatly. Since the process is improving greatly, it also just proves your other piece where you say he wastes investor money learning and never actually succeeds. My understanding is that this time is that they aren’t even making these boosters anymore. All of the flights are reused. They also produce these boosters cheaper so even on the first flight there is substantial savings over the traditional providers. If what you said was true and they made it to 3 it would be about a 10x cost reduction.

          All-in-all your post was basically, completely flawed.

          1. Don’t mind him. He does this on every SpaceX related article. Sometimes he even fabricates outright lies like that “SS is using toxic hypergolics or igniter fluid”. And as a true troll he’ll carefully choose which posts to reply to so he can maximize FUD spreading.

            I wish he could just keep hating Elon without hating SX. That company keeps pushing the envelope and doing so with much smaller government subsidies compared to the “old space” which are de-evolving with projects like SLS whereas SX:

            – Managed to have 2nd highest launch cadence in the world in 2020 (only behind the whole China)
            – For almost only half a price of Boeing’s managed to complete unmanned and manned demo flights, one operational mission and started second one before them.
            – Managed to launch 4 times more reused boosters than new one in 2020.

            Btw. booster landings provides (probably very) valuable feedback to the development process and allows for both better reliability and cost optimization. I’d not be surprised if such optimization is what is gradually driving the cost down and not the actual reuse at least at first.

            Btw. the analysis Dude linked somewhere above
            – is from 2015 which is like an ancient history, really
            – does not mention benefits of having a feedback to the development process (which I’d be not surprised if it were enormous)
            – is from SX’s direct competitor ULA *facepalm*.

          2. No,
            I’ll let the article speak for itself regarding the 90% success rate, “Over the last three years SpaceX has managed to maintain a success rate of around 90% for booster recoveries at sea…”
            It has nothing to do with how many flights a specific F9, it’s simply, success÷attempts.

          3. > fabricates outright lies like that “SS is using toxic hypergolics

            That is a lie on YOUR part. Last time we were discussing this, we were talking about the Falcon 9 rockets being launched off of a sea pad. The Raptor engine is spark torch ignited, while the Merlin uses a hypergolic igniter, which presents the problem that I was pointing out: poisoning the waters around the platform.

          4. > It means, on average, every booster will make it to 10.

            That’s not how the probabilities go. A 90% success or 10% random failure rate does not mean an individual booster will fly 10 times and then fail. It means it has a 90% probability of surviving a flight.

            So you have 90% chance to survive the first flight, and 90% for the second, which is 90% x 90%… and so on for the third. It’s 0.9^n for n flights.

        5. This is seriously the most ass-backwards take I’ve ever seen on recoverable boosters.

          What SpaceX actually does:
          1) Build booster for customer 1. Charge them $FULL_PRICE booster, actual cost of booster + launch costs + profit margin, still lower than legacy space. There’s your breakeven.
          2) Launch, then recover booster.
          3) QA booster, then send back to flight.
          4) Launch booster for customer 2. Price is $FULL_PRICE – $POSSIBLE_DISCOUNT, which is more than the cost of recovery/QA/fuel
          5) Actual profit, no underpants involved.

          Have you not been watching? Have you missed those 100+ launches over the past decade? Multiple boosters with 7+ re-flights? Or are you just upset that your $TSLA shorts didn’t work out?

        6. >It is unlikely that any booster will reach even 10 flights before it HAS to be expended for other reasons

          Booster B1051 will fly its tenth mission this weekend.

          >Same thing with the Tesla Autopilot, Hyperloop, etc. promising a pie in the sky that looks plausible on the face of it, but cannot actually be reached.

          I think it’s a bit of a stretch to compare landing F9 boosters to autopilot (FSD). They’ve actually demonstrated that landing the boosters *can* be done. It’s not “a product that fundamentally does not work”.

          I’m not a mathematician, but as a layperson I always hear his example about throwing a rocket away as being the equivalent of a NY to LA flight, and then throwing the jumbo jet in the ocean after each flight.

          Can the math be applied to that analogy that would show that it makes sense to expend jumbo jets? Because nobody has that as a business model afaik.

          1. But the “can be done” part isn’t what the whole thing is about – that had already been done before SpaceX. It’s the economics of being CHEAPER than a single use rocket that SpaceX set out to prove, which they haven’t done.

            It’s a bit of the same thing with Tesla. At first they set out a mission to “prove” that electric cars can be better than conventional ones. They made a bunch of wild promises, set out plans and goals, FAILED to meet any of those, and 20 years later they’re almost able to produce the mass-market car they promised in the beginning. It’s only twice too expensive for the role instead of 5-6x.

            And just now if you read the news, Elon Musk is saying they’ll reach autonomy level 5 by the end of the year, while the actual Tesla engineers are going “whoa whoa we’re only barely at level 2!”.

          2. >It’s a bit of the same thing with Tesla.

            It’s not, really. You described it as a product that “fundamentally does not work”. It sure looks like it works to me. That’s why it can’t be compared to a FSD Tesla. It’s a straw man. You can be flabbergasted that people are paying up front for not-yet-proven FSD on their Teslas, (and I am too, btw) and maybe one day we’ll have a nice big class action get-together to sort out the details, but that doesn’t mean SpaceX is selling snake oil.

            As far as whether they make money on it and are able to bring the cost down, unless they go public or you somehow have access to their consolidated financials or tax returns, it’s all just speculation. Outside of unofficial soundbites from Shotwell or Musk there’s no way for us to be certain of their break-even point per rocket. ULA says it’s 10, Musk says it’s 2. It’s probably somewhere in between, but we can’t be certain.

            >SpaceX runs on investor cash and forward payments, not on profitable business. Musk goes on begging campaigns every year to raise more cash.

            Show me another aggressive growth company that doesn’t seek out new investors. Courting new investors doesn’t necessarily make it a ponzi scheme. Again, how do you know for certain SpaceX’s profit margin?

            >that had already been done before SpaceX.

            Who else has previously landed and reused an orbital class first stage booster rocket? And don’t say the shuttle, because that’s another terrible comparison. The technical complexity and cost to refurbish the entire vehicle to make it safe for human flight each and every time made it outrageously expensive. No vehicle from any provider compares to the shuttle.

            >You need to catch them almost perfectly every time, and also launch them almost without failure, or else it just does not offer any advantage.

            Block 5 F9 already has a 100% successful launch rating (61/61). And a landing success rate of 92% (59/64 attempts). They haven’t had a launch failure of an F9 of any configuration since AMOS-6 blew up on the pad in 2016. I hate to rain on your parade, but they’re flirting with perfection.

        7. By your math, spacex & tesla are losing money hand over fist. Musk must be embezzeling every penny he can find in both companies. You should report him to our trusty govt.

          1. They are. The only reason Tesla posts profits is because of the clean car credit payments – otherwise they’re bleeding. SpaceX runs on investor cash and forward payments, not on profitable business. Musk goes on begging campaigns every year to raise more cash. The problem: the government is in it with them.

  2. Speaking as a (former) skydiver, this failure to catch the fairings really surprises me. It’s not difficult for a human to land within a few meters of a desired point with a parachute — I’ve done it many times. Good ones (not me) can often land on a frisbee.

    I would have thought it almost trivial to put the guidance and intelligence in the fairings, and simply have them autonomously aim for the desired catching net. This notion of trying to get a big, ungainly boat in the right spot at the right time seems like the wrong way to approach the problem.

    1. Perhaps the size of the fairing and its unpredictable aerodynamic properties makes this a lot harder. A human skydiver is pretty small compared to the parachute.

      1. It’s actually stable and predictable in flight, and I’d argue the large size makes the job easier (scale size).

        Human skydivers also need to flare at landing to slow the descent rate, and thus make those last few meters really susceptible to wind gusts. The fairings could just slam into the net under full control at 10 m/s, with little effect from local wind changes.

        1. Its the windage, the fairing is enormous and for its size pretty light so even a tiny breeze will push it downwind significantly.

          Its also pretty much a static shape with rather little aerodynamic control – the human skydiver can flex their entire body and control how much of whatever wind there is they catch, the fairing has some ability to steer and not much else…

          Scale for scale those catch ships are only a few meters, to allow for the relatively poor control surfaces to land them at a very specific static point you’d either need a fleet of flying craft in the flight path to gather wind speed data at every elevation in real time and in advance of fairing separation so its course can be calculated to deal with the winds even before its released (clearly a daft idea), or to scale up the size of the point you are trying to hit significantly… Perhaps cover Texas with gelatine to act as a cushion… As you really can’t give the fairings huge control surfaces they would just add too much weight, while also not really being able to deal with the huge shape’s wind catching abilities, and if you could change its shape to be less problematic it wouldn’t make a good fairing anymore… Perhaps a 6-8 way split fairing could work, as being so much flatter under the ‘chute they wouldn’t catch near as much wind, but again the separation and structural edges for it would add mass…

        2. “i’d argue the large size makes the job easier”

          I think the problem was, as stated, the amount of authority the system had to steer the parafoil. A large, massive ship with a tiny rudder might turn but it’ll do so slowly and with a large radius.

          Frankly given the way SpaceX handles the assembly of the stack and rollout to launch, I’m surprised they didn’t sea-proof the fairings from Day1. They have to have been designed to be waterproof and protect their contents from salt fog. That’s got to be far down the path needed to make recovery from the sea workable.

          1. I think some of the difficulty lies in the mechanical stresses that it’s subjected to upon landing. Without the rocket and other fairing half to support its edges, the things are quite wobbly — you can see this in some of the videos, where they’re conspicuously waving about just after being jettisoned from the second stage. They flex a *lot* upon hitting the water, and that probably creates some new entry points for seawater to go places it’s not welcome. It sounds like they’ve been tweaking the design to move weak points above the waterline.

      1. Skydiving is a fair-weather sport, to be sure. Even a 10-knot wind will spoil your day. But SpaceX isn’t exactly known for all-weather flying either. Just look at how many of their flights get postponed due to weather, even just down-range weather. They aren’t exactly Soyuz-class in that respect.

        1. SpaceX can’t control the weather. They have to consider the weather at the launch site and the landing site, and waiting till the two are just perfect leaves them with precious few launch windows and too many last minute cancellations.

          They pretty much need to launch when they can go up, and whether they can come down is a lesser concern.

        2. One needs to count the Z axis too. It is a common problem when trying to rescue humans from vessels into a helicopter. The elevation of the deck is constantly changing and it is not uncommon to encounter waves with over ten meter amplitude. It may not look like much if the point of view is fixed on the vessel but I challenge you to land on a vessel with a parachute without breaking about every bone in your body during average Atlantic weather conditions.

  3. > With each fairing half estimated to weigh approximately 950 kilograms (2094 pounds), having one break lose presents a clear danger to the crew and equipment aboard the recovery vessel, to say nothing of the fairing itself.

    weren’t they droneships? without crew?

    1. The unmanned drone ships are basically motorized barges for catching the 1st stage of the rocket. The fairings are caught by crewed vessels with very large nets.

  4. All part of the learning curve. I like watching this company because they aren’t afraid to try new things win/fail. So what is some of the ideas don’t pan out. We all go through that even in are fun projects. Point is though, we move through it and go ‘forward’ and that is what SpaceX is doing. Like they finally nailed the landing for the Starship 15 yesterday. Cool beans. Process!

    1. You can climb up a learning curve to infinity, but if the asymptote of the curve that you’re following doesn’t reach the threshold of success, all you’re doing is wasting time and money with superficial “progress”.

    2. But there is a goal in mind, Not a waste of time or money at all. Think of all the inventors back in the Wright Brother days that failed, and failed again, even died… But they tried, and finally someone succeeded. Conquering space is HARD. The moon, the planets, eventually the stars. A worthy place to spend time and money. I know if I was a zillionaire this is where I’d put my efforts too. You are doing something new and innovative to try to reach that goal — the stars. They aren’t just trying stuff for kicks, they are looking for solutions to the hard problems…. That is what is great about the last frontier. It is forcing us to come up with new technologies in computers, computer control, materials, propulsion, etc. A very good thing and sometimes that technology can be used back here on Earth too. Otherwise we stagnate as a people. It is also putting a lot of creative people to work and a lot of support workers, the builders, machinists, etc. to make it happen. It is all good. Also a good morale booster for planet Earth (well maybe not for the flat Earthers) too when all said and done.

      For the fairings, maybe they just need a special big floating foam/air cushion platform to land on to keep them out the water until pulled onto a ship :) Or a helicopter to snag them as they make their way down. Let the ideas flow….. And if they can’t make it back to platforms… You can still fish them out of the salt water, but require a bit more time to clean-up for next mission.

      1. >But there is a goal in mind

        Of course, if people shovel me money in through the doors and windows, I can invent some lofty goal and spend the rest of my life “approaching” that goal. That’s the basic point behind all over-unity, anti-gravity, cold fusion, etc. scams. The only difference with Musk is that he isn’t scamming with something that is patently impossible by the laws of physics.

          1. >It was said it one time that man would never fly

            Interestingly enough, we still can’t. Climbing into a winged tube full of chairs isn’t exactly man flying.

  5. Funny I haven’t seen much media hype over SN15 success. Only completely erroneous hyped up BS about the Tesla Model S crash in Texas. And now this nonsense. This is journalism at It’s worse, and reeks of bias or negative publicity for financial gain of some sort.

    1. I’m confused by your interpretation of the article. I don’t see any “shade” being thrown around. It seems like a rather objective statement of fact with some discussion as to whether or not such risks outweigh the benefits. Isn’t that the ideal way to analyze and learn from observation?

    2. I agree, I wish there was an objective way to perceive/express negative bias. Big bold fonts that say statements like “SpaceX Drops the ball…”, “Deadliest Catch” seem to point towards a non-objective view point. Even when the author doesn’t have all the facts they throw in statements that reveal their bias like “What, if any, steps were taken to prevent water from seeping into the fairing’s aluminum/composite construction is currently unknown.” It seems like the author starts with the assumption that SpaceX can’t engineer themselves out of a wet paper bag and then takes pride in pointing out the ways the author assumes they have failed. Wondering how they solved a problem is far different than just assuming they forgot about that part of the problem. The post is full of non-objective expressions many of which have nothing to do with the discussion of the fairing. In the big picture it doesn’t really matter. It has no effect on SpaceX. It is just a weird perspective to see on Hackaday.

  6. I personally think this change in direction is extremely positive. Being vested in a single solution is a recipe for disaster. They tried several times, failed, while applying alternatives to resolve other concerns. This is A/B testing at its finest.

    Sure, I hope they get a cooler solution in the future, same as their drop ship. They don’t have to stay this way forever just because that’s how it was done in the past. This tides them over for now though.

  7. As for the past Starship launches… “This is the Captain. There’s a little problem with our entry sequence so we may experience slight turbulence and then….. explode..” – Malcolm Reynolds – Serenity

  8. I wouldn’t be at all surprised if SpaceX comes up with something far more “radical” in a year or two like autonomous jet or rocket powered hydrofoil watercraft that increase the fairing catch rate to 90% or more within a few years.

    1. Their plan for the next few years is to quit dropping fairings at all as they shift payloads on to Starship. No need for fishing boats or catcher’s mitts if the whole upper stage just lands back at the pad. Any new systems for catching fairings would have to pay off before then, which is why I don’t think we’ll see them trying much of anything else on that front.

      1. Those were a lot smaller and lighter. Not sure about snagging and dragging large 2000 pound “flimsy” sail-like objects with plane….any volunteer test pilots? I’m not an aeronautical engineer but it seems that it would be a very difficult thing to do if possible at all.

      2. That’s just what Rocketlabs is proposing to recover their boosters: mid-air capture. They’ve already proven it possible (with a helicopter-dropped dummy booster).

        Reminds me of a fun game back in my flying days: Toss a roll of paper towel out the window at 10k ft or so, and see how many times you can cut the ribbon on its way down. It’s a wild ride, basically pulling 3 g’s (the airframe limit) the whole way down. (Another game was bombing targets on the field with lunchbags full of flour.) Moral: don’t tell a bunch of young punks in airplanes to loiter and wait for the rest of the team. They’ll find something stupid to do.

        1. Ah yes…that tiny 4 foot diameter booster without a rentry heat shield and no apparent mechanism to control rentry attitude and slow it down for parasail deployment. Dropping from helicopter and catching with another staged below is the easy part. How much will recovery reduce their paltry 660 lb payload to LEO 440 lb SSO

  9. would be much easier to catch them mid flight using a small fleet of airships and rig the net between them. The wind would affect the catchers and the catch alike. One could then lower the catch on land or on a reserve ship with much more controllable manner. Those ships could be autonomous, and operate on solar power lowering the operating costs.

    1. There are a lot of problems to solve there. First you’d have to be able to catch them from above, because otherwise the parachute would basically get tangled on whatever flight controls you had on the aircraft. The weight of the fairing probably gets rid of any type of solar aircraft. A miss means tangling the parachute around the aircraft and a loss of both vehicles. Once you’ve grabbed the fairing it would have to be aerodynamic in such a way that it wouldn’t impact being towed, which goes against being the proper aerodynamics to float on air.

      These, and so many more problems that I see. It might work with a helicopter, but even a helicopter would collapse the parachute on approach.

      1. Think of a pouch suspended from four hot air balloons, The point was not to haul but catch and gently lower the catch. One thing you got right, it would be a massive headache to attempt that using helicopters of any other vessel where the lifts is produced using power. Hence the airship.

    2. Sounds fun, but think about the amount of sideways force the airships would need to exert even on an empty net to keep it stretched… not to mention the ~1000kg fairing inside.

  10. There’s an expression for the kind of nit picking the title of this article is based on: “naysayers”. The funny thing is that a site called hackaday provides a forum for theoretical engineers like @Dude to bash this great hacker and Nay Say.

    1. Yeah, this series has definitely jumped the shark. Dull. No booms any more. Downright boring. Not going to watch any more.

      Though ‘boring’ is exactly what you want when riding a controlled explosion to orbit.

      Fun fact: the power density at the F9’s rocket nozzle exit is about 50 times the surface of the sun. The only reason it doesn’t promptly melt is because virtually all that heat is carried away by the burnt fuel.
      And the total power at launch is about the same as the whole state of Florida’s electrical grid.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

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